ABSTRACT
The study of in vitro morphogenesis is fundamental to understanding neoplasia since the dysregulation of morphogenic pathways that create multi-cellular organisms is a common hallmark of oncogenesis. The in vitro culture of human breast epithelial cells on reconstituted basement membranes recapitulate some features of in vivo breast development, including the formation of a three-dimensional structure termed an acinus. Importantly, the capacity to disrupt in vitro acinar morphogenesis is a common property of human breast oncogenes. In this report, we find that phosphatidylinositol 4-kinase IIIß (PI4KIIIß), a lipid kinase that phosphorylates phosphatidylinositol (PI) to PI(4)P, disrupts in vitro mammary acinar formation. The PI4KIIIß protein localizes to the basal surface of acini created by human MCF10A cells and ectopic expression of PI4KIIIß induces multi-acinar devlopment. Furthermore, expression of an oncogenic PI4KIIIß activator, eEF1A2 (eukaryotic elongation factor 1 alpha 2), phenocopies the PI4KIIIß multi-acinar phenotype. Ectopic expression of PI4KIIIß or eEF1A2 alters the localization of PI(4)P and PI(4,5)P(2) within acini, indicating the importance of these lipids in acinar development. Our work shows that PI4KIIIß, eEF1A2 and PI(4)P likely play an important role in mammary neoplasia and acinar development.
Subject(s)
Breast Neoplasms/genetics , Carcinoma, Acinar Cell/genetics , Peptide Elongation Factor 1/genetics , Phosphotransferases (Alcohol Group Acceptor)/genetics , Breast Neoplasms/metabolism , Breast Neoplasms/pathology , Carcinoma, Acinar Cell/metabolism , Carcinoma, Acinar Cell/pathology , Female , Humans , Minor Histocompatibility Antigens , Peptide Elongation Factor 1/metabolism , Phosphotransferases (Alcohol Group Acceptor)/metabolism , Tumor Cells, CulturedABSTRACT
OBJECTIVE: To determine whether eukaryotic elongation factor 1 alpha 2 (eEF1A2), a transforming gene previously shown to be highly expressed in primary human ovarian tumours, is a prognostic marker. METHODS: We have used an antibody specific for eEF1A2 to measure eEF1A2 protein expression in 500 primary ovarian tumours in a tissue microarray. We have also ectopically expressed eEF1A2 in SK-OV-3 cells, a clear cell carcinoma line that does not normally express eEF1A2. RESULTS: We have shown that eEF1A2 has high expression levels in approximately 30% of all primary ovarian tumours. 50% of serous tumours, 30% of endometrioid, 19% of mucinous and 8% of clear cell tumours highly express eEF1A2. Ectopic expression of eEF1A2 in the SK-OV-3 clear cell carcinoma line enhances their in vitro proliferative capacity and ability to form tumour-like spheroids in hanging drop culture. Expression of eEF1A2 did not alter sensitivity to anoikis, cisplatin, or taxol. In serous cancer, eEF1A2 is an independent prognostic marker for survival and high eEF1A2 protein expression was associated with increased probability of 20-year survival. CONCLUSIONS: eEF1A2 is highly expressed in ovarian carcinomas. Its expression enhances cell growth in vitro, and eEF1A2 expression is likely to be a useful ovarian cancer prognostic factor in ovarian cancer patients with serous tumours.
