Canine Adenovirus

Overview and Taxonomy of Canine Adenovirus

Canine adenoviruses (CAdVs) are globally significant, non-enveloped, linear double-stranded DNA viruses belonging to the genus Mastadenovirus within the family Adenoviridae [7, 11, 12]. These pathogens are responsible for a spectrum of clinical diseases in domestic and wild canids, and their importance extends beyond veterinary medicine into comparative oncology, gene therapy, and vaccine vector development. The taxonomic framework distinguishing the two principal serotypes, Canine adenovirus type 1 (CAdV-1) and Canine adenovirus type 2 (CAdV-2), is founded upon differences in genomic architecture, virulence, tissue tropism, hemagglutination patterns, and clinical manifestations [11, 42]. This dichotomy, while historically defined by serological neutralization assays, is now underpinned by comprehensive molecular characterization that reveals a more nuanced evolutionary and epidemiological landscape.

Taxonomic Classification and Serotypic Divergence

The genus Mastadenovirus encompasses a wide array of mammalian adenoviruses, and CAdV-1 and CAdV-2 represent two distinct serotypes within this group. Despite their antigenic cross-reactivity, a feature exploited in vaccination strategies where CAdV-2 vaccines confer protection against CAdV-1, they differ markedly in pathogenicity and organ tropism [11, 29]. CAdV-1 is the etiological agent of infectious canine hepatitis (ICH), a systemic, often fatal disease characterized by hepatic necrosis, renal lesions, and widespread vascular endothelial damage [7, 9, 33]. In contrast, CAdV-2 is primarily associated with infectious tracheobronchitis (ITB, "kennel cough") and, to a lesser extent, viral enteritis, particularly in puppies and co-infected animals [2, 4, 11, 35]. The genomic basis for these differences is substantial; comparative analyses reveal that while the two serotypes share the same overall genome structure, capsid diameter, and morphological features, they diverge significantly in specific coding regions, including the E3, fiber, hexon, and penton genes [11, 27, 42].

Whole-genome sequencing has become instrumental in refining the taxonomic relationships among CAdV strains. For instance, the complete genome sequence of a CAdV-2 strain from Türkiye (OQ596341) demonstrated nucleotide similarity exceeding 99.04% with other global CAdV-2 genomes, yet harbored unique point mutations in genes such as E1A, Pol, pTP, and E3 ORFA [1]. These mutations, including the first report of an E250K substitution in the E3 ORFA gene in Türkiye, highlight ongoing viral microevolution that may influence tropism and host interactions [1]. Similarly, the CAdV-1 F1301 strain isolated from a fox with encephalitis has a complete genome of 30,535 bp, and its phylogenetic positioning reveals close relationships with strains from Norwegian Arctic foxes and red foxes, underscoring the genetic continuity between wild and domestic canid reservoirs [5, 30].

Genomic Organization and Structural Proteins

The CAdV genome is a linear dsDNA molecule of approximately 30–32 kb, flanked by inverted terminal repeats (ITRs) and organized into early (E1A, E1B, E2, E3, E4) and late (L1–L5) transcription units [1, 11]. The viral capsid is composed of three major structural proteins: hexon, penton base, and fiber. The hexon protein is the primary target for neutralizing antibodies and forms the bulk of the capsid surface. Monoclonal antibody mapping has identified two linear neutralizing epitopes within the hexon, 634RIKQRETPAL643 and 736PESYKDRMYS745, which are highly conserved across all CAdV isolates and represent promising targets for epitope-based vaccines and diagnostic assays [22, 32]. The penton base mediates internalization via integrin interactions, while the fiber protein is a key determinant of receptor binding. Unlike human adenovirus type 5 (HAdV-5), which uses the coxsackievirus and adenovirus receptor (CAR) as its primary attachment receptor, CAdV-2 exhibits a CAR-independent transduction mechanism, capable of infecting cells with minimal or undetectable CAR expression [17]. This unique tropism expands its utility as a gene therapy vector for neuronal and respiratory epithelial cells, where CAR levels are often low [17, 24, 43].

The fiber protein itself shows substantial genetic variation between CAdV-1 and CAdV-2, and even among circulating CAdV-2 strains. Bioinformatics analysis of the CAdV-1-JL2021 fiber revealed six glycosylation sites and 107 phosphorylation sites, with a predicted B-cell linear epitope "VATTSPTLTFAYPLIKNNNH" and a shared MHC-I binding peptide "KLGVKPTTY" present in both serotypes [14]. The fiber’s head domain, in particular, is a hotspot for amino acid substitutions that can alter receptor affinity and antigenicity [27]. Such structural differences, while subtle, have profound implications for viral entry, host range, and the design of recombinant vectors for vaccine delivery or oncolytic therapy.

