Skip to main content

Vector-borne Risks for Europe: Risk assessment and control of West Nile and Chikungunya virus (VECTORIE)

Final Report Summary - VECTORIE (Vector-borne Risks for Europe: Risk assessment and control of West Nile and Chikungunya virus (VECTORIE))

Executive Summary:
West Nile virus (WNV) emerged in North America in 1999 and its emergence was associated with high numbers of neuroinvasive disease in humans and horses. Each WNV outbreak was preceded by mass mortality in birds, especially birds belonging to the family Corvidae proved to be particularly susceptible. WNV outbreaks have been reported in Europe since 1950s. These outbreaks were small and remained focal. In contrast to USA, bird mortality is not seen in Europe. Therefore, a couple of hypotheses have been proposed and investigated within VECTORIE. (1) European birds are not susceptible to lethal WNV infection, (2) WNV strains in Europe are less virulent compared to the American viruses, (3) Culex mosquitoes in Europe are not competent to transmit WNV to birds, (4) The feeding behavior of WNV infected Culex mosquitoes change. Another issue studied within VECTORIE was the ability of Aedes albopictus to transmit Chikungunya virus (CHIKV) at lower temperatures. This knowledge allows for efficient preparedness against a possible CHIKV incursion in Western Europe. The results obtained within VECTORIE provided valuable answers, which allows for more tailored surveillance programs. Specifically, VECTORIE has shown that carrion crows are susceptible to lethal infection with WNV and carrion crows could be considered potential amplifying hosts in Europe. However, the experiments in VECTORIE clearly showed that not all WNV strains in Europe can cause lethal infection in birds. Due to this difference in virulence, VECTORIE recommends a surveillance system which is based on identifying the WNV genotype and antibodies in birds. Active surveillance may only reveal circulation of WNV strains that are virulent to birds. But, virulence in birds do not correlate with virulence in mice and possibly also not humans. Furthermore, Culex pipiens pipiens are competent to transmit WNV in Europe and once infected by WNV their feeding behavior for birds do not change. Therefore lack of bird mortality in Europe cannot be explained by the four hypothesis tested within VECTORIE. Studies with Aedes albopictus and CHIKV indicated that Aedes albopictus mosquitoes can transmit CHIKV at lower temperature, suggesting that emergence of CHIKV in western Europe is not unlikely.
To date there is no specific treatment of WNV neuroinvasive disease. Our limited knowledge of the pathogenesis at the cellular and molecular level still hampers the development of intervention strategies to reduce mortality and long-term functional deficits in survivors of encephalitis. We hypothesized that understanding the pathogenesis of WNV neuroinvasive disease using state-of-the art technology will allow the identification of biomarkers and leads for novel treatment protocols. VECTORIE has identified potential biomarkers using the mouse model. Some of these markers have been validated using samples of human clinical cases. In this way, VECTORIE has paved to way for future discovery of prognostic markers as well as effective intervention strategies. Finally, several potential candidate vaccines against WNV were developed and the first preclinical testing proved promising results. Furthermore, one effective candidate vaccine against CHIKV was developed. Hereby, VECTORIE has provided novel vaccine strategies for these vector-borne diseases, since an efficient approach to control outbreaks caused by WNV and CKIKV is vaccination of people at risk.

Project Context and Objectives:
Project context
In August 2007, a Chikungunya outbreak, caused by the infective agent Chikungunya virus (CHIKV), was declared for the first time in Europe. Two hundred human cases were recorded in two small towns of Italy. The vector of the Chikungunya virus (CHIKV) – namely the carrier that transfers an infective agent from one host to another – was the Aedes albopictus, commonly known as the Asian tiger mosquito, which appeared to have become abundant in the area. The virus was apparently introduced by a traveller who came back from south-west India where the virus was circulating. This outbreak was the first one documented in a temperate climate country.

Although WNV has circulated in Europe for decades, the first major outbreak in humans was reported from Romania in 1996. That outbreak was characterized by a high incidence of neuroinvasive disease in humans, possibly representing the first manifestations of the changing epidemiology of WNV in Europe. The resurgence of West Nile virus in the US (2012, Texas) and a large WNV outbreak in Greece (2010, 35 initial deaths) illustrate the highly unpredictable nature of arboviral outbreaks and stress the importance of mitigation strategies, including the development of vaccines and antiviral drugs, preferably to be used in concert with targeted vector control. In July 2011, the latest West Nile virus (WNV) season started in Europe. By the end of November 2011, 93 confirmed human cases of West Nile fever had been reported in the European Union: 69 cases in Greece, 14 in Italy and 10 in Romania. In the neighbouring countries, 189 cases in total had been reported.

Both West Nile virus (WNV) and Chikungunya virus (CHIKV) are emerging vector-borne viruses, transmitted by mosquitoes. CHIKV is indigenous to tropical Africa and Asia. WNV was first found in a tropical region in Africa, and has since been described in Africa, Southern Europe, the Middle East, west and central Asia, Oceania and North America. Northward migration and increasing numbers of air transports, combined with the ability of the mosquitoes to adapt to different ecological situations is allowing for the introduction of these mosquito species in larger parts of Europe. Climate scenarios for the near future predict that many parts of Europe will become warmer and more humid. These changes are likely to increase the risks of WNV and CHIKV outbreaks in these parts of Europe, also because of vectors’ longer activity seasons, larger geographical areas becoming affected, and difficulties of eradicating the disease in Europe with an already established vector population.

In light of the increased risk of WNV and CHIKV outbreaks in Europe and the increasing notifications of the West Nile disease and Chikungunya fever cases in Europe, it has become imperative to assume a proactive approach to anticipate and deal with threats of mosquito-borne infections, such as WNV and CHIKV at the EU level. This is exactly what VECTORIE aims to do.

The viruses
West Nile virus (WNV)
West Nile virus is a virus belonging to the family Flaviviridae, genus Flavivirus. It can cause West Nile disease in humans. This condition ranges from subclinical to mild and severe. Mild disease is manifested by fever, generally called West Nile fever (WNF). More severe forms of disease, which can be life threatening, are manifested by meningitis (inflammation of the protective membrane of the brain and spinal cord), encephalitis (acute inflammation of the brain), and flaccid paralysis (muscle weakness or paralysis). The most typical manifestations for WNV infections are neurological, and development of WNV neuroinvasive disease (WNND) is observed most frequently in the elderly and immuno-compromised individuals (individuals with a weakened immune system).

Chikungunya virus (CHIKV)
CHIKV is a virus of the genus Alphavirus, that is spread by Aedes mosquitoes. It is known to cause Chikungunya disease in humans, resulting in fever, and in severe cases in poly-arthralgia (multiple joint pain), meningitis, encephalitis, and flaccid paralysis. Although the most typical manifestations for CHIKV infections are rheumatic, an increasing incidence of neurological complications is reported during recent CHIKV infections as well. CHIKV neuro-invasive disease appears to represent the most common clinical manifestation among young children.

Transmission of WNV and CHIKV
Traditionally, the main transmission mode of the viruses has been through mosquito bites - the Culex pipiens mosquito mainly responsible for transmitting WNV, the Aedes aegypti and Aedes albopictus for CHIKV. CHIKV is transmitted to humans by the bite of infected mosquitoes. Humans are the only known host for CHIKV. For WNV, humans and birds are both hosts, of which only birds are amplifying hosts. This means that in birds the virus is able to amplify to produce sufficient viral levels to be transmitted to other biting mosquitoes which then infect other birds and also humans. However, new modes of transmission through blood donations and organ transplants have now been reported.

