Orthoflavivirus nilense surveillance in the State of Piauí, northeastern Brazil

The Orthoflavivirus nilense (West Nile virus - WNV), a member of the Flaviviridae family, Orthoflavivirus genus, primarily infects mammals and birds. Humans and equines may serve as dead end hosts. The virus was detected in the Americas at the end of the 20th century, and was first discovered in Brazil in 2005.^{(1, 2, 3, 4, 5)} The first human case in Brazil was reported in 2014 in Piaui State, northeast Brazil⁽⁴⁾ and up until 2022 all WNV human cases in Brazil were confirmed in this state. Molecular detection has previously been performed in equines,^{(5, 6, 7, 8, 9)} but the vectors and birds involved in the epidemiological cycle in Brazil remain largely unknown. Therefore, the aim of this study was to investigate the presence of the WNV in birds, equids, and mosquito vectors in municipalities where human cases and/or unexplained mortality of free-living birds have been reported in Piauí State.

SUBJECTS AND METHODS

Study area and characterization of Orthoflavivirus nilense surveillance in Piauí - A total of 11 municipalities were sampled between the years 2019 and 2022, from north to the south Piauí State, Brazil: Água Branca, Amarante, Barro Duro, Bom Jesus, Buriti dos Montes, Juazeiro do Piauí, Lagoa Alegre, Parnaíba, Piripiri, Teresina, and Valença do Piauí. These municipalities were selected based on the confirmation of human cases for WNV or high mortality of free-living birds (Figure).

Capture and collection of biological material - The collection of bird, equid and mosquito samples followed the Epizootic Surveillance Guide of the Brazilian Ministry of Health,⁽¹⁰⁾ and recommendations from Tolsá et al.⁽¹¹⁾ For bird collection, ornithological capture nets (mist nets) were employed,⁽¹²⁾ and birds were identified according to published guides;^{(13, 14, 15)} up to 500 µL of blood was collected from the brachial vein using microhematocrit tubes.^{(2, 11)} In equids, we collected up to 20 mL of blood from the jugular vein of each animal. The animals were identified by species, sex, and age, and an informed consent form was provided to the owner, granting permission for the collection. Mosquitoes were collected in all municipalities, except in Bom Jesus, Buriti dos Montes, and Juazeiro do Piauí. CDC light traps were used for collection,⁽¹⁶⁾ identification was conducted following Consoli et al. and Forattini,^{(17, 18)} and pools were formed with up to 15 mosquitoes of the same species per collection point. All collected samples were refrigerated in liquid nitrogen until laboratory processing.

Molecular biology tests for viral detection

Bird and mosquito samples - Genetic material extraction was carried out using silica column kits (Bioclin ^® ) and magnetic beads (Zymo ^® ), following the manufacturer’s recommendations. The samples subjected to a reverse transcription reaction following Kuno and Catenacci et al.,^{(19, 20)} and the polymerase chain reaction (PCR) following Lanciotti et al.⁽²¹⁾ The reaction products were observed following electrophoresis on a 1% agarose gel stained with ethidium bromide.

Equid samples - Extraction of genetic material was performed using magnetic bead kits (Zymo ^® ). For real-time PCR (RT-PCR), the samples were grouped into pools, each containing 10 distinct samples from the same location. A set of specific primers was used to amplify the complete viral genome of the WNV.⁽⁵⁾ The PCR result was observed using an E-Gel (Thermo Fisher ^® ).

Ethical considerations - The research complied with the relevant laws governing research conducted in Brazil. Approval for the study was obtained from the Ethics Committee for the Use of Animals in Research of the Federal University of Piauí (CEUA/UFPI, Approval No. 605/19). Additionally, authorization was granted by the Institute Chico Mendes de Conservação da Biodiversidade (ICMBio), a division of the Brazilian Environmental Agency (ICMBio-SISBIO, Permit No. 75734-1), and the National System for the Management of Genetic Heritage and Associated Traditional Knowledge (SISGEN, Registration No. AF40BCA).

