Using simulations of protein dynamics to understand BCL2 binding with Venetoclax

Wanyue Zou

Mentor: Dr. Mathew McCoy, Innovation Center for Biomedical Informatics (ICBI) and Department of Oncology, Georgetown University Medical Center.

Date/Time: August 24, 2021 at 1:40pm

Abstract: BCL2 belongs to the BCL2 Apoptosis Regulator and regulates the cell death.The overexpression of BCL2 can be a reason why cancer cells have a longer cell life. Venetoclax is the medicine used in cancer therapy to target and bind with BCL2 to allow the cell to undergo apoptosis normally. BCL2 has several mutant forms that have different levels of drug resistance, and it is a challenge to understand mechanisms of drug resistance with current methods. BCL2 has five variants with known different extents of resistance. The 104C is the most resistant variant and the 152A is the least resistant variant. The analysis protein dynamics of these two variants compared with the other variants would be a way to understand how mutations impact the dynamics of the BCL2 protein structure, and generate hypotheses about how mutations impact Venetoclax binding.

The root-mean-square fluctuation (RMSF) is the distance traveled by an atom during a molecular dynamic simulation and can be used to quantify the flexibility of a protein structure. RMSF focuses on the differences of atomic positions during a simulation with selected protein residue range, and was applied to each of the five BCL2 variants. dFLux, the difference of RMSF between a variant and wildtype simulation, is introduced to understand the protein dynamics of the binding process of Venetoclax with BCL2 better. The RMSF is first calculated for each variant to understand the conformation of each atom compared to its initial position. Then the dFlux is the comparison of conformation between the variant and the wild type BCL2. The dFLux plots were used to analyze motion in different regions of the protein strucutre to propose a hypothesis for variant specific resistance mechanisms in BCL2.