Project objectives
The Vector-borne Risks for Europe: Risk assessment and control of West Nile and Chikungunya virus project (VECTORIE) is a focused research project that aims to enhance Europe’s preparedness for vector-borne emerging diseases. VECTORIE will contribute to the development of a surveillance system and response plan appropriate for European countries to detect, predict and deal with emerging WNV and CHIKV outbreaks. In addition, the studies will lead to novel knowledge that will enable Europe to improve its ability to monitor the spread of these infections in Europe, and will lead to diagnostic and prognostic tools as well as novel strategies to treat infected patients. Finally, VECTORIE addresses Europe’s need for novel vaccine strategies for these vector-borne diseases, since an efficient approach to control outbreaks caused by WNV and CKIKV is vaccination of people at risk. VECTORIE’s integrated approach combines three main areas of expertise: vector biology; viral virulence and pathogenesis; as well as vaccines.

The vector biology part aims at generating new knowledge on host susceptibility of birds for WNV, and vector competence (ability of the vector to transmit the virus) of the two main vectors of WNV and CHIKV. Host susceptibility of European birds for the different WNV strains circulating in Europe will be determined, thereby providing an important piece of information regarding the risk of the spreading potential of WNV and potential to cause outbreaks of human disease. Birds are not only susceptible to WNV, but also participate in maintaining the transmission cycle. In North America, WNV outbreaks were preceded by mortality in birds. One of the most important questions that European countries face with an incursion of WNV is how a virulent WNV strain will behave in European crows and jackdaws and if mortality in these birds can also serve as an early warning signal for WNV. If European crows turn out not to be susceptible to fatal infection, the results of these studies will reveal whether or not crows and/or jackdaws are relevant as sentinel hosts.
The European Centre for Disease Prevention and Control (ECDC) states that further spread of WNV and favourable environmental conditions for vectors would lead to larger geographical areas becoming affected and/or longer seasons of WNV activity. Furthermore, according to the ECDC, the risk for CHIKV is serious because with an already established Ae. albopictus vector population, larger outbreaks would increase the difficulty of eradicating the disease in Europe. In light of climate change and the associated risk of vector introduction and establishment in Europe, vector competence studies will greatly assist in the assessment of newly emerging infectious disease threats to Europe including those caused by WNV and CHIKV. The risk of WNV and CHIKV outbreaks in humans is not only determined by vector competence but also by their feeding behaviour. These factors have so far not been investigated for European vectors. Therefore, VECTORIE will study vector competence and feeding behavior of Culex pipiens mosquitoes, which is essential in formulating and focusing a prevention plan. With regard to CHIKV, the effect of temperature on vector competence of Aedes albopictus mosquitoes for CHIKV will be determined. This knowledge is necessary to be able to address the question whether Aedes albopictus will be able to efficiently transmit pathogenic CHIKV among the human population in temperate climate zones.
Related to this part are the following sub objectives:
• To compare the susceptibility of European crows to several WNV lineages circulating in Europe
• To define vector competence of Culex pipiens mosquitoes for WNV
• To define vector competence of Aedes albopictus mosquitoes for CHIKV at low temperature

The virulence and pathogenesis part addresses the need for an increased understanding of how the different WNV strains that circulate in Europe differ in their capacity to cause severe disease and the pathways involved in WNV and CHIKV neuroinvasive disease. The different WNV strains in Europe are assumed to be associated with differences in virulence, i.e. the disease causing capacity of the respective virus strains. So far, two of the four circulating WNV strains, one of which has emerged in Italy and one in Hungary, are known to cause WNV neuroinvasive disease, and therefore may pose a significant threat to humans. Furthermore, virulence factors (factors that predict the spreading and disease causing capacity of a strain) that are associated with increased transmission and morbidity of different WNV strains will be identified. In addition, the WNV strains that are relevant for Europe will be defined and targeted for subsequent vaccine development. Finally, this part will study the pathogenesis of WNV and CHIKV neuroinvasive disease using state-of-the art technology in order to allow for the identification of biomarkers, which are traceable biologic substances that can be used to measure the presence or progress of a disease or the effects of a treatment, and leads for novel treatment protocols. The knowledge obtained from these studies will allow for the development of novel therapies that can be tailored to resolve specific clinical manifestations of the diseases.
Related to this part are the following sub objectives:
• To determine markers of virulence and the pathogenicity of the four WNV lineages in mice
• To identify pathways involved in WNV and CHIKV neuroinvasive disease

The third part, Vaccines, addresses Europe’s need for novel vaccination strategies for populations at risk. VECTORIE will develop and subsequently perform rational improvements of vaccine candidates, resulting in safer and more efficient vaccine combinations for clinical development that are tailored specifically to the situation in Europe. VECTORIE brings together a platform of technologies using either virus-like particles (VLP) or Modified vaccinia Ankara (MVA) vaccine candidates that are among today’s most promising vaccine development systems. This will be combined with supporting studies on cross-protection of different WNV lineages, to guide in the development of (one) vaccine candidate(s) that can protect against the different circulating WNV strains. Furthermore, standardized neutralization and other immune response assays will be developed, which will allow accurate comparison of the protective capacity of different vaccines, which is a critical step forward in determining the most suitable vaccine candidates.
Related to this part are the following sub objectives:
• To determine the degree of cross-protection induced by a vaccine based on WNV-99 against the European strains of WMV
• To standardize read-out systems to measure for immunogenicity of WNV and CHIKV vaccines
• To develop and evaluate several candidate vaccines against WNV and CHIKV
Project Results:
This part describes the main science and technology results obtained from the project and are described per task, in line with the objectives as set out in the section above.

Task 1.1 To compare the susceptibility of European crows and jackdaws to infection with several lineages of WNV that circulate in Europe
Contrary to the invasive epidemic pattern observed in the US, where the virus was introduced into naïve immune populations of susceptible vertebrate hosts and vectors, there is evidence that WNV circulates in Europe as early as the 1950’s. Since then, relatively large outbreaks of West Nile neuroinvasive disease (WNND) have been recorded in humans and horses in Southern France, Romania, Czech Republic, Southern Russia, Hungary, Italy, and Greece. The WNV strains that have been characterised in Europe are very heterogeneous. This heterogeneity of WNV, both at lineage and genotype level, together with the appearance of point mutations potentially affecting virulence and/or transmissibility, and the co-circulation of other flaviviruses infecting birds and humans, have important consequences for understanding their ecology and pathogenicity. Introduction of the virus into Europe may be initiated by migratory birds while residential juvenile bird contribute to dispersal of WNV into larger areas or even within different European countries.
Mass mortality of highly susceptible bird species (as corvids) is frequently observed in North America, something not observed in Europe. Therefore, we first studied the susceptibility of the European (jackdaws and carrion crows) and North American corvids to lethal infection with WNV strains circulating in Europe. Our results indicate that carrion crows are susceptible to lethal infection with several WNV strains circulating in Europe and the level of viremia achieved in carrion crows suggest that they could function as amplifying hosts in Europe. In contrast, jackdaws which were less susceptible to lethal infection developed lower peak viremia. From these studies we can conclude that European birds are as susceptible to WNV infection than their North American counterparts. Therefore, intrinsic resistance of European birds cannot explain why mass mortality among birds are not seen in Europe. During the experiments, all captured birds were first screened for presence of antibodies to WNV. Since WNV specific antibodies were not detected in any of the collected birds, the hypothesis of herd immunity is unlikely and therefore not sufficient to explain lack of bird mortality in Europe during WNV outbreaks. Another hypothesis tested within VECTORIE was the existence of differential virulence between WNV strains. It is possible that the WNV strains circulating in Europe are less virulent than the viruses circulating in North America. The strains used in VECTORIE confirmed existence of virulent and less virulent strains when evaluated in birds. The Ita09 and Gr-10 strains (responsible for the 2009 outbreak in Italy and the 2010 outbreak in Greece respectively), both virulent in birds during experimental infections, caused big outbreaks of WNND in humans and horses, however bird mortality was not reported. In conclusion, although non-virulent strains circulates in Europe (e.g FIN strain), their low-frequency cannot explain the lack of mass mortality amongst birds.

