West Nile Seroprevalence Study in Bolivian Horses,2011

Introduction

West Nile virus (WNV) is a mosquito-borne virus belonging to the family Flaviviridae included with Japanese encephalitis virus (JEV), St. Louis encephalitis virus (SLEV), Murray Valley encephalitis virus (MVEV), and Usutu virus (USUV) in the Japanese encephalitis antigenic complex (JEAC). All members of this complex are related antigenically. WNV is one of the most widely distributed encephalitic flaviviruses (Weaver and Reisen 2010) and is maintained in an enzootic cycle involving several species of mosquitoes and birds (Hubálek and Halouzka 1999). The virus can also affect mammals, reptiles, and amphibians, but these, in contrast to birds, are generally dead-end hosts due to low viremia (Komar 2000). Infections are generally characterized by mild clinical signs, but severe disease and deaths can be observed in birds, humans, and horses (Hayes et al. 2005). Regarding South America, WNV and various viruses of the JEAC have been isolated in Argentina (Adrián Diaz et al. 2008, Tauro et al. 2012), Brazil (Medeiros et al. 2007), and Ecuador (Johnson et al. 2007). Currently, no studies have been published on the circulation of WNV or other members of the JEAC in Bolivia, although, as reported in a recent PROMED-mail post (November 1, 2012), health authorities of Beni Department suspected a case of SLEV infection in a young man, but due to serological cross-reactions, WNV infection cannot be excluded. No other information has been added so far to clarify the etiology of this case and no other cases have been reported. In the absence of clinical evidence of WNV infection both in humans and horses in Bolivia, the detection of specific antibodies might be an indication of WNV circulation. In this report, the presence of specific antibodies against WNV was investigated in horse sera to obtain evidence of WNV infections.

Materials and Methods

During July, 2011, a convenience sampling of 160 horse sera was conducted; the sera were randomly selected among those sent to LIDIVET (St. Cruz, Bolivia) to be subjected to a Coggins test for equine infectious anemia control program. Sera were tested for the presence of WNV antibodies by a commercially available competitive enzyme-linked immunosorbent assay (C-ELISA) (ID screen West Nile competition, IDvet Innovative Diagnostics, Montpellier, France). Horses were located in the Departments of St. Cruz, Beni, and Tarija. No information was available on vector-borne infection risk factors. The C-ELISA kit was designed to detect multispecies antibodies directed against an epitope of envelope E protein common to WNV and other members of the JEAC. Sixty-six out of 95 C-ELISA–positive sera, randomly chosen, were subsequently sent to the Istituto zooprofilattico sperimentale (IZS) of Teramo (Italy) to be tested for the presence of IgM and WNV neutralizing antibody. These sera belonged to horses located in the following provinces: Andrés Ibanez, 44; Velasco, 6; Florida, 2; Chiquitos, 5; Nuflo de Chavez, 4; Wames, 1; Marban, 3; Gran Chaco, 1. The presence of IgM was detected by the IgM-WNV ELISA (IDEXX, Montpellier, France). C-ELISA and IgM-WNV ELISA were performed and interpreted according to the manufacturer's instruction.

For the detection of WNV-specific neutralizing antibody, the sera were tested by the plaque reduction neutralization test (PRNT) and virus neutralization test (VN) in microtiter plates, using WNV strain Eg-101 kindly supplied by the Unité des Arbovirus et des Fièvres Hémorragiques, Institut Pasteur, Paris (France).

Two-fold serial dilutions of heat-inactivated sera were mixed with an equal volume of 100 median tissue culture infective dose (TCID50) of virus to perform the VN assay. After an incubation at 37°C for 1 h, Vero cells were added to the wells. Plates were read after 6 days of incubation at 37°C. The antibody titer was defined as the reciprocal of the highest serum dilution, which showed at least 90% neutralization.

The PRNT procedure to detect WNV antibodies was performed as described by the OIE Manual of Standards (World Organisation for Animal Health 2012). The titer was determined as the reciprocal of the highest serum dilution with a 90% reduction of the number of plaques (PRNT90). For both the aforementioned tests, sera with a titer ≥10 were considered positive. To ascertain the concordance between PRNT and VN assays, the K measure of agreement (Cohen 1960) was calculated.

Results

Ninety-five out of 160 tested sera (59.4%, 95% confidence interval [CI] 51.8–67.0%) were positive via C-ELISA. None of the 66 randomly selected C-ELISA–positive sera were positive to the IgM-WNV ELISA, whereas 21 and 18 of the same panel of sera tested positive, respectively, to WNV PRNT and VN with titers ranging from 10 to 80. The K measure of agreement between PRNT and VN was 0.926. All of the sera negative to PRNT were negative also to VN, and only three sera were positive to PRNT at titers of 10, giving a result just below the cutoff at VN. The frequency distribution of titers to WNV PRNT and VN is shown in Table 1. Figure 1 shows the geographical location of horses positive to WNV PRNT on a map on which the average of Bolivian annual rainfall is also reported.

OPEN IN VIEWER

FIG. 1.

Geographical location of horses positive to West Hile virus (WNV) plaque reduction neutralization test (PRNT) on Bolivian annual rainfall a map.

Discussion

A high proportion of the tested horses were found positive to C-ELISA. The C-ELISA kit, although designed to detect antibody against WNV, suffers from low specificity due to cross-reactions with other members of the JEAC. Therefore, according to these findings, a high number of sera reacted against JEAC viruses. Nevertheless, more than one-third of the selected horses showed specific WNV neutralizing antibodies. The high specificity of PRNT and VN, also confirmed by their high concordance, provides evidence that these horses experienced WNV infection. The infections were not recent because no IgM-specific antibodies were detected. Horses with WNV neutralizing antibodies were located mainly in some provinces of the St. Cruz Department. These data clearly reflect the bias of sampling and therefore cannot be used for prevalence estimates. Nevertheless Figure 1 shows that these horses were located in areas with climatic conditions suitable for transmission of viral diseases by mosquito vectors. Positive results to C-ELISA not confirmed by PRNT and VN might be due to SLEV a virus already detected in Argentina (Tauro et al. 2012), or other viruses belonging to JEAC, detected also in Equador and Brazil (Johnson et al. 2007, Medeiros et al. 2007).

Conclusions

Considering that WNV and most viruses of the JEAC have zoonotic potential and may have an impact on public health, we think that these preliminary results should be taken into consideration by Bolivian health authorities. It will be important to start a long-term surveillance system for vector-borne infections to define and understand the epidemiology of WNV and of others JEAC members in Bolivia and to better evaluate their impact in public health. This should be achieved by different coordinated actions: (1) Serological surveys based on probability sampling designed to have information on the prevalence estimates in different areas and species; (2) specific virological analysis followed by viral identification on vectors and vertebrates; and (3) active data collection on suspected cases of disease followed by laboratory investigation.