Occurrence of avian reovirus and picobirnavirus in wild birds from an environmental protection area in the Brazilian Amazon

Wild birds have great prominence in the transmission of diseases to humans, mainly due to their ease of access to human populations, raising concerns about the potential impact of that proximity on public health. The present study reports ARV and PBV occurrence in wild birds from an environmental protection area in the Amazon biome, in Pará state, Brazil. We obtained 155 fecal specimens from 47 species of wild birds for RNA extraction, ARV and PBV detection utilizing molecular methods, nucleotide sequencing, and phylogenetic analysis. ARV prevalence was 0.6% (1/155), being positive in an individual of Myrmotherula longipennis , and PBV prevalence was 1.29% (2/155), affecting two individuals of Guira guira . The PBV strains were assigned to genogroup I based on phylogenetic analysis, and they shared a strong phylogenetic link with strains isolated from different geographic locations. The ARV strain was more closely related to strains that had previously circulated in the same region. The presence of ARV and PBV in this habitat suggests that infection cycles of these agents occur naturally in the wild ecosystem, potentially triggering transmission events between bird species and humans. This is the first study on ARV detection in wild birds in Brazil and the first report on the occurrence of PBV in wild Guira guira . Additional studies are required to determine the epidemiology, origin, evolution, and emergence of new potentially pathogenic viruses in the Amazon.


INTRODUCTION
Wild birds are among the animals with high prominence in the context of disease transmission to humans.The ability to fly propitiates birds ease of access to human populations, increasing the transmission risk of pathogens to humans, and the potential impact on public health (Morais et al. 2019).
Viruses are the most important clinical and epidemiological pathogens in birds.Infections that occur in the first weeks of avian life are usually from viral etiology (Luz et al. 2014;Santos et al. 2015).Rotaviruses (RV), avian reoviruses (ARV), picobirnaviruses (PBV), avian influenza viruses (AIV), astroviruses (AstV), coronaviruses (CoV) and West Nile viruses (WNV) are examples of the most significant viruses for global public health transmitted by birds.This is due to the potential for dispersal of wild birds, especially those that have migratory habits (Bezerra et al. 2014;Lu et al. 2015;Noh et al. 2018;El Taweel et al. 2020;Hassan et al. 2020;Vidaña et al. 2020;Rahman et al. 2021).
ARV and PBV are frequently reported in poultry infections, associated with clinical or subclinical diseases, which causes serious economic impacts to the poultry industry (Day and Zsak, 2016;Pankovics et al. 2018).ARV are described as important agents of gastroenteric diseases, viral arthritis, and tenosynovitis in birds (Davis et al. 2013;Assunção et al. 2018).PBV can be detected in excrements with both normal and diarrheic aspect, both from domestic and wild birds, therefore its role as a primary agent of acute gastroenteritis remains unclear (Silva et al. 2014;Verma et al. 2015).
In Brazil, there are few studies on the occurrence of ARV and PBV in free-living wild birds, especially in natural environments close to urban centers, where the risk of zoonotic transmission of infectious agents by wild animals to the human populations is higher.Studies on molecular epidemiology are essential to describe the prevalence of infectious agents, characterize variants and estimate the potential impact of wild reservoirs on public health.
This study aimed at describing the occurrence of ARV and PBV in wild bird samples collected in an environmental protection area in the Amazon biome.

Ethical and legal aspects
This study is in accordance with the ethical principles of animal experiments and was approved by the Animal Ethics Committee of Universidade Federal Rural da Amazônia (protocol # 025/2018 CEUA/UFRA), and by the Biodiversity Information and Authorization System (SISBIO) (license # 63488-1).

Study area
The study area included forest areas close to deforested areas for pasture and/or urbanization in an area in the campus of Universidade Federal Rural da Amazônia (UFRA) (1º27'21"S, 48º26'12"W), within the environmental protection area of the metropolitan region of the city of Belém (Área de Proteção Ambiental da Região Metropolitana de Belém, or APA Belém) (Figure 1), in Pará state (Brazil), a protected area of 5,647 ha that harbors a wide variety of wild animals.

Bird capture and clinical specimen collection
From February to October 2019, birds were collected in the months of February and March (rainy season), June and ACTA AMAZONICA July (dry season), and October (transition season).Seven mist nets were used (Figure 2), fixed to the ground with metal wattle, opened from 5:00 am to 10:00 am, and checked every thirty minutes.Captured birds were weighed and identified to species level, sex (male or female), and life stage (young or adult) (Gwynne et al. 2010;Sigrist 2014).The birds were kept individually in cardboard boxes lined with aluminum foil paper.Fecal specimens deposited in the boxes were collected or obtained directly from the bird's cloaca using sterile swabs, kept in cryogenic tubes, and stored at -20ºC until processing.After collecting the fecal samples, the birds were marked with non-toxic ink (Raidex®) to identify in case of recapture and released back into the environment.Due to stress, some birds died during handling, so that the collection of feces was not possible.The birds that died were submitted to necropsy and their intestine was stored for future studies.

