Unveiling the Genetic Landscape of Canine Papillomavirus in the Brazilian Amazon

1. Introduction

Papillomaviruses (PVs) are small, circular, double-stranded DNA viruses known to cause both pre-neoplastic and neoplastic diseases across various species, including humans, cats, horses, and dogs [1,2]. Classified within the Papillomaviridae family, PVs are categorized based on the nucleotide sequence similarity of the L1 gene, biological properties, and phylogenetic tree topology [1,2]. In dogs, most infections with Canis familiaris Papillomavirus (canine papillomavirus - CPV) are asymptomatic [3]. However, CPVs have been implicated in causing conditions such as hyperplastic lesions [4] and, less frequently, neoplastic diseases [5]. Currently, 33 CPV types have been identified and are divided into three genera Lambdapapillomavirus, Taupapillomavirus, and Chipapillomavirus according to the Papillomavirus Episteme (PaVE) database (https://pave.niaid.nih.gov), with some types still unclassified [6].

CPVs are associated with various lesions, including oral and cutaneous papillomatosis, pigmented plaques, and squamous cell carcinomas (SCC) [7,8,9,10,11]. While most canine pigmented plaques are small and localized, cases of extensive and disseminated plaques have been reported, with rare instances progressing to SCC [7]. Specific CPV types, such as CPV16, are believed to have a higher potential for neoplastic transformations, although other types have also been found in lesions undergoing malignant changes [3,7,10,11,12]. In contrast, according to PaVE, over 400 human papillomavirus (HPV) types have been characterized. Despite recent reports of new viral types and co-infections, CPV diversity remains notably low. Factors such as viral evolution, anthropogenic landscape changes, and global climate shifts significantly influence the incidence and geographic distribution of viral agents affecting both animal and human populations.

The Amazon region, with its unparalleled biodiversity and complex ecosystems, highlights the critical need for virology research. Understanding how environmental factors contribute to the emergence and transmission of various viral pathogens is essential. Therefore, there is an urgent need for studies analyzing the genetic diversity of CPVs, particularly on Amazon. Thus, this study aims to identify the papillomavirus types present in oral and cutaneous papillomatous lesions in dogs from the Western Amazon region of Brazil. By expanding our knowledge of CPV diversity and its potential health implications, this research will provide valuable insights into canine virology and inform veterinary practices in biodiversity-rich areas.

2. Materials and Methods

2.1. Clinical Sampling and Histopathological Evaluation

This study included 61 domiciled dogs clinically presenting papillomatous lesions—38 with cutaneous and 23 with oral lesions—originating from Acre (n = 18) and Rondônia (n = 43) states in the Western Brazilian Amazon. To ensure a minimally invasive procedure, lesion samples were collected under local infiltrative anesthesia using 2% lidocaine without vasoconstrictor (Ceva, São Paulo, SP, Brazil). Sterile scalpels and forceps were used for excision. Following collection, each lesion was divided: one fragment was stored at –20 °C for subsequent DNA extraction, while 54 lesion samples, depending on tissue availability, were fixed in 10% buffered formaldehyde for histopathological examination.

Detailed clinical data were recorded for all dogs, including sex, age, breed, anatomical location, and the macroscopic morphology of lesions. Based on visual examination, lesions were classified into six distinct types: (1) cauliflower-like—irregular with a broad base; (2) filiform—thin projections resembling grains of rice; (3) nodular—raised, 1–2 cm in diameter; (4) dome-shaped—small, endophytic, approximately 4 mm; (5) papular—small (≈2 mm) flat lesions; and (6) pigmented plaques—hyperpigmented patches of skin.

To minimize animal suffering, all procedures were carried out in accordance with the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes (2010/63/EU, revised 2010) and in accordance with Colégio Bra-sileiro de Experimentação Animal (COBEA). The project was approved by the Comissão de Ética no Uso de Animais da Universidade Federal do Acre (protocol number 23107.005499/2018-96).