Genetic Diversity and Global Phylogenetic Patterns

Molecular epidemiological studies have delineated multiple phylogenetic subgroups within CAdV-2, reflecting both geographic segregation and evolutionary pressure. Early classification described two major groups: an America-Europe subgroup and an Asia subgroup [27, 35]. However, subsequent analyses have refined this schema. For instance, Turkish and Chinese CAdV-2 isolates form a distinct China-Turkey subgroup based on E3 gene sequences, differing from the America-Europe cluster by nine amino acid substitutions [35]. More recently, Indian CAdV-2 isolates have been shown to constitute a novel third group, characterized by a unique "G" nucleotide insertion at position 1077 of the E3 gene. This frameshift mutation results in an extension of the E3 polypeptide by 11 amino acids at its C-terminal end, a feature not observed in any other global isolate [21, 23]. This insertion was first identified in 2020 and subsequently corroborated in additional Indian strains, suggesting endemic circulation of a distinct CAdV-2 lineage on the Indian subcontinent [21, 25, 28].

Recombination is a recognized driver of adenovirus evolution, and CAdV is no exception. Analysis of the Turkish CAdV-2 genome revealed two recombination events, and selective pressure signatures have been detected in the 100K protein of CAdV-1 strains from mink [1, 5]. Furthermore, the 100K protein, which is not exposed on the viral surface and can therefore serve as a marker to differentiate vaccine from wild-type infection, exhibits evidence of recombination and negative selection, emphasizing the dynamic nature of the CAdV genome [5]. The E3 gene region, in particular, is a hotbed of genetic variation, as it encodes proteins involved in immune evasion and host modulation. Differences in E3 ORFA sequences have been used to trace transmission patterns and identify region-specific variants [23, 25, 35].

Host Range, Tissue Tropism, and Cross-Species Transmission

CAdV-1 and CAdV-2 exhibit overlapping yet distinct host ranges. Both serotypes infect domestic dogs (Canis lupus familiaris), but CAdV-1 is particularly virulent in wild canids and has been documented in a wide array of species including red foxes (Vulpes vulpes), Arctic foxes (Vulpes lagopus), wolves (Canis lupus), raccoon dogs (Nyctereutes procyonoides), and even non-canid carnivores such as the European brown bear (Ursus arctos arctos) and American black bear (Ursus americanus) [8, 18, 31, 33, 36, 40]. Serosurveys in free-ranging wolves from northern Canada revealed a 1% prevalence of CAdV-1 DNA, while studies in Norwegian Arctic foxes documented seroprevalence rates increasing from 25–40% in the late 1990s to 68% by 2002–2003, indicating enzootic circulation within these populations [8, 36]. Similarly, 14% of free-ranging wolves in Asturias, Spain tested positive for CAdV-1 DNA, with prevalence highest in juveniles (<2 years old) and during the years 2010–2011, coinciding with fatal infections in endangered Cantabrian brown bears [18, 33].

CAdV-2, while primarily a pathogen of domestic dogs, has also been detected in wild raccoon dogs in Korea, where isolated strains showed over 99% nucleotide identity with the Toronto A26/61 vaccine strain, suggesting spillover from vaccinated companion animals [10]. The oncological and therapeutic implications of CAdV-2’s broad in vitro tropism are substantial; it can infect canine, swine, and human cells, including HeLa (human cervical cancer) cells, although replication efficiency declines with serial passage in human lines [3]. This property positions CAdV-2 as a promising candidate for oncolytic virotherapy and gene delivery to the central nervous system, where its ability to undergo retrograde axonal transport and transduce projection neurons has been extensively documented in rodent and non-human primate models [15, 20, 24, 34, 44].

