Low West Nile Virus Circulation in Wild Birds in an Area of Recurring Outbreaks in Southern France

Abstract

West Nile virus (WNV) has a history of irregular but recurrent epizootics in countries of Mediterranean and of Central and Eastern Europe. We have investigated the temporal enzootic activity of WNV in free-ranging birds over a 3-year period in an area with sporadic occurrences of WNV outbreaks in Southern France. We conducted an intensive serologic survey on several wild bird populations (>4000 serum samples collected from 3300 birds) selected as potential indicators of the WNV circulation. WNV antibodies were detected by seroneutralization and/or plaque reduction neutralization in house sparrows, black-billed magpies, and scops owls, but these species appeared to be insufficient indicators of WNV circulation. Overall seroprevalence was low (<1%), including in birds that had been potentially exposed to the virus during recent outbreaks. However, the detection of a seroconversion in one bird, as well as the detection of seropositive birds in all years of our monitoring, including juveniles, indicate a constant annual circulation of WNV at a low level, including in years without any detectable emergence of WN fever in horses or humans.

Introduction

W est Nile virus (WNV) is the most widely distributed arbovirus that affects humans. Its original Old World range includes Southern Europe, Africa, South Asia, and Oceania (Komar 2000). This virus is transmitted between birds by ornithophilic mosquitoes, many of which rarely bite mammals. It is pathogenic for horses and humans who are incidental to zoonotic transmission—dead-end hosts. Phylogenetic relationships among viruses isolated from distinct continents (Charrel et al. 2003) indicate that the virus is highly mobile, potentially disseminated by asymptomatically infected migratory birds. In temperate regions there is evidence that it can survive winter in hibernating mosquitoes (Nasci et al. 2001).

Transmission has been confirmed at erratic intervals since the mid-20th century in several countries of the Mediterranean basin (e.g., Abdelhaq 1996, Autorino et al. 2002) and in Central and Eastern Europe (e.g., Hubalek and Halouzka 1999, Platonov et al. 2001), including very recent reports (ProMED-mail 2008, Savini et al. 2008). In France, outbreaks occurred on the Mediterranean coast in the delta of the river Rhône—the Camargue—(Fig. 1) in 1962–1963 affecting humans and horses (Joubert et al. 1970), and in 2000 (Murgue et al. 2001) and 2004 (Zeller et al. 2004) in horses. Origin of the virus responsible for these sporadic WNV outbreaks in this area remains uncertain, suspected to be either maintained in resident birds or re-introduced by migratory birds originating from Africa in spring or Eastern Europe in late summer.

FIG. 1.

Recent West Nile outbreak areas in the Camargue region and location of sampling sites. Hatched areas indicate communes (the smallest French administrative subdivisions) where West Nile equine and bird cases had been reported in the previous decade (Murgue et al. 2001, Languille et al. 2004). Dot size is proportional to number of tested birds. Black dots indicate sampling sites without any positive birds; open dots, sampling sites with some positive birds; star, additional sampling place.

We investigated the temporal enzootic activity of WNV in free-ranging birds over a 3-year period in the Camargue. We hypothesized that there is a local circulation of WNV in wild birds that sometimes lead to detectable cases in horses or humans. We first assessed seroprevalence in resident birds after a recent outbreak, and then monitored the potential persistence or re-occurrence of WNV in wild bird populations during successive years without outbreak.

Materials and Methods

Field sampling

We conducted a serological survey in species selected as potential indicators of WNV circulation according to ecological and epidemiological criteria. Two resident terrestrial species, the black-billed magpie (Pica pica) and the house sparrow (Passer domesticus), were selected for their susceptibility to WNV infection in laboratory (Komar et al. 2003) and field studies (Juricová et al. 1998, Komar et al. 2001, Jourdain et al. 2008). We also selected the barn swallow (Hirundo rustica) and some aquatic warblers (Acrocephalus species) as some of the most abundant trans-Saharan migratory species that breed or transit the study area. Black-billed magpies were captured in corvid baited traps. Sparrows, barn swallows, and Acrocephalus warblers were captured by mist net. Nestlings from two additional aquatic breeding species (yellow-legged gull [Larus michaellis] and white stork [Ciconia ciconia]) and few raptors (common kestrel [Falco tinnunculus], scops owl [Otus scops], and tawny owl [Strix aluco]) were also sampled. All birds were ringed and aged when possible by plumage characteristics. Some barn swallows and house sparrows were recaptured (up to nine times), and were re-sampled to test for seroconversion. Blood was collected from the jugular vein, except for storks and gulls, which were bled from the tarsal vein. The volume of blood collected did not exceed 1% of the bird's body weight.

