Screening and Evaluation of Natural Product Derivative Library for Anticancer Activity in Human Prostate Cancer Cells

Abstract

Natural products have been a source for almost all major FDA approved drug categories. Presently, almost 50% of all small molecule agents in clinical use are either directly or indirectly derived from natural products. Screening of natural products and their derivatives has gained prominence due to the advances made in combinatorial chemistry and cheminformatics. Both have facilitated high throughput screening methods for lead identification and lead optimization in the drug discovery process. For the purpose of our study we utilized Tim-Tec's Natural Product Derivative Library (NDL-3000). We screened 3000 compounds for their anti-cancer activity in the prostate cancer cell line, PC-3. First, we employed the MTS cell-viability assay for screening and those that caused a decrease in cell-viability by 50% or higher were considered “hits” and potential candidates for further evaluation. Second, we employed the cheminformatics software tools from Molinspiration to calculate important physicochemical properties such as LogP, number of hydrogen bond donors and acceptors and molecular planarity for the ten “hit” compounds obtained from the screen. Lastly, we used the SwissTragetPrediction web tool as well as Molinspiration to predict the biological activity of these ten compounds. We proceeded to conduct a preliminary evaluation of the anti-cancer activity of one of the “hit” compound, ST-985 in three prostate cancer cell lines: PC-3, DU-145 and LNCaP. As part of this evaluation, we calculated the GI50 value of ST-985 in the three prostate cancer cell lines and assessed the mechanism of cell death by examining the markers of apoptosis (caspase-7 and PARP) and autophagy (LC3-B). We have identified one lead compound, ST023985, from our screening of the NDL-3000 library on the basis of the MTS assay results followed by utilization of cheminformatics web tools. GI50 values of ST-985 in the three prostate cancer cells were found to be 8.31 +/- 0.6384 µM in PC-3, 5.492 +/- 1.124 µM in LNCaP, and 9.148+/- 0.0915 µM in DU-145 cells. Results from Molinspiration showed that ST-985 did not violate the “Lipinski five rule” for oral drugs. This compound has molecular weight (MW) 360.41 (less than 500), partition coefficient (logP) 3.75 (less than 5), the numbers of hydrogen bond donors 1 (less than 5) and acceptors 5 (less than 10). Both SwissTargetPrediction and Molinspiration web tools have predicted ST-985 to be a GPCR ligand. Finally, ST-985 induced apoptosis as evidenced by cleaved caspase-7 and PRAP as well as increased autophagy marker, LC3-B in a dose dependent manner. Future studies are aimed at further understanding the mechanism of anti-cancer activity of ST-985 and other lead compounds obtained from the screening of NDL-3000 and studying their molecular targets.