For histopathological processing, formalin-fixed tissues were kept in solution for at least 72 h, processed routinely, and sectioned at 3 μm thickness. The sections were stained with hematoxylin and eosin (H&E) and examined microscopically by a veterinary pathologist. A visual overview of the clinical sampling and histopathological workflow is presented in Figure 1.

2.2. Molecular Analyses: DNA Extraction, Sequencing, and Phylogenetic Inference

Approximately 25 mg of each lesion was used for DNA extraction using the PureLink® Genomic DNA Mini Kit (Invitrogen, Carlsbad, CA, USA), following the manufacturer’s protocol. Extracted DNA was stored at –20 °C.

To detect papillomaviral DNA, partial amplification of the L1 gene was performed using degenerate primers FAP59 and FAP64, as previously described [13]. PCR products were purified using the PureLink® Quick PCR Purification Kit (Invitrogen, Carlsbad, CA, USA) and sequenced using the ABI PRISM 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA) with the BigDye Terminator v3.1 Cycle Sequencing Kit. The resulting data were processed with Data Collection 3 software and converted to FASTA format using Sequence Analysis Software v6 (Applied Biosystems, Foster City, CA, USA). Consensus sequences were assembled using Geneious Prime software, and sequence identity was assessed via BLASTn and BLASTx against GenBank databases.

To obtain the complete viral genomes, rolling circle amplification (RCA) was applied to three samples with higher DNA quality and concentration using the TempliPhi™100 Amplification Kit (GE Healthcare, Burlington, MA, USA) [14,15]. The RCA products were purified, and their concentration and purity were measured using NanoDrop™ and Qubit™ systems (Invitrogen, Carlsbad, CA, USA). DNA libraries were prepared using 50 ng of purified DNA with the Nextera XT Sample Preparation Kit (Illumina, San Diego, CA, USA) and sequenced on an Illumina MiSeq platform using the v2 reagent kit (300 cycles).

Sequence data quality was evaluated with FastQC. Reads were trimmed at the 3′ end using a Phred quality threshold of 20 and assembled into contigs with SPAdes v3.5. Chimeric sequences were excluded based on RDP4 software analysis [16]. Assembled sequences were annotated and analyzed using Geneious Prime, and alignment with known papillomavirus sequences was performed using MAFFT [17]. The final phylogenetic trees were constructed using the maximum likelihood method with the HKY model and 1000 bootstrap replicates in IQ-TREE [18].

3. Results

3.1. Overview of Clinical and Histopathological Findings

Among the 61 dogs evaluated, clinically evident papillomatous lesions were primarily distributed in the oral cavity and limbs. The lesions varied in morphology, including cauliflower-like, nodular, filiform, and pigmented plaque presentations. Table 1 compiles all individual case data, including PCR results (including phylogeny findings), breed, age, sex, lesion location, and detailed macroscopic and histological characterizations.

Histopathological analysis was performed on 54 samples (88.5%), of which 20 (37%) were diagnosed as squamous papilloma (SP). These cases exhibited hallmark features such as epidermal hyperplasia, marked hyperkeratosis, and hypergranulosis (Figure 2a). In some instances, there was fibroblast proliferation within the dermis (Figure 2b), as well as lymphoplasmacytic inflammatory infiltrates, keratohyalin granules, and intranuclear inclusions—consistent with papillomavirus etiology.

3.2. Papillomavirus Detection and Lesion Classification

Of the 61 samples, 30 (49.2%) tested positive for papillomavirus DNA using PCR, including both cutaneous and oral lesions. Among these PCR-positive samples, 11 (36.7%) were also histologically confirmed as SP. Conversely, nine PCR-negative lesions (29%) were histologically consistent with SP, suggesting either low viral load or the presence of unamplified PV types. The overlap and divergence between molecular and histopathological diagnoses are summarized in Figure 3.

