Physic-based computational estimation of binding affinities

This will allow researchers to estimate protein-ligand binding affinity

Examining experimentally the "millions" of available chemical compounds with drug-like properties is a formidable task. Researchers analyze drug databases computationally, filter them based on structure, chemical characteristics, and affinities towards target proteins, and then test the final, high-affinity molecules in the laboratory. Existing scoring systems find it difficult to account for entropy and desolvation effects, despite improvements in processing power. Even though computational approaches for incorporating protein flexibility in scoring are challenging and under development, they remain a potential option for protein-ligand binding affinity estimation. Quantitative agreement between computational and experimental estimates will result from the application of a binding affinity framework based on physics and employing unique free energy calculation methodologies. A method for estimating binding affinities that takes into account the conformational dynamics of both proteins and ligands and employs improved sampling methods within a unified framework could expedite the process of locating new drug-like molecules with accurate target binding affinities in enormous databases.
To calculate binding affinities, the proposed method employs cutting-edge technologies such as enhanced sampling generated from biased computer simulations, a theoretical framework based on Riemannian geometry, and an orientation quaternion formalism in lieu of the conventional three Euler angles.
Our method generalizes and simplifies previously suggested stratification strategies that employ umbrella sampling or other enhanced sampling simulations with additional collective-variable-basedrestrictions. This method employs a versatile scheme that can be easily adapted to any system of interest. We estimate the binding affinity of human fibroblast growth factor 1 to heparin hexasaccharide using the crystal structure of the complex as the initial model and four variations of the proposed method to compare with the experimentally determined binding affinity obtained from isothermaltitration calorimetry experiments.

Research article:
Binding affinity estimation from restrained umbrella sampling simulations

Nanoparticle Synthesis

Designing Recombinant Proteins for Peptide-Directed Nanoparticle Synthesis

In this study, we investigate in a comparative manner how a specific peptide known as the Pd4 and its two known variants may form nanoparticles both in an isolated state and when attached to the green fluorescent protein (GFPuv). More importantly, we introduce a novel computational approach to predict the trend of the size and activity of the peptide-directed nanoparticles by estimating the binding affinity of the peptide to a single ion.

I used molecular dynamics (MD) simulations to explore the differential behavior of the isolated and GFP-fused peptides and their mutants.
Some questions I seek to address with this work:

1) How Mutations in peptide effect the size and shape of the Nanoparticle?

2) How GFP would control the peptide conformational dynamics?

3) Can GFP overcome the mutational effect of peptide on Nanoparticle Size?

4) Is secondary important for catalytic reaction activity?

5) Estimating peptide amino acid and nanoparticle binding free energy helps understand Nanoparticle formation?
The findings reveal that whereas isolated Pd4 and its two known variations (A6 and A11) form nanoparticles of various sizes, fusing these peptides to the GFPuv protein yields nanoparticles of identical size and activity. To put it another way, the GFPuv decreases the nanoparticles' peptide sequence sensitivity. This research develops a computational framework for creating free and protein-attached peptides, which aids in the creation of nanoparticles with well-controlled characteristics.

Research article:
Developing a rational approach to designing recombinant proteins for peptide-directed nanoparticle synthesis

Mechanosensitive Channel of Large Conductance (MscL)
Spontaneous Activation

Applying Biased and Unbiased Molecular Dynamics

This work paves the way for a computational framework for engineering more efficient pH-sensing mechanosensitive channels.

Understanding the chemical basis of an engineered mechanosensitive channel's spontaneous activation. The orientation-based strategy used to generate and optimize an open model of modified MscL in this research is a promising tool for creating unknown protein states and exploring ion channel activation mechanisms. This study aids research into pH-triggered drug delivery liposomes (DDL), which use MscL as a nanovalve.

