Recent Circulation of West Nile Virus and Potentially Other Closely Related Flaviviruses in Southern France

Introduction

I n recent years, veterinary and medical scientists have recognized numerous (re)emergences of flaviviral zoonoses worldwide (Weissenböck et al. 2010). In western Europe, there have been numerous reports on the increase of West Nile virus (WNV) cases in horses and humans (ECDC 2012) and the emergence of Usutu virus (USUV) in Austria in 2001, and its subsequent expansion to Hungary, Spain, Italy, Switzerland, the United Kingdom, and Germany (Becker et al. 2012). USUV and WNV are maintained through enzootic cycles involving wild birds as reservoir hosts and ornithophilic mosquitoes as vectors. WNV principally infects wild birds, but spillover infections occasionally occur in humans and horses, sometimes leading to severe neurological disorders (Weissenböck et al. 2010). In addition, a wide diversity of flaviviruses has recently been identified in Spain in birds and mosquitoes (Vasquez et al. 2012).

In southern France, WNV infections were reported in horses, humans, and birds in the 1960s and between 2000 and 2006 (Jourdain et al. 2007, Jourdain et al. 2008, Balança et al. 2009). Since 2006, no WNV case has been detected, although multiple outbreaks have occurred throughout southern Europe (ECDC 2012). Despite frequent detection of WNV and USUV in neighboring countries, the French flavivirus surveillance network has been limited since 2007 to reporting important mortality events in wild birds and clinical cases in horses and humans. Therefore, two major issues remain unresolved: (1) Does WNV regularly circulate in southern France in the absence of human and equine cases? (2) Do other flaviviruses circulate in southern France without being detected? We have investigated these questions on the basis of a serological study conducted on wild birds in the Camargue area.

We chose to focus on the magpie (Pica pica) because this species is a sensitive sentinel to detect WNV enzootic activity (Jourdain et al. 2008) and because USUV has already been isolated from this species (Savini et al. 2011). Furthermore, the sedentarity of the magpie offers some guarantee that positive individuals have been infected within the study area or within a few kilometers of it. We chose the Camargue as a study area because WNV circulation has repeatedly been reported there, the last local WNV case detection dating from 2004 (Jourdain et al 2008, Balança et al. 2009). Moreover, this region, located at the crossroads of several bird migration routes, is a hotspot for the potential introduction of bird pathogens.

Materials and Methods

The study was conducted from November, 2009, to December, 2010, in two sites, i.e., A1, which includes dry and wet habitats, and A2, which is a wetland area (Fig. 1). WNV had been reported at both sites in 2004 (Jourdain et al. 2007) in horses (A1) or wild birds (A2), and seropositive magpies had been detected in 2005 (A1 and A2) (Jourdain et al. 2008). Circular multicatch magpie traps were set every working day during three capture sessions: November–December 2009, May–June 2010, and November–December 2010. On the basis of plumage criteria, magpies were classified as juveniles (i.e., born in the preceding spring), second year (i.e., born in spring of the previous year), or adults (Svensson 1992). They were ringed and sampled for blood analysis before being released. Blood samples were centrifuged and resulting serum samples were stored at −20°C. Neutralizing antibody (NAb) titers for WNV (IS-98-ST1 and France00 strains) and USUV (SAAR-1776 strain) were determined using a 96-well plate neutralization test as described (Figuerola et al. 2007). Serum samples with a neutralizing activity (absence of cytopathic effect) at dilution ≥1:10 were considered positive. When there was sufficient volume remaining, positive samples were further tested for NAbs specific for Japanese encephalitis virus (JEV, Nakayama strain) and tick-borne encephalitis virus (TBEV, Hypr strain), a flavivirus belonging to the tick-borne flavivirus group, which is not closely related to WNV. To assess the specificity of the methods, we tested serum samples collected on magpies in the Camargue area during and after the 2004 WN epizootic in horses (Jourdain et al. 2008), for which we strongly expected seropositivity for WNV. These samples were tested for antibodies against WNV, USUV, and TBEV (Hypr strain). All strains were provided by the National Reference Centre for arboviruses of the Institut Pasteur of Paris.

OPEN IN VIEWER

FIG. 1.

Localization of the sampling sites and proportion of seropositive samples (either to Usutu virus and/or to West Nile virus) detected in each site for each sampling session.

Results

Among the 65 magpies sampled, a second-year bird trapped in spring 2010 at site A1 was seropositive for WNV with NAb at titer 160 (Table 1). In addition, three adult magpies, trapped in June 2010, and a first-year bird trapped in December, 2009, all of them at site A2 were seropositive for USUV with NAb at low titer (10–20; Table 1). Finally, two adult birds caught at site A2 presented NAbs against both WNV and USUV. NAbs specific for TBEV or JEV were not detected in the five and three USUV- and/or WNV-positive samples, respectively, that could be tested (Table 1). It remained, however, unclear whether USUV-seropositive birds had been exposed to USUV, WNV, both, or a closely related virus. Cross-reaction between WNV and USUV antibodies can be visualized at low titers in virus neutralization tests because WNV and USUV are closely antigenically related.

The aim of the tests implemented on samples collected in 2004–2005 was to investigate this cross-reactivity. Out of 49 sera, 12 were positive for WNV NAbs only and 22 were positive for both WNV and USUV NAbs (see Supplementary Data available at www.liebertonline/vbz/). In the latter, WNV NAb titers were greater than or equal to USUV NAb titers, whereas in five of our 2009–2010 samples USUV NAb titers were higher than WNV NAb titers. No sample was positive for TBEV or JEV, suggesting that, although cross-reactivity can occur between WNV and USUV, our sero-neutralization tests were specific for flaviviruses antigenically close to WNV.

Discussion and Conclusions

This serosurvey shows that WNV and potentially other closely related flaviviruses have recently circulated in the Camargue area, although no human or equine cases have been reported since 2004. These results indicate the need for further investigations into flavivirus circulation in southern France. However, serology does not allow determining when the sampled individuals were infected. In addition cross-reactivity can occur between antibodies that are specific to closely related viruses. Thus, such future investigations should couple serological and virological detection approaches to isolate the flavivirus strains that are circulating in the region and determine their geographical origin, following the example of the recent study by Valiakos et al., which isolated a WNV lineage 2 strain from the tissues of a magpie in Greece (Valiakos et al. 2011). Furthermore, our findings highlight the fundamental importance of undertaking epidemiological studies on a long-term basis, rather than over short periods following public health crises, to gain insight into viral dynamics within natural reservoirs. They also suggest that the magpie is a suitable sensitive indicator species for the detection of WNV circulation. This point is in accordance with a recent study, which underlined the geographical association between WNV seropositivity in magpies and WNV human cases in Greece (Valiakos et al. 2012). We suggest that a long-term multifocus serosurvey undertaken on magpies throughout the Mediterranean basin would be an efficient tool that could be added to active and passive surveillance focused on humans and horses. The simultaneous use of such complementary tools could greatly help in predicting and dealing with future outbreaks linked to flaviviruses.