Rift valley fever (RVF) is one of numerous zoonotic diseases affecting human and livestock health in Africa, and has previously spread to the Arabian Peninsula [1, 2]. The disease is caused by a mosquito-borne, negative-stranded RNA virus of the Bunyaviridae family, RVF virus, that was first isolated in 1930 from sheep on a Kenyan farm . RVF virus can infect a wide range of domestic and wild animals, but pathology is most severe in sheep where almost 100% mortality and abortion rates occur in newborn lambs and pregnant ewes, respectively . In humans, RVF primarily occurs following close contact with infected animal tissue or body fluids and presents as a mild febrile illness that sometimes progresses to more severe, fatal manifestations such as encephalitis and hemorrhage. Although a highly effective live-attenuated vaccine known as Clone 13  is available for livestock use in RVF-endemic countries, no licensed livestock vaccines are available for use in RVF-free areas such as Europe and there is currently no licensed human RVF vaccine.
Humans and animals recovering from infection with RVF virus develop long-lasting immunity that is attributable to the acquisition of virus-neutralizing antibodies [3, 5–8]. These virus-neutralizing antibodies mainly target the Gn and Gc envelope glycoproteins (of which there is only one serotype) encoded in the M segment of the RVF virus genome [9–11]. Subunit vaccines incorporating one or both glycoproteins can induce a virus-neutralizing response that may confer complete protection from experimental RVF viral challenge in rodents and livestock (reviewed in ). Thus, development of Gn and Gc-based vaccines utilizing vectors with an established human safety profile could be a promising strategy for a future human RVF vaccine.
Replication-deficient adenovirus vectors have so far been used in human clinical trials of vaccines against Plasmodium falciparum, human immunodeficiency virus , Mycobacterium tuberculosis, hepatitis C virus  and influenza virus  in many thousands of adults, children and infants in Europe and Africa, including countries that are prone to frequent RVF epizootics. These studies have highlighted the excellent safety and immunogenicity profile of adenovirus vectors in humans, with similar properties observed in several animal species (including those susceptible to RVF) where adenovirus-vectored vaccines have been tested against multiple diseases [18–27]. Development of an effective adenovirus-vectored RVF vaccine may therefore provide a prophylactic tool that could be used not only in humans, but also in the animal species susceptible to RVF.
To this end, we evaluated the immunogenicity and efficacy of a replication-deficient chimpanzee adenovirus vector, ChAdOx1 , encoding the RVF virus glycoproteins Gn and Gc (ChAdOx1-GnGc) in BALB/c mice. The ChAdOx1 vector, unlike the widely tested human adenovirus type 5 (HAdV5) vector, is not affected by significant pre-existing anti-vector immunity that may limit vaccine performance in the human population . Though ChAdOx1 is phylogenetically classified as a Human adenovirus E, its serotype is distinct from HAdV5 (a Human adenovirus C) [28–30], meaning that anti-vector immunity to HAdV5 cannot dampen the potency of a ChAdOx1-vectored vaccine. Furthermore, ChAdOx1 is currently in clinical development for human influenza and tuberculosis vaccines, making it a good candidate vector for a human RVF vaccine.
We therefore assessed the immunogenicity and efficacy of the ChAdOx1-GnGc vaccine in comparison with a HAdV5 vector encoding Gn and Gc (HAdV5-GnGc), a strategy previously shown to induce protective immunity against RVF in mice . In addition, we examined the effect of two commercially available adjuvants, Matrix-M™ and AddaVax™, on the RVF virus neutralizing antibody response elicited by the ChAdOx1-GnGc and HAdV5-GnGc vaccines. Matrix-M™ is a saponin-based adjuvant developed for human use , whereas AddaVax™ is a squalene-based oil-in-water emulsion whose formulation is similar to the MF59® adjuvant licensed for human influenza vaccines . Both these adjuvants were selected for use in this study based on their ability to enhance antibody responses induced by candidate human influenza vaccines [32, 34–36].