Research article:
Elucidating the molecular basis of spontaneous activation in an engineered mechanosensitive channel

An Investigation of the YidC-Mediated Membrane Insertion of Pf3 Coat Protein

Mechanistic Study of Protein Mechanism Pathway



YidC is a membrane protein that facilitates the insertion of newly synthesized proteins into lipid membranes. Through YidC, proteins are inserted into the lipid bilayer via the SecYEG-dependent complex. Additionally, YidC functions as a chaperone in protein folding processes. Several studies have provided evidence of its independent insertion mechanism.
The mechanistic details of the YidC independent protein insertion mechanism remain elusive at the molecular level. In this study, we looked at the local and global conformational changes of YidC associated with Pf3 insertion into the hydrophilic groove, Pf3 interactions with YidC and the membrane, and conformational changes in Pf3 that occurred during the insertion process.
The incoming Pf3 coat protein will first get intact with the hydrophilic groove located in the trans-membrane region forming a salt-bridge with ARG72. The positive charged arginine will form a salt bridge with the negative charged amino acid of Pf3 coat protein D7. The salt bridge formation play a very big key role in the insertion mechanism of YidC. The Pf3 coat protein then move towards the periplasmic side of the membrane acted by a force of hydrogen bond attraction with the E2 region of the YidC protein this interaction will help the protein movement towards periplasmic side which is also assisted by the salt-bridge between D18 of Pf3 and ARG72 of YidC this salt-bridge and hydrogen bond combination will stabilize the Pf3 protein position in the membrane. On other hand for pushing Pf3 protein in to the hydrophilic groove of the YidC the loops of YidC on the cytoplasmic side of the membrane are very crucial. Since there are interaction involved with this region we believe that cytoplasmic loops form a strong contact with the Pf3 coat. The protein infiltrates across the membrane within water-filled cleft , leaving the adjoining hydrophobic TM region in the lipid bilayer.

Conference Poster:
Membrane Insertion Of A Pf3 Coat Protein Using MD Simulations

Developing Efficient Transfer Free Energy Calculation Methods
For Hydrophobicity Predictions

Molecular dynamics for Hydrophobicity Prediction

The interaction of peptides with membrane lipids is significant in the biological processes. Short peptides are an excellent alternative to the immune response antibodies, and they play a very crucial role in binding, insertion, and folding of membrane proteins. The characterization of solvent dependent conformational ensemble of the peptides is required for a molecular-level understanding of the thermodynamic hydrophobicity scale. To characterize the solvent-peptide interactions, we have developed a computational procedure that allows us to accurately model the peptides in both aqueous and organic solvent conditions and determine their properties at a thermodynamic level. This study evaluates the peptide conformational dynamics at different temperatures using molecular dynamics (MD) in the explicit solvent of water and octanol to estimate the transfer free energies accurately and to predict the partition coefficients. We have used a series of equilibrium MD simulations, and alchemical free energy calculations to measure the transfer free energies within various approximations. This study sheds light on the efficiency and accuracy of several different computational strategies for the study of transfer free energies.

Conference Poster:
Developing Efficient Transfer Free Energy Calculation Methods For Hydrophobicity Predictions

Influenza Hemagglutinin (HA) is a Paradigm for
Protein-mediated Membrane Fusion

Molecular dynamics of Membrane Fusion

The proposed research project aims to study the conformational dynamics of the complete trimeric influenza HA ectodomain for the first time at an atomic level. In particular, we will study the confor- mational changes of HA that are triggered by protonation of a highly-conserved histidine residue in the HA2 hinge region. This will be followed by an investigation of the conformational transitions of HA2 upon the full exposure to solvents due to removal of HA

Conference Poster:
Molecular Dynamics Investigation of The Ph-Dependent Influenza Hemagglutinin Conformational Changes

Conformational free energy landscapes of
SARS coronavirus spike glycoproteins

The state-of-the-art enhanced sampling molecular dynamics simulations

The proposed research project aims to study understanding how coronavirus spike glycoproteins undergo conformational changes to bind to host ACE2 receptors is key to the development of coronavirus vaccines and therapeutics, which requires a dynamic rather than a static picture to provide a reliable structure-based drug design framework. The virus that causes COVID-19, SARS-CoV-2, is more stable and slower changing than the previous form that caused the SARS outbreak in 2003, according to new computational simulations of the behavior of SARS-CoV-1 and SARS-CoV-2 spike proteins before fusion with human cell receptors.

Conference Poster:
Characterizing The Roles Of Chemomechanical Couplings In The Differential Behavior Of Sars-cov-1 And Sars-cov-2 Spike Glycoproteins

Contact

Adithya Polasa

University of Arkansas

Department of Chemistry and Biochemistry

apolasa@uark.edu