Diagnostic and Taxonomic Implications

The accurate differentiation of CAdV-1 and CAdV-2 is critical for clinical diagnosis, epidemiological surveillance, and vaccine deployment. Traditional methods such as virus neutralization and hemagglutination inhibition remain valuable, but molecular assays offer superior sensitivity and specificity. Real-time PCR assays, including SYBR Green-based melting curve analysis and multiplex RT-qPCR platforms, now enable simultaneous detection and discrimination of both serotypes from clinical samples [4, 42]. For example, a quadruplex RT-qPCR targeting the CAdV-2 hexon gene achieved a detection limit of 10² copies/reaction, with a positivity rate of 2.84% among 1,688 clinical samples from Chinese pet hospitals [4]. Point-of-care immunochromatographic strips incorporating monoclonal antibodies against the hexon protein (e.g., 2C1 and 7D7) provide rapid visual detection of both CAdV serotypes, with a limit of detection of approximately 2.0 × 10² TCID₅₀/mL [32]. Similarly, CRISPR-Cas13a-based SHERLOCK technology combined with recombinase-aided amplification (RAA) allows for naked-eye detection of CAdV-2 within 90 minutes, achieving a sensitivity of 10² copies/µL and a 95% concordance with conventional PCR [2]. Such tools are indispensable for field surveillance in resource-limited settings and for monitoring the emergence of new genetic variants.

From a taxonomic perspective, the accumulation of whole-genome sequences, now numbering over 20 complete CAdV genomes in GenBank, has enabled robust phylogenetic analyses that challenge simplistic serotype classifications [1, 27, 38]. The World Organisation for Animal Health (WOAH) recognizes ICH as a notifiable disease, and the genetic characterization of CAdV-1 and CAdV-2 is essential for understanding its epidemiology, particularly in wildlife where clinical signs may be absent but viral shedding occurs [31, 40]. Red foxes, for instance, can harbor CAdV-1 in the kidney, liver, spleen, brain, and lung without overt histopathological lesions, suggesting a carrier state that facilitates environmental dissemination [31, 40]. The detection of CAdV-1 in the urine of inapparently infected red foxes underscores the potential for indirect transmission via contaminated water or fomites [40].

The Role of CAdV in Vaccine Development and Vector Technology

The taxonomic distinction between CAdV-1 and CAdV-2 has practical ramifications for vaccination. Modified live CAdV-2 vaccines are widely used to protect against both serotypes, capitalizing on antigenic cross-reactivity while avoiding the adverse effects, such as the "blue eye" phenomenon (corneal edema and keratouveitis), historically associated with some CAdV-1 vaccines [6, 11]. Despite decades of vaccination, cases of ICH and keratouveitis still occur, and serological evidence indicates that contemporaneous wild-type CAdV-1 infection may be responsible for a proportion of these cases [6, 13]. Immunity surveys across diverse regions, from Ecuador to South Korea, reveal variable seroprevalence; for example, 77.8% of Korean dogs had virus-neutralizing antibodies against CAdV-2, but only 44.8% exhibited protective titers, highlighting gaps in vaccine-induced immunity [19, 26]. In Zimbabwe, only 13% of unvaccinated communal dogs had detectable CAdV antibodies, placing wildlife populations at risk [46].

Beyond immunization, CAdV-2 has emerged as a preferred vector for gene therapy and recombinant vaccine delivery due to its low seroprevalence in humans, efficient transduction of respiratory epithelial and neuronal cells, and capacity for long-term transgene expression [16, 24, 37, 39, 43]. Helper-dependent (HD) CAdV-2 vectors, devoid of all viral coding sequences, offer a 30-kb cloning capacity and minimal immunogenicity, making them suitable for treating neurodegenerative disorders and corneal clouding in mucopolysaccharidosis VII [43, 45]. Oncolytic CAdV-2 constructs, such as ICOCAV15, have shown safety and partial efficacy in canine carcinomas, with evidence of immune cell infiltration and stable disease in a majority of treated dogs [41]. The taxonomic knowledge of CAdV, its genome structure, recombination potential, and receptor usage, directly informs the rational design of these vectors, ensuring that modifications enhance safety and efficacy without compromising genetic stability.

