Molecular characterization of calnexin and FABP5 proteins to design strategies for novel effective inhibitors targeting secondary brain cancer

Blood brain barrier (BBB) dysfunction is a hallmark of many Central Nervous System (CNS) disorders. Secondary metastatic cancer of CNS lies on the lack of selective permeability and the consequent trafficking of immune cells through the BBB. At present, research has not yet developed a strategy to prevent this access. It has recently been discovered that Calnexin (Canx) cytoplasmic tail and Fatty Acid Binding Protein 5 (FABP5) interact forming a complex which it is suggested to be responsible for T-cell passage into the CNS. However, there are many issues concerning the mapping of the interaction and its mechanism that have not been explored yet. The purpose of this study is the characterization of this interaction by means of an in-silico model based on Molecular Mechanics computational techniques. To do that, we previously performed a structural atomistic model of the two proteins. On one hand, FABPs family were explored in order to highlight different electrostatics features among its members, on the other hand we adopted a homology strategy for the building of the Canx tail model. We refined this model through a multiscale approach. Initially, a Coarse-grained force-field was assigned to the structure in order to extend the simulation time scale and to obtain an effective conformational sampling. At a later stage, the model was mapped back and simulated in a nanoscale level. The computational flow was entirely conducted in an explicit aqueous environment which included also a portion of phospholipid bilayer. Thus, we dealt with the mechanism of the protein aggregation. In this context we investigated new binding sites located in regions of potential interaction and performed a geometric Virtual Screening of the National Cancer Institute (NCI) database to identify putative inhibitors of the complex. Binding energy calculations were then performed using an MD application known as MM-GBSA (Molecular Mechanics General Born Surface Area). This made a meaningful comparison with experimental complex Kd possible. The results indicated which residues could take part in the formation of the complex and suggested how we could stop these interactions through a ligand obstruction. Hence, we provided new insights to in-vitro studies. The last part of the work was inspired by the well-known “portal” region in FABPs. It is assumed that fatty acids reach the protein binding site by penetrating through this dynamic area. Thus, we supplemented the all-atom MD with a second approach based on a Principal Component Analysis (PCA) in order to investigate the principal modes of the protein. We verified for the first time the consistency of the opening-closing motion with one of the main vibration modes of the protein, corroborating the “portal” hypothesis. Subsequently, we wanted to find out if there was any relation between the “breath” movement and protein’s residues. In conclusion, we employed in-silico techniques to analyse Calnexin and FABP5 separately and then as a complex, proposing a way to inhibit their interaction. This could result in future advances in therapy design for metastatic secondary brain cancer and other conditions.