GA’s universal inhibition of viral protein-mediated cell-cell fusion indicates that its inhibitory effect is by a common fusion mechanism. LPC also universally blocks membrane fusion; it does so by conferring spontaneous positive curvature, which prevents hemifusion. This block can be relieved, regardless of fusion protein, by the addition of the negative spontaneous curvature agent OA25,26. The finding that OA relieves the GA-induced inhibition of EBOV GP-mediated fusion implies that, similar to LPC, GA acts by producing positive spontaneous curvature and this prevents hemifusion. A number of rigid amphipathic fusion inhibitors (RAFI) with positive spontaneous curvature have been shown to inhibit fusion induced by unrelated viral fusion proteins27. In the future, it would be interesting to measure the values of the spontaneous curvatures of GA and RAFI and to relate them to the concentration of OA necessary to relieve fusion inhibition. For the inhibition of non-enveloped adenovirus, we suggest that since GA affects lipid bilayer curvature, it would be predicted to affect the endocytic entry of a non-enveloped virus such as adenovirus. In addition, as we report here, GA appears to have potential secondary mechanisms of viral DNA and protein synthesis inhibition, and these would be predicted to be targeted in both enveloped and non-enveloped viruses.