Molecular Pathogenesis of Canine Adenovirus

Genomic Architecture and Molecular Determinants of Serotype-Specific Pathogenesis

The molecular pathogenesis of canine adenovirus (CAdV) is fundamentally anchored in the distinct genomic configurations of its two serotypes, CAdV-1 and CAdV-2, which, despite sharing a common Mastadenovirus ancestry and over 70% genomic homology, orchestrate profoundly different disease phenotypes through subtle but critical genetic variations [11]. The CAdV genome, a linear double-stranded DNA molecule approximately 30–31 kb in length, is organized into early (E1A, E1B, E2, E3, E4) and late (L1–L5) transcriptional units, with the E3 region serving as a primary hotspot for serotype-specific divergence and host-range determination [1, 11]. Complete genome characterization of contemporary CAdV-2 strains from Türkiye has revealed novel mutations across multiple genes, including H34Y in E1A, P55A in E1B 55K, and K679R, V934I, and K989N in the DNA polymerase gene, that may subtly alter replication kinetics and host-cell interaction dynamics [1]. These mutations, while maintaining overall nucleotide similarity exceeding 99% with reference genomes, underscore the ongoing molecular evolution of CAdV and its potential impact on viral fitness and pathogenesis. The E3 gene, in particular, exhibits remarkable genetic plasticity; Indian CAdV-2 isolates harbor a unique guanine nucleotide insertion at position 1077, resulting in a frameshift that extends the E3 polypeptide by eleven amino acids at its C-terminal end, a molecular signature that defines a novel phylogenetic group and may influence the protein's immunomodulatory functions within infected cells [21, 23].

Molecular Determinants of Cellular Tropism and Entry

The initial step in CAdV pathogenesis, viral attachment and entry, is mediated by the fiber protein, a homotrimeric projection that extends from the penton base at each vertex of the icosahedral capsid. Contrary to long-held assumptions that CAdV-2 utilizes the coxsackievirus and adenovirus receptor (CAR) in a manner analogous to human adenovirus type 5 (HAdV-5), rigorous comparative infection studies have demonstrated that CAdV-2 transduction efficiency does not correlate with CAR expression levels across a panel of canine and human cell lines [17]. This finding represents a paradigm shift in understanding CAdV-2 tropism: while CAR is indeed expressed on the surface of many target cells, including respiratory epithelium, hepatocytes, and neurons, CAdV-2 can productively infect cells with minimal or undetectable CAR expression, suggesting the existence of alternative, yet-to-be-identified primary receptors that govern its broader host range [17]. In contrast, CAdV-1 fiber protein exhibits a distinct tertiary structure with 6 glycosylation sites and 107 phosphorylation sites, and while it shares the same putative CAR-binding motif, computational modeling predicts conformational differences in the fiber knob that may contribute to the differential tissue tropism observed between the two serotypes [14]. Specifically, CAdV-1 demonstrates a marked predilection for vascular endothelium, hepatocytes, and renal parenchyma, whereas CAdV-2 preferentially targets respiratory epithelium and, notably, neuronal cells, the latter property being exploited extensively for retrograde tracing and gene therapy applications in neuroscience research [14, 24, 43]. The identification of the predicted MHC-I binding peptide KLGVKPTTY, conserved across both CAdV-1 and CAdV-2 fiber proteins, provides a molecular target for understanding T-cell-mediated immune recognition during infection [14].

The Pathogenic Cascade: From Cell Death to Systemic Disease

At the molecular level, CAdV-1 pathogenesis is characterized by a devastating induction of programmed cell death pathways that transcend classical apoptosis. Groundbreaking investigations have revealed that CAdV-1 triggers PANoptosis, a unique, multifaceted inflammatory cell death program integrating pyroptosis, apoptosis, and necroptosis, in canine splenocytes through activation of the absent in melanoma 2 (AIM2) inflammasome [48]. Mechanistically, the CAdV-1 double-stranded DNA genome, released into the cytoplasm during viral replication, serves as a direct ligand for AIM2, a cytosolic pattern recognition receptor that assembles into an inflammasome complex with apoptosis-associated speck-like protein containing a CARD (ASC) and pro-caspase-1. This molecular platform catalyzes the cleavage and activation of caspase-1, which in turn processes pro-interleukin-1β and pro-interleukin-18 into their bioactive forms, while simultaneously cleaving gasdermin D to generate membrane pores that drive pyroptotic cell death [48]. Remarkably, CAdV-1 selectively upregulates AIM2 expression in immune organs, spleen, tonsils, and lymph nodes, rather than in the primary target organs of liver and kidney, suggesting that the virus orchestrates a targeted immunosuppressive strategy by eliminating T and B lymphocytes in lymphoid tissues, thereby blunting the adaptive immune response and facilitating systemic dissemination [48]. This PANoptotic mechanism explains the profound lymphopenia, hemorrhagic lymphadenitis, and splenic necrosis observed in fatal cases of infectious canine hepatitis (ICH). Concurrently, CAdV-1 infection induces transcriptional reprogramming of infected cells, with transcriptomic analyses of CAdV-1-infected MDCK cells revealing differential expression of 592 long non-coding RNAs and 11,215 microRNAs that converge on the PI3K-AKT, Wnt, herpes simplex, hepatitis C, and Epstein–Barr virus infection pathways, indicating that CAdV-1 hijacks multiple host signaling cascades to create a permissive environment for replication [30].

