The Ebola virus (lat. ebolavirus) is a (-)ssRNA filovirus known for its high transmission and mortality rates. The mortality rate of the epidemic that erupted in 2014 in West Africa, for example, ranged between 50-70%. The disease caused by this virus is transmitted through contact with bodily fluids of the infected. The infection is followed by a severe weakening of the immune system, an inflammatory response, vascular system failure and multi-organ failure. Death occurs after 6-16 days from the first signs of infection. There are currently no effective treatments available. Some vaccines that have been proven to protect against Ebola infection are currently in the final stages of clinical trials.
The ebolaviruses RNA genome has 7 genes coding for at least 10 different proteins among which are glycoprotein (GP) and viral proteins VP24, VP30, VP35 and VP40. Like many other viruses, ebolavirus utilizes host cells for its replication. Knowledge of the viral structure gives us insight into the functioning of the virus and allows us to predict the interactions of individual viral proteins with selected molecules. With specialized software tools we can predict whether a molecule will bind to a specific protein, where it will bind, and how strongly and efficiently it will bind to the selected binding site on the protein. This determines whether the selected molecule is a potential inhibitor of the chosen viral protein and thus a potential cure for infection with virus Ebola.
In this master's thesis, we searched for potential Ebola virus inhibitors using virtual screening methods. We have created a collection of compounds that have presumed therapeutic effect against infection with virus Ebola. Of the compounds collected, many are already used to treat various diseases. As part of this thesis, we looked for compounds that are already available on the market and could also be used in the treatment of Ebola. This approach is called drug repurposing. Using the molecular docking method in the program SeeSAR, we selected a handful of compounds that showed promising results for in silico binding to chosen binding sites of different ebolavirus proteins.
We prepared a second collection of compounds and their corresponding data on the binding efficiencies to the viral protein 35 (VP35). We used this data for generation of two predictive models for identification of potential viral inhibitors. Predictive models QSAR (quantitative structure- activity relationship) and QSPR (quantitative structure-property relationship) were created using the multiple linear regression method. With these predictive models, we can calculate with some certainty, what the binding efficiency of a selected molecule to the VP35 protein binding site, is going to be.
|
