Design, synthesis and biological evaluation of substituted naphthalene diimides as anticancer agents

It has been proved that naphthalene diimide (NDI) derivatives display anticancer properties as intercalators and G-quadruplex-binding ligands, leading to DNA damage, senescence and down-regulation of oncogene expression. This thesis deals with the design and synthesis of disubstituted and tetrasubstituted NDI derivatives endowed with anticancer activity, interacting with DNA together with other targets implicated in cancer development. Disubstituted NDI compounds have been designed with the aim to provide potential multitarget directed ligands (MTDLs), in order to create molecules able to simultaneously interact with some of the different targets involved in this pathology. The most active compound, displayed antiproliferative activity in submicromolar range, especially against colon and prostate cancer cell lines, the ability to bind duplex and quadruplex DNA, to inhibit Taq polymerase and telomerase, to trigger caspase activation by a possible oxidative mechanism, to downregulate ERK 2 protein and to inhibit ERKs phosphorylation, without acting directly on microtubules and tubuline. Tetrasubstituted NDI compounds have been designed as G-quadruplex-binding ligands endowed with anticancer activity. In order to improve the cellular uptake of the lead compound, the N-methylpiperazine moiety have been replaced with different aromatic systems and methoxypropyl groups. The most interesting compound was 1d, which was able to interact with the G-quadruplexes both telomeric and in HSP90 promoter region, and it has been co-crystallized with the human telomeric G-quadruplex, to directly verify its ability to bind this kind of structure, and also to investigate its binding mode. All the morpholino substituted compounds show antiproliferative activity in submicromolar values mainly in pancreatic and lung cancer cell lines, and they show an improved biological profile in comparison with that of the lead compound. In conclusion, both these studies, may represent a promising starting point for the development of new interesting molecules useful for the treatment of cancer, underlining the versatility of the NDI scaffold.

Study of novel natural compounds as potential drugs in the treatment of acute leukemias

Introduction: Among all cancer types leukemia represents the leading cause of cancer death in man younger than 40 years. Single-target drug therapy has generally been highly ineffective in treating complex diseases such as cancer. A growing interest has been directed toward multi-target drugs able to hit multiple targets. In this context, plant products, based on their intrinsic complexity, could represent an interesting and promising approach. Aim of the research followed during my PhD was to indentify and study novel natural compounds for the treatment of acute leukemias. Two potential multi-target drugs were identified in Hemidesmus indicus and piperlongumine. Methodology/Principal Findings: A variety of cellular assays and flow cytometry were performed on different cell lines. We demonstrated that Hemidesmus modulates many components of intracellular signaling pathways involved in cell viability and proliferation and alters gene and protein expression, eventually leading to tumor cell death, mediated by a loss of mitochondrial transmembrane potential, raise of [Ca2+]i, inhibition of Mcl-1, increasing Bax/Bcl-2 ratio, and ROS formation. Moreover, we proved that the decoction causes differentiation of HL-60 and regulates angiogenesis of HUVECs in hypoxia and normoxia, by the inhibition of new vessel formation and the processes of migration/invasion. Clinically relevant observations are that its cytotoxic activity was also recorded in primary cells from acute myeloid leukemia (AML) patients. Moreover, both Hemidesmus and piperlongumine showed a selective action toward leukemic stem cell (LSC). Conclusions: Our results indicate the molecular basis of the anti-leukemic effects of Hemidesmus indicus and indentify the mitochondrial pathways, [Ca2+]i, cytodifferentiation and angiogenesis inhibition as crucial actors in its anticancer activity. The ability to selectively hit LSC showed by Hemidesmus and piperlongumine enriched the knowledge of their anti-leukemic activity. On these bases, we conclude that Hemidesmus and piperlongumine can represent a valuable strategy in the anticancer pharmacology.

Design and Synthesis of Naphthalene Diimides Based Small Molecules as Anticancer Agents: Targeting the Polyamine Transporter, G-quadruplex Structures and HDAC

Cancer is a multifactorial disease characterized by a very complex etiology. Basing on its complex nature, a promising therapeutic strategy could be based by the “Multi-Target-Directed Ligand” (MTDL) approach, based on the assumption that a single molecule could hit several targets responsible for the pathology. Several agents acting on DNA are clinically used, but the severe deriving side effects limit their therapeutic application. G-quadruplex structures are DNA secondary structures located in key zones of human genome; targeting quadruplex structures could allow obtaining an anticancer therapy more free from side effects. In the last years it has been proved that epigenetic modulation can control the expression of human genes, playing a crucial role in carcinogenesis and, in particular, an abnormal expression of histone deacetylase enzymes are related to tumor onset and progression. This thesis deals with the design and synthesis of new naphthalene diimide (NDI) derivatives endowed with anticancer activity, interacting with DNA together with other targets implicated in cancer development, such as HDACs. NDI-polyamine and NDI-polyamine-hydroxamic acid conjugates have been designed with the aim to provide potential MTDLs, in order to create molecules able simultaneously to interact with different targets involved in this pathology, specifically the G-quadruplex structures and HDAC, and to exploit the polyamine transport system to get selectively into cancer cells. Macrocyclic NDIs have been designed with the aim to improve the quadruplex targeting profile of the disubstituted NDIs. These compounds proved the ability to induce a high and selective stabilization of the quadruplex structures, together with cytotoxic activities in the micromolar range. Finally, trisubstituted NDIs have been developed as G-quadruplex-binders, potentially effective against pancreatic adenocarcinoma. In conclusion, all these studies may represent a promising starting point for the development of new interesting molecules useful for the treatment of cancer, underlining the versatility of the NDI scaffold.

Anticancer drug discovery using artificial intelligence: an application in pharmacological activity prediction

Hematological cancers are a heterogeneous family of diseases that can be divided into leukemias, lymphomas, and myelomas, often called “liquid tumors”. Since they cannot be surgically removable, chemotherapy represents the mainstay of their treatment. However, it still faces several challenges like drug resistance and low response rate, and the need for new anticancer agents is compelling. The drug discovery process is long-term, costly, and prone to high failure rates. With the rapid expansion of biological and chemical "big data", some computational techniques such as machine learning tools have been increasingly employed to speed up and economize the whole process. Machine learning algorithms can create complex models with the aim to determine the biological activity of compounds against several targets, based on their chemical properties. These models are defined as multi-target Quantitative Structure-Activity Relationship (mt-QSAR) and can be used to virtually screen small and large chemical libraries for the identification of new molecules with anticancer activity. The aim of my Ph.D. project was to employ machine learning techniques to build an mt-QSAR classification model for the prediction of cytotoxic drugs simultaneously active against 43 hematological cancer cell lines. For this purpose, first, I constructed a large and diversified dataset of molecules extracted from the ChEMBL database. Then, I compared the performance of different ML classification algorithms, until Random Forest was identified as the one returning the best predictions. Finally, I used different approaches to maximize the performance of the model, which achieved an accuracy of 88% by correctly classifying 93% of inactive molecules and 72% of active molecules in a validation set. This model was further applied to the virtual screening of a small dataset of molecules tested in our laboratory, where it showed 100% accuracy in correctly classifying all molecules. This result is confirmed by our previous in vitro experiments.