Molecular Basis of Tissue Tropism and Neuropathogenesis

The capacity of CAdV to infect and cause pathology in diverse organ systems is governed by molecular interactions at the level of viral gene expression and host restriction. CAdV-1, the archetypal cause of ICH, exhibits a biphasic pathogenesis: initial replication in tonsillar and lymph node macrophages is followed by viremic dissemination to the liver, kidney, spleen, and, critically, the central nervous system (CNS) [38, 51]. The neurological manifestations of CAdV-1, including circling, ataxia, vocalization, and obtundation, are mediated by viral-induced vasculitis and endothelial cell necrosis within the brain parenchyma, with histopathological hallmarks including intranuclear inclusion bodies in endothelial cells and hepatocytes [51, 53]. This neurotropism is not exclusive to CAdV-1; a paradigm-shifting report documented CAdV-2 infection in a vaccinated dog presenting with both respiratory and neurological signs, confirmed by virus isolation from rectal swabs and phylogenetic characterization of the E3 gene [21]. The detection of CAdV-2 in CNS tissues challenges the conventional dichotomy that restricts CAdV-2 pathogenesis to the respiratory tract, suggesting that under certain conditions, possibly related to immune status or viral dose, CAdV-2 can cross the blood-brain barrier or access neural tissue via retrograde axonal transport along the olfactory or trigeminal pathways [21]. This property is molecularly encoded in the CAdV-2 capsid; indeed, replication-defective CAV-2 vectors are routinely employed for retrograde transduction of projection neurons in rodent and primate models, with demonstrated efficiency in labeling locus coeruleus noradrenergic neurons, striatal cholinergic interneurons, and craniofacial motoneurons following intramuscular injection [20, 24, 34, 50]. The molecular determinants of this neurotropism likely reside in the fiber knob's interaction with neuronal surface receptors distinct from CAR, enabling efficient entry into presynaptic terminals and microtubule-dependent retrograde transport to the soma [17, 24, 43]. In the context of natural infection, this neuroinvasive capacity contributes to the pathogenesis of "blue eye" (corneal edema) through immune-mediated endothelial damage, and in fatal cases, to meningoencephalitis with intranuclear inclusion bodies in glial cells and neurons [6, 51, 52].

Immunomodulation and the E3 Region: A Molecular Arms Race

The E3 transcription unit of CAdV encodes a repertoire of immunomodulatory proteins that are dispensable for viral replication in vitro but critical for subverting host antiviral defenses in vivo. The E3 region exhibits the highest genetic variability between CAdV-1 and CAdV-2, and this divergence underpins their distinct pathogenic profiles. Comparative genomic analyses have identified a unique substitution (W126S) in the E3 protein of a CAdV-1 strain from India, while Turkish CAdV-2 strains harbor four distinct mutations in the E3 ORFA gene (T14A, E250K, D287N, I293T), with the E250K substitution reported for the first time in Türkiye [1, 7]. The functional consequences of these mutations are beginning to emerge: E3 proteins, particularly those of the ORFA family, are known to inhibit tumor necrosis factor (TNF)-mediated apoptosis and downregulate major histocompatibility complex (MHC) class I surface expression, thereby protecting infected cells from cytotoxic T lymphocyte (CTL) recognition and killing. The frameshift insertion in the E3 gene of Indian CAdV-2 isolates, which extends the open reading frame by eleven amino acids, may alter the subcellular localization or interaction partners of the E3 protein, potentially enhancing its ability to modulate the host immune response [21, 23]. This molecular arms race extends to the 100K protein, a non-structural late protein that is not exposed on the viral surface at any stage and therefore serves as an ideal marker for differentiating vaccine-induced immunity from natural infection. Bioinformatics analyses of the 100K protein from CAdV-1 isolates of mink origin reveal it to be an unstable, hydrophobic protein with 60 dominant B-cell epitopes, 9 MHC-I and MHC-II binding peptides, and evidence of selective pressure and recombination, indicating ongoing molecular adaptation to host immune environments [5]. The 100K protein's primary binding partners, the DNA-binding protein (DBP) and the 33K protein, suggest it plays a scaffolding role in the assembly of viral replication compartments, and its subcellular localization to the endoplasmic reticulum and mitochondria may indicate additional functions in modulating host cell stress responses and apoptosis [5].

