123
Modeling of Biomolecular Systems Interactions, Dynamics, and Allostery Poster Session II
70-POS
Board 23
Computational Investigation of the Energy Exchange Pathways in Peptide-loaded Major
Histocompatibility Complex Proteins
Onur Serçinoglu, Gülin Özcan,
Pemra Özbek
.
Marmara University, Istanbul, Turkey.
The dynamic behavior of peptide-loaded Major Histocompatibility Complex (pMHC) plays a
role in defining the immunogenicity of the pMHC-T-cell Receptor interaction. However, due to
the highly polymorphic nature of the antigen chain of the complex (Human Leukocyte Antigens,
HLA) and the limitation of experimental and computational methods, studying the association
between disease states and pMHC dynamics is difficult. In this work, we aim to perform an
overall functional characterization of HLA alleles by employing computational methods with an
emphasis on identifying energy exchange pathways within the residue interaction network. As an
initial step, a dataset was constructed consisting of a total of 389 HLA proteins in complex with
peptide epitopes using the Protein Data Bank and manually-curated Immune Epitope Database as
sources. In addition, disease associations of alleles were identified using relevant information
available from the literature. As a fast and efficient screening method to find energy exchange
pathways, we employed a coarse-grained method based on the perturbation of Kirchhoff matrix
of the Gaussian Network Model (GNM). In order to test the effectiveness and validity of this
method, we performed initial 20 ns long atomistic molecular dynamics (MD) simulations of B-
27:05 and B-27:09 subtypes which are differentially associated with Ankylosing Spondilitis
using NAMD molecular dynamics software and CHARMM27 force-field. A Principal
Component Analysis of the cross-correlations of residue fluctuations showed a large overlap
between the predictions of the method and dynamics simulations. Our approach of GNM-based
screening followed by MD simulations establishes a framework for our upcoming studies, in
which we plan to extend our analysis to other allele and subtypes in our dataset and complement
it with more exhaustive methods such as atomic packing analysis.