Sampling sites and period

Sampling sites were distributed in the western part of the Camargue (Rhône river delta and neighboring wetlands) and adjacent dry areas (Fig. 1). We sampled most house sparrows in the Camargue, during a 2-year longitudinal survey (>100 birds per month on average from March to October in 2005–2006; few samples were also collected in 2007, n = 119) at a site located within the 2004 outbreak area. In addition, some sparrows (n = 84) were sampled at another site on the Mediterranean coast about 150 km southwest of the Camargue, 2 weeks after an outbreak of West Nile fever in horses in October 2006. Adult breeding and juvenile Barn Swallows were sampled from May to August (>100 birds per month on average) from 2005 to 2007. Magpies were captured and sampled over a range of habitats, from coastal wetlands to inland cultivated areas, from March to December. Migratory Warblers were sampled at one site from late July to late September 2005.

Laboratory analyses

We used plaque reduction neutralization tests (PRNT) and virus neutralization in microtiter plates (SN) to have a standardized diagnostic method for all species. For small-volume samples we used only the PRNT or virus neutralization test in microtiter plates. Positive samples (titer ≥1:10) were tested by PRNT for antibodies to tick-borne encephalitis virus and USUTU virus to rule out possible cross reactions. The West Nile strain Eg-101 was supplied by Institut Pasteur, Lyon; USUTU virus strain was from Vienna 2001-blackbird (939/01); tick-borne encephalitis virus Hypr strain was kindly provided by Prof. Zdenek Hubalek (Medical Zoology Laboratory, Institute of Vertebrate Biology, Academy of Sciences, Valtice, Czech Republic).

Results

A total of 3978 serum samples from 3250 birds were collected in the Camargue and 84 in the area of the small outbreak, mostly from house sparrows (47%) and barn swallows (31%) (Table 1). Eleven birds from three species (house sparrow, magpie, and scops owl) had detectable WNV antibodies (overall seroprevalence 0.3%; Table 2), with titers ranging from 1:10 to 1:160. All WNV-positive samples were negative for TBE and USUTU viruses. These few seropositive birds were found in almost all months from February to October; hence, no seasonal pattern could be detected.

Table 1.

Birds and Serum Samples Tested for WN Virus Antibodies in Different Species in the Camargue

Species	Birds	Samples	Positive
House sparrow (Passer domesticus)	1526 (47.0%)	1780 (45.4%)	7
Barn swallow (Hirundo rustica)	1006 (31.0%)	1473 (36.3%)	0
Black-billed magpie (Pica pica)	256 (7.9%)	263 (6.5%)	2
Acrocephalus Warblers^a	233 (7.2%)	233 (5.7%)	0
Yellow-legged gull (Larus michaellis)	175 (5.4%)	175 (4.3%)	0
White stork (Ciconia ciconia)	36 (1.1%)	36 (0.9%)	0
European kestrel (Falco tinnunculus)	9 (0.3%)	9 (0.2%)	0
Scops owl (Otus scops)	6 (0.2%)	6 (0.1%)	2
Tawny owl (Strix aluco)	3 (0.1%)	3 (0.1%)	0
Total	3250	3978	11

Reed warbler (Acrocephalus scirpaceus) (n = 184), moustached warbler (A. melanopogon) (n = 35), great reed warbler (A. arundinaceus) (n = 8), sedge warbler (A. schoenobaenus) (n = 5), and aquatic warbler (A. paludicola) (n = 1).

Table 2.

Annual Neutralizing Antibodies Seroprevalence of WN Virus Per Age-Class in House Sparrows, Barn Swallows, and Magpies, in the Camargue Area

		2005			2006			2007
Species	No. of birds	Ad.	Juv.	Unk.	Ad.	Juv.	Unk.	Ad.	Juv.	Unk.
House sparrow	Tested	216	329	85	339	472	30	114	1	4
House sparrow	Positive	2 (0.9%)	0	0	3 (0.9%)	0	0	2 (1.7%)	0	0
Barn swallow	Tested	125	265		188	139		121	300
Barn swallow	Positive	0	0		0	0		0	0
Magpie	Tested	32	44		37	68		31	46
Magpie	Positive	1 (3.1%)	0		0	0		1 (3.2%)	0

Juv., first-year birds (i.e., born during sampling year); Ad., >first-year birds (i.e., born before sampling year); Unk., sparrows of unknown age, captured in autumn, after complete molt.