Task 1.2 Study of vector competence of Culex pipiens mosquitoes for several lineages of WNV (Confidential)

Task 1.3 Study of vector competence of Aedes albopictus mosquitoes for CHIKV at low temperature
The establishment of Aedes albopictus in Europe is a public health concern as this mosquito species is experimentally competent to transmit 26 arboviruses, with the status of secondary vector for chikungunya (CHIKV) and dengue (DENV) viruses. We evaluated the transmission efficiency of French Ae. albopictus population to transmit CHIKV. To be a competent vector, the virus ingested through an infectious blood-meal must overcome different physical barriers to infection and dissemination in the host. Thus, vector competence, i.e. is the ability of a vector to acquire a pathogen and to transmit it to another susceptible host, depends on both intrinsic factors and extrinsic factors. Temperature is one of the most important extrinsic variables affecting arbovirus transmission via the vector. Variations of environmental temperatures can notably impact mosquito life history traits (as survival) and also, vector competence. However, to our knowledge, few/no data on vector competence of temperate Ae. albopictus populations are available in climatic conditions found in the field.
Orally infected mosquitoes from France were submitted to different temperature regimes. We used two French CHIKV, one isolated from a patient in La Réunion outbreak in 2006 and one from an autochthonous case in Southeast France in 2010. We estimated (i) virus infection by measuring the number of infectious viral particles in mosquito bodies (thorax and abdomen), (ii) virus dissemination corresponding to the presence of virus in secondary organs such as heads, and (iii) virus transmission by analyzing the presence of virus in mosquito saliva and by measuring the number of infectious viral particles and (iv) survival of infected or uninfected mosquitoes.
Our principal findings showed that CHIKV can replicate, disseminate and be transmitted at a low temperature (20°C) in Ae. albopictus from France. Moreover, survival analysis of infected French females indicated that there were no effects or lesser effects on mosquito survival at low temperatures depending on the CHIKV used. However, vector competence studies of other European and tropical Ae. albopictus populations (e.g. France, Italy, Montenegro, La Reunion, Vietnam and La Reunion) at different temperatures (20°C and 28°C) revealed the role of the three-way interactions between mosquito population, viral strain and temperature on CHIKV transmission. Indeed, CHIKV transmission by Ae. albopictus strongly depends on the three-way combination of mosquito population, viral strain and temperature. These findings highlight the significant role of Ae. albopictus in potential emergence of chikungunya in Europe and establish the importance of G x G x E interactions for the transmission of an emerging human pathogen.

Task 2.1.1 Determine the pathogenicity of four WNV lineages that circulate in Europe and markers of virulence for the European lineage 2 strain
To further study the virulence of the respective strains, experiments were conducted in vitro and in mice. To this end, replication characteristics of the different WNV strains were studied in vitro. In addition, the tropism for cells in the brain, the LD50 and mortality percentage, were determined in mice. The experiments in vitro suggested differences between WNV strains. These in vitro differences were confirmed in the mouse model in that all the strains tested were virulent, but subtle differences between strains were observed. The results suggested that the 578/10 strain (a lineage 2 WNV strain isolated from the brain of a dead horse in Hungary in 2010) is more virulent than the other strains of WNV tested. Taken together, task 1.1 and 2.1 indicate that European WNV strains are as virulent as the North American viruses and differences in virulence are more pronounced in birds than in mice.

Task 2.1.2.: Study the correlates of WNV virulence
The most serious consequence of WNV infection is the development of WNND, the ability of the virus to invade the central nervous system (neuroinvasiveness) is a significant virulence factor. WNV strains are considerably different regarding their neuroinvasive property. Two main genetic lineages of WNV are known. Previous studies identified several potential genetic marker loci of lineage 1 WNV strains, influencing virus multiplication in vitro and in vivo, in mammal and avian models. However, an exotic strain of WNV, belonging to lineage 2 emerged in 2004 in Central Europe (in Hungary) and caused several encephalitis cases in animals and in humans. Genetic markers of virulence and neuroinvasiveness of WNV 2 strains have not been investigated in details previously. Therefore the aim of this task was to investigate the effect of certain nucleotide substitutions on the phenotype of a lineage 2 WNV strain in vitro (in cell cultures) and in vivo (in mouse model). To this end a full-length infectious clone of the virus RNA genome was constructed. Thereafter certain, selected nucleic acids were changed with site-specific mutagenesis. The sites chosen to be manipulated were deduced from the literature regarding correlates of virulence of the North American lineage 1 WNV strain.
The candidate virus strain (designated 578/10) chosen for pathogenic studies was a lineage 2 WNV strain isolated from the brain of a dead horse in Hungary in 2010. Titration and growth curves were determined in infected Vero cells. Growth properties showed mild to severe differences between mutated viruses, moreover the clone virus containing C102S amino acid alteration in NS4B was unable to grow up at a sufficient titre in BHK-21 cells nor after passage into Vero cells. Finally the propagation and neuroinvasiveness of wild type virus infectious clone and the mutant constructs were compared in vitro and in vivo. In vivo characterisation was performed in 5-week old C57/Bl6 mice. Groups of eight female mice were inoculated intraperitoneally (ip.) with high (= 103) and low (=101.1) plaque forming unit (PFU) per mice of wild type infectious clone derived virus and each mutant strains. The abovementioned NS4B C102S clone was excluded from in vivo studies due to insufficient in vitro growing properties. Observation period lasted 14 days, with daily monitoring. Quantitative PCRs and virus titration was performed. Survival results varied between 0% and 100% among the different mutated viruses.
The results of the studies indicate that the NS4B C102S significantly reduced the fitness of the virus, therefore it could not multiply even in cell culture. The NS1 P250L mutant clone has shown lower virulence (0% mortality in mice both in low and high dose infections) compared to the wild type virus and to the other mutants. Because the back-titrations of the virus inoculums revealed minor variance in the infectious doses, the data must be extrapolated by the determination of lethal dose 50% (LD50) values. Unfortunately, due to delays with some administrative processes in national and institutional levels connected to authorisations of mutagenesis and animal experiments, and because of unforeseen technical problems resulting temporary unavailability of BSL-3 biology safety facilities, the in vivo experiments were accomplished by the end of January, 2014. Therefore, the final analysis of the results is ongoing, and publications are planned within a few months.

