Clear cell renal cell carcinoma (ccRCC) is the most common form of kidney cancer, comprising of about 80 percent of all cases. A hallmark of disease progression is a loss of function of the von Hippel-Lindau tumor suppressor gene, which leads to a highly vascularized tumor microenvironment. Additionally, virtually all patients develop progressive drug-refractory disease limiting the effectiveness of antiangiogenic therapies as a treatment plan. With the role of cytoskeleton dysregulation in tumor progression, we suggest actin-binding proteins as an alternate treatment route. In particular we show that profilin1 (Pfn1), a known actin-binding protein, is important to ccRCC migration and proliferation in vitro. We developed two novel computationally designed Pfn1-actin inhibitors, which reduce proliferation and migration of ccRCC cells in vitro and ccRCC tumor growth in vivo. The first inhibitor was identified via a computational screen. The second inhibitor was identified by using the first compound as a seed for a secondary virtual screen. Notably, as the final step of the secondary screen we utilized gnina, a convolutional neural network based scoring function, to rank the molecules. This work provides both a proof-of-concept of both Pfn1 as a treatment target and an experimentally validated application of neural networks in the drug discovery context.

Convolutional neural networks identify novel antiangiogenic cancer treatment

Where thousands of chemistry professionals meet to share ideas and advance scientific and technical knowledge. Spring 2023 abstracts, posters, and presentations will launch at the end of March.

ACS Spring 2023

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.