- Email: [email protected]
Biomarker detection in pancreatic ductal adenocarcinoma
EACR24 Poster Sessions / European Journal of Cancer 61, Suppl. 1 (2016) S9–S218 777 Proteome and phosphoproteome analysis of stratified meningiomas S. Ferluga1 , J. Dunn1 , E. Lasonder2 , V. Sharma3 , D.A. Hilton4 , C. Adams1 , C.O. Hanemann1 . 1 Peninsula College of Medicine and Dentistry, Neurobiology, Plymouth, United Kingdom, 2 Plymouth University, Proteomics, Plymouth, United Kingdom, 3 Plymouth University, School of Biomedical & Healthcare Sciences, Plymouth, United Kingdom, 4 Plymouth Hospitals, Neuropathology, Plymouth, United Kingdom Introduction: Meningiomas are the most common primary tumours of the central nervous system in adults and despite the majority of them displaying benign features they can cause mild to severe morbidity. The current main therapeutic approach is complete resection commonly with adjunct radiotherapy. No pharmaceutical treatment is available and surgical resection is not always possible, so new therapeutic options are needed. In the context of molecular targeted therapy, we aim to identify novel targets/biomarkers by deciphering both the proteome and phosphoproteome in stratified meningiomas. Material and Methods: Meningiomas were stratified based on their histological grade defined by the World Health Organization (WHO). A total of 5 grade I, 5 grade II and 4 grade III meningioma specimens were analysed and compared to 3 meninges used as normal control. Equal amounts of total proteins were loaded on a gradient SDS-PAGE and evenly separated bands were fractionated and tryptic digested. Extracted peptides were purified and analysed by liquid chromatography–tandem mass spectrometry (LC-MS/MS). Phosphoprotein purification was performed using Qiagen® PhosphoProtein Purification Kit. Phosphoproteins were processed as before. Raw mass spectrometry files were analysed using MaxQuantTM and data were processed by using DAVID 6.7 and the Ingenuity® Pathway Analysis (IPA) software. Results: We have identified approximately 3000 proteins across the 14 samples with over 2000 quantified; similar numbers were obtained for phosphoproteins. Over 600 candidates were commonly identified across the grades for the total proteome, and over 800 for the phosphoproteome, which were not quantified in normal meningeal tissue. Additional proteins and phosphoproteins were found to be grade-specific. Gene ontology (GO) enrichment analysis of upregulated proteins (fold change 1.5; p-value <0.05) revealed several up-regulated biological processes. After uploading the datasets to IPA, comparative analysis of primary meningiomas vs. normal meningeal tissue showed several dysregulated pathways. Among the most affected pathways identified were the mTOR, integrin and ERK/MAPK signalling. Notably, using this tool we identified several novel targets associated with an already available drug, either in clinical practice or trial. These targets will be validated first as well as the efficacy of specific repositioned drugs in an effort of making this study time and cost effective. Conclusions: Proteome and phosphoproteome analysis of different grade meningiomas revealed several potential novel targets/biomarkers and dysregulated pathways, demonstrating the effectiveness of the proteomic approach. Our studies increased the overall knowledge of protein expression and regulation in meningiomas and will give the rationale for future drug testing on common or stratified targets. No conflict of interest. 778 Investigation of the functional significance of YAP1 in prostate cancer cells and tissues F. Kisaayak Collak1 , F. Sagir ˘ 1 , U. Demir1 , S. Ozkanlı2 . 1 Istanbul Medeniyet University, Molecular Biology and Genetics, Istanbul, Turkey, 2 Istanbul Medeniyet University, Medicine, Istanbul, Turkey Background: YAP1 (Yes associated protein 1) is a transcription co-activator of the Hippo tumor suppressor pathway mediating gene regulation. Its function is controlled by phosphorylation on S127 site leading to the sequestration in the cytoplasm and degradation. The functional significance of YAP in prostate cancer has remained elusive. Material and Methods: Prostate cancer cell lines (LNCaP, PC3, DuCaP) and prostate epithelial cells (PrPec) were used. The protein amount of YAP1 and pYAP1(S127) was investigated using western blotting. Cellular fractionation was performed to determine cellular localization of YAP1 and pYAP1. Functional assays such as