Use this URL to cite or link to this record in EThOS:
Title: Targeting genomic instability using a genomic and proteomic approach to identify molecular drivers of poor prognosis in cancer
Author: Menezes, Karen Abigail
ISNI:       0000 0004 6496 1652
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
Cancer is a disease of the genome whereby mutational events that confer a survival advantage to cells are selectively retained. Consequently, genome instability and evasion of cell death are fundamental hallmarks of cancer. Certain cancers, including epithelial ovarian cancers (EOC), pancreatic ductal adenocarcinomas (PDAC) and a subset of sarcomas with complex karyotypes are characterized as being very genomically unstable suggesting that such tumours are driven by defects in DNA damage recognition and repair. Additionally, these tumours are also characterized by a high frequency of p53 mutation, extensive intratumoral heterogeneity and resistance, acquired or intrinsic, to DNA damaging chemotherapeutic agents: these tumour types consequently carry a poor prognosis. This highlights the need for detailed molecular characterization to identify therapeutic targets and stratification biomarkers. The aim of this study is to use a bioinformatic approach to identify commonly amplified genes, that function in DNA damage response and apoptotic processes, across the three tumour types, and which confer a poorer prognosis in these patients. For target identification, datasets for EOC and PDAC (obtained from TCGA) and sarcoma (obtained from GEO, NCBI) were analysed for common copy number aberrations (CNA) in 734 genes relating to DNA damage response and apoptosis and confer a poor prognosis in progression-free survival (PFS) data in EOC patients. Genes with prognostic significance (p < 0.05) using the log rank test were picked from the top 100 most frequently amplified loci. Next, the targets were functionally validated by siRNA-mediated knockdown, overexpression and/or pharmacological inhibition using apoptosis, proliferation and migration assays. Four genes, SGK3, c19orf40, MRPS12 and ZBTB32 were highlighted as being commonly amplified across all three tumour types, of which only SGK3 and c19orf40 were statistically significant in circular binary segmentation CNA calling when looking at PFS data in ovarian cancer patients. SGK3, a member of the serum/glucocorticoid regulated kinase (SGK) family was interesting as it has similar functions and substrates to the AKT kinase family, which we have previously shown to have key roles in tumour cell survival in response to therapy. Intra-patient paired sensitive (PEA1) and resistant (PEA2) and SKOV3 ovarian cancer cell lines, and Aspc1 and Panc-1 pancreatic cancer cell lines were used for validation studies. siRNA-mediated knockdown of SGK3 did not alter induction of caspase 3/7 activity in response to chemotherapy, relative to control treatments in all cell lines. In order to account for compensatory effects by the other SGK members in the presence of SGK3 knockdown, all three SGK members (SGK1, SGK2 and SGK3) were knocked down by siRNA and this also did not increase induction of caspase 3/7 activity. Migration, invasion and adhesion assays however, revealed that overexpressing SGK3 increases cell motility, invasion and adhesion respectively suggesting SGK3’s role in prognosis is via migration/tumour spread rather than response to therapy. Furthermore, there is evidence to suggest that SGK3 may act through the CXCR4-CXCL12 signalling axis and inhibiting CXCR4 may overcome SGK3 oncogenic effects. Additionally, recent data indicated that non-canonical subcellular localisation of key proteins alters treatment response. The oncogenic serine/threonine kinase, AKT, has been reported by us to be activated by DNAPKcs (PRKDC) in response to DNA damage in the nucleus of DNA damage resistant cancer cells and it’s inhibition restores therapeutic response in vitro, in vivo and clinically (Stronach et al, 2011; Blagden et al 2014). Inhibition of DNAPKcs has also shown to restore cisplatin sensitivity in platinum resistant ovarian cancer cell lines. Here, I also show that inhibition of DNAPKcs enhances sensitivity to doxorubicin in all ovarian, pancreatic cancer and sarcoma cell lines studied. Further cytoplasmic/nuclear signalling changes have been observed indicating the importance of nuclear signalling in chemoresistance. We describe here a proteomic approach to capturing novel nuclear signalling changes associated with chemo-resistance. Stable isotope labelled populations of intra-patient paired ovarian cancer cell lines PEA1 (platinum sensitive) and PEA2 (resistant) and the chemo-resistant pancreatic cancer cell line Panc-1 were prepared, seeded overnight and treated with cisplatin (25μM) or doxorubicin (1.5μM) for 24 hours with and without an inhibitor of DNA-PK (10µM NU7441; prevents DNA damage induced AKT activation). Treatment and control cell populations were mixed and fractionated to collect nuclear and cytoplasmic fractions for Stable Isotope Labelling of Amino acids in Cell culture (SILAC)/mass spectrometry analysis. Following established statistical filtering methods, common alterations in protein families were observed in nuclear fractions of drug treated resistant cells. Seven proteins were differentially expressed, some belonging to serine/arginine rich splicing factor protein family (1.3-2.4-fold increase), nuclear ribonucleoprotein family (1.2-fold increase) and DEAD box protein families. Protein changes in PRKDC were also observed in the resistant cell lines, which were absent in the sensitive PEA1 cells. The application of SILAC proteomics has identified novel protein changes in subcellular compartments that correlate with response/resistance to chemotherapy. Functional validation of identified some interesting candidate proteins, however further characterization is necessary to assess their roles in chemoresistance. The work presented in this thesis has highlighted several candidates that could have good therapeutic potential to either act alone (i.e. SGK3 inhibitors could be effective in preventing certain dynamics of the metastatic process, which is a major contributor of poor prognosis in patients) or in combination with existing agents (i.e. DNAPKcs inhibitor with doxorubicin in platinum resistant patients). Additionally, several targets have also been identified through RPPA and SILAC proteomics, that may have functional relevance in chemoresistance, however further characterization of these candidates is needed to fully understand their roles.
Supervisor: Stronach, Euan ; Cunnea, Paula ; Gabra, Hani Sponsor: Clement-Wheeler Bennett Memorial Trust
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral