Serologic evidence of West Nile virus and Saint Louis encephalitis virus in horses from Southern Brazil

Abstract

Flaviviruses as West Nile virus (WNV), Saint Louis encephalitis virus (SLEV), Ilhéus virus (ILHV), and Rocio virus (ROCV) are previously reported in different Brazilian regions, but studies in Southern Brazil are still scarce. To improve the information regarding flaviviruses in Southern Brazil, horse serum samples were analyzed using RT-qPCR and a commercial ELISA-Ab against WNV followed by PRNT₇₅. All 1000 samples analyzed by real-time RT-PCR resulted negative. The 465 subsampled samples were analyzed by a commercial ELISA-Ab against WNV, and the 18.5% (86/465) positive samples were further analyzed by PRNT₇₅. In the PRNT₇₅, 13/86 and 2/86 horses were positive for SLEV and WNV, respectively. It was observed that 5.8% (13/226) of the farms presented at least one positive animal for SLEV in PRNT₇₅, whereas 0.9% (2/226) for WNV. Apart from the lower seroprevalences identified when compared to data previously reported in other Brazilian regions, our results suggest that public health professionals must be aware of the presence of these potential zoonotic pathogens.

Supplementary Information

The online version contains supplementary material available at 10.1007/s42770-021-00474-7.

Introduction

The Flavivirus genus, within the Flaviviridae family, contain arthropod-borne RNA viruses with zoonotic potential, as Zika virus (ZIKV), dengue virus (DENV), yellow fever virus (YFV), tick-borne encephalitis virus (TBEV), West Nile virus (WNV), Saint Louis encephalitis virus (SLEV), Ilhéus virus (ILHV), Rocio virus (ROCV), and Japanese encephalitis virus (JEV) [1].

Particularly, the SLEV, WNV, ROCV, and ILHV were reported in different Brazilian regions [2–5]. They present cross-reactive immunity [6] and are transmitted by a bird-mosquito cycle, with birds acting as amplifying hosts [3, 5, 7–9]. These flavivirus species infect mostly Culicidae mosquitoes which can potentially transmit these viruses to a wide range of vertebrates on which they feed [2, 3, 8–14]. Mammals, primarily horses and humans, are unable to contribute to the transmission cycle and thus are considered “dead-end” hosts [3, 9, 11, 15, 16].

An increasing number of WNV and SLEV infections reported in humans and animals confirms the increasing spread to areas with moderate climate in Europe, Americas, and Mediterranean [7, 8, 10, 12, 17–21]. Its rapid spread is credited to migratory birds [7, 15, 18, 20], changes in the world’s climate, long-distance travelers, and economic trade globalization [22].

In Brazil, where suspicious cases of WNV, among other arboviral diseases, are of mandatory notification for animal and public health professionals (see https://www.saude.gov.br/o-ministro/942-saude-de-a-a-z/febre-do-nilo-ocidental/21160-vigilancia-nilo), recent clinical cases have been identified in both humans and horses from the northeast [23, 24] and southeast regions [24, 25]. The virus was isolated in a horse’s brain from Espírito Santo state, Southeast Brazil [25]. In 2015, the first documented human case of WNV in Brazil was reported in Piauí state, northeast Brazil, in a rural worker [23].

The continuous monitoring of flaviviruses as SLEV, WNV, ROCV, and ILHV in hosts and reservoirs is important as public health strategy due the consequence of these viruses infection in human hosts [2, 26–31] SLEV, WNV, ROCV, and ILHV positive serology in special was previously reported in horses from Brazil, where the majority of studies had a descriptive design and were conducted in the central-west region [4, 5, 9, 17, 18, 30, 32]. In horses from this region, these works showed positive serology for WNV ranging between 3 and 8% [4, 17], for SLEV between 5.4 and 12.3% [4, 5], 1.8% for ILHV [4], and 6.1% for ROCV [5]. However, despite the evidence of positive serology for WNV detected by ELISA-Ab but not confirmed in PRNT in wild birds collected in Southern Brazil in two different studies [18, 33], there is a lack of knowledge available about this virus infection status in horses from this region. Moreover, the Brazilian surveys performed in horses only analyzed a small number of samples from a limited number of farms [17, 18, 32]. Thereby, the present study aimed to verify the presence and geographical distribution of flaviviruses in horses from the State of Rio Grande do Sul.

Materials and methods

Study design and descriptive analyses

The 1000 horse serum samples analyzed in this study were previously collected by the Official Veterinary Service (SEAPDR) of the State of Rio Grande do Sul (RS), between September and October of 2013, during the equine infectious anemia virus (EIAV) surveillance campaign as a step of National Equine Health Program (Programa Nacional de Sanidade Equina) of the Ministry of Agriculture, Livestock and Supply (Ministério da Agricultura, Pecuária e Abastecimento) from Brazil. During the sample collection, a questionnaire designed to access potential risk factors for EIAV and containing immunization information was applied in each randomly selected farm. In addition, geographical coordinates were gathered.

Environmental and farm-level available determinants were descriptively analyzed with Microsoft Office Excel ©, while descriptive maps were elaborated in QGIS 3.6 ©.

RNA isolation and RT-qPCR

All 1000 samples were analyzed for RT-qPCR to verify the presence of flavivirus specific-RNA. Total RNA was extracted using TRIzol LS reagent (Thermo Fisher Scientific, Waltham, MA, USA), according to manufacturer’s instructions. The cDNAs were synthetized using GoScript™ Reverse Transcription System (Promega, Madison, WI, USA) using random primers (Invitrogen™, Carlsbad, CA, USA). The RT-qPCR was performed using panflavivirus primers [34] and PowerUp™ SYBR® Green Master Mix (Thermo Fisher Scientific). This primer pair is able, in silico, to detect WNV, SLEV, ROCV, and ILHV. ZIKV strain MR766 (GenBank accession number MK105975) was used as the positive control.

ELISA-Ab and PRNT₇₅

The sample size was calculated to find at least one seropositive case with 95% confidence level assuming a prevalence of 1%. The methodology applied was that recommended by Cannon and Roe [35] and Martin and others [36]. The evaluation of flavivirus-specific antibodies presence was made in a subsample of 465 animals. An ELISA-Ab against glycoprotein E of WNV (Anti-West Nile virus ELISA horse IgG, EUROIMMUN Medizinische Labordiagnostika AG, Lübeck, SH, Germany) was used for screening.

Positive and ambiguous samples in ELISA-Ab assay were further evaluated using a PRNT₇₅ in BHK-21 cells (ATCC® CRL-1201™), as previously described [37]. For WNV, SLEV, ILHV, and ROCV PRNT assays, viral chimeras build on a YFV backbone (where prM and E genes of YFV 17D vaccine strain were exchanged by those of each virus) were used (S. Arenhart, A. G. Oliveira, and L. H. V. G. Gil, unpublished data). Serial dilutions of previously positive ELISA-Ab sera (1:10 to 1:5,120) were mixed with 100 PFU of selected flaviviruses (YFV-prM/E-WNV, YFV-prM/E-ILHV, YFV-prM/E-SLEV, and YFV-PrM/E-ROCV) in the respective working dilution, incubated for 1 h at 37°C and then added to the cells and incubated again at 37°C. After l h, MEM containing 1.5% carboxymethyl cellulose (CMC), 5% FBS, and 1% penicillin-streptomycin were added to each well, and plates were incubated at 37°C, 5% CO₂, for 96 h. After this incubation, plates were fixed with 10% formaldehyde for 1 h and subsequently stained with 1% crystal violet for 15 min. Plaques were counted, and the neutralization capacity was estimated as the reciprocal of the sera dilution that caused a 75% reduction in PFU. This assay was performed in duplicate and repeated once. A serum sample was considered to contain antibodies to one of the flaviviruses tested if the PRNT₇₅ titer was at least 4-fold greater than the other putative titer.