RNA extraction and electrophoresis
Suspensions were prepared at 10% by diluting the feces and/or intestinal samples in Tris/HCl/CaCl 2+ buffer (pH 7.2 0.01M), clarified by centrifugation at 4,000 rpm/10 minutes.The supernatant was submitted to RNA extraction according to the protocol by Boom et al. (1990).
The products of RNA extraction were submitted to polyacrylamide gel electrophoresis (PAGE) for ARV and PBV detection by electrophoretic profiles according to Pereira et al. (1983).

RT-PCR for ARV detection
RT-PCR was performed targeting the ARV S2 gene to amplify a partial fragment of 625 bp of the S2 gene, using forward primer PAF (5' -ACT TCT TYT CTA CGC CTT TCG -3') and reverse PAR (5' -ATY AAW DCW CGC ATC TGC TG -3') (Zhang et al. 2006).To complementary DNA strain (cDNA), 4 µL of extracted dsRNA and 2 µL of pair of primers (20 mM) were used.The reaction followed an incubation of 5 minutes at 97ºC for denaturation, followed by a heat shock of 5 minutes at 0ºC.
The amplicons obtained by PCR were performed using agarose gel electrophoresis at a concentration of 1.5% in Tris/Borate/EDTA (TBE) buffer, and gel stained with SYBR® Safe DNA Gel Stain (Invitrogen®).GEL DOC 1000 image processor (Bio-Rad Laboratories, Inc., Hercules, CA) performed photo documentation.
The amplicons obtained by PCR were performed using agarose gel electrophoresis at a concentration of 1.5% in Tris/Borate/EDTA (TBE) buffer, and gel stained with SYBR® Safe DNA Gel Stain (Invitrogen®).GEL DOC 1000 image processor (Bio-Rad Laboratories, Inc., Hercules, CA) performed photo documentation.

Nucleotide sequencing
Amplicons were purified using ExoSAP-IT™ kit (Applied Biosystems™) according to the manufacturer's recommendations.After purification, products were subjected to nucleotide sequencing using RT-PCR/nested-PCR primers, and Big Dye Terminator® v.3.1 kit (Applied Biosystems™) according to the manufacturer's recommendations.The final reaction was submitted through ABI PRISM 3130 Automated Genetic Sequencer (Applied Biosystems™).

Phylogenetic analysis
The sequences were edited and aligned with programs BioEdit v.7.2 and MEGA v.10.0.537 (Tamura et al. 2013), respectively, and compared with other sequences deposited in GenBank (www.ncbi.nlm.nhi.gov)through the Basic Local Alignment Search Tool (BLAST).Phylogenetic trees were constructed with MEGA v.10.0.537 program using the neighbor-joining method and the Kimura two-parameter model (Kimura 1980).A bootstrap of 2000 replicates was used for phylogenetic grouping.Nucleotide similarity was calculated with Geneious v.10.0.7 (Kearse et al. 2012).

Accession numbers
The nucleotide sequence accession numbers are available at www.ncbi.com/nucleotideunder the codes: OM287555, OM287556, and OM287557.

RESULTS
A total of 155 clinical specimens from 47 different species of wild free-living birds were collected (Table 1) from February to October 2019, including 144 fecal specimens and 11 cloacal specimens.
The PAGE test was negative for all the samples tested, with no electrophoretic profile consistent with ARV or PBV.There was a prevalence of 0.6% (1/155) for the ARV S2 gene through RT-PCR.The positive sample belonged to a longwinged-antwren (Myrmotherula longipennis).The prevalence for PBV GI was 1.29% (2/155), both samples belonging to guira-cuckoo (Guira guira).There were no positive samples for PBV GII.
The partial sequences of the PBV RdRp gene of the two strains obtained from our samples were compared with other prototype PBV sequences isolated in Brazil and other countries and deposited in GenBank.Phylogenetic analysis grouped the two strains from this study into PBV GI, however, the sequences were heterogeneously related, grouping divergently (Figure 3).One strain, GI/PBV/Guira-cuckoo/ BRA/UFRA-115/2019, grouped with a sequence from a duck in Australia in 2018, with a bootstrap of 94%.The group was phylogenetically related to other strains isolated from toucan in Brazil and chicken in Brazil and South Korea (bootstrap of 72%).The other strain, GI/PBV/Guira-cuckoo/ BRA/UFRA-114/2019, grouped with a strain isolated from swine in the USA, with a bootstrap of 85%.This cluster was phylogenetically related to other PBV isolated from primates.
Regarding nucleotide identity, the S2 gene sequence showed 51.9% to 86.4% similarity when compared with prototype ARV sequences.The highest similarity was observed with sequences obtained from chicken in Brazil (KY783741.1,KY783743.1,KY783742.1),with values of 86.4,86.0, and 85.8%, respectively.The lowest similarity was obtained for a sequence acquired from brown-eared bulbul (AB914767.1)isolated in Japan (51.9%) (Figure 4).  ) represents a prototype of mammalian reoviruses (MRV) and was used as an outgroup to better understand the phylogenetic relationships between the strains.The phylogenetic tree was constructed using the neighbor-joining method and Kimura two-parameter model, with bootstrap of 2000 replicas to give consistency to the phylogenetic groups.