Task 2.2 Study the pathways involved in WNV and CHIKV neuroinvasive disease using protein profiling
This task consisted of two parts: the proteomic analysis of brain samples from mice infected with WNV or CHIKV (part A), and the identification of biomarker candidates associated with human WNV neuroinvasive disease (part B).
A- Proteomic analysis of brain samples from mice infected with WNV or CHIK
To gain insight into the physiopathological processes in severe virus infection, a kinetic analysis of protein expression profiles in the brain of WNV- and CHIKV-infected mice was conducted on half brain hemisphere samples collected before (early time-point, E) and after (late time-point, L) the onset of clinical symptoms. Non infected mice (mock, M) were used as control. Brain tissue samples were analysed by microarray (transcriptomics and proteomics analyses). The proteomics study overlapped significantly with the proteomic studies and therefore proteomics was used to validate the transcriptomics results. Combined 2D-DIGE and gel-free iTRAQ labeling proteomic approaches followed by protein identification by mass spectrometry was used for the proteomics analyses. Differential expression of selected proteins during the time-course of infection was further validated by fluorescent-based Western Blotting (WB).
1- Altered Protein Networks and Cellular Pathways in Severe West Nile Disease in Mice
The combined analysis of the two complementary quantitative proteomic approaches gave rise to a total of 148 unique host proteins that were found to be differentially expressed in brain tissue samples after WNV infection at the early and/or late time-points. To understand and highlight the relationship of these differentially regulated proteins and the consequences of these modifications in the context of their cellular function, during the course of WNV infection in the brain, a bioinformatics analysis using a web-based entry tool developed by Ingenuity Systems, Inc. (http//www.ingenuity.com) was performed for each comparison (E vs M, L vs M and L vs E). Collectively, the analysis of each protein dataset originating from the different time-point comparisons revealed 4 major functions altered during the course of WNV-infection in mouse brain tissue: i) modification of cytoskeleton maintenance associated with virus circulation; ii) deregulation of the protein ubiquitination pathway; iii) modulation of the inflammatory response. This study hereby provides novel insights into the in vivo kinetic host responses following WNV infection and the physiopathologic processes involved, according to clinical symptoms. This work offers useful clues for anti-viral research and further evaluation of early biomarkers for the diagnosis and prevention of severe neurological disease caused by WNV.
2- Kinetic analysis of mouse brain proteome alterations following Chikungunya virus infection before and after appearance of clinical symptoms. To decipher the mechanisms of CHIKV infection processes in the nervous system, a kinetic analysis of host proteome modifications in the brain of CHIKV-infected mice sampled before and after the onset of clinical symptoms was performed, taking into account the 2 late symptoms observed (“paralytic”, LP and “tetanus-like”, LT). The combination of 2D-DIGE and iTRAQ proteomic approaches, followed by mass spectrometry protein identification, revealed 177 significantly differentially expressed proteins, considering the comparisons E vs M and L vs E, taking into account the two late symptoms. Classification of these proteins according to their GO biological functions was performed indicating that CHIKV perturbed numerous cellular processes such as the modulation of gene expression (14%), the nervous system development (12%), and the maintenance of cytoskeleton organization (12%). Other molecules related to apoptosis or ubiquitination were also found differentially expressed. The most striking result was that CHIKV infection induces an early dramatic shut-off of host protein expression (80% proteins down-regulated), followed subsequently by their up-regulation in association with the clinical symptom onset. The in silico analysis of proteins that were differentially expressed among the three compared groups (i.e. E vs M, LP vs E and LT vs E) revealed that the main functions and processes altered during the course of CHIKV infection in the mouse brains were: i) integrin signalling and cytoskeleton dynamics, ii) endosome recycling machinery and synapse function, iii) regulation of host gene expression, and iv) modulation of the ubiquitin-proteasome pathway. This work gives new information on putative mechanisms that could be associated with severe neurological CHIKV infection-mediated disease. It also describes possible markers or targets that would need further evaluation as tools for diagnosis and/or antiviral strategies.
B- Identification of biomarker candidates associated with human WNV neuroinvasive disease.
1- Based on results from WNV-infected mice models, HMBG1 and PRDX6 were evaluated as potential biomarkers for detection of WNV disease severity in humans. The brain proteomic study from a mouse model of WNV infection with neuronal involvement showed the kinetic up-regulation of high-mobility group box-1 (HMGB1) and peroxiredoxin-6 (PRDX6), before and after onset of clinical symptoms, respectively. We therefore evaluated at the human serum level whether these two proteins could be associated with WNV infection and/or neurological disease. HMBG1 and PRDX6 concentrations in serum from WNV-infected patients (n=49) diagnosed either for WNF (n=22) or WNND (n=27), were measured by ELISA and compared to concentrations in serum from uninfected healthy individuals (n=30). HMGB1 serum concentration in WNV-infected patients was found to be significantly higher than in healthy controls. Furthermore, HMGB1 serum levels were found to be significantly elevated in patients with WNND than in those diagnosed for West Nile Fever (WNF). In contrast to HMGB1, PRDX6 serum concentration in WNV-infected patients was found to be significantly decreased compared to controls. In contrast, PRDX6 levels were significantly elevated in healthy individuals relative to WNV-infected patients, regardless of clinical symptoms. This study provided candidate biomarkers associated with WNV infection or disease severity. Further investigation in larger cohorts could determine the usefulness of measuring HMBG1 and PRDX6 concentrations as WNV infection diagnostic or prediction of detrimental clinical evolution.
2- Identification of cerebrospinal fluid (CSF) biomarker candidates associated with human WNV neuroinvasive disease by proteomic analysis. Due to the limited volume of the CSF samples, pooled CSF samples from each experimental group (i.e. group WNND, group AH (acute headache), group IIH (idiopathic intracranial hypertension) were generated by mixing an equal volume of each sample per group, before iTRAQ labelling and submission to mass spectrometry. Forty-seven proteins were found modified in the CSF of WNND patients as compared to control groups, and most of them are reported for the first time in the context of WNND. Among them, Defensin-1 alpha (DEFA1), a protein reported with anti-viral effects, presented the highest increasing fold-change (FC>12). The augmentation of DEFA1 abundance in patients with WNND was confirmed at the CSF, but also in serum, compared to the control individual groups. Furthermore, the DEFA1 serum level was significantly elevated in WNND patients compared to subjects diagnosed for WNF.
Among the 47 differentially expressed proteins, in silico analysis (IPA) highlighted that 15 of them were significantly associated with viral infection. Among them, LCN2, TIMP1, S100A8/9, and PRDX2 related to host response and/or inflammation in neurological disorders, could be used as potential biomarkers of brain pathology. The present study provided the first insight into the potential CSF biomarkers associated with WNV neuroinvasion. The higher abundance of DEFA1 at the peripheral level in WNND patients compared to those with WNF, suggests that this peptide could serve as potential biomarker of WNV neuroinvasion. Further investigation in larger cohorts with kinetic sampling could determine the usefulness of measuring DEFA1 as diagnostic or prognostic biomarker of detrimental WNND evolution.

Task 3.1 Determine the level of cross-immunity and cross-protection induced by WNV-NY99 strain against the WNV lineages circulating in Europe
In order to study the level of cross-immunity and cross-protection between NY99 strain and the WNV lineages circulating in Europe, we cloned, express and purified domain 3 of the envelope protein and subsequently immunized mice with the recombinant protein of WNV-NY99 adjuvanted with ISCOM. Immunized mice were challenged with the following strains: WNV-NY99 (homologous antigen), WNV-FIN, WNV-Ita09, WNV-578/10. All vaccinated animals were protected against challenge with the different WNV lineages indicating that domain 3 derived from NY-99 is enough to confer protection against the European WNV strains. In addition, sera from hyper-immunized mice were used to determine the serum-neutralization antibody titers against homologous and heterologous WNV. The results of the cross-neutralization assays indicate low levels of neutralizing antibodies against all strains. However, in-vivo data suggest that such low cross-neutralizing titers are sufficient to confer protection against lethal challenge with the respective strains of WNV. In conclusion, it was shown that a vaccine based on the E-protein of the NY99 strain is able to protect against the other strains circulating in Europe, in addition to NY99.
Furthermore, this study aimed to estimate in vitro cross-reaction and in vivo cross-protection of three related flaviviruses, simultaneously circulating in Central Europe. WNV, Tick-borne encephalitis virus (TBEV), and Usutu virus (USUV) positive sera were simultaneously tested with ELISA, in-house indirect immunofluorescence (IIF), haemagglutination-inhibition (HAI) and plaque reduction microneutralisation test (PRNT) assays to assess serological cross-reactivity. The results clearly confirm that extensive cross-reactivity exist between the different flavivruses. Next cross-protection between flaviviruses was investigated in mouse model, in order to reveal whether immunity induced by infection of one flavivirus (USUV 939/01 strain) will protect the host from the development of disease after challenge infection with another flavivirus (WNV strain 578/10). Clinical signs and mortality was observed neither before, nor after WNV challenge in any of the USUV-infected mice. All animals that were not primed first with USUV died as a result of WNV challenge. The results of the study indicate that USUV infection can induce protective cross-immunity against WNV infection in the mouse model.