Results

All the 1000 horse serum samples were analyzed for the presence of flavivirus specific-RNA by RT-qPCR using a panflavivirus primer pair [34] and resulted negative.

A subsample of 465 horse serum, representing 226 different farms, were tested using an ELISA-Ab against glycoprotein E of WNV for screening putative sera containing flavivirus-specific antibodies, and 18.5% (86/465) were positive (Fig. 1). These samples were reanalyzed using a PRNT₇₅ in order to classify the antibody specificity as SLEV, WNV, ROCV, and/or ILHV. The assay pointed to 13/86 samples positive for SLEV-Ab, 2/86 for WNV-Ab, but no sample was positive for ROCV and ILHV. The SLEV PRNT₇₅ titers ranged from 40 to 320, whereas the WNV PRNT₇₅ from 80 to 320 (Table 1).

Fig. 1.

Scatter plot showing results for ELISA-Ab for WNV in 465 horse serum samples. The results were expressed as the ratio between optical density (OD) observed in tested sample and the OD detected in the kit calibrator, according manufacturer’s instructions. Samples presenting ration >1.1 were considered as positive

Table 1.

Results obtained in ELISA-Ab for WNV detection and PRNT₇₅ against WNV, SLEV, ROCV, and ILHV

Sample	Municipality	ELISA-Ab for WNV*	PRNT75 (antibody titer)
			WNV	SLEV	ROCV	ILHV	Result
007-04	Alegrete	3.614407	<20	80	<20	<20	SLEV
042-04	Bagé	1.404545	320	<20	<20	<20	WNV
056-05	Bom Jesus	1.786145	<20	80	<20	<20	SLEV
081-01	São Pedro das Missões	4.072289	<20	320	<20	<20	SLEV
133-03	São José do Ouro	3.336364	<20	80	<20	<20	SLEV
158-01	São José do Norte	2.772727	<20	80	<20	<20	SLEV
217-01	Capão do Leão	2.243671	<20	40	<20	<20	SLEV
220-01	Caguçu	1.759494	<20	40	<20	<20	SLEV
253-02	Candelária	2.789174	80	<20	<20	<20	WNV
256-02	Amaral Ferrador	1.888889	<20	80	<20	<20	SLEV
276-01	Encruzilhada do Sul	1.513636	<20	40	<20	<20	SLEV
311-04	Pirapó	1.794872	<20	80	<20	<20	SLEV
326-02	Santana do Livramento	3.046613	<20	80	<20	<20	SLEV
327-02	Uruguaiana	2.216102	<20	80	<20	<20	SLEV
331-09	Uruguaiana	3.182203	<20	80	<20	<20	SLEV

Only samples positive for PRNT₇₅ are represented

*ELISA-Ab results were expressed as the ratio between OD observed in sample and OD of the kit calibrator, according manufacturer’s instructions

It was observed that 5.8% (13/226) of the farms presented at least one positive animal for SLEV in the confirmatory assay (PRNT₇₅), whereas 0.9% (2/226) for WNV (Table S1). All the positive farms presented only one animal positive for WNV or SLEV.

Figure 2 presents the geographical distribution of the 226 farms according to its serological results. Farms were localized all over the Rio Grande do Sul state and characterized by a population ranging from 1 to 260 horses, with a median of 2. In 85% of these enterprises, the presence of mosquitos was confirmed by local workers, and in 95% of them, some sort of water resource, such as a river, a lagoon, or a dam, was reported (Table S1).

Fig. 2.

Geographical distribution of sampled farms investigated for the presence of flaviviruses in Rio Grande do Sul state, Southern Brazil. The location of each negative and positive farm is represented.

Discussion

Of interest of the present study, especially WNV but also SLEV, ILHV, and ROCV display a bird-mosquito cycle with birds acting as amplifying hosts [3, 5, 7–9], where humans and horses are dead-end hosts [2]. Human cases were previously reported for SLEV, WNV, ILHV, and ROCV in Brazil [2, 9, 23–25, 31], and positive serology was obtained in horses from north, northeast, central-west, and southeast regions of the country [3, 5, 9, 17, 18, 30, 32]. In order to solve the lack of information regarding the presence of these four flaviviruses in Southern Brazil, we used serology and real-time RT-PCR in horse serum.

