Since its initial identification in the late 1970’s, canine parvovirus (CPV) has been evolving, generating new genetic and antigenic variants worldwide. Since its emergence, CPV has remained a common and important cause of morbidity and mortality in young dogs worldwide (Decaro et al., 2009; Decaro et al., 2011; Miranda and Thompson, 2016). The CPV evolution has given rise to three genetic/antigenic variants, CPV-2a, CPV-2b and CPV-2c (Vieira et al., 2008; Decaro and Buonavoglia, 2012). Molecular and epidemiological data indicate that the newest type, CPV-2c, is becoming prevalent in several geographic areas and is often associated with severe disease in adult dogs, including those with complete vaccination record (Decaro and Buonavoglia, 2012). In Brazil, the CPV-2c variant has already been described, especially in southern Brazil (Costa et al., 2005; Castro et al., 2010; Oliveira et al., 2017).
The furthermost important diseases transmitted from dogs to wild carnivores, e.g. rabies, canine distemper and parvovirosis, are named “ The Big Three” due to the strong impact over the health of wild species (Curi et al., 2016). Thus, infectious diseases driven-mortality is one of the major causes of population decline and, in some situations, even extinction of wild mammal carnivores worldwide (Curi et al., 2016). Although evidence on the role of domestic dogs as reservoirs of viral diseases for wild carnivores is becoming a global pattern, few such reports are described in South America (Acosta-Jamett et al., 2015; Curi et al., 2016). In this respect, a pioneer study has recently described the prevalence and risk factors for viral exposure in rural dogs around protected areas of Atlantic forest (Curi et al., 2016). For this purpose, virus isolation, PCR detection and sequencing analysis of isolated CPV were performed in 100 faecal samples collected from dogs living in a rural area belonging to Atlantic forest biome
Samples and virus isolation
Faecal samples were collected per rectum from apparently healthy dogs and stored at −86 °C prior to virus isolation. The studied region is localized 21°12′32″S 50°25′58″W and was divided into 4 different areas (50 Km2). The collection was performed at the same period (one week) corresponding to each area, during 2014 to 2016 and corresponded to 100 different individuals. This study was approved by the ethical committee (CEEA) of Universidade Estadual Paulista “Júlio de Mesquita Filho”. The study was performed under all applicable institutional guidelines for the care and ethical use of animals were followed as recommended by Animal Brazilian Experimentation Committee protocol number 2015/09754. Approximately 2 g of faeces was homogenized in 1 volume of sterile phosphate-buffered solution (PBS), clarified by centrifugation at 2,500 x g for 10 min. The supernatant was filtered through a 0.75 μm filter (Millipore™) and treated with amphotericin and penicillin (concentrated at 100X; Sigma-Aldrich®, St. Louis, MO, USA). AF-72 cells (ATCC, CRL 1542) were cultured in MEM (Sigma-Aldrich®) supplemented with antimycotic/antibiotic 1X solution, 10% foetal calf serum (Sigma-Aldrich®), 2 mM L-glutamine (Sigma-Aldrich®), and non-essential amino acids (100x, Invitrogen®, Life Technologies, Carlsbad, CA, USA). Cultures were incubated at 37 °C in 5% CO2 with 95% humidity. After AF-72 cells reached 80% confluence, 1 ml of faecal preparation was added to 4.7 × 105 cells/ml and the culture supernatant was submitted to three blind passages at 5-day intervals. Inoculated and control cells were monitored under phase-contrast using an Olympus IX–70 microscope for production of cytopathic effect (CPE) (Olympus®, Tokyo, Japan). Approximately 10 fields were analysed in each condition, and photographs were taken at 200 x magnification by using cell Sens™ software (Olympus®). CPV VR-2016™ (ATCC), Cornell strain, which is present in all Brazilian vaccines, were used as controls. Mock-infected AF-72 cells were used as negative controls nd cells inoculated with CPV VR-2016™, strain Cornell were used as positive controls. Other enteric viruses, e.g., canine adenoviruses 1 and 2 (CAV-1, CAV-2) and canine distemper virus (CDV), were not detected in the examined samples by molecular analysis (data not shown).
For detection of the CPV genome, total DNA from infected/uninfected AF-72 cells was extracted using DNAzol™ according to the manufacturer’s instructions (Invitrogen®). An average of 100 ng of genomic DNA was used for PCR as described previously. Polymerase chain reaction (PCR) was performed to amplify a fragment of 583 base pairs (bp) of the VP2 gene, a region surrounding position 426, using primers described previously (Decaro et al., 2009; Mohan Raj et al., 2010). CPV amplicons were purified using the NucleoSpin Extract II kit and sequenced with an ABI PRISM 3100 Genetic Analyser (Applied Biosystems™) using the BigDye Terminator v. 3.1Cycle Sequencing Kit (Applied Biosystems™). Sequences were aligned using BioEdit Sequence Alignment Editor V.126.96.36.199 (Tamura et al., 2007). The evolutionary history was inferred using the Neighbour-Joining method (Saitou and Nei, 1987). The percentages of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree (Hall, 1999). All positions containing gaps and missing data were eliminated from the dataset. Phylogenetic analyses were conducted in MEGA4 (Tamura et al., 2007). A phylogenetic tree was constructed based on amino acid alignments using BLAST analysis (www.blast.ncib.nlm.nih.gov/Blast.cgi), and sequences generated for CPV were submitted to GenBank and assigned.
