West Nile Virus Monitoring in Migrating and Resident Water Birds in Iran: Are Common Coots the Main Reservoirs of the Virus in Wetlands?

Introduction

W est Nile virus (WNV) is one of the most widespread flaviviruses in birds and mosquitoes, with geographic distribution in Africa, America, Australia, Southern Europe, West Asia, and the Middle East. WNV infections have been described in a wide variety of vertebrates. The main reservoir of the virus is supposed to be wild birds, and the virus has been found in >150 species of wild and domestic birds (van der Meulen et al. 2005). The high and long-term WNV viremia in some species of wild birds allows transmission of the virus to mosquitoes, which maintain the transmission cycle in nature. Spring migration of wild birds is responsible for the introduction of the WNV into previously uninfected areas (van der Meulen et al. 2005), and wild bird surveillance has proven effective for tracking the spread of the virus in North America (Gibbs et al. 2005).

WNV infections in humans and animals have been reported from several West and Central Asian countries, and it seems that the virus in a few of those countries is endemic. This includes southwestern part of Russia (Astrakhan), Armenia, Azerbaijan, Kazakhstan, Tajikistan, Uzbekistan, and Turkmenistan (Lvov et al. 2000). A high risk of exposure to the WNV has been observed in the northwestern part of the Caspian Sea, where a WNV outbreak occurred in 1999 (Lvov et al. 2000). Recently, several cases of WNV infection have been reported in Greece and Turkey. WNV isolation or WN fever infection or clinical symptoms have never been reported in Iran, but serological investigations revealed 10%, 26.6%, and 5% antibody prevalence rates in human blood samples in 1974, 1976, and 2005, respectively (Saidi 1974, Saidi et al. 1976, Sharifi et al. 2010).

The aim of the present study was to investigate the prevalence of WNV antibodies and genome in migratory and resident water birds. Here we report the results for 519 serum samples and 400 oropharyngeal/cloacal swab samples from 26 species of wild water birds and indicate the importance and potential role of the common coot as a natural WNV reservoir.

Materials and Methods

A total number of 519 serum samples from wild water birds belonging to 26 species were investigated (10 families, Table 1). Samples were collected between October and March, from 2004 to 2007, at 11 sites located in six Iranian provinces: Mazandaran, Gilan, West Azerbaijan, Tehran, Fars, and Khuzestan. The samples were mainly obtained from captured birds, or during ringing activities. The sampling sites comprise the most important wetlands of Iran and serve as wintering sites for migratory water birds. The majority of samples (90%) were collected from birds staging in the wetlands along the southern shores of the Caspian Sea—wetlands that form an important ecological site for migratory wild birds along the Central Asian Flyway (Fereidouni et al. 2010).

All serum samples were tested for antibodies to WNV by virus neutralization (VNT) assay and 213 of them also additionally by an indirect immunofluorescence test (IFT). These methods have been previously described (Linke et al. 2007, Seidowski et al. 2010, Ziegler et al. 2010). Samples were initially screened in duplicate at dilutions from 1:10, up to 1:80. Samples with neutralizing antibodies (≥1:10) were retested using serial serum dilutions of up to 1:2560. Serum controls for cytotoxicity were included for every sample. To exclude cross reactions with other members of the Japanese encephalitis virus (JEV) serogroup, a representative batch of serum samples with high WNV antibody titers was also assayed in an Usutu virus (USUV) and JEV neutralization tests under the same conditions as for WNV except using the USUV strain Vienna_2001 (kindly provided by N. Nowotny) and JEV strain Nakayama, respectively.

In addition, 400 oropharyngeal and cloacal swab samples were screened by quantitative real-time RT-polymerase chain reactions (RT-qPCRs), which targeted NS1 and/or E protein open reading frame-specific nucleotide sequences (Eiden et al. 2010). The swab samples were processed by automated RNA extraction (Freedom Evo 3000, Tecan) using the NucleoSpin 96 Virus Core kit (Macherey & Nagel).

To exclude the WNV viremia in serologically negative common coots, all sufficient volume available serums (n=28) were pooled (seven serum samples per pool) and tested by the same RT-qPCRs. RNA was extracted using the QIAamp Viral RNA kit (Qiagen) according to the manufacturer's instructions.

Results

Serum samples from 78 out of the 519 water birds were positive for antibodies to WNV using the VNT and IFT (Table 1), which indicates a seroprevalence of 15% overall. Common coots alone constituted 71 out of the 78 positives. In total, out of 131 tested common coots, 71 (54.2%) revealed positive results, whereas out of 388 other birds belonged to 25 species (except common coots), only 7 were positive. Other positive species included great cormorant (Phalacrocorax carbo), great white egret (Egretta alba), gray heron (Ardea cinerea), and common teal (Anas crecca), respectively (Table 1). The VNT titers ranged from 10 to >240. In total, 28 out of 78 WNV-positive serum samples were checked using Usutu-VNT, 16 of which were negative and the remaining WNV titers ≥3 to 12× higher than USUV. For analyzing cross reaction with JEV, 24 WNV-positive serum samples were tested in VNT, 12 of which were negative and the remaining WNV titers in average 5× higher than JEV.

The highest numbers of WNV antibody-positive common coot samples were found during February and March (Table 2). In addition, the proportions of highly positive samples were more prevalent during these 2 months. Nearly 53% of samples were collected during November and December, whereas only 38% of positive results were found in these 2 months (Table 2).