Task 3.2 Standardization of read-out systems to measure vaccine immunogenicity
In this task, we aimed to produce recombinant WNV and CHIKV viruses expressing Fluc genes as marker genes. First of all, CHIKV infectious clones were constructed with marker gene insertions (Rluc, Fluc or EGFP) at different positions. The second objective was to development a WNV/CHIKV neutralization test. Since the recombinant viruses with marker genes inserted displayed high instability of transgene expression, we decided to develop an alternative neutralization test based upon recombinant expressed viral protein. We successfully expressed soluble WNV-E protein in Sf21 insect cells using recombinant baculoviruses. Western blot detection confirmed that the protein produced in large quantities is indeed WNV-E as it reacts with both anti-his and anti-E (cocktail of 3 monoclonal) antibodies. In addition, WNV-E is also found as secreted protein in the medium fraction, which enables Talon purification if desired. These cell lysates and supernatants containing soluble WNV-E subunit protein were used for ELISA testing to measure vaccine immunogenicity of our developed vaccine candidates (ELISA for WNV E Protein binding antibodies).
All standardized protocols to monitor WNV-specific CD8+ T-cell immune responses by ELISPOT system and flow cytometry (intracellular cytokine staining; dextramer multimer staining) have been established and validated. In addition, all necessary standard protocols to determine the level of WNV-specific antibodies (ELISA for WNV E Protein binding antibodies; WNV neutralizing antibodies) have been established. ELISpot assays for monitoring WNV and CHIKV vaccines were established. Standard protocols for different MVA vector vaccination experiments using intramuscular and subcutaneous routes and various vaccine dosages have been established. All experimental protocols for the assessment of the protective capacity of immunizations against WNV challenge infections have been established.

Task 3.3.1 Design of MVA-based vaccine and immunogenicity testing
Two different bivalent MVA-WNV vector vaccine candidates (delivering WNV E and M structural antigens) and three different monovalent recombinant MVA-WNV vaccines (delivering the WNV E envelope antigen) have been generated and clonally isolated. All five experimental MVA vaccine candidates have been established as working virus stocks and have been submitted to large scale amplification in tissue cultures for master seed virus production. In addition, high titre virus preparations suitable for purification and formulation of experimental vaccines have been successfully generated with all five vector viruses. All five MVA-WNV recombinant viruses were fully quality tested. Genetic homogeneity, genetic stability, and genetic identity have been confirmed upon analysis viral genomic DNA. WNV-specific gene expression, antigen synthesis, antigen stability and antigen release from vector virus infected cells have been confirmed upon in vitro infection of relevant tissue cultures including cells of human and equine origin. The replication deficiency in tissue of human origin was confirmed for all recombinant MVA-WNV vectors. All the MVA-WNV vectors/vaccines qualified for use under conditions of biosafety level 1 (for GMOs in Germany) and with regard to quality aspects for use as seed material for preparation of clinical evaluation. Experiments for immunogenicity testing in mice have been performed in normal and transgenic mouse models humanized for HLA molecules. Vaccines derived from all five different recombinant MVA induced WNV antigen specific humoral as well as cellular immune responses upon evaluation in these mouse models. The analyses of vaccine induced antibody responses included the confirmation of WNV neutralizing antibodies. The analysis of T cell responses included the detailed measurement of acute and memory phase CD8 T cells recognizing an immunodominant WNV-specific epitope identified in human patients. Overall, the levels of the induced immune responses strongly suggested the protective capacity of all MVA-WNV vaccines against WNV challenge in mouse models. Importantly, four of these highly promising MVA-WNV vector vaccines are based on novel strategies of WNV antigen delivery and require thorough preclinical testing. Therefore, the comparative evaluation of the protective efficacy of MVA-WNV candidate vectors is still ongoing in vaccination/challenge experiments. However, it can be concluded that several novel MVA-WNV candidate vaccines have been successfully developed and qualify for further evaluation in preclinical and also clinical research. Similarly, recombinant MVA expressing the CHIKV 6KE1, E3E2, and E3E26KE1 were generated. To generate the vaccine batches for the pre-clinical animal studies, all recombinant MVA were propagated in a multistep amplification process on CEF cells, purified by ultracentrifugation through 20% sucrose, and reconstituted in TE buffer, pH 9.0. Expression of the respective genes was confirmed by Western blot analysis of lysates of infected BHK-21 cells. All the candidate vaccines proved to be stable after 20 passages in vitro. The respective proteins were expressed when cloned into the MVA backbone. The immunogenicity and efficacy studies were performed and results are described in task 3.4.

Task 3.3.2 Design of VLP-based vaccine and immunogenicity testing
The potentially virus-neutralising domains of West Nile virus (WNV) and Chikungunya virus (CHIKV) envelope proteins have been selected for the exposure on VLPs. In the case of WNV, the domain III (DIII, 111 amino acid (aa) residues in length) of the WNV glycoprotein E was selected. In the case of CHIKV, three domains were selected: domain III from the E1 protein (E1-DIII, 91 aa) and A and B domains from the E2 protein (E2A,133 aa, and E2B, 60 aa, respectively).
The selected domains were exposed on the surface of virus-like particle (VLP) carriers represented by two RNA phage coat protein (CP)-derived VLPs: AP205 and Qß. Two alternate strategies were used in parallel for the exposure of potential virus-neutralising domains: (i) generation of mosaic virus-like particles, (ii) chemical coupling of separately expressed domain sequences to the VLP surface. Immunisation of BALB/c mice (6-8 weeks of age) with AP205- and Qß-based mosaic and chemically conjugated VLPs led to induction of substantial specific anti-epitope responses. In the case of the most studied mosaic AP205-DIII VLPs, the level of anti-WNV DIII antibodies was comparable to that of the anti-AP205 antibodies generated against the VLP carrier. After immunization of mice without any adjuvant, the AP205-DIII VLPs induced remarkable anti-DIII response (titers to 103 level), which is about four times weaker than the anti-AP205 carrier response. Formulation of the AP205-DIII VLPs in the Alhydrogel adjuvant enhanced preferably anti-DIII (to the level of 104), but, to less extent, the anti-AP205 response. The DIII polypeptide alone appeared as a weak immunogen after immunization of mice without any adjuvant (102). Formulation of the DIII polypeptide in the Alhydrogel or silicon dioxide enhanced anti-DIII response to the level of the anti-DIII response (104) induced by the AP205-DIII VLPs in Alhydrogel. Isotyping of anti-DIII response revealed remarkable preference of the AP205-DIII VLPs over DIII polypeptide in the ability to induce anti-DIII antibodies of the IgG and especially of the IgG2 isotype. Whereas the DIII polypeptide in Alhydrogel induced predominantly IgG1 antibodies, the AP205-DIII VLPs demonstrated clear competence to induce antibodies of the IgG2 isotype. Without Alhydrogel, the DIII polypeptide was unable to switch antibody production from the IgM to the IgG isotype. The AP205 carrier within the AP205-DIII VLPs induced broad spectrum of IgG1, IgG2a, and IgG2b antibodies. The level of anti-CHIKV antibodies in the case of mosaic AP205-CHIKV VLPs was lower, but these results need further elucidation and improvement. High specific anti-E2A were obtained after immunisation of mice with Qß VLPs carrying chemically coupled CHIKV domain E2A. In total, four immunizations have been performed in parallel with Qß simply mixed with E2A and Qß VLPs chemically coupled with E2A. The immunisation results indicated that chemical coupling of E2A to Qß VLPs had indeed enhanced the immune response quite significantly. Both subcutaneous (sc) and intraperitoneal (ip) injections and Alum as an adjuvant were used in all cases. Therefore, both mosaic and chemically coupled chimeric VLPs demonstrated definite potential to induce specific anti-epitope antibodies. To evaluate the protective efficacy of the vaccine candidate, AG129 mice were immunized twice and challenged 6 weeks post booster vaccination. None of the animals were protected. Concluding, although first direct CHIKV protection experiments failed, both mosaic and chemically coupled RNA phage CP-based VLPs could be regarded as promising prototypes for the generation of WNV and CHIKV vaccines. Overall, an efficient VLP production and purification technology is generated as a direct result of the project.

Task 3.4 Study the protective efficacy of candidate vaccines in mice
Mice lacking the IFN-a/ß/? receptor (A129 mice) have been described previously to be susceptible to CHIKV infection. Mice were immunized with the MVA-6KE1, MVA-E3E2 and E3E26KE1 twice at 3-week intervals. Wildtype (wt) MVA was used as a negative control. Immunization with MVA-6KE1 and MVA-E3E2 induced low levels of neutralizing antibodies against both CHIKV-S27 and CHIKV-IND/NL10, whereas MVA-E3E26KE1 induced significantly higher levels of neutralizing antibodies compared to the other two candidate vaccines (P < 0.05) with titers ranging from 40 - 160. In order to study the protective efficacy of the different recombinant MVA candidate vaccines, all the mice were challenged intraperitoneally with a lethal dose of CHIKV-S27 (103 TCID50). The survival rates of the eight animals were monitored in each group after challenge. All mock-vaccinated (MVA-wt) mice died within five days post infection. In the groups of mice vaccinated with MVA-E3E2 and MVA-E3E26KE1 respectively, all animals were protected against severe disease. In contrast, 75% (6/8) of the animals immunized with 6KE1 were protected against severe disease caused by challenge with CHIKV-S27.
The candidate vaccines based on E3E2 and E3E26KE1 also protected the animals against high viral replication in the spleen, liver and brain. Although viral RNA was detected in several organs, the vaccines based on E3E2 and E3E26KE1 protected the immunized animals against pathology. Taken together, MVA-E3E26KE1 is an effective candidate vaccine against CHIKV.

Potential Impact:
The main objective of VECTORIE is to enhance Europe’s preparedness for vector-borne emerging diseases, by contributing to the development of a surveillance system and response plan appropriate for European countries to detect, predict and deal with emerging West Nile virus (WNV) and Chikungunya virus (CHIKV) outbreaks. In order to do so, VECTORIE’s integrated approach combined three main areas of expertise: vector biology; viral virulence and pathogenesis; as well as vaccines. The studies in the vector biology part have led to novel knowledge that will enable Europe to improve its ability to monitor the spread of these infections in Europe. The viral virulence part has provides valuable insights for the development of effective clinical diagnostic and prognostic tools. Finally, the vaccine part has provided novel vaccine strategies for these vector-borne diseases, since an efficient approach to control outbreaks caused by WNV and CKIKV is vaccination of people at risk. The potential impact of VECTORIE is described in more detail below for each strand.

Vector biology studies
The vector biology part aims at generating new knowledge on host susceptibility of birds for WNV, and vector competence (ability of the vector to transmit the virus) of the two main vectors of WNV and CHIKV. In North America, WNV outbreaks were preceded by mortality in birds. One of the most important questions that European countries face with an incursion of WNV is how a virulent WNV strain will behave in European crows and jackdaws and if mortality in these birds can also serve as an early warning signal for WNV. Furthermore, the risk of WNV and CHIKV outbreaks in humans is not only determined by vector competence but also by their feeding behaviour. These factors have so far not been investigated for European vectors. Therefore, VECTORIE studied vector competence and feeding behavior of Culex pipiens mosquitoes, which is essential in formulating and focusing a prevention plan. With regard to CHIKV, the effect of temperature on vector competence of Aedes albopictus mosquitoes for CHIKV was determined. This knowledge is necessary to be able to address the question whether Aedes albopictus will be able to efficiently transmit pathogenic CHIKV among the human population in temperate climate zones and thus assess the risks of future CHIKV outbreaks in Europe.

Contrary to the invasive epidemic pattern observed in the US, where the virus was introduced into naïve immune populations of susceptible vertebrate hosts and vectors, there is evidence that WNV circulates in Europe as early as the 1950’s. Since then, relatively large outbreaks of West Nile neuroinvasive disease (WNND) have been recorded in humans and horses in Southern France, Romania, Czech Republic, Southern Russia, Hungary, Italy, and Greece. This extraordinary heterogeneity of WNV, both at lineage and genotype level, together with the appearance of point mutations potentially affecting virulence and/or transmissibility, and the co-circulation of other flaviviruses infecting birds and humans, have important consequences for understanding their ecology and pathogenicity. Introduction of the virus into Europe may be initiated by migratory birds while residential juvenile bird contribute to dispersal of WNV into larger areas or even within different European countries. We first studied the susceptibility of the corvids (jackdaws and carrion crows) to lethal infection with WNV strains circulating in Europe. Our results indicate that carrion crows are susceptible to infection with several WNV strains circulating in Europe. Furthermore, carrion crows could function as amplifying hosts in Europe, in contrast to jackdaws which were less susceptible to lethal infection and developed lower peak viremia. From these studies we can conclude that European birds are not more resistant to WNV infection than their North American counterparts. Therefore, intrinsic resistance of European birds cannot explain why mass mortality among birds are not seen in Europe. During the experiments, all captured birds were first screened for presence of antibodies to WNV. Since WNV specific antibodies are not detected in juvenile birds, the hypothesis of herd immunity is unlikely and therefore not sufficient to explain lack of bird mortality in Europe during WNV outbreaks. Another hypothesis tested within VECTORIE was differential virulence. It is possible that the WNV strains circulating in Europe are less virulent than the viruses circulating in North America. The Ita09, and Gr-10 strains, caused big outbreaks of WNND in humans, however bird mortality was not reported. The strains used in VECTORIE confirmed existence of virulent and less virulent strains for birds. However, the majority of strains tested were virulent for birds. In conclusion, although non-virulent strains circulates in Europe, their low-rank presence cannot explain the lack of mass mortality amongst birds.

Together with a very close relative to WNV, JEV, these viruses are ranked within the top 5 of most imminent threats of emerging zoonoses in the Netherlands and surrounding European countries (RIVM report 2010). The outcomes of future experiments will identify the most competent Culex (sub)species involved in WNV transmission and will answer whether or not natural WNV transmission in the field can readily be expected. These outcomes are important to even better assess the risk for WNV and related arboviral infection in humans and are essential input for mitigation strategies (i.e. targeted vector control) once these viruses invade the Netherlands or neighboring countries. The results of this project are of prime significance for public health in the Netherlands. Knowledge on the ability of indigenous mosquitoes to transmit WNV and other arboviral diseases will aid greatly to early warning and risk assessment of these zoonotic vector-borne viral diseases in the Netherlands and neighboring countries. Our results will be immediately communicated with responsible institutions including the Dutch Ministry of Public Health. This ensures that in case of future zoonotic WNV circulation in the Netherlands, the knowledge from this project is directly used for efficient mosquito control to target the right mosquito species.

