Unusual nucleic acid structures in the RAS genes and design of anti-cancer strategies

The ras genes are promising therapeutic targets in anticancer strategies. Somatic mutations on the genes have been found in 30 % of all human cancer. More than 90 % of pancreatic tumours are driven by mutations in the Kirsten (KRAS) gene, while about 30 % of bladder tumours are characterized by mutations in Harvey (HRAS) gene. G-quadruplexes (or G4) are non-canonical structures that can be adopted by guanine-rich DNA and RNA sequences. Bioinformatic tools have shown a non-random distribution of G4 in the human genome. A high level of G4 motifs have been found in gene promoters (including the ras genes), upstream of the transcription start site (TSS) and in the 5’-UTR and 3’-UTR untranslated regions. This finding strongly supports a role of G4 DNA in several biological processes, including a regulatory function in gene expression. Starting from this consideration my work focused on the promoter region of the HRAS and on the KRAS 5’-UTR mRNA. My research has characterized the unusual structures, G-quadruplex and i-motif (or iM), formed by these genes, in order to design new anti-ras strategies to inhibit the oncogenes in bladder and pancreatic cancer cells. Specifically, we investigated the role of the i-motif formed by the C-rich strand of the critical CG region controlling the expression of HRAS. I have characterized the i-motif under near-physiological conditions and discovered that it is recognized and unfolded by hnRNP A1, one of the most abundant eukaryotic protein. We proposed a model in which HRAS transcription is regulated by a G4/iM switch interacting with proteins that recognize non B-DNA conformations. As MAZ and hnRNP A1 unfold the G4 and iM, respectively, and are essential for the HRAS expression, we developed a decoy strategy to reduce the expression of the oncogene in bladder T24 cancer cells. We designed G4 decoy oligonucleotides mimicking the critical HRAS quadruplex structure in order to sequester the transcriptional factors (hnRNP A1, MAZ, SP1) that normally promote HRAS transcription. The G4 oligonucleotides were engineered with anthraquinone insertions and LNA residues to increase their bioactivity. They significantly suppressed the expression of HRAS and showed a strong antiproliferative effect, suggesting that they are promising agents in cancer therapy. The last part of my PhD work focused on KRAS: the gene responsible of the malignant transformation of pancreatic cancer cells. We considered the 5’- and 3’-UTR sequences of the gene. As the 5’-UTR of KRAS is very rich in G4 motifs, we investigated the formation of RNA G-quadruplexes (RG4s) in this particular region of mRNA. By using enzymatic and spectroscopic techniques, we first demonstrated that RG4s are present in the 5’-UTR. We then proposed an anti-KRAS strategy based on the used of small molecules recognizing RG4s. The small molecules used are anthrafuranediones and anthrathiophenediones as they are taken up efficiently by cancer cells and, upon binding to RG4s, they inhibit translation, as demonstrated by luciferase, RT-PCR and western blot assays. Reduced KRAS protein resulted in apoptosis and inhibition of cell growth. A second approach for KRAS was based on targeting the 3’-UTR region. This regulatory region is recognized by endogenous microRNAs (miRNAs). By examining datasets reporting miRNA expression profiles in normal and pancreatic cells, we found that miRNA 216b is downregulated pancreatic tumors. We designed single-stranded miR-216b mimics, engineered with unlocked nucleic acid (UNA) modifications, and found that they strongly suppressed KRAS in human pancreatic cell lines. Moreover, we tested a new delivery strategy for miR-216b, which was based on the use of palmitoyl-oleyl-phosphatidylcholine (POPC) liposomes functionalized with lipid-modified miR-216b and lipid-modified cell penetrating TAT peptide. The modified miRNAs showed to be very strong anti KRAS effector molecules.