Seroprevalence in magpies was low every year, ranging in adults from 0% to 3.2% (95% confidence interval [CI] 0–9.4%). Seroprevalence was also low (3.1%, CI 0–9%, n = 32 in magpies potentially exposed to the 2004 outbreak (i.e., adult birds sampled in 2005 in the outbreak area; Table 2). The only two seropositive birds we found had been sampled near a 2004 WNV horse case. For comparison, none of 179 magpies sampled outside the 2004 epizootic area were seropositive.

Overall, a low seroprevalence was also detected in house sparrows (0.46% [7/1526]; CI 0–0.8%), but some seropositive birds were detected every year. Seroprevalence in birds exposed to the 2004 outbreak (i.e., adults sampled in 2005 in the 2004 outbreak area) was low (0.9%, CI 0–2.2%) and similar to seroprevalence measured in adults in successive years at the same site (0.9% in 2006, CI 0–1.9%; 1.7% in 2007, CI 0–4.2%) (Table 2). Similarly, none of the birds sampled in the center of the 2006 epizootic area (64 house sparrows and 20 tree sparrows [Passer montanus] from the additional southern site) were seropositive. From the 239 birds we recaptured and re-sampled during the 3 years, one had seroconverted between March and October 2007, revealing circulation of the virus.

The two other seropositive birds were juvenile scops owls sampled in late August 2006. All barn swallows (292 adults and 507 juveniles) sampled in an urban breeding colony from May to August each year were seronegative, as well as barn swallows (n = 207) and Acrocephalus warblers (n = 233) sampled in a reed-bed in August and September 2005 and 2007. Finally, all nestlings from other local breeding birds tested (white storks and yellow-legged gulls) were seronegative.

Discussion

Results from our 3-year intensive serologic survey indicate a low but regular circulation of WNV in wild birds in an area with a history of recurrent epizootics. Overall seroprevalence measured in several bird species was low (<1%), including in birds that had been potentially exposed to the virus during recent outbreaks. However, the detection of a seroconversion in one house sparrow, as well as the detection of seropositive birds in all years of our monitoring, including two juveniles, suggests a constant annual circulation of WNV at a low level in this area.

Seroprevalence measured in magpie was low, including in birds that had been potentially exposed to WNV during the outbreak in 2004. Previous surveys of magpie in the study region indicated high seroprevalence in birds sampled within epizootic areas (22% in 2000, n = 18 [Hars et al. 2004]; 34% in adults in 2005, n = 76 [Jourdain et al. 2008]). During our study, the only two seropositive magpies were both captured 5–8 km from the site of a confirmed equine case in 2004. Magpies are sedentary birds, with a range generally limited to a few kilometers (Cramp and Perrins 1994), and are widespread throughout the study area. However, none of the birds we sampled outside the 2004 epizootic area or birds born after the 2004 outbreak were seropositive. Our findings suggest that infection of magpie by WNV is associated with epizootic events and that WNV did not circulate, at least in an extensive manner, beyond the area revealed by infected horses.

Seroprevalence in house sparrows was also very low, even in adult birds that had been potentially exposed to WNV either during the 2004 outbreak in the Camargue or during the 2006 outbreak in our additional southern site. A similar low seroprevalence had been reported in this species during the 2004 outbreaks in the Camargue (0.9%, n = 144 [Jourdain et al. 2007]) or in Spain (0%, n = 79 [Lopez et al. 2008]). The house sparrow has been proposed as a species involved in WNV amplification in the Camargue after virus isolation in 2004 (Jourdain et al. 2007). Experimental infections demonstrate that it is susceptible to different WNV strains (Langevin et al. 2005), and has good host competence (Komar et al. 2003). We sampled a large number of sparrows (1780 samples) over the 3 years in a wetland area with abundant putative vectors (especially Culex modestus) located <1 km of a confirmed horse case. The very low seroprevalence we found, together with the absence of seropositive birds in our additional outbreak site, consistent with other studies, suggests that house sparrow plays a minor role in the WNV transmission cycle in southern France, related to either a low exposition to WNV vectors or a low susceptibility to the circulating strain, and therefore is not a sufficient indicator of WNV circulation.