A discrepancy between the results was observed in the ELISA-Ab for WNV (Fig. 1) and PRNTs (Table 1), where only 15 out the 86 ELISA-Ab-positive sera could be classified as positive for SLEV or WNV, and all were negative for ROCV and ILHV. It is important to highlight that it was used a commercial ELISA-Ab for WNV, and despite cross-reactions do occur among flavivirus, they do not happen 100% of the time, and some may be missing out on other positive samples. However, the discrepancies between ELISA-Ab for WNV and PRNTs presented a higher magnitude when compared with other works [17, 18, 38]. The Brazilian territory contains a plethora of flaviviruses [2, 4, 6, 29, 39] which present cross-reactive serology and made difficult to apply completely specific assays. Moreover, the presence of flaviviruses not included in the PRNTs and even unreported or yet unknown flaviviruses must be considered. It is of interest that two previous studies performed in the Rio Grande do Sul state also detected flavivirus antibodies by ELISA-Ab in wild birds, but was also not able to identify the origin of the antibodies as a result of WNV or SLEV infection by PRNT [18, 33]. Both studies analyzed migratory birds, where the first one analyzed common terns (Sterna hirundo) [18], while the second study evaluated a variety of birds comprising the orders Acciptriformes, Falconiformes, and Strigiformes [33]. Migratory birds are of special interest since they can carry flaviviruses along distant regions [2, 7]

All the horse sera tested negative in the panflavivirus RT-qPCR assay. The results may be due to that horses and other mammals present low viremia, which made them unable to contribute to the transmission cycle [3, 9, 11, 15, 16], besides the viremia generally longs 1 to 6 days post-infection [16].

We found, in PRNT tests, serologic evidence of SLEV and WNV exposure (Table 1). Farms with at least one positive sample were spread all over state boundaries (Table S1), with some small areas tending to a cluster pattern (Fig. 2). It is important to highlight that the practice of moving horses to gather in events inside Rio Grande do Sul State boundaries is quite common. So, if we are not able to specify the location where some animals were infected, our results strength the hypothesis that WNV and SLEV circulate in Southern Brazil.

Apparently the frequency of SLEV-positive serology in horses from the Pampa biome is higher than the observed for WNV, which agrees with previous works performed in Pantanal region [4, 17, 18, 30]. The lower frequencies of SLEV and WNV observed in RS state agree with the lower frequencies sporadically reported for flaviviruses, as reported for DENV in humans [40], ZIKV [41], and YFV [42], both probably due a less favorable climatic conditions for the arthropod hosts.

Previous studies reported the presence of SLEV serology alone in horses from Uruguay [43], whereas both SLEV and WNV were reported by isolation and serology in horses, humans, and birds from Argentina [20, 44, 45]. The vicinity between these countries and the region analyzed in the present work may favor the entrance of these viruses in Southern Brazil, carried by wild bird reservoirs that are able to access these different areas.

Conclusions

To the best of our knowledge, this is the first evidence of positive serology for SLEV and WNV in horses from Southern Brazil. These results are of extremely importance to public health professionals in Brazil, since infectious disease surveillance programs, by rule, are conducted by distinct agencies and do not integrate, in a formal and continuous manner, data from human and animal populations in the country. Entomological and sentinel animal surveillance must be continuously carried out aiming to the early identification of viral circulation to reduce the risk of enzootic cycles’ establishment and to protect public health and economy in Brazil. Said that, systematical studies should be performed to uncover the presence of other flaviviruses in Southern Brazil.

Funding

This study was funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul (FAPERGS), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)—Finance Code 001, and Propesq/UFRGS.

Declarations

Conflict of interest

The authors declare no competing of interests.