proliferation, colony formation and soft agar colony formation were applied after silencing of YAP1 by siRNA and overexpression of YAP1 and YAP1S127A constructs. The effect of YAP1 on cisplatin resistance was investigated by overexpression. Expression and localization of YAP1 and pYAP1 in prostate cancer and normal tissues were assesed by immunohistochemistry. Results: Expression level of YAP1 was not increased in LNCaP and PC3 compared to normal cells PrPec. The subcellular localizations of YAP1 and pYAP1 were consistent with the literature. YAP1 protein was located both in cytoplasm and nucleus. pYAP1 was located in the cytoplasm. YAP1 was notexpressed by DuCaP cells. Upon YAP1 silencing the proliferation ability was decreased in LNCaP and PC3 cells and colony forming ability was decreased in PC3 cells. Overexpression of FlagYAP and FlagS127A stimulated the
S177
proliferation of LNCaP cells compared to vector control. The proliferation ability of PC3 cells was not affected by YAP overexpression whereas YAPS127A stimulated the proliferation ability of PC3 cells. Colony forming ability of PC3 cells were increased by YAP and YAPS127A overexpression. The resistance of PC3 cells to cisplatin was overcomed by YAPS127A overexpression. Immunostaining demonstrated that YAP expression was relatively week in prostate cancer tissues compared to benign tissues. Conclusions: Our results identify YAP1 as a potent regulator in prostate cancer cell proliferation, colony formation, anchorage-independent growth and drug resistance. Further in vivo studies are needed to better understand the regulation of YAP1 in prostate cancer. No conflict of interest. 779 The role of the HER2 and HER3 in prostate cancer and their potential as therapeutic targets K. Rao1 , L. Gaughan1 , C. Robson1 , S. McCracken1 . 1 Newcastle University, Northern Institute for Cancer Research, Newcastle upon Tyne, United Kingdom Introduction: HER family members (EGFR, HER2, HER3 and HER4) and their ligands have been associated with the development of castrate-resistant prostate cancer. Previous research has focussed on targeting the EGFR and HER2 receptors, however the role of HER3 in driving PI3K/Akt signalling and mediating drug resistance has shifted the attention towards this signalling partner. This project focuses on further understanding the role of the HER receptor family in prostate cancer and the therapeutic potential of a novel pan-inhibitor against EGFR, HER2 and HER3. Materials and Methods: We have performed immunohistochemistry on human prostate samples, plus immunoblotting with nuclear-cytoplasmic extractions, luciferase reporter assays and proliferation/migration/invasion assays on prostate cancer cell lines. Utilized prostate cell line models included: androgen sensitive disease (LNCaP); an androgen independent derived line (LNCaP-AI); and cell line models of acquired resistance to Lapatinib (LNCaPLapR) and Enzalutamide (LNCaP-EnzR). We have also developed HER2 and HER3 overexpressing stable cell lines using lentiviral technology. Results: We have analysed the expression of HER2 and HER3 receptors using prostate cancer tissue microarrays. We have shown strong correlations between HER2 and HER3 cytoplasmic and nuclear expression with poor prognostic disease. These results are of particular interest as part of the analysis consists of matched paired patient tissue samples (before and after relapse on conventional anti-androgen treatment). We have also shown in our cell line models that HER2/HER3 signalling: increases androgen receptor activity; increases downstream MAP Kinase and PI3 Kinase signalling activity; and increases cell proliferation, migration and invasion. The addition of a panHER family inhibitor inhibited all of these activities. It was also observed that pan-HER inhibition reduced nuclear translocation of HER2 and HER3, with observed prevention of chromatin binding. Finally, both the Lapatinib and Enzalutamide resistant cell lines demonstrated increased HER2 and HER3 activity, which was subsequently shown to be sensitive to pan-HER inhibition. Conclusions: The analysis of clinical samples has established that HER2 and HER3 play a prominent role in advanced prostate cancer. We have also demonstrated the reduced activity of androgen receptor, plus PI3 kinase and MAP kinase pathways downstream of the HER2/3 signalling pathways by the use of a novel pan-HER receptor family inhibitor. This pan-HER inhibitor has also reduced cellular activities important for cancer progression. The increased expression of HER2 and HER3 in both the Lapatinib and Enzalutamide resistant lines suggests that combination therapy of novel anti-androgens with pan-HER inhibitors may delay cancer progression in prostate cancer. No conflict of interest. 