DISCUSSION
ARV and PBV are included in a group of important enteric viruses.They have been widely reported to infect poultry and wild birds, consequently producing impacts on the economy and wildlife (Silva et al. 2014;Lu et al. 2015;Verma et al. 2015;Assunção et al. 2018;Wang et al. 2019;Duarte-Júnior et al. 2021).The characteristic of the ARV and PBV segmented genome is an important factor that contributed to the rapid dispersion, evolution, and adaptation of these viruses in different hosts (Ganesh et al. 2014).Wild birds are known to be important reservoirs of several infectious agents (Ogasawara et al. 2015).
In the present study, there were no electrophoretic profiles of ARV and PBV in PAGE migration, corroborating previous studies involving wild bird specimens from northern Brazil (Chagas 2018;Guerreiro et al. 2018;Duarte-Júnior et al. 2021).However, previous studies in Brazil and other countries have already characterized electrophoretic profiles of PBV and ARV in chicken (Gallus gallus domesticus, Linnaeus 1758), domestic ducks (Cairina moschata, Linnaeus 1758) and wild birds (Hypsipetes amaurotis, Temminck 1830) (Yun et al. 2013;Silva et al. 2014;Ogasawara et al. 2015).
The high prevalence of PBV in broilers, especially in poultry farming, suggests this pathogen may be associated to confined systems, facilitating viral dispersion through direct among birds (Silva 2012).As our study dealt with different free-ranging bird species, the probability of virus spread among individuals is reduced, even though species live in flocks.
The present study is the first to report the occurrence of ARV in wild free-ranging birds in Brazil.The low prevalence of ARV in our study (0.6%) is in disagreement with previous studies, that reported a prevalence of 33.3% (5/15) for poultry in Egypt (Al-Ebshahy et al. 2019) and 30.2% (58/192) in wild birds in Poland (Styś-Fijoł et al. 2017).In Brazil, two previous surveys that assessed the ARV S2 gene in wild birds fecal found no positive results (Chagas 2018;Guerreiro et al. 2018), but a prevalence of 32.9% (28/85) was detected in fecal specimens of farm poultry from the region of Belém (Silva 2012).
In studies that reported a high prevalence of ARV infection, specimens were from birds with clinical signs of infection, such as arthritis, tenosynovitis, and enteric syndromes, in addition to more severe cases, when the central nervous system is compromised or the bird was dead (Styś-Fijoł et al. 2017).In an analysis of tendons, synovial tissue, and viscera of 311 poultry with clinical simptoms of infection in the USA, specimens were positive for ARV (Lu et al. 2015).In our study, we did not observe any external sign of disease while handling birds, but we did not measure clinical parameters nor analyzed internal organs, therefore we could not formally evaluate the health condition of the captured birds.
PBV strains isolated in different geographic regions and from different hosts have been shown to be phylogenetically closer than strains isolated in the same region (Silva et al. 2014;Verma et al. 2015;Malik et al. 2018;Wille et al. 2018).The similarity of our PBV strains with those of disjunct geographical regions and/or other species corroborates the capacity of these organisms for rapid spread, evolution, and adaptation, which may be related to anthropic dispersion of ancestor sharing strains (Ganesh et al. 2014;Ribeiro et al. 2019).
The two PBV strains in this study were isolated from two individuals of the same species (Guira guira) captured on the same day and location, but showed high genetic diversity, which agrees with the heterogeneous nature of PBV described in the literature (Silva et al. 2014;Verma et al. 2015).Several factors are considered relevant to explain the high genetic diversity of PBV, such as short fragments analyzed, genetic variability, multiple interspecies transmissions, and genetic rearrangement events between segments of different PBV strains (Ganesh et al. 2014;Silva et al. 2014;Ribeiro et al. 2019).
PBV GI has a worldwide distribution pattern and has been reported to infect a variety of hosts such as mammals, birds, reptiles, and even fish (Kumar et al. 2020).The phylogenetic grouping of PBV GI isolated from different hosts indicates these viruses are not species-specific, i.e., they can be transmited from one host to another (Verma et al. 2015;Chagas 2018;Malik et al. 2018).The great genetic similarity observed between PBV strains isolated from different hosts raises increasing concerns about the zoonotic potential of this virus in the context of public health (Ganesh et al. 2014;Kumar et al. 2020).
Among 15 ARV S2 strains isolated from farm poultry in the surroundings of Belém, 13 were more phylogenetically related to each other, while two were more closely related to ACTA AMAZONICA ARV prototypes from poultry around the world (Silva 2012).All 15 strains showed a nucleotide homology of 90.1-100% among them, and 90.9-94.4% with the reference prototypes from other production birds around the world, further demonstrating that the degree of nucleotide similarity that among strains of ARV is independent of geographic proximity.
Contrary to this notion, however, the ARV strain isolated in the present study was more related to ARV strains that have been circulating for some years in poultry in the same region than to strains from other countries.This result suggests that this ARV strain has adapted to the local environment in the Belém region over the years, and that transmission occurs between domestic and wild birds in the region.Commercial poultry farms in the Belém region are normally located in rural environments close to forest fragments, which can facilitate the transmission of pathogens between poultry and wild fauna.
The results of this study further expand the occurrence of ARV and PBV in wild birds, and confirm the occurrence of these viruses in wild birds in the Amazon region.The documented hosts, Guira guira and Myrmotherula longipennis, are not long distance migrants, but their proximity to a large urban center and several poultry farms in the Belém region poses a risk as they may act as possible dispersion agents for these viruses (Pacheco et al. 2021).The presence of these viruses in wild birds inhabiting an environmental protection area with limited human activity suggests ARV and PBV circulation in natural environments.Yet APA Belém is also partially interspersed with several low-income neighborhoods with poor basic sanitation infrastructure, creating conditions for possible avian to human transmission.