RESULTS AND DISCUSSION

Between October 2019 and August 2022, we collected a total of 2,706 samples, including 636 birds belonging to 99 species, 420 equids, including donkeys, mules, and horses, and 1,650 mosquitoes grouped into 346 pools, totaling 18 species (Table I). All 2,706 samples underwent reverse transcription followed by WNV RT-PCR testing, but none yielded positive results. All the procedures followed the recommendations of the Ministry of Health of Brazil, both for the diagnosis and for the epidemiological surveillance of birds, equids, and vectors.⁽²²⁾

The most well-represented bird families were Thraupidae (24.69%; n = 157), Phasianidae (11.79%; n = 75), and Columbidae (11.79%; n = 75) (Table II). Despite the sampling effort and the diversity of birds collected, only 3.3% considered potential sentinels for WNV based on previous studies with confirmed detection:^{(23, 24)} Passer domesticus (Passeridae), with 13 specimens sampled (Água Branca n = 6, Teresina n = 6, and Amarante n = 1); Cyanocorax cyanopogon (Corvidae), with four specimens sampled (Valença n = 2, Bom Jesus n = 1, and Piripiri n = 1); Glaucidian brasilianum (Strigidae), with two specimens sampled (Piripiri n = 1, Valença n = 1); Megacops choliba (Strigidae), with one (1) specimen collected (Barro Duro); and Rupornis magnirostris (Accipitridae), with one (1) specimen collected (Piripiri). Based on the lists of dead birds detected with WNV in the United States,⁽²⁵⁾ we can also add the Anatidae family, the most abundant among the 332 species reported; only one representative of the family was collected in Piauí: Anas platyrhynchos domesticus, with five specimens, all collected in Barro Duro, which does not significantly alter the percentage of potential sentinels collected in the state, which would remain below 5%.

However, 6.1% (n = 6) of bird species collected in this study were previously reported with antibodies to WNV in Brazil:^{(26, 27, 28, 29)} 76 specimens of Gallus gallus domesticus (Água Branca n = 8, Amarante n = 9, Barro Duro n = 9, Lagoa Alegre n = 9, Parnaíba n = 20, Teresina n = 12, and Valença n = 8), 38 specimens of Columbina talpacoti (Água Branca n = 16, Barro Duro n = 7, Lagoa Alegre n = 2, Piripiri n = 4, Teresina n = 7, and Valença n = 2), 15 specimens of Furnarius figulos (Água Branca n = 5, Barro Duro n = 1, Bom Jesus n = 1, Lagoa Alegre n = 1, Piripiri n = 3, Teresina n = 2, and Valença n = 2), 13 specimens of Passer domesticus (Água Branca n = 6, Teresina n = 6, and Amarante n = 1), five specimens of Coryphospingus pileatus (Lagoa Alegre), and one (1) specimen of Arenaria interpres (Parnaíba).

Among the potential wild bird hosts for WNV collected, all have a wide distribution across Brazil. Passer domesticus, for example, is an exotic bird present throughout the Americas, with records of occurrence in all Brazilian states, generally associated with anthropogenic environments and highly adaptable.^{(30, 31)} The species C. cyanopogon, on the other hand, is one of the eight species of corvids that occur in Brazil, mainly related to the Northeast region but also present in the North and Central-West regions.^{(32, 33)} The representatives of the Strigidae family collected have different characteristics. Glaucidian brasilianum has a diurnal habit and is the largest representative of its genus in Brazil; its distribution, according to occurrence records, is more associated with the Northeast region, while M. chiliba has a nocturnal habit.^{(34, 35)} Its distribution is similar to that of P. domesticus, but it prefers areas with more vegetation cover and is a species difficult to spot.^{(36, 37)} Meanwhile, Rupornis magnirostris, the only species of the Accipitridae family, is considered the largest hawk in Brazil, with a wide distribution across the Americas, mainly in the South; it is present throughout the country, inhabiting open areas (with sufficient tree cover), forest edges, and urbanized environments.^{(38, 39)}

Regarding the characteristics and distribution of the birds collected that have already been reported with WNV exposure,^{(26, 27, 28, 29)} we have Columbina talpacoti, distributed throughout the Cerrado, being a typical species of the Cerrado and could be considered a bioindicator of environmental quality due to its abundance in areas with little anthropogenic alteration;^{(40, 41)} F. figulos, an endemic species of Brazil;⁽⁴²⁾ Coryphospingus pileatus, found in South America, in countries like Venezuela, Colombia, and French Guiana, in addition to Brazil, in the Northeast, Central-West, and Southeast regions;⁽⁴³⁾ and Arenaria interpres is one of the few migratory species collected, nesting in the far north of North America, Europe, and Asia, migrating to the southern Hemisphere and almost all of the Southern Hemisphere, with reports of occurrence in all regions of Brazil.⁽⁴⁴⁾ The migration of birds in Brazil, coming from the Northern Hemisphere, is one of the most accepted hypotheses for the introduction of WNV in the country, since the WNV epidemiological cycle has not yet been elucidated in the country. Thus, it is still unknown which vectors and hosts exist for the virus in Brazilian territories.^{(45, 46)} Therefore, there is great importance and necessity for investigating bird populations, vectors, and equids, whether through direct or indirect diagnostic techniques, in order to understand the transmission dynamics and the main agents involved.

