Structural modelling of the Resistin family proteins and their cognate antibodies

Embargo until
Journal Title
Journal ISSN
Volume Title
Johns Hopkins University
Pulmonary Hypertension (PH) is associated with the increase in blood pressure in the lung vasculature leading to fatigue, dizziness, chest pain and ultimately death. The general prognosis is survival of 2-3 years and then death caused by right ventricular failure. Research is focused on developing therapeutic methods to cure this disease. Recent studies showed that the Resistin and Resistin-like molecule (RELM) family of proteins (adipocyte-specific hormone) are involved in the vascular remodeling and cardiac dysfunction seen in animal and human pulmonary arterial hypertension (PH). These results suggested human-resistin (hresistin) and human-RELMβ (hRELMβ) are important for the etiology of human PH and are potential biomarkers and therapeutic targets for this disease. This led to development of a series of human antibodies that target distinct and common epitopes of hresistin and hRELMβ by the members of Dr. Roger Johns’ research group. My goal is to create 3-d models of these antibodies in complex with hresistin + hRELMβ to gain insight into their functions. Using homology modelling techniques such as SWISS-MODEL and MODELLER, I predicted the three-dimensional structures of hresistin and hRELMβ, using the crystallographic structures of mouse-resistin and mouse-RELMβ. Using RosettaAntibody protocol, I modelled the three-dimensional structures of a series of 32 human antibody sequences designed specifically to target hresistin and hRELMβ. These models confirmed the antibody’s stability and suitability. The predicted structures of these 32 antibodies validated that they were designed conforming to the standard structural parameters of naturally occurring human antibodies. Experimental studies showed that hresistin and hRELMβ exist in multimer states but that the monomer form is the most functional. Therefore, using the SnugDock protocol in Rosetta, I docked each of the antibodies to hresistin and hRELMβ monomeric forms. From the docking results, I identified the most-likely antibody-antigen docked state using model energies. From the docked structure of each antibody-antigen complex, I extracted the epitope regions of hresistin and hRELMβ. The models suggest that the lead antibody, antibody AntiRes13 binds strongly to the head region of hresistin and strongly to the tail region of hRELMβ. The docking results of antibody AntiRes-13 also validated its corresponding antigen binding experiments. Out of the set of 32 antibodies, antibody AntiRes-2, 3, 9, 11 and 41 also showed positive binding affinities to the antigens. In the models, AntiRes-2 binds hresistin at the head and hRELMβ at the head. Antibody AntiRes-3 binds the tail region of hresistin and head region of hRELMβ. Antibodies AntiRes-9 and AntiRes-11 bind to the head regions of both hresistin and hRELMβ whereas AntiRes-41 prefers to bind the tail region of both antigens. This project gives insight into design for potential therapeutic applications for curing PH.
Pulmonary Hypertension, antibody, antibody-antigen docking, resistin, Rosetta