Nevertheless, our intensive monitoring of house sparrows over the 3 years of the study suggests a low but recurrent WNV circulation in the study area. Although no juvenile house sparrows tested positive in the study period, the fact that seroprevalence did not appear to decline in adult birds (although it was at very low levels) and that one adult did seroconvert in 2007 suggest that some transmission occurs every year. This is supported by evidence of WNV circulation during previous years without outbreaks, with two seroconversions reported in sentinel birds in the Camargue in 2001 and 2002 (Hars et al. 2004) and the detection of virus in a dead magpie in 2005 (Jourdain et al. 2008). The antibodies detected in two juvenile scops owls in our study in 2006 may have a maternal origin. Maternal antibodies to WNV have been found to persist in birds for 3–4 weeks (Gibbs et al. 2005, Hahn et al. 2006, Nemeth and Bowen 2007). However, the WNV-seropositive owls were at least 2 months old at the time of sampling in late August, which also coincided with the beginning of the usual WNV epizootic period in the region (Murgue et al. 2001), which suggests that WNV was circulating in the area in 2006 too.

WNV is typical of many arboviral zoonoses that are characterized by long periods of invisibility, followed by irregular recrudescences and epizootics. Results from various serological surveys conducted in the Camargue since 2000 suggest that WNV circulated among birds almost every year. As pointed out by Komar (2000), however, such evidence is insufficient to deduce whether the virus is reintroduced annually by migratory birds, or persists in the mosquito population, or if both mechanisms occur. Evidence for WNV-overwintering capacities in mosquitoes exists (Nasci et al. 2001), but phylogenetic relationships among WNV isolated in both Europe and Africa also suggest that an intercontinental transfer of WNV via migratory birds does occur (Charrell et al. 2003).

The Camargue is a major route for birds migrating to and from Africa, being both an important breeding and staging area (Isenmann 1993). Migratory birds we tested (barn swallows and Acrocephalus warblers) originate in spring from their sub-Saharan wintering quarters, and in late summer from their breeding grounds in Central Europe. These birds had been potentially exposed to the WNV in both their wintering and breeding regions where WNV is commonly reported (Hubalek and Halouzka 1999, Murgue et al. 2002, Hubalek et al. 2008), and where they use wetland habitats where ornithophilic mosquitoes are abundant (e.g., C. modestus, one of the major WNV vector) (Balenghien et al. 2006). The absence of seropositive migratory bird in our study (n > 1100) does not support the assumption of a WNV introduction by migratory birds in Camargue, though this has to be interpreted in relation to the millions of migratory birds visiting this region.

In conclusion, our study reveals the existence of an enzootic activity of WNV in the Camargue during a 3-year period without any detectable emergence of WNV epizootic. Unfavorable conditions, including virus strain, vector and reservoir competences, or climate constraints, may prevent WNV from circulating at a high level for more than one summer and lead to only sporadic and geographically limited outbreaks in southern France.

Footnotes

Acknowledgments

We thank the Lamouroux family, owner of the Parc Ornithologique de Pont de Gau; Denis Reudet, Manager of the Réserve Naturelle de l'Estagnol; and Laurence Colas, Manager of the Zoological Park in Montpellier, who allowed us to catch birds on their respective grounds. We also thank T. Baldet for his coordination work within the French West Nile group of the EDEN project. This research was partially funded by EU Grant GOCE-2003-010284EDEN, and the paper is catalogued by the EDEN Steering Committee as EDEN0113 (http://www.eden-fp6project.net/).

Disclosure Statement

The contents of this publication are the responsibility of the authors and do not necessarily reflect the views of the European Commission. No competing financial interests exist.

References

Abdelhaq

AT.

West Nile fever in horses in Morocco. Bulletin de l'Office International des Epizooties, 1996; 108:867–869.

Autorino

, Battisti

, Deubel

, Ferrari

et al. West Nile virus epidemic in horses, Tuscany region, Italy. Emerg Infect Dis, 2002; 8:1372–1378.

Balenghien

, Fouque

, Sabatier

, Bicout

et al. Horse, bird and human-seeking behavior and seasonal abundance of mosquitoes in a West Nile focus of southern France. J Med Entomol, 2006; 43:936–946.

Charrell

, Brault

, Gallian

, Lemasson

et al. Evolutionary relationship between Old World West Nile virus strains: evidence for viral gene flow between Africa, the Middle East, and Europe. Virology, 2003; 315:381–388.

Cramp

. Perrins

. The Birds of the Western Palearctic, VIII. Oxford: Oxford University Press, 1994.

Gibbs

, Hoffman

, Stark

, Marlenee

et al.

Persistence of antibodies to West Nile virus in naturally infected rock pigeons (Columba livia)

Clin Diagn Lab Immunol, 2005; 12:665–667.

Hahn

, Nemeth

, Edwards

, Bright

et al. Passive West Nile virus antibody transfer from maternal Eastern screech-owls (Megascops asio) to progeny. Avian Dis, 2006; 50:454–455.

Hars

, Augé

, Chavernac

, Balança

et al. Surveillance de l'infection de l'avifaune camarguaise par le virus West Nile. Faune Sauvage, 2004; 261:54–58.