780 Biomarker detection in pancreatic ductal adenocarcinoma 4 , M. Werner3 , ¨ H. Klett1 , M. Klose2 , S. Kowar2 , P. Bronsert3 , S. Kusters P. Christoph5 , H. Busch1 , M. Boerries1 . 1 German Cancer Research Center DKFZ, Institute for Molecular Medicine and Cell Research- DKTK, Freiburg, Germany, 2 Institute for Molecular Medicine and Cell Research, DKTK, Freiburg, Germany, 3 Institute of Pathology, University Medical Center Freiburg, Freiburg, Germany, 4 University Medical Center Freiburg, Clinic for General and Visceral Surgery, Freiburg, Germany, 5 Institute for Molecular Medicine and Cell Research, Comprehensive Cancer Center Freiburg, Freiburg, Germany
Background: With a 5-year survival of less than 6% is pancreatic ductal adenocarcinoma (PDAC) among the most lethal cancers of all types. Early detection biomarkers are in urgent need to improve diagnosis and survival. Here, we present a robust gene signature based on extensive meta-analysis of six different transcriptome data sets to classify PDAC and non-tumor tissue. Furthermore, we identified three different classes (tumor, stroma and pancreas specific) that show different risk dependencies on TCGA survival data. Materials and Methods: We normalized samples within each data set and combined results from each data set in the meta-analysis structure. Briefly,
S178
EACR24 Poster Sessions / European Journal of Cancer 61, Suppl. 1 (2016) S9–S218
all but one data set were used for feature selection. Here, we ranked genes according to their log fold-change (tumor vs. non-tumor) on each data set and then aggregated scores. The resulting gene set was tested on the left out data set by calculating the cross-validation performance of a support vector machine as measured by the area under the curve (AUC) of the receiveroperating characteristic (ROC). Consensus clustering was based on spearman correlations and Kaplan–Meier curves were used for survival analysis. Results: The meta-analysis resulted in one gene set for each data set yielding prediction performances of AUC = 0.92–1.00. The union identified 132 unique genes as potential biomarkers for PDAC among which we found multiple established PDAC markers, e.g., LAMC2, S100P. Consensus clustering of genes showed 3 distinct groups that could be labeled precisely as tumor (57), stroma (25) and pancreas (50) specific according to literature knowledge. Investigating the risk of the different groups in independent TCGA samples led to 5-year survival rates of >90% of patients with low gene expressions in the tumor and stroma group whereas survival rates of intermediate and high gene expressions decreased rapidly over time. Lastly, we tested our gene set on an independent microarray data set consisting of non-tumor, pancreatitis, tumor and PanIN samples With a prediction performance of AUC= 0.96–1.00 was it possible to classify non-tumor and pancreatitis from PanIN and tumor samples. Conclusion: We established a robust meta-analysis to obtain a gene signature, which is able to classify PDAC from non-tumor samples. The need of >100 unique biomarkers confirms the previously found heterogeneity in PDAC. We identified genes specific associated with tumor but also stroma and pancreas suggesting the importance of the tumor microenvironment and functional genes in classification. Survival was tremendously higher for PDAC patients with relative low expression in tumor and stroma specific genes proposing the fact that these genes may be already regulated at the onset of the disease. Ongoing work consists of further reducing the biomarker genes and experimental and clinical validation. No conflict of interest. 781 Role of microRNAs in signal transduction pathways of the inflammatory cytokine interleukin-6 in hepatocellular carcinoma cell lines and primary hepatocytes F. Servais1 , M. Kirchmeyer1 , M. Casper2 , M. Hamdorf1 , C. Haan1 , P. Nazarov3 , L. Vallar3 , C. Rubie4 , M. Glanemann4 , F. Lammert2 , S. Kreis1 , I. Behrmann1 . 1 University of Luxembourg, Life Sciences Research Unit, Belvaux, Luxembourg, 2 Saarland University Medical Center, Department of Medicine II, Homburg, Germany, 3 Luxembourg Institute of Health, Oncology Department, Luxembourg, Luxembourg, 4 Saarland University Medical Center, Department of Surgery, Homburg, Germany Introduction: IL-6 plays important roles in the regulation of liver functions and promotes hepatocarcinogenesis. As the contribution of IL-6-induced miRNAs to these effects is largely unknown, we investigate the effects of IL-6 on the miRNome of hepatoma cells and non-transformed hepatocytes. Moreover, little is known about miRNAs regulating key players of IL-6 signal transduction. Therefore, our aim is to identify such miRNAs in a miRNA mimics screen and to elucidate their effects on this pathway. Material and Methods: Experiments were carried out with HCC cell lines (HepG2, HuH-7 & Hep3B) as well as with primary human hepatocytes (derived from liver metastasis surgery), which were challenged with hyperIL-6 (20 ng/mL) for 10 min up to 72 hours. Microarray analyses (Affymetrix GeneChip human transcriptome and miRNA arrays) were validated by qPCR, including further cell lines. Luciferase assays were performed with the supernatant of HEK293 cells stably expressing the reporter gene under the control of a STAT3-regulated promoter. Results: We have observed previously that IFN-g has a profound effect on the miRNome of melanoma cells (Schmitt, Cell Commun. Signal. 2012). However, in HepG2 and HuH-7 hepatoma cell lines, only few miRNAs change their expression level significantly in response to IL-6 (and hyperIL-6). In contrast to HCC cell lines, a stronger response could be observed in array analyses of primary hepatocytes (68 and 27 differentially expressed miRNAs in two samples, respectively); the regulation and roles of selected miRNAs are currently further analysed. Moreover, we have set-up a mimic library screen based on a STAT-regulated luciferase read-out to identify miRNAs involved in the regulation of the IL-6/JAK/STAT pathway, and analysed the effect of 538 mimics on reporter gene activity and cell viability. Candidate miRNAs will now be tested in secondary screen. Conclusion: The miRNA response to cytokines seems to be highly dependent on the cell type and on the stimulus. Our work may contribute to the identification of novel regulatory circuits regarding the IL-6/JAK/STAT3 signaling pathway. No conflict of interest.
783 Indirect control of the p53-induced apoptosis based on RNA interference R. Jaksik1 , A. Lalik1 , K. Puszynski1 . 1 Silesian University of Technology, Institute of Automatic Control, Gliwice, Poland Introduction: NFkB and p53 proteins are considered master regulators of various forms of cell death, thus their appropriate functioning is important for the cell survival. The ability to control processes, that are associated with NFkB and p53 signaling pathways, provides the possibility for the development of new, targeted anticancer therapies. The complexity of regulatory interactions limits the advancement in this field, however the incorporation of mathematical models into the research provides new possibilities for both selecting appropriate molecular target and the treatment type, in order to achieve a desirable response. In this work we introduce methodology and software that allows identifying new targets for the control of cellular viability, which we used to sensitize cancer cells to ionizing radiation with RNA interference. Materials and Methods: We used extended version of our p53-NFkB signalling pathway model, to identify appropriate targets for the RNAinterference-based regulation. Model response simulations were conducted using our custom Solvary application, based on a new method that allows defining some of the model variables as deterministic and some as stochastic, reflecting the heterogenic nature of cellular response to external stimuli. In order to validate the model we transfected human cancer cells with various doses of selected siRNAs and assessed changes in the cell viability using clonogenic and MTS-based assays after radiation treatment. Results and Discussion: In this work we present two examples of a single gene regulation inside NFkB and p53 pathways. Based on Mdm2-specific siRNA we were able to sensitize MCF7 cells (with inactivated PTEN gene) to ionizing radiation, allowing to achieve similar apoptosis level as with higher radiation doses. We also show that a comparable effect can be achieved by the use of IkBa-specific siRNA, which is a part of the NFkB pathway. We believe that this crosstalk is a result of a positive feedback loop, which exists between NFkB and p53 pathways, in which NFkB is a transcription factor for the p53, which in turn inhibits production of NFkB inhibitors such as IkBa. By silencing NFkB’s inhibitors we allow to increase p53 expression to the level that triggers apoptosis. Conclusion: By controlling various elements of the NFkB and p53 signaling pathways, based on knowledge derived from in silico studies, we show how cancer cells can be sensitized to radiation treatment, enhancing its efficiency. Acknowledgement: This work was supported by the Polish National Science Center grant DEC-2012/05/D/ST7/02072 and BKM/227/Rau1/2015. RJ is supported by the Foundation for Polish Science (FNP). No conflict of interest. 