Overall, 39 cases were confirmed by either PCR and/or histopathology as consistent with CPV infection. These lesions predominantly exhibited cauliflower-like macroscopic morphology (61.5%), followed by nodular (20.5%), filiform (10.3%), and pigmented plaques (7.7%). Macroscopically, the lesions identified as papillomatosis or CPV-positive were predominantly cauliflower-like, followed by nodular, filiform, and pigmented plaques (Figure 4).

3.3. Histological Analysis of Non-Papillomatous Lesions

Lesions that did not meet the criteria for SP diagnosis (n=24) were histologically classified into various categories. These included sebaceous adenomas, melanocytomas, histiocytomas, chronic otitis, and trichoblastomas, among others. Several samples showed non-specific hyperkeratosis or inflammatory changes, and some were deemed inconclusive or lacking significant histopathological alterations (Table 2).

3.4. Phylogenetic Characterization of CPV

Of the PCR-positive samples, 16 yielded high-quality L1 sequences suitable for phylogenetic analysis. Most sequences showed >99% nucleotide identity with CPV1, the prototype of the Lambdapapillomavirus 2 species (GenBank NC_001619). One notable exception was sample 28AC18BR, which exhibited 99% identity with CPV8, a member of the Chipapillomavirus 3 species, and 98% identity with a Swiss CPV8 strain (GenBank YP_004857848) associated with pigmented plaques.

The phylogenetic tree based on the partial L1 region (Figure 5) confirmed the taxonomic placement of CPV1 and CPV8 among known Canis familiaris papillomaviruses, demonstrating strong bootstrap support for the respective clades.

3.5. Whole-Genome Sequencing of Selected Isolates

Three samples (16AC18BR, 20AC18BR, and 34AC19BR) with high DNA quality were selected for whole-genome sequencing via high-throughput sequencing. The complete genomes ranged from 8,607 to 8,626 bp, and all included the canonical PV gene set: E1, E2, E4, E5, E6, E7, L1, L2, and a URR (upstream regulatory region). Genome coverage depth ranged from 20x to 35x, ensuring high confidence in base calling and genomic structure. All three genomes clustered within Lambdapapillomavirus 2 (CPV1), supporting the partial L1-based findings.

4. Discussion

Papillomavirus infections in dogs most commonly manifest as oral and cutaneous papillomas, both of which are typically recognizable through clinical examination due to their characteristic exophytic, verrucous appearance [19,20]. In this study, 61 lesions with macroscopic features consistent with papillomatosis were collected from domiciled dogs in the Western Brazilian Amazon. The most frequently affected regions were the oral cavity (37.7%) and limbs (14.8%), consistent with the anatomical predilections reported in previous literature [20,25].

Although papillomatosis is classically considered a disease of young dogs with immature immune systems [20], our findings challenge this paradigm. Among the 39 dogs confirmed by PCR and/or histopathology, only eight were ≤ 2 years of age, while 28 animals were older, 10 of which were over 10 years. These results suggest that factors beyond age-related immune status, such as individual immunogenetics, co-infections, or environmental pressures, may influence disease expression. This notion aligns with studies demonstrating that many dogs may harbor papillomavirus DNA asymptomatically, with lesion development being contingent on immune dysregulation [19,21].

The detection of papillomavirus DNA by PCR in 49.2% of cases underscores the challenges inherent to molecular diagnosis in PV infections. Despite using the broad-range FAP59/64 primers, nearly half of the lesions tested negative. This is not unexpected, given that these degenerate primers preferentially amplify a limited spectrum of CPV types (1–5, 7) and may fail to detect divergent or low-copy-number variants [21,22]. Furthermore, the “hit-and-run” oncogenic model provides a compelling explanation for histologically confirmed squamous papillomas (SPs) lacking detectable viral DNA. In this model, the viral genome is required to initiate—but not maintain—neoplastic transformation [23].

Interestingly, of the 20 SPs identified by histopathology, nine were PCR-negative. These cases reinforce the importance of combining histopathological evaluation with molecular tools, as reliance on PCR alone could lead to underdiagnosis. Our data also support previous observations that papillomavirus infections in dogs may follow heterogeneous clinical courses, influenced by complex virus–host–environment interactions [3,10].

