Role of CA125 in ovarian cancer biology

The cancer antigen 125 (CA125) is a cell-surface mucin which is over-expressed by the majority of ovarian cancers. However, its biology and the role it plays in ovarian cancer is largely unknown, although other cell-surface mucins have been shown to play a role in apoptosis, cell growth and tumour immune evasion. To analyse the function of CA125 in ovarian cancer, we initially knocked down the expression of CA125 using RNA interference. Knocking down CA125 expression using in vitro transcribed short interfering RNAs (siRNAs) induced a potent cell death response, which has been well characterised in the literature as an induction of an interferon response and resulting in cell apoptosis. Subsequently, using the short hairpin RNA expression vector, pSUPER, which has been shown to knock down genes with high efficiency with reduced off-target affects, we generated stable sub-lines of the ovarian cancer cell line, OVCAR-3, which had been transfected with pSUPER constructs targeting CA125. Intriguingly, these sub-lines had a range of abnormal mitotic events and nuclear defects. However, there was no clear association with the level of CA125 knock down. This could be either due to clonal selection from the parent OVCAR-3 cell line or in addition to CA125 knock down, additional genetic changes are required to occur to favour a state of survival. Similar to the in vitro data, xenografts of the sub-clones into SCID mice generated inconclusive results as to whether CA125 knock down contributes to tumour growth, invasion and metastasis in vivo. More recently, we have been able to achieve high levels of short-term CA125 knock down using synthetic siRNAs designed to reduce off-target affects. These preliminary in vitro and in vivo experiments conducted with pSUPER sub-lines should be repeated using synthetic siRNAs to confirm the role of CA125 in this context. Given the role which the cytoplasmic tail of cell-surface mucins plays in its function, we generated a polyclonal antibody recognising the CA125 cytoplasmic tail, designated M16.1. Immunofluorescence imaging of CA125 in ovarian cancer cell lines, OVCAR-3 and PEO-1, using the OC125 extracellular domain antibody indicated cell-surface localisaton of CA125. However, in addition to the cell-surface localisation, the M16.1 antibody localised to the cell cytoplasm, indicating cleavage and release of the CA125 cytoplasmic tail into the cytosol. Additionally, M16.1 co-localised with α-tubulin at perinuclear sites and to areas resembling microtubule organising centres. However, M16.1 did not co-localise with γ-tubulin at the centrosome, indicating association with non-centrosomal microtubules. Furthermore, depolymerisation of microtubules on ice for 1 hour resulted in loss of diffuse cytoplasmic M16.1 staining but co-localisation between M16.1 and α-tubulin at non-centrosomal sites remained. Intriguingly, when microtubules were allowed to reform at 37oC in PEO-1 cells which had CA125 knocked down by synthetic siRNAs, the ability to reform radial asters was impaired, possibly indicating the CA125 cytoplasmic tail involvement in anchoring microtubules to non-centrosomal sites. Furthermore, we also cloned a portion of CA125 encompassing the cytoplasmic tail, transmembrane domain and 9 tandem repeats. When this construct was transfected into COS-1 cells, the CA125 cytoplasmic tail localised to microtubule bundles during metaphase. Mitotic involvement of the endogenous CA125 cytoplasmic tail was confrmed in OVCAR-3 and PEO-1 cells using M16.1. Given this association and also the results from the pSUPER sub-lines with CA125 knockdown, CA125 may be involved in controlling the fidelity of mitosis, which is grossly altered during tumourigenesis. More recently, it was identified that galectin-1 (Gal-1), an S-type lectin, is a ligand for CA125. Gal-1 is a potent inducer of T cell apoptosis and has been implicated as playing a major role in immune evasion for cancer cells. Consequently, we analysed the expression of CA125 and Gal-1 in ovarian cancer and confirmed the two molecules were expressed concurrently at the mRNA level by RT-PCR. Moreover, immunofluorescence studies also confirmed that CA125 and Gal-1 interacted with each other at the cell-surface of 27/87 cells, an ovarian cancer cell-line. Therefore, we hypothesised that CA125 presents Gal-1 to the immune system, which then induces T cell apoptosis and allows the tumour to escape the immune system. However, CA125 did not protect tumour cells from recognition or killing by T cells, which was shown by no differences in IFN-γ secretion or tumour lysis by cytotoxic T cells using influenza peptide pulsed pSUPER sub-lines with CA125 knockdown. The work described in this thesis suggests that CA125 plays a major role in the aetiology and progression of ovarian cancer through its actions on mitosis, microtubule organisation and immune evasion