784 Regulation of the p53 pathway by siRNA and antisense oligonucleotides A. Lalik1 , R. Jaksik1 , K. Puszynski1 . 1 Silesian University of Technology, Systems Engineering Group- Faculty of Automatic Control- Electronics and Informatics, Gliwice, Poland Introduction: Antisense oligonucleotides (single-stranded DNA molecules) can induce the degradation of target mRNA either via an RNase H-dependent mechanism, or by acting as the steric-blocker oligonucleotides, which physically prevent or inhibit the progression of splicing or translation machinery [Mol. Cancer Ther. 2002 March; 1: 347–355]. siRNA (double-stranded RNA) silence the expression of a target gene by transcript degradation through the RNA-inducing silencing complex (RISC). In this work we want to compare the results obtained by both of those techniques, and assess the possibility of increasing Mdm2 level in the cells, by the use of specific oligonucleotides, which role is to prevent Mdm2 transcript degradation by miRNA. Material and Method: HCT116 (human cancer cells) cells were transfected with Mdm2 siRNA (Santa Cruz Biotechnology) or antisense oligonucleotides, and expression level changes of both Mdm2 and p53, at the level of transcript (qPCR) and protein (Western Blotting), were estimated. Analysis of cell cycle (cytometric methods) and cell viability (MTS assay) was performed 24 h after transfection of the cells. Results and Discussion: We compared the biological effects of Mdm2 siRNA and antisense oligonucleotides, complementary to the 3 -UTR of Mdm2, using HCT116 cells. The results obtained confirm that the use of antisense oligonucleotides leads to an increased level of the Mdm2 mRNA, while Mdm2 siRNA treatment has the opposite effect. Since Mdm2 is responsible for the p53 regulation this corresponds to changes in the percentage of viable cells that correspond to the concentration of siRNA/antisense oligonucleotides used. Conclusion: The mechanisms of both Mdm2 siRNA and antisense oligonucleotides enhance the possibilities of cancer treatment: cancer cells can be transfected with Mdm2 siRNA, while the neighboring (healthy) cells can be protected by the antisense oligonucleotides, increasing the efficiency of radiation treatment, and at the same time decreasing the side-effects e.g. of the radiation based therapy. Acknowledgement: This work was supported by grant DEC-2012/05/D/ST7/ 02072 from the Polish National Science Centre. The experiments were
S177
proliferation of LNCaP cells compared to vector control. The proliferation ability of PC3 cells was not affected by YAP overexpression whereas YAPS127A stimulated the proliferation ability of PC3 cells. Colony forming ability of PC3 cells were increased by YAP and YAPS127A overexpression. The resistance of PC3 cells to cisplatin was overcomed by YAPS127A overexpression. Immunostaining demonstrated that YAP expression was relatively week in prostate cancer tissues compared to benign tissues. Conclusions: Our results identify YAP1 as a potent regulator in prostate cancer cell proliferation, colony formation, anchorage-independent growth and drug resistance. Further in vivo studies are needed to better understand the regulation of YAP1 in prostate cancer. No conflict of interest. 779 The role of the HER2 and HER3 in prostate cancer and their potential as therapeutic targets K. Rao1 , L. Gaughan1 , C. Robson1 , S. McCracken1 . 