Aedes albopictus, originating from South-East Asia, has expanded over the world, including temperate areas. Presently, its presence was reported in 17 European countries. This mosquito species is an efficient vector of 26 arboviruses. It was the principal vector in the first European outbreak of chikungunya (CHIK) in the province of Ravenna (Italy) in 2007 and was incriminated in the autochthonous cases of CHIK and dengue (DEN) in Fréjus (France) in 2010. Thus, European countries have to be considered at risk for the transmission of arboviruses as the result of globalization. Indeed, imported cases of CHIK allowing the initiation of local transmission have been reported periodically in Europe. To be a competent vector, the virus ingested through an infectious blood-meal must overcome different physical barriers to infection and dissemination in the host. Thus, vector competence, depends on both intrinsic factors and extrinsic factors. Temperature is one of the most important extrinsic variables affecting arbovirus transmission via the vector. Variations of environmental temperatures can notably impact mosquito life history traits (as survival) and also, vector competence. However, to our knowledge, few/no data on vector competence of temperate Ae. albopictus populations are available in climatic conditions found in the field. Our results showed that European populations are as efficient to transmit CHIKV at low temperature (20°C) than at optimal conditions (28°C). Moreover, the period between mosquito infection and the transmission to a susceptible host is short. Virus was found in mosquito saliva only three days after virus infection of females submitted to different temperature regimes. Additionally, we showed that a low temperature has a negligible effect on infected mosquitoes. These findings highlight the role of Ae. albopictus in emergence of CHIK in Europe. Furthermore, these results underline the need to consider the different factors that can lead to arboviruses emergence in Europe. First, a high density of Ae. albopictus can enhance the risk of infection of mosquitoes through imported cases. So, the efficient monitoring of both Ae. albopictus density and imported cases returning from CHIK-endemic areas is essential to prevent CHIK epidemics in Europe; and this survey has to be extended over the warm season. On the other hand, communication to general public on the risk of arboviral diseases are indispensable and can help in improving vector control with the help of local populations. As infected mosquitoes can transmit the virus 3 days after having ingested an infectious blood-meal, the detection of imported or/and autochthonous cases should be very rapid. To conclude, the vector competence studies of European populations at temperate environmental conditions may help to better assess the risk of emergence and implement appropriate control measures to limit CHIKV outbreaks in Europe.

Viral virulence studies
The different WNV strains in Europe are assumed to be associated with differences in virulence, i.e. the disease causing capacity of the respective virus strains. To further study the virulence of the respective strains circulating in Europe, experiments were conducted in vitro and mice. The results indicate that all the strains tested were virulent in mice. Taken together with the host susceptibility studies in the vector biology part, we can conclude that virulence has been shown to be species dependent, in that virulence in mice does not correlate with virulence in birds. Furthermore, as all strains tested were virulent in mice and the majority of strains were virulent in birds, the strains circulating in Europe may indeed pose a significant threat to humans.
Furthermore, virulence factors (factors that predict the spreading and disease causing capacity of a strain) that are associated with increased transmission and morbidity of different WNV strains were identified. Identification of the virulence markers of WNV i) may improve the accuracy and predictive value of diagnostic methods, ii) may provide valuable tools for risk assessment of WNV infections, and iii) may provide a basis for the development of attenuated vaccine strains for the protection of vertebrate hosts. This is described in more detail below.
i) The laboratory diagnosis of acute WNV infection is mainly based on the specific detection of viral RNA in sample material, using PCR-based techniques. If the target regions of WNV RNA amplifications include the predicted marker loci investigated in this study, sequence analysis of the amplified genome regions can give additional information on the expected virulence of the strain, and hence suitable treatment options for patients can be identified.
ii) WNV activity is usually monitored in potentially affected regions with passive and active surveillance methods. One of the most sensitive and useful surveillance techniques is detection of WNV RNA in mosquito samples. However, in mosquitoes no clinical manifestation of WNV infection is observable. Therefore, by the investigation of the virulence markers (as some of them are described in this study) could give valuable information on the potential neuroinvasive character and virulence of the virus strain circulating in the affected area. The virulence data could also be taken into consideration during the risk assessment of human and domestic animal infections, as well as when preventive and control measures are implemented.
iii) The most effective and widespread method in the combat against viral diseases is preventive vaccination of susceptible hosts. The development of human vaccines was in the focus of the third strand of this project. Against WNV, inactivated and canarypox virus vectored vaccines are developed and marketed for the protection of horses. However, there are limitations on the application of inactivated vaccines in certain hosts, particularly in wild animals. The identification of genetic markers of virulence gives scientific background for the rational design of attenuated WNV strains, and hence, provides opportunities for development of novel vaccines for large scale applications (i.e. perioral administration).

Furthermore, this part studied the pathogenesis of WNV and CHIKV neuroinvasive disease using state-of-the art technology in order to allow for the identification of biomarkers, which are traceable biologic substances that can be used to measure the presence or progress of a disease or the effects of a treatment, and leads for novel treatment protocols. Understanding the pathogenesis of severe disease is a priority in order to develop effective preparedness measures. This study enabled the identification of functional signaling networks in samples collected during early and late infection. The results translated into biological processes provides novel insights into the in vivo kinetic host reactions against WNV infection and the pathophysiologic processes involved, according to clinical symptoms. Finally, this work offers useful clues for anti-viral research and further evaluation of early biomarkers for the diagnosis and prevention of severe neurological disease caused by WNV.

Besides the understanding of the physiopathogenesis of severe WNV infection, the lack of human vaccine or specific treatment against WNV infection also imparts a pressing need to characterize indicators associated with neurological involvement. Based on kinetic brain proteomic analysis from a mouse model of WNV infection with neuronal involvement, two inflammatory molecules HMGB1 and PRDX6 were associated for the first time to WNV neurological infection. The evaluation of HMGB1 and PRDX6 human serum level in serum from WNV-infected patients diagnosed for either West Nile Fever (WNF) or West Nile neuroinvasive disease (WNND) underlined that both candidate biomarkers appeared associated with WNV infection. Additionally, HMGB1 seems also associated with disease evolution and neurological disorders. Thus, it would be particularly interesting to measure kinetic human serum concentration of HMGB1 in a longitudinal study and determine whether a relationship exists between high early serological levels of this protein and evolution of clinical symptoms. Additionally, an investigation is needed to determine whether a relationship exists between high serological levels of HMGB1 and brain damage in WNF patients. Finally, to distinguish WNV patients from healthy individuals, measurement of serological HMGB1 levels could be helpful for differentiating severe cases from fever. By its intimacy with central nervous system (CNS) structures, modifications in the cerebrospinal fluid (CSF) composition could indeed accurately reflect CNS pathological process. Until now, few studies investigated the association between imbalance of CSF elements and severity of WNV infection. Thus, in the aim to identify potential biomarkers associated with neuroinvasion, an unsupervised approach based on the comparison of CSF protein profiles between patients with WNND and control individuals was performed for the first time. Among the 47 proteins found significantly differentially expressed, Defensin-1 alpha (DEFA1), presented the highest increasing fold-change (FC>12). The augmentation of DEFA1 abundance in patients with WNND was next confirmed at the CSF, but also in serum, compared to the control individual groups. Furthermore, the DEFA1 concentration in serum was significantly more elevated in WNND patients compared to subjects diagnosed for WNF. This study thus provided the first insight into the potential CSF biomarkers associated with WNV neuroinvasion.
Further investigation in larger cohorts with kinetic sampling could determine the usefulness of measuring DEFA1, HMGB1 and PRDX6, as well as the other candidate proteins identified, as diagnostic or prognostic biomarker of detrimental WNND evolution. Moreover, as serum sampling is less invasive than CSF collection, the detection of these proteins at the serum level could be useful and much easier for the physicians to follow and to predict risk of WNV detrimental evolution.