To date, 33 distinct CPV types have been described, each demonstrating varying tissue tropism and pathogenic potential [6,9]. In our study, CPV1 was the predominant type detected, aligning with its known association with oral and cutaneous papillomas and its wide global distribution across five continents [24,25,26,28,29,30,31,32,33,34,35]. CPV1-positive lesions in this cohort included nodular, filiform, cauliflower-like, and even pigmented plaque morphologies—suggesting phenotypic versatility of this genotype. This reinforces the concept that CPV1 remains the most ubiquitous and clinically relevant papillomavirus in dogs.

Notably, one sample (28AC18BR) revealed CPV8, a virus traditionally associated with pigmented plaques and cutaneous squamous cell carcinoma, particularly in Chipapillomavirus infections [9,26,27]. However, in this case, CPV8 was detected in a filiform SP, which closely resembled classical benign papillomas rather than pigmented plaques. This finding raises important questions regarding the pathogenic spectrum of CPV8 and supports its potential to induce diverse lesion types. Its high sequence identity with a CPV8 strain from Switzerland [YP_004857848] further suggests that this genotype is conserved across distant geographic regions, despite differences in lesion phenotype.

The histological features of confirmed SPs in this study—epidermal hyperplasia, marked hyperkeratosis, fibroblast proliferation, and eosinophilic intranuclear inclusions—were consistent with those described in the literature [28,29,30,31]. Lesions were most commonly localized to the lips, oral mucosa, and head, a distribution pattern also reported in studies from southern and northeastern Brazil [28,29].

The sequencing of partial L1 genes from 16 PCR-positive samples revealed strong identity (>99%) with CPV1, and the three complete genomes obtained confirmed their classification within the Lambdapapillomavirus 2 species. The low genetic divergence among these CPV1 isolates—even across geographically distinct Brazilian states—highlights the remarkable genomic stability of this virus. These findings are in agreement with previous molecular surveys conducted in South America [28,29], North America [32,33], Asia [11], Africa [34], and Europe [35,36].

The detection of CPV1 and CPV8 in dogs from the Western Amazon represents the first molecular and phylogenetic characterization of canine papillomaviruses in this biome. Despite the Amazon's unique ecological complexity, our CPV1 strains were nearly identical to isolates from southern and northeastern Brazil [28,29,37], suggesting either broad viral dissemination or limited mutational drift. This genetic homogeneity raises intriguing questions about the evolutionary constraints and selective pressures acting on CPVs in different environments.

While bovine papillomaviruses (BPVs) have been widely studied and are known for their oncogenic potential and capacity to induce malignant lesions, even in cattle from the amazon region [38], studies on CPVs remain limited—especially in remote, biodiverse areas such as the Amazon. Unlike BPVs, CPVs are typically associated with benign cutaneous and mucosal lesions in dogs; however, the detection of CPV8 in an atypical lesion phenotype in this study challenges existing assumptions and expands the known pathogenic spectrum of these viruses.

Given the Amazon's exceptional biodiversity and ecological distinctiveness, further studies are warranted to assess the potential presence of unique or emerging CPV genotypes. Such research could elucidate virus–host dynamics under environmental conditions distinct from urbanized or temperate zones. It may also reveal whether co-infections, environmental stressors, or host genetic variability influence viral tropism, pathogenicity, or oncogenic potential in this biome.

5. Conclusions

This study reinforces the value of integrating molecular and histopathological approaches to improve the accuracy of CPV diagnosis. Our findings expand the epidemiological landscape of CPVs by documenting the presence of CPV1 and CPV8 in dogs from the Brazilian Amazon and underscore the need for continued viral surveillance in underrepresented regions. By illuminating the genetic stability of CPV1 and detecting CPV8 in an unusual lesion phenotype, this study provides new insights into papillomavirus diversity and evolution in companion animals. Understanding these dynamics is crucial not only for veterinary virology but also for comparative oncology and One Health strategies.