1 Newcastle University, Northern Institute for Cancer Research, Newcastle upon Tyne, United Kingdom Introduction: HER family members (EGFR, HER2, HER3 and HER4) and their ligands have been associated with the development of castrate-resistant prostate cancer. Previous research has focussed on targeting the EGFR and HER2 receptors, however the role of HER3 in driving PI3K/Akt signalling and mediating drug resistance has shifted the attention towards this signalling partner. This project focuses on further understanding the role of the HER receptor family in prostate cancer and the therapeutic potential of a novel pan-inhibitor against EGFR, HER2 and HER3. Materials and Methods: We have performed immunohistochemistry on human prostate samples, plus immunoblotting with nuclear-cytoplasmic extractions, luciferase reporter assays and proliferation/migration/invasion assays on prostate cancer cell lines. Utilized prostate cell line models included: androgen sensitive disease (LNCaP); an androgen independent derived line (LNCaP-AI); and cell line models of acquired resistance to Lapatinib (LNCaPLapR) and Enzalutamide (LNCaP-EnzR). We have also developed HER2 and HER3 overexpressing stable cell lines using lentiviral technology. Results: We have analysed the expression of HER2 and HER3 receptors using prostate cancer tissue microarrays. We have shown strong correlations between HER2 and HER3 cytoplasmic and nuclear expression with poor prognostic disease. These results are of particular interest as part of the analysis consists of matched paired patient tissue samples (before and after relapse on conventional anti-androgen treatment). We have also shown in our cell line models that HER2/HER3 signalling: increases androgen receptor activity; increases downstream MAP Kinase and PI3 Kinase signalling activity; and increases cell proliferation, migration and invasion. The addition of a panHER family inhibitor inhibited all of these activities. It was also observed that pan-HER inhibition reduced nuclear translocation of HER2 and HER3, with observed prevention of chromatin binding. Finally, both the Lapatinib and Enzalutamide resistant cell lines demonstrated increased HER2 and HER3 activity, which was subsequently shown to be sensitive to pan-HER inhibition. Conclusions: The analysis of clinical samples has established that HER2 and HER3 play a prominent role in advanced prostate cancer. We have also demonstrated the reduced activity of androgen receptor, plus PI3 kinase and MAP kinase pathways downstream of the HER2/3 signalling pathways by the use of a novel pan-HER receptor family inhibitor. This pan-HER inhibitor has also reduced cellular activities important for cancer progression. The increased expression of HER2 and HER3 in both the Lapatinib and Enzalutamide resistant lines suggests that combination therapy of novel anti-androgens with pan-HER inhibitors may delay cancer progression in prostate cancer. No conflict of interest. 780 Biomarker detection in pancreatic ductal adenocarcinoma 4 , M. Werner3 , ¨ H. Klett1 , M. Klose2 , S. Kowar2 , P. Bronsert3 , S. Kusters P. Christoph5 , H. Busch1 , M. Boerries1 . 1 German Cancer Research Center DKFZ, Institute for Molecular Medicine and Cell Research- DKTK, Freiburg, Germany, 2 Institute for Molecular Medicine and Cell Research, DKTK, Freiburg, Germany, 3 Institute of Pathology, University Medical Center Freiburg, Freiburg, Germany, 4 University Medical Center Freiburg, Clinic for General and Visceral Surgery, Freiburg, Germany, 5 Institute for Molecular Medicine and Cell Research, Comprehensive Cancer Center Freiburg, Freiburg, Germany
Background: With a 5-year survival of less than 6% is pancreatic ductal adenocarcinoma (PDAC) among the most lethal cancers of all types. Early detection biomarkers are in urgent need to improve diagnosis and survival. Here, we present a robust gene signature based on extensive meta-analysis of six different transcriptome data sets to classify PDAC and non-tumor tissue. Furthermore, we identified three different classes (tumor, stroma and pancreas specific) that show different risk dependencies on TCGA survival data. Materials and Methods: We normalized samples within each data set and combined results from each data set in the meta-analysis structure. Briefly,
S178
EACR24 Poster Sessions / European Journal of Cancer 61, Suppl. 1 (2016) S9–S218