Results and discussion
The cytopathic effect was characterized by cell rounding and/or partial or total lysis of the monolayers in comparison to control (Fig. 1A, B and C, respectively). To construct the phylogenetic tree, the sequences of four viruses were included. A fragment of 583 bp corresponding to the VP2CPV gene was amplified in 67 out of 100 (67%) analyzed samples. Nucleotide sequencing confirmed the CPV identity. Forty positive samples (59.7%) belonged to animals that had free access to fragmented forest. Interestingly, 33 out of 67 positive samples (49.2%), were obtained from dogs that have free access to forest, being considered an interface to human-wildlife. The amplified DNA fragments were directly sequenced. Sequence analysis showed that 90.2% of the samples carried the amino acid (aa) Glu at the position 426, characteristic of CPV-2c strains, and 65% of these CPV-2c strains showed the substitution. Despite of the fact that CPV-2c was first described in South America in 2007, the high prevalence of CPV-2c strains in Brazil continues nowadays (Pérez et al., 2007; Streek et al., 2009; Calderón et al., 2015; Fontana et al., 2013; Oliveira et al., 2017). Out of the 67 positive samples, 10% were identified as CPV-2a (KP694304) and 7% CPV-2b (KP694306) (Fig. 2). An additional 50 samples were related to CPV-2c sequences (KP694303 and KP694302). Sequence alignment showed that these sequences presented the amino acid substitution (Asp-426 to Glu-426) that characterizes the 2c type (data not shown). The phylogenetic tree was constructed by the use a representative strains which had been characterized as CPV2-c were subjected to full length VP2 amplification and cloning.
Although CVP-2c has been detected in several Brazilian regions, this is the first description of its circulation in rural/wild areas representing human/wildlife interfaces (Costa et al., 2005; Streek et al., 2009; Castro et al., 2010; Fontana et al., 2013; Oliveira et al., 2017). Despite of the availability and use of several commercial vaccines, viral enteritis − including that associated with CPV infection − remains an important problem for Brazilian dogs (Oliveira et al., 2017). In addition, our study showed that CPV isolation is feasible when faeces are processed shortly after collection. This procedure allows for enriching viral preparations for several purposes, including biological studies of viral replication, drug susceptibility and virus-host interactions (Decaro and Buonavoglia, 2012; Parthiban et al., 2011). In a recent study, rural dogs living in households around six Atlantic forests in southeastern Brazil were serologically tested for CPV antibodies and 97% were positive (Curi et al., 2016). Our study provides a confirmatory piece of information about the circulation of CPV in rural areas and reinforce the risk of potential transmission of this virus from domestic dogs to wild carnivores. In fact, it is usual see wild carnivores circulating in urban areas due to forest fragmentation in protected areas in Brazil (Curi et al., 2016). In this sense, increased risk of CPV in rural areas in Australia highlighted a need for vigilance in order to prevent virus spread to wild species (Zourkas et al., 2015). Moreover, in Argentina, a re-emergence of CPV-2a in domestic dogs was observed, corroborating the genetic diversity detected worldwide (Calderón et al., 2015). Accurate genotyping for field CPV strains is important for epidemiological issues. Moreover, evaluation of the serological responses suggests vaccination with CPV-2b would cross-protect against CPV-2a and CPV-2c, as well as against CPV-2 which is now extinct in the field (Wilson et al., 2014).
The findings described herein provide relevant information on the potential role of asymptomatic dogs as reservoirs of CPV in rural areas, mainly close related to wildlife carnivores in protected areas of Brazil.
Author contribution statement
Flavia V. Viieira, Daniel J. Hoffmann: Conceived and designed the experiments; Performed the experiments.
Carolina U.F. Fabri, Katia D.S. Bresciani, Roberto Gameiro, Eduardo F. Flores: Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data.
Tereza C. Cardoso: Conceived and designed the experiments; Analyzed and interpreted the data; Wrote the paper.
This work was supported by the Fundação Amparo à Pesquisa do Estado de São Paulo (Grants 2013/24951-4 and 2012/16715-6). Tereza C. Cardoso and Eduardo F. Flores are recipients of CNPq (Brazilian Council for Research) funding.
Competing interest statement
The authors declare no conflict of interest.
No additional information is available for this paper.