**3. Vaccines**

Despite the grea<sup>t</sup> e fforts invested in recent years in the development of prophylactic measures against this pathogen, there is currently no specific drug or therapy licensed for its treatment [18,19]. However, several candidate vaccines have been successfully developed, some of which have been licensed for use in horses but, despite the fact that no adverse events or safety concerns have been reported in the few clinical trials conducted, none has been authorized for humans. For the development of the WNV vaccine, all possible approaches available have been tested, from purified inactivated and live attenuated viruses, to candidates based on nucleic acids (DNA or RNA), virus-like particles, subunit elements, and recombinant viruses.

### *3.1. Animal Vaccines*

As commented before, equids are sporadically infected by WNV and, although in most cases they remain asymptomatic, around 20% can develop clinical signs that use to be more severe than in humans and have important health and economic consequences [20]. When present, signs can range from fever to infection and inflammation of the nervous system, and can cause ataxia, hind and forelimb weakness, quadriplegia, paresis, seizures, chewing and paralysis of the tongue, depression, and ophthalmologic manifestations. [20,21]. Therefore, grea<sup>t</sup> e fforts have been made to develop and implement equine vaccines. Four of the six licensed vaccines are currently on the market for use in horses. The WN-Innovator, with a classic inactivated whole virion-based approach, was the first to be developed and was licensed by the USDA in 2003 [22]. Live attenuated recombinant viruses have also been used (either based on canary poxvirus or yellow fever virus), as well as a plasmid DNA vaccine, which was the first licensed by the USDA [23], although it was subsequently withdrawn from the market by the manufacturers. All these vaccines are shown to be protective and their use has contributed greatly to reducing the incidence of the disease in horses in the US [23,24]. However, despite their proven e fficacy, these vaccines still exhibit some limitations, as the need for repeated administrations to ge<sup>t</sup> a solid initial immunization, and the relatively short duration of the induced immunity, which makes necessary annual boosters.

Birds are the natural hosts of WNV and play a key role in the epidemiology of the virus, being many species susceptible to the infection, particularly corvids [4,25]. The disease shows up due to virus invasion of di fferent organs: liver, spleen, kidney, heart, and mainly the central nervous system, and can lead to death within 24–48 h later [26,27]. Several commercial and experimental vaccine candidates have been assayed in wild and domestic birds, although not one has ye<sup>t</sup> been authorized for use on them [27]. Overall, they induced humoral and, although less analyzed, cellular responses, and reduced disease, injury, viremia, viral shedding, and mortality associated with WNV. Furthermore, if they induce herd immunity, they could help prevent outbreaks and the spread of the virus. For example, prospective vaccination of the entire population of California condors (*Gymnogyps californianus*), an endangered species, before the arrival of WNV would have helped prevent infection and its possible extinction [28]. Likewise, vaccination also greatly reduced virus incidence in domestic geese in Israel [29]. However, the implementation of bird vaccines faces several drawbacks, as the feasibility of access to the target host, mainly for wild species, and the administration route. In any case, its availability could benefit domestic populations (farm birds, including those for hunting and restocking activities), as well as wild ones (as those housed in rehabilitation centers and wildlife reserves, and in recreational facilities, like zoos) [27].

### *3.2. Human Vaccines*

As mentioned above, human WNV outbreaks can have serious health repercussions that the availability of licensed vaccines would help to minimize.

Human vaccines must be cost-e ffective, protective, and safe, especially for the most vulnerable populations, like the elderly and the immunosuppressed, whose numbers are increasing around the world. Ideally, vaccines should also be strongly immunogenic and long-lasting with a single dose. Furthermore, although neutralizing antibodies are currently the most reliable protective correlate for flavivirus infections, vaccines should also include determinants that stimulate a balanced T-cell response, essential for providing an e ffective protective response against infection [30].

Nowadays, e ffective licensed vaccines, either attenuated (yellow fever, Dengue, and Japanese encephalitis), or inactivated (Japanese encephalitis, tick-borne encephalitis, and Kyasanur forest disease), are available against several flaviviruses. Nevertheless, none has been licensed for human use against WNV, and none of the six vaccines assayed in humans have progressed further than to phase I/II clinical trials [31]. Even more, only two attenuated recombinant candidates expressing the WNV prM and E proteins, ChimeriVax (in a yellow fever virus backbone) and rWN/DEN4 Δ30 (in a truncated Dengue virus 4 backbone), induced strong immunity after a single dose [32,33]. Noteworthy, the immunogenicity of these vaccines has been analyzed based on seroconversion and detection of neutralizing antibody, and the development of a T cell-specific response has hardly been addressed.

Therefore, for the implementation of human vaccines, several factors must still have to be analyzed in depth. Among them, the possibility that they induce disease associated with pre-existing immunity to heterologous flaviviruses should be avoided, as the phenomenon of antibody-dependent enhancement (ADE) that can arise from the binding of antibodies that cannot neutralize the virus, and can lead to increased uptake of virus into host cells through Fc receptor-mediated endocytosis in macrophages. Consequently, although its role in the pathogenesis of flavivirus infections other than Dengue remains controversial [34], ADE might lead to more severe symptoms during a secondary, heterologous infection, as it may happen in dengue virus infections [35]. Although it has been reported that DIII does not induce ADE for other flaviviruses [36,37], this hypothetical drawback could be resolved by modifying protein E with mutations in its DIII domain, or nearby, to reduce the binding of antibodies induced against other flaviviruses [38,39]. In any case, addressing and excluding this possibility is necessary during WNV vaccine development.

### *3.3. Current Challenges for Human VACCINES Implementation*

Several issues hamper the implementation and introduction of WNV vaccines into the market, such as scientific challenges, safety considerations, difficulties in setting up clinical trials, and cost-effectivity issues, among others.

As mentioned, one of the main scientific questions to solve is the induction of a complete and long-lasting protective immunity that, although seroprevalence studies and experimentation with animal models sugges<sup>t</sup> that this is the case, still needs to be verified in real conditions. In addition, ideally, vaccines should also include determinants that induce a strong cellular response. Likewise, it is desirable that immunization will be induced after the administration of a single dose, and, if possible, that vaccines are marked, that is, that they are DIVA (Differentiating Infected from VAccinated individuals), which could have a considerable impact on blood donation.

Further, clinical trials are difficult to conduct because WNV outbreaks occur sporadically and often unpredictably in regions where other cross-reactive flaviviruses co-circulate. Hence, it becomes difficult to properly assess the impact of candidate vaccines and their implementation strategies.

In addition, another fact to take into account is the possibility of the imposition of new antigenic variants on a previous dominant one, as exemplified by the lineage 1 WN02 strain that has replaced the original lineage 1 NY99 strain in the USA [40]. Although WNV has to infect and replicate in different hosts, making it difficult for mutations to fixate on its genome because they can be beneficial for infecting one host (mosquitoes) but detrimental to others (birds), vaccine production must be conscious of the possibility that new antigenic variants appear and are imposed in order to minimize their possible consequences.

Ultimately, vaccines should be inexpensive, but keep in mind that, while still debated, the few studies conducted so far have concluded that current WNV vaccines are unlikely to save costs [41,42].
