Caratterizzazione di varianti localizzate in regioni "non coding" di geni clinicamente rilevanti nella patogenesi di tumori al seno.

Breast cancer is the most common female cancer, with an incidence of 25%; it is well known that the gene BRCA1 and BRCA2 if mutated in coding regions, may increase the incidence of cancer. Several studies have identified more than 25 genes involved in genome integrity related to BRCA1 and BRCA2 activity. Moreover, recent work has identified many variants in multiple genes located in coding and non-coding regions. Many reports suggest that “non-coding” variants located in promoters, 3’UTR, 5’ UTR, or splicing sites, may increase cancer risk. This work has characterized variants found in the 3’ UTR regions of 2 tumor suppressor genes: Ataxia-Telangiectasia Mutated (ATM) and Phosphatase and Tensin Homolog (PTEN), both associated with breast and ovarian cancers. ATM is a serine/threonine kinase playing a pivotal role in DNA double-strand break repair regulation through the up-regulation of factors required for DNA repair and control cell cycle checkpoints, such as BRCA1 and p53. The 3’ UTR regions have a crucial role in maintaining mRNA stability and interacting with regulatory factors, including microRNAs. We performed the “luciferase” assay to establish the effect of these 3’UTR variants on the interaction with specific miRNA. To accomplish this assay, 3’UTR is inserted at 3’ of the gene encoding “firefly” luciferase such that expression of the reporter gene is modulated by miRNA-3’UTR interaction. So, it is possible to observe if the miRNA investigated can bind the non-coding region and how this binding regulates protein expression. The host chosen to follow this interaction has been HEK293T cells since they are characterized by high transfection efficiency. In addition, the plasmid in which the 3’UTR was cloned (pMIR) was transfected with the plasmid (pRL-TK), which represents the internal control expressing the “renilla” luciferase in HEK293T cells. The cells were transfected with a scrambled siRNA characterized by a random sequence unable to bind sequences in the target genome as a negative control. To study the miRNA-3’UTR interactions of ATM, an 1887 bp fragment of the 3’UTR was cloned into the pGEM-Teasy vector, then site-specific mutagenesis was performed in position 26 of the WT UTR of ATM, leading to a base change A>G obtaining ATM c.*26A>G variant, that was identified in some patients. Recently, bioinformatics tools have been used to predict the binding of miRNAs hsa-miR-5681a and hsa-miR-488-3p to the 3’UTR of ATM. Results obtained by transfection with these miRNAs showed that both bind with good affinity the 3’UTR of ATM, compared to the negative control, decreasing in a statistically significant way the expression of “firefly”: about 40% in hsa-miR-488-3p and 30% in hsa-miR-5681a. Two variants in the 3’UTR of PTEN have been identified in previous studies: PTEN c.*1152C>T and PTEN c.*923T>A. Again, bioinformatics tools identified interaction with specific miRNAs. PTEN 3’UTR was placed in the pMIR plasmid, and by site-specific mutagenesis, vectors were built with the variants under study. The miRNAs chosen were for the 1152C>T variant: has-miR-519a-3p and hsa- miR-519c-3p while those for 923T>A were has-miR-193a-3p. The data collected on 1152C>T suggest a significant difference in regulation by both miRNAs over WT; in fact, the mutation results in a considerable reduction of inhibition with the expression value of the reporter gene that is like the negative control (scrambled siRNA). However, no satisfactory data were obtained to evaluate 923T>A and its interaction with hsa-miR-193a-3p. Also, in the context of the characterization of non-coding mutations, some BRCA1 splicing variants have been investigated, which result from mutations on intronic regions and cause in-frame deletions of exons, whose pathogenicity is still uncertain. The challenged variants are ∆9-10 (deletion of exons 9 and 10), ∆9-11 (deletion of exons 9,10, and 11), ∆11 (total deletion of exon 11), and ∆11q (partial deletion of exon 11). However, coding regions of these BRCA1 splicing variants were assembled in yeast by homologous recombination into the pYES2 plasmid of the GFP gene in frame at the C- terminus. S. cerevisiae was chosen as the host to observe differences in localization of BRCA1 variants because of heterologous expression in yeast since literature reported that pathogenic variants localize differently from BRCA1 wt. Based on the images obtained by fluorescence microscopy observations, variants ∆9-10, ∆11, ∆11q are characterized by a different localization than BRCA1 wt and ∆9-10. Indeed, ∆9-10 and BRCA1 wt variants localize predominantly at the nuclear level with few foci. In contrast, the variants ∆9-11, ∆11, and ∆11q have a uniquely cytoplasmic localization in multiple spots. So, it corroborates what has been shown in previous work that pathogenic variants have cytoplasmic localization in S. cerevisiae