MBP-1 expression strategies and identification of downstream targets in breast cancer cell lines

The Myc promoter Binding Protein -1 (MBP-1) is an important tumour suppressor which negatively regulates different protooncogenes, such as C-MYC and HER2, thereby playing a key role in carcinogenesis. MBP-1 shares 97% homology with the glycolytic enzyme α-enolase, since both are encoded by the same gene ENO1. ENO1 mRNA is subjected to alternative translation from different start codons (AUG); the first one is used for the synthesis of the 48 kDa α-enolase and the second ones, 94 and 97 codons downstream, are used for the synthesis of the 37 kDa MBP-1. Despite this, α-enolase and MBP-1 have different function: α-enolase acts mainly as a glycolytic enzyme in the cytoplasm, while MBP-1 acts as a transcription factor and has a prevalent nuclear localization. Both α-enolase and MBP-1 are involved in tumorigenesis, although as antagonists. In cancer cells, for the so-called Warburg Effect, a reprogramming of energy metabolism occurs, with the enhancement of glycolysis and lactic fermentation (aerobic glycolysis) at the expense of aerobic respiration. The transcription factor c-Myc and the Human Epidermal Growth Factor 2 (HER2) play a key role in promoting aerobic glycolysis, since they are involved in the overexpression and activation of the aerobic glycolysis enzymes, including α-enolase. The oncogene C-MYC, in turn, is amplified or overexpressed in most human cancers, while HER2 is amplified in 15% to 20% of breast cancers. Both genes are repressed by MBP-1. In breast cancer patients, α-enolase levels increase in almost all cancer tissues compared to the normal ones, while nuclear MBP-1 is present in almost all the normal tissues but retained in only 35% of the tumors. Overexpression of α-enolase is correlated with greater tumor size, poor nodal status, and a shorter disease-free interval, while MBP-1 expression in breast tumours turned out to be an independent favourable prognostic indicator for disease-free survival. The first aim of the project on which this thesis in based was the restoring of MBP-1 expression in breast cancer cell lines by using antisense oligonucleotides named ENO-Mut1, ENO-Mut2¸ ENO-Mut3 and ENO-Mut4. These oligonucleotides have been designed to pair, with imperfect complementarity, to ENO1 mRNA in the region containing the AUG in position 1, to block the canonical translation of α-enolase and verify whether the alternative translation of MBP-1, starting from the AUG in position 94 or 97, is promoted. Transient transfection of SkBr3 (a HER2+ breast cancer cell line) revealed that a dose-dependent MBP-1 increase in expression occurs in presence of ENO-Mut1 at 40nM, 80nM and 120 nM. As regards ENO-Mut2, ENO-Mut3 and ENO-Mut4, in the presence of the antisense oligonucleotide at 40 nM and 120 nM the MBP-1 levels are much higher than in the control and the highest levels are achieved in presence of ENO-Mut4. Since α-enolase acts as a dimer to catalyse the conversion of 2-phosphoglycerate (2-PG) to phosphoenolpyruvate (PEP) during glycolysis, the second aim was to verify whether a quote of MBP-1 in the cytoplasm can interact with α-enolase to form a non-functional dimer, thereby interfering with aerobic glycolysis. Therefore, SkBr3 cells have been transfected with a plasmid containing a sequence encoding MBP-1under the control of a strong constitutive promoter, to then perform an Enolase Activity assay. The latter revealed a significant reduction of enolase catalytic activity if compared with that of control samples. The third aim was the identification of novel MBP-1 targets and functional pathways. Therefore, whole genome microRNA (miRNA) expression analysis has been conducted on SkBr3, MCF-7 and MDA-MB-231 - i.e. three breast cancer cell lines that together represent the most common human breast cancer subtypes - overexpressing exogenous MBP-1. To do that, the microarray technology has been exploited, which revealed the existence of three miRNAs - i.e. miR-2861, miR-20a and miR-99a – differentially expressed in MCF-7, MDA-MB-231 and SkBr3. More specifically, each of them was upregulated in the MBP-1 overexpressing samples, compared to the relative control samples, in all the three cell lines. Targets prediction analysis revealed that six genes - i.e. NR6A1, NTRK3, RRAGD, SATB1, TRIM25 and TTC39A – are common target of miR-2861, miR-20a and miR-99a. A comparison with a dataset from previous whole genome expression analysis conducted on MCF-7, MDA-MB-231 and SkBr3 overexpressing exogenous MBP-1 showed that all the six genes are down-regulated in the MBP-1 overexpressing samples, compared to the relative controls, in each of the three breast cancer cell lines, confirming their negative regulation by miR-2861, miR-20a and miR-99a, which are, in contrast, up-regulated in consequence of MBP-1 overexpression. The selected six genes have been cumulatively correlated with prognosis in breast cancer by using the Kaplan-Meier plotter, which showed that the downregulation of the six genes together is positively correlated with a higher survival probability on all breast cancers. Moreover, the multivariate analysis shows a better hazard ratio for the selected genes, compared with that of Ki-67, ER and HER2, accompanied by a better significance. This suggests that the six genes together represent a better prognostic signature than the prognostic markers used to date. In conclusion, the possibility of restoring the expression of an important tumour suppressor, such as MBP-1, by using antisense oligonucleotides would justify, if these data were confirmed in vivo, the use of the same for therapeutic purposes. For example, the antisense oligonucleotides could be conveyed specifically to the tumor mass, by using functionalized nanoparticles for the recognition of a tumour marker, such as HER2 protein. Furthermore, although MBP-1 has been proposed as a promising breast cancer prognostic marker, the evaluation of its expression levels is affected by sequence homology with α-enolase. Nevertheless, the six selected genes NR6A1, NTRK3, RRAGD, SATB1, TRIM25 and TTC39A constitute a promising prognostic signature in breast cancers and may represent a potential target for therapy in primary Invasive Ductal Breast Carcinoma (IDC) patients.