Revealing the atomistic details behind the binding of B7–1 to CD28 and CTLA-4: A comprehensive protein-protein modelling study

• Published in: Biochimica et biophysica acta. General subjects Vol. 1862; no. 12; pp. 2764 - 2778
• Main Authors: Ganesan, Aravindhan, Moon, Tae Chul, Barakat, Khaled H.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.12.2018
• Subjects: B7-1 Antigen - chemistry; B7-1 Antigen - metabolism; B7–1; binding capacity; CD28; CD28 Antigens - chemistry; CD28 Antigens - metabolism; Cluster Analysis; CTLA-4; CTLA-4 Antigen - metabolism; energy; Humans; Immune checkpoints; Immunotherapy; Ligands; Lymphocyte Activation; Molecular dynamics; Molecular Dynamics Simulation; neoplasms; Protein Binding; Protein-protein docking; protein-protein interactions; receptors; Reproducibility of Results; synapse; T-lymphocytes; CD28; Molecular dynamics; B7–1; Protein-protein docking; CTLA-4; Immune checkpoints
• ISSN: 0304-4165; 1872-8006; 1872-8006
• DOI: 10.1016/j.bbagen.2018.08.010

CD28 and CTLA-4 are homologous T-cell receptors that bind with B7–1 and produce two opposing immunological signals required for T-cell activation and inactivation, respectively. It has been clinically proven that specific blockade of these key protein-protein interactions at the synapse can offer immunotherapeutic benefits for cancers and autoimmune treatments. Hence, there is a growing interest towards developing anti-CD28 and anti-CTLA-4 small molecule inhibitors. To achieve this goal, it is important to understand unique molecular level fingerprint interactions that stabilize CTLA-4/B7–1 and CD28/B7–1 complexes. However, until recently, the structure of the human CD28/B7–1 complex has not been resolved experimentally, which remains a significant setback in achieving specific inhibitors against CTLA-4 or CD28. Here, we employed a combination of advanced molecular modelling and extensive molecular dynamics (MD) simulations to model the CD28/B7–1 complex and characterize the key interactions that stabilize the complex. Ensemble protein-protein docking and MD-based binding-free energy calculations were used to obtain a comprehensive structural model of the CD28/B7–1 complex, which was validated with various mutation-based experimental data from literature. Our CD28/B7–1 model has much weaker binding affinity than the CTLA-4/B7–1 complex, which is in agreement with the results from our binding assay experiments and previous studies. Per-residue energy decomposition of the binding affinities of the two complexes revealed the unique fingerprint hot-spot sites in CTLA-4/B7–1 and CD28/B7–1 complexes. The results presented in this work will, on a long-run, be useful to develop new generation of specific CD28 and CTLA-4 inhibitors for targeted immunotherapy. •A model of human CD28/B7–1 complex is reported using rigorous in silico modelling approaches.•They key fingerprint interactions in the CD28/B7–1 and the CTLA-4/B7–1 complexes are analyzed.•Results propose a CTLA-4-like bivalent interactions between human CD28 and B7–1.•Key residues and potential hot-spots on the key immune checkpoints are reported.