Among the sampled equids, 76.67% (n = 322) were horses, followed by 9.76% donkeys (n = 41), and 5.71% mules (n = 24) (Table III). These species are considered excellent sentinels for WNV, since they often precede human cases.^{(6, 47, 48)} Unlike birds, there are reports of WNV circulation in equids, both by serological detection^{(26,27,49-52)} and molecular methods^{(6, 7, 8, 9, 29)} in the State of Piauí.⁽⁵⁾

Regarding vectors, all mosquito pools belonged to the potential vector of WNV: mosquitoes in the Culicidae family. The genera Culex spp. (47.40%, n = 164), Mansonia spp. (20.81%; n = 72), Coquillettidia spp. (9.25%; n = 32), Aedes spp. (8.67%; n = 30), and Anopheles spp. (7.23%; 25) were the most sampled (Table IV). There is still much to explore regarding potential WNV vectors in Brazil. There remains just one confirmed detection of WNV in vectors in Brazil from the Amazon region, and identified as a mosquito in the Culex genus,⁽⁵³⁾ which was the most well-represented in the present study.

Despite the extensive effort in sample collection, the non-detection of WNV in this study may be related to delayed diagnosis between human infections in the State of Piauí and the study (up to four years).

In conclusion - We strongly recommend carrying out local surveys in animals and vectors at the time of any suspected WNV infections in humans. Additionally, in case of a delay between a WNV diagnosis in humans, serological tests should be performed in animals instead of molecular tests, as antibodies will better inform of viral circulation as a monitoring tool. We also suggest that free-living birds with neurological symptoms rescued from wildlife rehabilitation centers be tested as part of any survey, as this will increase the sample size of potential natural hosts of WNV in Brazil. In addition to understanding the importance of using official notification systems, which can lead to early detection of outbreaks, we recommend the development of mitigation strategies, both of which will be strengthened through the collaboration between different government and health sectors in the prevention of zoonoses. Lastly, we recommend testing of any neurological equids that were negative on rabies testing since WNV should be a differential diagnosis.^{(54, 55, 56)}

ACKNOWLEDGEMENTS

To the Agência de Defesa Agropecuária do Estado do Piauí (ADAPI), Fundação Municipal de Saúde de Teresina (FMS), Secretaria de Estado da Saúde do Piauí (SESAPI), Laboratório Central de Saúde Pública do Piauí (LACEN), and Centro de Inteligência em Agravos Tropicais Emergentes e Negligenciados (CIATEN) for their assistance and support in the development of this research. We also extend our gratitude to all the collaborators of this work, GBIO - UFPI, GEAS - UFPI, Laboratório de Estudos de Vírus Emergentes (LEVE - UNICAMP), Departamento de Medicina Veterinária Preventiva (UFMG), especially the field teams and the partner laboratories and researchers throughout Brazil.

AUTHORS’ CONTRIBUTION

OLL conducted the molecular investigation, wrote the manuscript, and assisted in the collection of biological samples; TSN and TETL assisted in sample collection, processing, and manuscript translation; FJCA and MGGM assisted in the collection and processing of samples; RSP identified mosquito species and contributed to the manuscript review; JX, MRA, PPB, SCS and EAC assisted in sample processing and manuscript review; ASR identified bird species and participated in the manuscript review; LCJA, WMS, JLPM, ASLN, LCSF, MSPC, MACSV and SLD contributed to the manuscript review; SMSSS and SAFB provided laboratory support and participated in the manuscript review; SLD also provided funding through the Saint Louis Zoo - Institute for Conservation Medicine; LSC conceived the research project, participated in all stages of the research, and contributed to the writing and review of the manuscript. The authors declare no conflicts of interest.