To decipher the mechanisms of CHIKV infection processes in the nervous system, a kinetic analysis of host proteome modifications in the brain of CHIKV-infected mice sampled before and after the onset of clinical symptoms was performed. This work gives new information on putative mechanisms that could be associated with severe neurological CHIKV infection-mediated disease. It also describes possible markers or targets that would need further evaluation as tools for diagnosis and/or antiviral strategies. Interestingly, the comparison of brain proteome dataset from CHIKV-infected mice to WNV-infected mice indicated that very few proteins overlapped. These data underlined that the infectious processes and host proteome hijacked or responses are clearly distinct between these two arboviruses.

Vaccine studies
This part addresses Europe’s need for novel vaccination strategies for populations at risk. VECTORIE developed and subsequently performed rational improvements of vaccine candidates, resulting in vaccine combinations for clinical development that are tailored specifically to the situation in Europe. VECTORIE brings together a platform of technologies using either virus-like particles (VLP) or Modified vaccinia Ankara (MVA) vaccine candidates that are among today’s most promising vaccine development systems. This will be combined with supporting studies on cross-protection of different WNV lineages, to guide in the development of (one) vaccine candidate(s) that can protect against the different circulating WNV strains. Furthermore, standardized neutralization and other immune response assays were developed, which will allow accurate comparison of the protective capacity of different vaccines, which is a critical step forward in determining the most suitable vaccine candidates.

Flaviviruses share common surface antigen components, therefore certain antibodies produced after one flavivirus antigen impulse, can also react with antigens of other, related flaviviruses. This phenomenon on one hand influences the specificity of serological tests; on the other hand, may also have cross-protective value. Cross-protection is a rational basis of active immunisation, which has been successfully used for the control of human virus infections (i.e. vaccinia virus against smallpox virus) and in veterinary applications (i.e. turkey herpesvirus against Marek’s disease virus). The serological cross-reactivity of three related flaviviruses (WNV linage 2, TBEV and USUV) were compared in four different serological tests (ELISA, IIF, HAI and PRNT). These three viruses are simultaneously circulating in certain regions of Europe, and are able to infect and cause similar diseases in humans and some animal hosts. The serological surveillance is usually based on semi-automated methods (i.e. ELISA). Additionally, some diagnostic laboratories are using in-house developed tests (ELISA, IIF or HAI). PRNT is less frequently applied because its complicity and special requirements (BSL-3 laboratory conditions). The results of our studies revealed the differences in specificity and sensitivity of the different methods, and can help us for the interpretation of the results of ELISA-based surveys on the occurrence and prevalence of WNV infections in certain host species and regions. These data can improve the accuracy of surveillance, the applicability of risk assessment tools, and the timing and intensity of control measures.
The ecology of WNV is rather complex. The maintenance and circulation of the virus in a certain region is influenced by viral (i.e. virulence), host (species, susceptibility), vector (species, competence) and environmental (i.e. weather, habitat) factors. One important factor is the susceptibility of vertebrate hosts. If a significant amount of the host individuals are immune against the virus infection, the possibility of the virus amplification is limited in the host populations. The natural hosts of WNV are wild birds. They are also susceptible to USUV infections. The aim of our study was to estimate whether USUV-infection induced immunity has a protective power on subsequent WNV infection. The results of the study indicate that mice immune to USUV did not show clinical signs and mortality after experimental WNV infection. This observation may help to understand the interaction between USUV and WNV circulation in affected regions, and its impact on the epizootiology and epidemiology of WNV. Additionally, the current observation may serve a basis for the development of USUV-based heterologous vaccines against WNV-associated diseases, mainly in wild animals.

Furthermore, our work on the expression of WNV glycoproteins will aid the development of highly specific and cost effective diagnostic kits. Our scientific discoveries on the specific function of CHIKV nsP2 and nsP3 not only have advanced our general understanding of virus replication, they may also aid the development of more effective antiviral drugs and the rational design of live-attenuated vaccines with a much better safety profile.
In addition, our studies provided highly significant advances concerning the standardisation of read-out systems to measure vaccine immunogenicity enabling a more accurate and efficient comparison of candidate vaccines. This achievement is of utmost importance for the further development of the VECTORIE candidate vaccines in late preclinical and clinical research.

We successfully developed several (monovalent, bivalent, and multivalent) VECTORIE candidate vaccines (MVA-WNV and MVA-CHIKV vectors) that are ready to progress to late term preclinical studies. These important results from our vaccine studies substantially contribute to Europe’s capacity to develop novel vaccines against WNV and CHIKV infections, and to improve prevention of these vector-borne diseases in the near future.
In addition, the gained results contribute strongly to the generation of modern and safe VLP-based preventing tools against vector-borne diseases by rational development of potential WNV and CHIKV vaccines candidates. Since both WNV and CHIKV demonstrate a definite risk of human illness and death, generation of ecologically safe, free of potentially infectious genomes, and technologically attractive vaccines on the VLP basis remains among the first priorities of the nowadays healthcare.
In this sense, the performed studies have strong socio-economic impact on the prevention of emerging health risks by consolidated introduction of novel scientific and technological ideas and solutions generated with the project.

The gained results contribute strongly to the creation of an integrated research capacity on vector-borne emerging diseases in Europe, uniting researchers from different fields. The VECTORIE project presented an excellent example of multidisciplinarity and complementarity in the scientific investigations by covering a wide range of biological and medical disciplines and allowing broad exchange of ideas and results. Furthermore, VECTORIE has also significantly contributed to the training of young scientists by the organization of two training courses in the area related to tackling emerging vector-borne diseases in Europe. Finally, the addition of the East-European partner (BMC from Latvia) to the project has strong impact on the general development of that partner as a local scientific and scholar excellence centre in the field of vaccinology, with advanced possibilities for the bachelor, master, and PhD studies. The project and further development of cooperation with consortium partners may prevent scientific brain drain from Latvia to other, also non-European, countries.
The results obtained have been and will continue to be disseminated to the (inter)national scientific and professional (medical) community on (inter)national meetings, by press releases and by publication in international, peer-reviewed journals. Furthermore, VECTORIE has set up an overarching scientific advisory board together with other EU FP7 funded projects on vector-borne infectious diseases (EuroWestNile, ICRES and WINGS) in order to integrate the results of the different projects and to develop a coherent response plan for Europe on how to deal with emerging vector-borne infectious diseases in Europe. Public health authorities and policy makers have been reached through amongst others presentations held at relevant national and international conferences targeting the scientific community as well as policy makers, and through publication in The Parliaments Magazine’s Research review with an editorial focus on Biotechnology looking at the field Healthcare & Pharmaceutical Applications, and publication in European Commission Executive Agency for Health and Consumers (EAHC) brochure published electronically at EAHC webpage, the SANCO webpage as well as in the EU bookstore. In addition, we have disseminated our findings to a broad audience including the general public through our website (www.vectorie.eu) and the lateral FP7 project CommHERE (EU FP7 project on science communication; www.horizonhealth.eu).
Most notable is the final annual VECTORIE meeting. For this meeting we had invited stakeholders such as the European Centre for Disease prevention and control (ECDC) and the European Network for Diagnostics of Imported Viral Diseases (ENIVD) to engage in an interactive discussion on how the results from VECTORIE can be used by our stakeholders to enhance Europe’s preparedness for emerging vector-borne diseases. Furthermore, we had invited other EU funded consortia on vector-borne infectious diseases such as EuroWestNile, WINGS, and ICRES, for discussions to answer the question how the different EU FP7 funded projects on this topic complement each other, and how we can translate the research results of the different projects into adequate and effective response measures to emerging vector-borne diseases.

List of Websites:

www.vectorie.eu