Molecular Epidemiology and Host Range Determinants

The molecular pathogenesis of CAdV cannot be fully appreciated without understanding its remarkable capacity to breach species barriers and establish infection in a wide range of carnivores, bears, and even non-carnivore species. Serological surveys have detected anti-CAdV-2 antibodies in cattle (85.0%), horses (35.0%), sows (48.6%), and even cats (2.8%), indicating widespread exposure and suggesting that the molecular determinants of host range, primarily the fiber-receptor interaction and the ability to replicate in heterologous cells, are more permissive than previously appreciated [53]. The isolation of CAdV-2 from wild raccoon dogs (Nyctereutes procyonoides) in Korea, with full-genome sequences showing >99% identity to the vaccine strain Toronto A26/61, provides molecular evidence for the spillover of vaccine-derived viruses into wildlife populations, raising concerns about vaccination safety and the potential for reversion to virulence in non-target hosts [10]. Similarly, CAdV-1 has been identified as a cause of fatal hepatitis in free-ranging European brown bears (Ursus arctos arctos) in the Cantabrian Mountains of Spain, with histopathological lesions including petechial hemorrhages, friable livers, and basophilic intranuclear inclusions in hepatocytes and Küpffer cells, and wolves (Canis lupus) in the same region serve as sentinels and likely reservoirs, with 14% of sampled wolves harboring CAdV-1 DNA in spleen samples [18, 33]. The molecular mechanisms enabling cross-species transmission likely involve the conservation of the coxsackievirus and adenovirus receptor across mammalian species, combined with the ability of CAdV to replicate in a broad range of cell types, including those of swine, canine, and human origin, as demonstrated by the CAV-HN45 strain's efficient infection of HeLa cells and selective infectivity toward human cervical cancer cells [3]. This zoonotic potential, while not resulting in clinical disease in humans, has profound implications for the use of CAdV-2 as a vaccine vector and oncolytic agent, where pre-existing immunity or the potential for recombination with human adenoviruses must be carefully considered [3, 16, 49].

Molecular Signatures of Latency, Persistence, and Reactivation

A critical but often overlooked aspect of CAdV molecular pathogenesis is its ability to establish persistent infections in the urinary tract, tonsils, and lymphoid tissues, with intermittent shedding in urine and feces serving as a source of environmental contamination and transmission. Subclinically infected red foxes (Vulpes vulpes) in the United Kingdom and continental Europe harbor CAdV-1 DNA in multiple tissues, including liver, kidney, spleen, brain, and lung, with 18.8% of foxes exhibiting inapparent infections detected by nested PCR, and importantly, viral DNA was detected in the urine of three foxes, confirming the role of persistent renal infection in viral dissemination [31, 40]. The molecular basis for persistence likely involves the modulation of host cell cycle and apoptosis pathways by early region proteins: E1A proteins bind to retinoblastoma (Rb) family members to drive cells into S phase, creating a favorable environment for viral DNA replication, while E1B 55K and E4 ORF6 proteins cooperate to degrade p53 and inhibit apoptosis, allowing infected cells to survive and maintain viral genomes in a latent state [1, 11]. The detection of CAdV-1 in wolves, foxes, and bears across multiple European countries, often with no histopathological evidence of disease, suggests that these species serve as asymptomatic reservoirs, with stress, co-infection with other pathogens (such as canine distemper virus or Neospora caninum), or immunosuppression triggering reactivation and clinical disease [8, 9, 31, 40, 52]. Co-infection studies have revealed that up to 29.5% of dogs with parvoviral enteritis are simultaneously infected with CAdV-1, CAdV-2, or canine circovirus, and such polymicrobial interactions may potentiate pathogenesis through synergistic suppression of innate immune responses, disruption of mucosal barriers, or altered viral replication kinetics [13, 52]. The molecular interplay between CAdV and other pathogens, particularly the enhancement of CAdV replication by CDK4/6 inhibitors in cancer cells, and the induction of interferon-stimulated genes by adenovirus-vectored vaccines, highlights the complex networks of host-pathogen and pathogen-pathogen interactions that shape the outcome of CAdV infection [47, 49].