CONCLUSIONS
This study recorded, for the first time, the occurrence of ARV in any wild bird species in Brazil, and the occurrence of PBV in Guira guira.The study further confirms the circulation of these viruses in wildlife in close proximity of a large urban center in the Amazon region, prompting additional studies to determine the epidemiology of infectious agents in wild birds in the region, especially concerning segmented genome viruses, for which the processes of transmission, evolution, and adaptation to new environments and hosts occur faster.The molecular characterization and phylogenetic analysis support the notion that the ARV strain detected has been circulating among poultry and wild birds and adapting locally, while the PBV strains were more closely related to geographically disjunct strains, suggesting long distance dispersal through human activities.

Figure 1 .
Figure 1.Location of the APA Belém environmental protection area in the outskirts of Belém (Pará state, Brazil) and the campus of Universidade Federal Rural da Amazônia (UFRA) therein, where the wild birds used in this study were captured.

Figure 2 .
Figure 2. Location of the mist nets used to capture the wild birds used in this study in the campus of Universidade Federal Rural da Amazônia (UFRA) in Belém (Pará state, Brazil).

Figure 3 .
Figure 3. Phylogenetic tree based on partial sequence alignment of the PBV RdRp gene.The sequences of this study are represented in bold.The silhouettes represent zoological families.The numbers next to the nodes indicate bootstrap values >70%.The scale bar is proportional to the phylogenetic distance.The prototype strain GII/PBV/Human/USA/4-GA-91/2000 (AF246940.1)was used as an external group to better understand the phylogenetic relationships between the strains.The phylogenetic tree was constructed using the neighbor-joining method and the Kimura two-parameter model, with bootstrap of 2000 replicas to give consistency to the phylogenetic groups.

Figure 4 .
Figure 4. Phylogenetic tree based on the alignment of partial sequences of the ARV S2 gene.The sequence of the present study is represented in bold.The silhouettes represent zoological families.The numbers next to the nodes indicate boostrap values >70%.The scale bar is proportional to phylogenetic distance.The strain REO/ Bat/SLO/SI-MRV04/2009 (MG457105.1)represents a prototype of mammalian reoviruses (MRV) and was used as an outgroup to better understand the phylogenetic relationships between the strains.The phylogenetic tree was constructed using the neighbor-joining method and Kimura two-parameter model, with bootstrap of 2000 replicas to give consistency to the phylogenetic groups.