New small molecules for the Treatment of TNBC innovative Polypharmacological In Silico Approach

Abstract

In silico methods represent a crucial resource in the development of targeted therapies for breast cancer, particularly for tumor subtypes resistant to current treatments. Triple-negative breast cancer (TNBC) is notorious for its aggressiveness and high risk of recurrence, often due to the due to the lack of receptors for human epidermal growth factor 2 (HER2), estrogen (ER), and progesterone (PR). As a result, treatment options for TNBC are limited to the use of conventional chemotherapy, the efficacy of which is unfortunately generally modest. Therefore, research is focused on identifying new potential targets involved in the progression of this neoplasm. Studies conducted by Turdo and colleagues have demonstrated that the combination of olaparib and dinaciclib, inhibitors of PARP-1 and CDK-1, respectively, leads to a significant reduction in the growth of TNBC cells with BRCA mutations, while maintaining the health of normal breast cells. To this end, a ligand- and structure-based computational protocol was applied in the present study to identify novel heterocyclic molecules with dual-target activity. The polypharmacological approach with simultaneous inhibition of PARP-1 and CDK1 could be a successful strategy in the treatment of triple-negative breast cancer (TNBC). From a medicinal chemistry perspective, the development of a single molecule with dual CDK-1/PARP-1 inhibitory effect could be an attractive therapeutic strategy to overcome the pharmacokinetic/dynamic drawbacks of the co-administration regimes. Figure shows the research workflow. The experimental in silico protocol begins with the creation of a database of novel compounds analogues of flavonoid natural products, such as flavones, flavonols, flavanones, flavanols, anthocyanins, and isoflavonoids, which are known for their biological properties and ability to modulate pathophysiological processes associated with various types of cancer. In particular, the amentoflavone, a naturally occurring biflavonoid isolated from Selaginella moellendorfii Hieron, showed a potent inhibition activity on PARP-1 with IC50 = 198 nM and over 50-fold selectivity compared to PARP-2. However, like most flavonoids, amentoflavone presents several problems related to pharmacokinetic, pharmacodynamic, and physicochemical aspects, such as low metabolic stability due to the presence of endocyclic oxygen and free aromatic hydroxyl groups. To overcome these challenges, a scaffold-hopping approach was chosen to build the database. This approach, widely established in the literature, represents a valuable tool to obtain more effective and easily synthesizable molecules. Specifically, the oxygen atom at position 1 of the flavonoid was replaced by a nitrogen atom. Subsequently, using our proprietary "Biological Target Finder" (BPT) tool, which is based on molecular descriptors, it was possible to predict the binding affinity between our in-house aza-flavonoid database and the two biological targets of interest, PARP-1 and CDK-1. A set of molecules was selected according to the best multitarget score, also in relation to the ligand-affinity scores obtained for the reference compounds dinaciclib and olaparib, as modulators of CDK-1 and PARP-1. Finally, further molecular docking studies were performed on the selected compounds to assess their ability to interact with the binding sites of the target proteins. Conformational changes between ligands and proteins were explored by Induced Fit Docking (IFD) studies and molecular dynamics simulations (MDS) to investigate the selectivity of the identified structures towards PARP-1 and CDK-1.