J.R. "Keng" Pineda, MD

Personal Information
Cardiovascular Disease Fellow
Research projects: Simulation of molecular dynamics and ligand binding to the thin filament to study cardiotoxic effect of some chemotherapeutics

Research project goals at Sarver Heart Center:

  1. To continue to develop specialized skills in simulation of molecular dynamics and ligand binding to the thin filament. Many chemotherapeutic drugs are known to cause cardiomyopathy. We hypothesize that the cardiotoxic effect of some chemotherapeutics is in part due to interaction with the thin filament.

Human fetal cardiac TnT differs from the adult isoform though alternative RNA splicing. Inclusion of exon 5 in the fetal isoform adds ten amino acids, EEDWREDEDE, into the N-terminal region of the adult cardiac TnT (1). These are inserted between residue 23 (E) and residue 24 (Q) of the adult isoform.

The Tardiff lab previously published a full atomic model of the human thin filament which contains two adult isoforms of troponin T. I built the fetal TnT into the previously published atomic model. A homology model of the entire fetal TnT (accession NP_001263274) was generated using Phyre2. The Phyre2-generated model of human fetal TnT is a fully alpha-helical structure. I started with the assumptions that the structure of cTnT in the core troponin complex is conserved. As such, the first twenty three amino acids in the N-terminal of a TnT in the thin filament model were replaced with the first thirty three amino acids of the homology model of fetal TnT.

I performed physics-based refinement of the model using CHARMM. Using the Charmm36m forcefield, we performed energy minimization and subsequently carried out a 75 picosecond Langevin dynamics simulation at 298 K using the Generalized Born Molecular Volume (GBMV) implicit solvent utility in CHARMM. During the initial energy minimization, only the first thirty four N-terminal amino acids of the fetal TnT were allowed to move. Subsequently, the first one hundred N-terminal amino acids in the fetal TnT were allowed to move in the static field of the rest of the thin filament model during Langevin dynamics.

The amino acid difference between fetal cTNT and adult TNT involves insertion of a segment of acidic residues. These residues are highly charged and hydrophilic. It is noteworthy that there are also many highly charged acidic residues in neighboring proteins in the complex. It will be interesting to observed how this highly charge N-terminal tail of the fetal isoform will behave differently compared to the adult cTNT. This will be an objective in our long time scale molecular dynamics simulation of the fully solvated thin filament.

  1. To investigate the potential interaction of selected chemotherapeutic compounds with the thin filament model.

The overall objective of the computational biophysical component of my research is to carry out classical mechanical simulations of the interaction of the thin filament with selected small molecule ligands. In order to do this, one has to develop forcefield parameters for each of these compounds. To fulfill this objective, I will be using available online tools in conjunction with quantum chemical simulations to develop forcefield parameters for the small molecule ligands.

I will group the compounds in terms of severity of associated LVEF depression and conduct an analysis of molecular similarity (similar to defining a pharmacophore in drug development). A single agent from each group will be selected based on availability and cost. Further investigation of the interaction of the representative compound with the cardiac thin filament will be conducted utilizing a hybrid of solution-based experiment (isothermal titration calorimetry) and computational biophysics.


Fellowship Years: 
2015 - 2018