Detection and Outcome Markers . Metastasis . EMT . Aberrant Glycosylation, Drugs and Treatment . Therapeutic Targets

Ovarian Cancer Research

Identification of novel molecular signatures to improve epithelial ovarian cancer outcome

Gynecological cancers in general and ovarian cancer (OC) in particular are the main focus of our research. OC is the fifth most common cause of death from all cancers in women and the leading cause of death from gynecological malignancies, with a poor prognosis (5-year survival <20%). Major issues with OC are its heterogeneity, its unclear genetic origin, its diagnosis at advanced stages due to the lack of accurate biomarker, its ability to rapidly metastasize into other organs,and its insufficient treatment due to disease recurrence (therapy resistance)and missing personalized therapy regimens. Our particular focus is to: (i) identify molecular/genetic signatures that unequivocally discriminate OC patients by their clinico-pathological parameters (e.g. histotype, grade, stages) and identify molecular targets for the better prediction of disease outcome and for the design of targeted therapies. (ii) elucidate the molecular basis and biological functions of aberrantly glycosylated proteins and lipids in cancer initiation, progression, and dissemination. (iii) evaluate in clinical studies novel means to improve diagnosis and treatment of gynecological cancers.

Identification of discriminating signatures in the era of omics (Prof. Viola Heinzelmann, Research Group Leader)
Identification of novel “discriminating” diagnostic and prognostic markers and therapeutic targets is an urgent need in the fight against cancer. Employing various glycan-based immunoassays we identified sets of specific markers (glycans) that enable us to discriminate healthy controls from ovarian cancer patients with sensitivity and specificity comparable to the current biomarker CA125 (Jacob et al,2012, 2014; Pochechueva et al, 2011, 2014, 2016). We also found particular glycan based signatures (glycans and glyco-genes) which distinguish serous ovarian from peritoneal cancers: this, together with transcriptomic and epidemiological data,provides further evidence that these two cancers are clearly different diseases and hence should no longer be clinically treated as one (manuscript in preparation). We also identified two protein kinases with significance in ovarian cancer. ROR2,a Wnt-signaling-associated receptor tyrosine kinase, is over expressed in ovarian cancer patients (tissue microarray) and is implicated in proliferation, migration,and invasion (Henry et al, 2015). MELK (maternal embryonic leucine-zipperkinase) was identified as an interesting candidate in ovarian cancer (Heinzelmann-Schwarz et al, 2004) already in 2004 when still known as KIAA0175. We now show in broad transcriptomic/bioinformatical data analyses that MELK expression is elevated in ovarian cancer patients, increases with ascending aggressiveness, and correlates with poor disease outcome, and that its inhibitor OTSSP167 is highly active against (including drug-resistant) ovarian cancer cells (Kohler et al., 2017).

Regulation and function of specific glycan motifs in ovarian cancer (Glyco-Oncology: Dr. Francis Jacob, Project Manager)
It is acknowledged that glycosylation is crucial to the proper function(s) of glyco proteins and glycosphingolipids but the molecular and biological details are poorly understood. Owing the expertise of Francis Jacob as Glycobiologist the “Glyco-Oncology” branch within Ovarian Cancer Research was launched in 2015. In collaboration with Nicki Packer (Sydney), membrane protein glycan features (“bisecting GlcNAc” type N-glycans) unique to ovarian cancer cells were identifiedand experimental evidence is provided that expression of MGAT3 (enzyme for bisecting GlcNAc synthesis) is epigenetically regulated by DNA-methylation and correlates with presence of bisecting GlcNAc on glycoproteins (Anugraham et al.,2014, Kohler et al., 2016). Likewise, evaluation of 18 TCGA cancer types (6118 samrecepples)demonstrated a generally poor overall survival in cancer patients with highMGAT3 expression but interestingly identified among ovarian cancer a subgroup of “long-term survivors” with low MGAT3-expression (Figure). We also identified by glycan-based immunoassays, mass spectrometry, and flow cytometry a subset of glycosphingolipids with possible roles in ovarian cancer, including cell migration(Jacob et al., 2014, Alam et al., 2015, Anugraham et al., 2015). Further in vitro and in vivo investigations address the role(s) of glycosphingolipids in epithelial-mesenchymal transition (EMT) and drug responses using specifically designed (CRISPR/Cas9-edited) “glyco-gene knockout” ovarian cancer cells: one intriguing result indicates that glycosphingolipids Play a key role in mediating EMT and ist reverse MET in ovarian cancer metastasis (Jacob et al., 2018).

Drug resistance, hypoxia, and tumor acidosis – function and significance of glycans (PD Dr. André Fedier, Project Manager)

Drug resistance, tumor hypoxia and tumor acidosis are additional critical obstacles in cancer management. Questions addressed include: Do particular glycosphingolipids affect cellular responses to chemotherapeutic agents, to hypoxia, and to acidosis of ovarian cancer cells? Do drug-resistant and –sensitive cells have different glycosphingolipids profiles? Does sustained exposure to drugs, hypoxia, and acidosis stably alter this profile? The current results indicate that particular glycosphingolipids promote chemosensitivity but hamper responses to hypoxia various ovarian cancer cell lines, suggesting that these biomolecules modulate cellular responses to stressors.

Identification of the molecular events during ovarian cancer metastasis to the omentum (Dr. Ching-Yeu Liang, Project Manager)

An interdisciplinary and collaborative project (“Sinergia”, funded by SNF) initiated in 2017 aims at designing and establishing a semi in vivo 3D-omentum model to study how, i.e. by which mechanisms, ovarian cancer metastasizes with high preference  to the omentum, a still poorly-studied organ localized in the peritoneal cavity. In this multidisciplinary project we will use these 3D-models to identify molecular events and key molecules implicated in ovarian cancer metastasis to the omentum by using CRISPR-Cas9 Technology.









Figure: MGAT3 transcription start site is hypermethylated (black circles) in OVCAR8 and poorly methylated (white) in OVCAR4 and BG1 cells (A) Demethylation by 5-Aza in hypermethylated OVCAR8 recovers MGAT3 expression (B) and bisecting GlcNAc expression (C: red arrow in insert). Boxplot of MGAT3 expression in TCGA PANCAN12 data set (n=3587): lowest expression in head and neck squamous cell cancer (top) to highest in ovarian serious cystadenocarcinoma (red frame) (D). Kaplan-Meier: low MGAT3 expression (grey line) associates with poor survival (various cancers combined) (E). Some high-grade serous ovarian cancer patients have low MGAT3 expression (grey) and are “long-term survivors” (F). Kohler et al., 2016.