all but one data set were used for feature selection. Here, we ranked genes according to their log fold-change (tumor vs. non-tumor) on each data set and then aggregated scores. The resulting gene set was tested on the left out data set by calculating the cross-validation performance of a support vector machine as measured by the area under the curve (AUC) of the receiveroperating characteristic (ROC). Consensus clustering was based on spearman correlations and Kaplan–Meier curves were used for survival analysis. Results: The meta-analysis resulted in one gene set for each data set yielding prediction performances of AUC = 0.92–1.00. The union identified 132 unique genes as potential biomarkers for PDAC among which we found multiple established PDAC markers, e.g., LAMC2, S100P. Consensus clustering of genes showed 3 distinct groups that could be labeled precisely as tumor (57), stroma (25) and pancreas (50) specific according to literature knowledge. Investigating the risk of the different groups in independent TCGA samples led to 5-year survival rates of >90% of patients with low gene expressions in the tumor and stroma group whereas survival rates of intermediate and high gene expressions decreased rapidly over time. Lastly, we tested our gene set on an independent microarray data set consisting of non-tumor, pancreatitis, tumor and PanIN samples With a prediction performance of AUC= 0.96–1.00 was it possible to classify non-tumor and pancreatitis from PanIN and tumor samples. Conclusion: We established a robust meta-analysis to obtain a gene signature, which is able to classify PDAC from non-tumor samples. The need of >100 unique biomarkers confirms the previously found heterogeneity in PDAC. We identified genes specific associated with tumor but also stroma and pancreas suggesting the importance of the tumor microenvironment and functional genes in classification. Survival was tremendously higher for PDAC patients with relative low expression in tumor and stroma specific genes proposing the fact that these genes may be already regulated at the onset of the disease. Ongoing work consists of further reducing the biomarker genes and experimental and clinical validation. No conflict of interest. 781 Role of microRNAs in signal transduction pathways of the inflammatory cytokine interleukin-6 in hepatocellular carcinoma cell lines and primary hepatocytes F. Servais1 , M. Kirchmeyer1 , M. Casper2 , M. Hamdorf1 , C. Haan1 , P. Nazarov3 , L. Vallar3 , C. Rubie4 , M. Glanemann4 , F. Lammert2 , S. Kreis1 , I. Behrmann1 . 1 University of Luxembourg, Life Sciences Research Unit, Belvaux, Luxembourg, 2 Saarland University Medical Center, Department of Medicine II, Homburg, Germany, 3 Luxembourg Institute of Health, Oncology Department, Luxembourg, Luxembourg, 4 Saarland University Medical Center, Department of Surgery, Homburg, Germany Introduction: IL-6 plays important roles in the regulation of liver functions and promotes hepatocarcinogenesis. As the contribution of IL-6-induced miRNAs to these effects is largely unknown, we investigate the effects of IL-6 on the miRNome of hepatoma cells and non-transformed hepatocytes. Moreover, little is known about miRNAs regulating key players of IL-6 signal transduction. Therefore, our aim is to identify such miRNAs in a miRNA mimics screen and to elucidate their effects on this pathway. Material and Methods: Experiments were carried out with HCC cell lines (HepG2, HuH-7 & Hep3B) as well as with primary human hepatocytes (derived from liver metastasis surgery), which were challenged with hyperIL-6 (20 ng/mL) for 10 min up to 72 hours. Microarray analyses (Affymetrix GeneChip human transcriptome and miRNA arrays) were validated by qPCR, including further cell lines. Luciferase assays were performed with the supernatant of HEK293 cells stably expressing the reporter gene under the control of a STAT3-regulated promoter. Results: We have observed previously that IFN-g has a profound effect on the miRNome of melanoma cells (Schmitt, Cell Commun. Signal. 2012). However, in HepG2 and HuH-7 hepatoma cell lines, only few miRNAs change their expression level significantly in response to IL-6 (and hyperIL-6). In contrast to HCC cell lines, a stronger response could be observed in array analyses of primary hepatocytes (68 and 27 differentially expressed miRNAs in two samples, respectively); the regulation and roles of selected miRNAs are currently further analysed. Moreover, we have set-up a mimic library screen based on a STAT-regulated luciferase read-out to identify miRNAs involved in the regulation of the IL-6/JAK/STAT pathway, and analysed the effect of 538 mimics on reporter gene activity and cell viability. Candidate miRNAs will now be tested in secondary screen. Conclusion: The miRNA response to cytokines seems to be highly dependent on the cell type and on the stimulus. Our work may contribute to the identification of novel regulatory circuits regarding the IL-6/JAK/STAT3 signaling pathway. No conflict of interest.
