Engineering cyanovirin-N for enhanced viral neutralization

Abstract

Cyanovirin-N (CVN) is an 11-kDa lectin originally isolated from the cyanobacterium Nostoc ellipsosporum during a high-throughput screen for novel anti-HIV activities. In addition to having anti-HIV activity, CVN has since been shown to neutralize a number of other enveloped viruses including influenza and Ebola. This antiviral activity is attributed to two homologous carbohydrate binding sites that specifically bind α(1-2)-linked oligomannose glycosylation sites present on many envelope glycoproteins. Because of its broad ability to neutralize enveloped viruses, CVN is a promising target as a potential therapeutic or prophylactic. In this work, we oligomerized CVN to determine whether an increase in the number of carbohydrate binding sites has an effect on its viral neutralization activity. To create obligate dimers, we covalently linked multiple copies of CVN through flexible polypeptide linkers. Using HIV-1 as our viral system, we found that a tandem repeat of two CVN molecules (CVN2) increased the efficacy of HIV-1 neutralization by up to 10-fold. An additional benefit was not seen when CVN was trimerized. We also show here that CVN and the CVN2 variants show extensive cross-clade reactivity and higher neutralization efficacy than the most broadly reactive neutralizing antibodies. To determine whether any major structural changes or differences in domain swapping occurred because of the linkage, we solved the crystal structures of three dimeric variants and showed that all variants are intramolecularly domain-swapped. Additionally, we present in this thesis a novel CVN-Fc chimera, a “lectibody,” which shows antiviral activity similar to wild-type CVN. This variant is dimerized through the Fc region of an antibody and has the additional benefit of incorporating Fc-mediated effector functions, which may be therapeutically advantageous. Initial results on the lectibody indicate that domain swapping of CVN has an integral role in the antiviral function as well as in the overall folding and stability of the molecule. Future work on this variant to assay the effector functions as well as create a monodispersive, stable variant are underway. Although CVN is already a promising candidate for antiviral therapeutics, we show here that engineering CVN to add additional functionalities or creating variants with an increased number of active sites can significantly enhance the potential benefit of these molecules.