Background: Ebola virus causes a hemorrhagic fever syndrome that is associated with high mortality in humans. In the absence of effective therapies for Ebola virus infection, the development of a vaccine becomes an important strategy to contain outbreaks. Immunization with DNA and/or replication-defective adenoviral vectors (rAd) encoding the Ebola glycoprotein (GP) and nucleoprotein (NP) has been previously shown to confer specific protective immunity in nonhuman primates. GP can exert cytopathic effects on transfected cells in vitro, and multiple GP forms have been identified in nature, raising the question of which would be optimal for a human vaccine.
Methods and findings: To address this question, we have explored the efficacy of mutant GPs from multiple Ebola virus strains with reduced in vitro cytopathicity and analyzed their protective effects in the primate challenge model, with or without NP. Deletion of the GP transmembrane domain eliminated in vitro cytopathicity but reduced its protective efficacy by at least one order of magnitude. In contrast, a point mutation was identified that abolished this cytopathicity but retained immunogenicity and conferred immune protection in the absence of NP. The minimal effective rAd dose was established at 10(10) particles, two logs lower than that used previously.
Conclusions: Expression of specific GPs alone vectored by rAd are sufficient to confer protection against lethal challenge in a relevant nonhuman primate model. Elimination of NP from the vaccine and dose reductions to 10(10) rAd particles do not diminish protection and simplify the vaccine, providing the basis for selection of a human vaccine candidate.