No WNV-specific RNA was detected by RT-qPCR in cloacal, oropharyngeal, or serum samples of these birds.

Discussion

Birds are the main reservoir of WNV and many related flaviviruses in nature. WNV infections have never been reported in Iran and serological investigations in the 1970s indicated a high sero-prevalence among the human population in several provinces (Naficy and Saidi 1970, Saidi 1974, Saidi et al. 1976). Recent real-time RT-PCR and ELISA studies on blood donors revealed negative results for viral RNA detection, but 5% serological positive results (Sharifi et al. 2010). The data presented here highlight the presence of WNV in wild birds in particular Iranian provinces and also point to the important role of the common coot as a natural reservoir for WNV. More than 50% of tested common coots were seropositive by IFT and VNT; conversely, <2% of other species were seropositive. Although all investigated water birds inhibit close or common habitats, the prevalence of WNV antibodies was significantly different. Compared to Anseriformes, coots are supposedly more exposed to mosquitoes because of their long skinny legs and the bald spots in the plumage on their heads (Hubalek et al. 2008). The alternative hypothesis is that there may be specific WNV receptors in common coots, compared to other water birds. A high prevalence of WNV antibodies in common coots has been also shown before in Spain and Czech Republic (Figuerola et al. 2007b, Hubalek et al. 2008, Lopez et al. 2011).

The common coot and most of the other water birds investigated are migratory birds wintering in Iran. The wetlands in the southern part of the Caspian Sea represent major wintering and stopover sites during migration for many wild water birds from Siberia and northern Russia. Several million migratory birds normally arrive in October and either remain until February/March or migrate even further southward. It seems that the wetlands around the northern part of the Caspian Sea (belong to Russia, Azerbaijan, Kazakhstan, and Turkmenistan) and India are the main breeding or alternate wintering sites for coots wintering in Iran (Argyle 1975). However, there are only small pieces of data originating from ring findings in these countries, and real mapping of the migratory routes would need satellite telemetry studies. At least one ringed coot was found in Kalmyk, Russia—an area close to Volgograd and Astrakhan, the regions where WNV outbreaks occurred in 1999 (Platonov et al. 2001). Other ringed coots were found in or originated from northern countries (Russia, Azerbaijan, Kazakhstan, and Turkmenistan) in which WNV was reported as endemic infections among humans (Lvov et al. 2000). Therefore, it cannot be excluded that the seroreactive coots contracted WNV in Russia, other central Asian countries, India, or any other neighboring countries, although there are less favorable climatic conditions for sustaining endemic cycles compared to those in Iran.

Clinical manifestations of WN fever in humans usually start in July or August, depending on how high temperatures get and how active the resulting mosquito population is (Zeller and Schuffenecker 2004). With this in mind, the high proportion of positive common coot samples during February and March in our study is hard to explain. This phenomenon could be due to annual fluctuations of WNV activity, because samples that collected during November and December belonged to 2003, whereas most of the February and March samples were collected in 2005 and 2007. However, a recent capture/recapture study in Spain proved a seroconversion of juvenile coots even during December (Figuerola et al. 2007b). The same scenario could also apply to Iran during the winter months. An alternative hypothesis could involve long-lasting antibody titers after late summer/early autumn infections or a recent infection in a more temperate country such as India, which is one of the migration sites for common coots wintering in Iran (Argyle 1975). However, in this study, we found positive resident birds like the great cormorant, gray heron, and white great egret, which may indicate an endemic WNV circulation in Iran. The high WNV antibody titer observed in a great cormorant in December could be indicative for a recent local infection.

This study was focusing on water birds, and no sample from terrestrial birds was available for analysis. However, several studies in the United States and Europe indicate the role of corvids and birds of prey in the ecology of WNV (Anderson et al. 1999, Dawson et al. 2007, Angelini et al. 2010). It would be worthwhile to study also such birds, especially resident species, in future. Two species dominate among the samples (common teal, n=140; common coot, n=131). Relatively high positivity rate was also found in great cormorants (three positive of seven investigated); therefore, higher sample number might have also indicated a potential role of this species.

The presence of Culex pipiens, Culex modestus, Anopheles messeae, Aedes vexans, and other important mosquito vectors for WNV (Azari-Hamidian 2007) in Iran could also contribute to the transmission of the virus among different vertebrate hosts. Further studies are needed on the seasonal distribution of vector mosquitoes in different regions of Iran in addition to studies to determine whether other particular arthropod vector species can also play an important role in the transmission of WNV.

Although common coots are competent hosts for WNV, their role in the perpetuation of the virus in its ecological niche remains elusive. For active virus transmission during stopovers between long-haul migratory flights, birds must remain persistently infected and develop a high viremia before being bitten by an appropriate vector (Figuerola et al. 2007a). However, all RT-qPCR investigations on cloacal and oropharyngeal swab samples and WNV antibody-negative common coot serums collected from these water birds revealed negative results, which exclude a carrier state. More detailed studies on the pathogenesis and immunology of WNV infections in common coots are therefore necessary.

Conclusions

This investigation provides the first evidence of WNV antibody seroprevalence in resident and migratory water birds in Iran. Our results demonstrate that common coots are highly susceptible to WNV. An endemic situation or a regular seasonal introduction of WNV into Iran must be assumed based on the broad antibody prevalence in humans, wild birds, and horses (personal communication), even though the virus has not yet manifested itself.