783 Indirect control of the p53-induced apoptosis based on RNA interference R. Jaksik1 , A. Lalik1 , K. Puszynski1 . 1 Silesian University of Technology, Institute of Automatic Control, Gliwice, Poland Introduction: NFkB and p53 proteins are considered master regulators of various forms of cell death, thus their appropriate functioning is important for the cell survival. The ability to control processes, that are associated with NFkB and p53 signaling pathways, provides the possibility for the development of new, targeted anticancer therapies. The complexity of regulatory interactions limits the advancement in this field, however the incorporation of mathematical models into the research provides new possibilities for both selecting appropriate molecular target and the treatment type, in order to achieve a desirable response. In this work we introduce methodology and software that allows identifying new targets for the control of cellular viability, which we used to sensitize cancer cells to ionizing radiation with RNA interference. Materials and Methods: We used extended version of our p53-NFkB signalling pathway model, to identify appropriate targets for the RNAinterference-based regulation. Model response simulations were conducted using our custom Solvary application, based on a new method that allows defining some of the model variables as deterministic and some as stochastic, reflecting the heterogenic nature of cellular response to external stimuli. In order to validate the model we transfected human cancer cells with various doses of selected siRNAs and assessed changes in the cell viability using clonogenic and MTS-based assays after radiation treatment. Results and Discussion: In this work we present two examples of a single gene regulation inside NFkB and p53 pathways. Based on Mdm2-specific siRNA we were able to sensitize MCF7 cells (with inactivated PTEN gene) to ionizing radiation, allowing to achieve similar apoptosis level as with higher radiation doses. We also show that a comparable effect can be achieved by the use of IkBa-specific siRNA, which is a part of the NFkB pathway. We believe that this crosstalk is a result of a positive feedback loop, which exists between NFkB and p53 pathways, in which NFkB is a transcription factor for the p53, which in turn inhibits production of NFkB inhibitors such as IkBa. By silencing NFkB’s inhibitors we allow to increase p53 expression to the level that triggers apoptosis. Conclusion: By controlling various elements of the NFkB and p53 signaling pathways, based on knowledge derived from in silico studies, we show how cancer cells can be sensitized to radiation treatment, enhancing its efficiency. Acknowledgement: This work was supported by the Polish National Science Center grant DEC-2012/05/D/ST7/02072 and BKM/227/Rau1/2015. RJ is supported by the Foundation for Polish Science (FNP). No conflict of interest. 784 Regulation of the p53 pathway by siRNA and antisense oligonucleotides A. Lalik1 , R. Jaksik1 , K. Puszynski1 . 1 Silesian University of Technology, Systems Engineering Group- Faculty of Automatic Control- Electronics and Informatics, Gliwice, Poland Introduction: Antisense oligonucleotides (single-stranded DNA molecules) can induce the degradation of target mRNA either via an RNase H-dependent mechanism, or by acting as the steric-blocker oligonucleotides, which physically prevent or inhibit the progression of splicing or translation machinery [Mol. Cancer Ther. 2002 March; 1: 347–355]. siRNA (double-stranded RNA) silence the expression of a target gene by transcript degradation through the RNA-inducing silencing complex (RISC). In this work we want to compare the results obtained by both of those techniques, and assess the possibility of increasing Mdm2 level in the cells, by the use of specific oligonucleotides, which role is to prevent Mdm2 transcript degradation by miRNA. Material and Method: HCT116 (human cancer cells) cells were transfected with Mdm2 siRNA (Santa Cruz Biotechnology) or antisense oligonucleotides, and expression level changes of both Mdm2 and p53, at the level of transcript (qPCR) and protein (Western Blotting), were estimated. Analysis of cell cycle (cytometric methods) and cell viability (MTS assay) was performed 24 h after transfection of the cells. Results and Discussion: We compared the biological effects of Mdm2 siRNA and antisense oligonucleotides, complementary to the 3 -UTR of Mdm2, using HCT116 cells. The results obtained confirm that the use of antisense oligonucleotides leads to an increased level of the Mdm2 mRNA, while Mdm2 siRNA treatment has the opposite effect. Since Mdm2 is responsible for the p53 regulation this corresponds to changes in the percentage of viable cells that correspond to the concentration of siRNA/antisense oligonucleotides used. Conclusion: The mechanisms of both Mdm2 siRNA and antisense oligonucleotides enhance the possibilities of cancer treatment: cancer cells can be transfected with Mdm2 siRNA, while the neighboring (healthy) cells can be protected by the antisense oligonucleotides, increasing the efficiency of radiation treatment, and at the same time decreasing the side-effects e.g. of the radiation based therapy. Acknowledgement: This work was supported by grant DEC-2012/05/D/ST7/ 02072 from the Polish National Science Centre. The experiments were