Hubalek

, Halouzka

West Nile fever-a reemerging mosquito-borne viral disease in Europe. Emerg Infect Dis, 1999; 5:643–650.

10.

Hubalek

, Halouzka

, Juricová

, Sikutova

et al.

Serologic survey of birds for West Nile flavivirus in Southern Moravia (Czech Republic)

Vector Borne Zoonot Dis, 2008; 8:659–666.

11.

Isenmann

Oiseaux de Camargue. The Birds of Camargue. Brunoy: Société d'Etudes Ornithologiques, 1993.

12.

Joubert

, Oudar

, Hannoun

, Beytout

et al. Epidémiologie du virus West Nile: étude d'un foyer en Camargue. IV. La méningo-encéphalomyélite du cheval. Annales de l'Institut Pasteur, 1970; 118:239–247.

13.

Jourdain

, Schuffenecker

, Korimbocus

, Reynard

et al. West Nile virus in wild resident birds, Southern France, 2004. Vector Borne Zoonot Dis, 2007; 7:448–452.

14.

Jourdain

, Gauthier-Clerc

, Sabatier

, Grège

et al. Magpies as hosts for West Nile virus, Southern France. Emerg Infect Dis, 2008; 14:158–160.

15.

Juricová

, Pinowski

, Literák

, Hahm

et al. Antibodies to alphavirus, flavivirus, and bunyavirus arboviruses in house sparrows (Passer domesticus) and tree sparrows (P. montanus) in Poland. Avian Dis, 1998; 42:182–185.

16.

Komar

West Nile viral encephalitis. Revue scientifique et technique. Office international des épizooties, 2000; 19:166–176.

17.

Komar

, Panella

, Burns

, Dusza

et al. Serologic evidence for West Nile virus infection in birds in the New York city vicinity during an outbreak in 1999. Emerg Infect Dis, 2001; 7:621–625.

18.

Komar

, Langevin

, Hinten

, Nemeth

et al. Experimental infection of North American birds with the New York 1999 strain of West Nile virus. Emerg Infect Dis, 2003; 9:311–322.

19.

Langevin

, Brault

, Panella

, Bowen

et al.

Variation in virulence of West Nile virus strains for house sparrows (Passer domesticus)

Am J Trop Med Hyg, 2005; 72:99–102.

20.

Languille

, Zientara

, Zeller

, Hendrikx

et al. Bilan de la surveillance West Nile en France en 2004. http://agriculture.gouv.fr/IMG/pdf/article_wn_2004_5.pdf. August 18 2008.

21.

Lopez

, Jimenez-Clavero

, Gomez Tejedor

, Soriguer

et al. Prevalence of West Nile virus neutralizing antibodies in Spain is related to the behavior of migratory birds. Vector Borne Zoonot Dis, 2008; 8:615–622.

22.

Murgue

, Murri

, Zientara

, Durand

et al. West Nile outbreak in horses in Southern France: the return after 35 years. Emerg Infect Dis, 2001; 7:692–696.

23.

Murgue

, Zeller

, Deubel

The ecology and epidemiology of West Nile virusAfricaEuropeAsia [Review]. In: MackenzieJSBarettADTDeubelV,Japanese Encephalitis and West Nile Viruses. Heidelberg, Germany: Springer, 2002; 195–221.

24.

Nasci

, Savage

, White

, Miller

et al. West Nile virus in overwintering Culex mosquitoes, New York City, 2000. Emerg Infect Dis, 2001; 7:742–744.

25.

Nemeth

, Bowen

RA.

Dynamics of passive immunity to West Nile virus in domestic chickens (Gallus gallus domesticus)

Am J Trop Med Hyg, 2007; 76:310–317.

26.

Platonov

, Shipulin

, Shipulina

, Tyutyunnik

et al. Outbreak of West Nile virus infection, Volgograd region, Russia, 1999. Emerg Infect Dis, 2001; 7:128–132.

27.

ProMED-mail. West Nile virus–Romania: (Braila). ProMED-mail 2008: 26-SEP-2008:20080926.3042. http://www.promedmail.org. 2008 Oct 22.

28.

Savini

, Monaco

, Calistri

, Lelli

Phylogenetic analysis of West Nile virus isolated in Italy in 2008. Eurosurveillance, 2008; 13. http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=19048. Dec 11 2008.

29.

Zeller

, Zientara

, Hars

, Languille

et al. West Nile outbreak in horses in Southern France: September 2004. Eurosurveillance Weekly, 2004; 8.