The Hexon Protein: A Nexus of Neutralization and Immune Evasion

The hexon protein, the major capsid component, is the primary target of neutralizing antibodies and a key determinant of ser

Epidemiology and Transmission of Canine Adenovirus

Global Prevalence and Geographic Distribution

Canine adenovirus (CAdV) exists as two distinct serotypes, CAdV-1, the etiological agent of infectious canine hepatitis (ICH), and CAdV-2, primarily associated with infectious tracheobronchitis (ITB) and canine infectious respiratory disease (CIRD), and exhibits a truly global distribution, with documented circulation across all continents where canid populations exist [11, 12]. The epidemiological landscape of CAdV infection is complex, shaped by vaccination practices, wildlife reservoirs, and viral evolutionary dynamics. Despite widespread vaccination programs in domestic dog populations, compelling evidence indicates that CAdV-1 and CAdV-2 continue to circulate endemically in both domestic and wild canid populations, with prevalence rates varying dramatically by geographic region, host species, and diagnostic methodology employed [8, 9, 27].

In domestic dog populations, prevalence estimates for CAdV-2 have been reported across numerous countries. A comprehensive study from central China (2017–2019) utilizing PCR detection of CAdV-2 in rectal swabs from diarrheic dogs revealed a positivity rate of 2.84% (48/1688) in Guangxi province [4], while a separate investigation in Henan, Hubei, and Jiangsu provinces identified persistent circulation of CAdV-2 among pet dogs with diarrhea [27]. In Turkey, molecular surveillance of shelter dogs with respiratory signs detected CAdV-2 nucleic acid in 2.5% (4/155) of nasal swab samples [35], and a broader study of diarrheic dogs found no CAdV-1 or CAdV-2 positivity among 62 animals tested [54], suggesting significant temporal and geographic variation. In India, a study of 302 dogs with gastroenteritis identified CAdV-2 in 4.9% of samples [23], while another investigation of 216 faecal samples from northern India reported CAdV-1 in 5.56% of dogs and co-infections with canine parvovirus-2 (CPV-2) in 12.04% [28]. A separate Indian study of 26 nasal swabs from dogs with respiratory and gastrointestinal signs detected CAdV-2 in only one sample (3.8%) [25]. In southern Italy, CAdV-1 was detected in 2.1% (6/291) of deceased stray dogs suspected of infectious gastrointestinal disease [9], and a retrospective analysis of 95 dogs with parvoviral enteritis in Italy (1995–2017) revealed CAdV-2 DNA in 20% of samples and CAdV-1 in 2.1% [13]. In Korea, analysis of 3,179 dog samples from 2000–2017 confirmed canine adenovirus infection in 38 animals (1.2%) [53], while serological surveillance from 2019–2022 demonstrated an overall CAV-2 seropositivity rate of 77.8% among 400 dogs, though only 44.8% exhibited protective antibody titers [19]. In the United Kingdom, a study of red foxes found 64.4% seropositivity for canine adenovirus by ELISA [40], and in Ecuador, 60% of 154 community dogs had detectable antibodies against CAV [26]. In Zimbabwe, only 13% of 225 unvaccinated communal dogs had antibodies to CAV [46], highlighting the stark contrast between vaccinated and unvaccinated populations.

Host Range and Interspecies Transmission

The host range of CAdV extends far beyond domestic dogs (Canis lupus familiaris), encompassing a remarkable diversity of wild carnivores and even non-carnivore species, a factor of critical epidemiological significance. CAdV-1 has been documented in gray wolves (Canis lupus) across Europe and North America [8, 18, 38, 56], red foxes (Vulpes vulpes) in the United Kingdom, Italy, Germany, and Norway [31, 36, 40], Arctic foxes (Vulpes lagopus) in Svalbard [36], raccoon dogs (Nyctereutes procyonoides) in Korea [10], and European brown bears (Ursus arctos arctos) in Spain [18, 33]. Perhaps most alarmingly, CAdV-1 has been identified as a cause of fatal infectious hepatitis in free-ranging Cantabrian brown bears, an endangered population, with three confirmed deaths between 1998 and 2018 [33]. This finding underscores the potential for CAdV-1 to act as a conservation threat to vulnerable wildlife populations. Additionally, CAdV-1 has been isolated from mink in China, with phylogenetic analysis revealing that the mink-derived strains cluster separately from canine strains, suggesting host-specific adaptation [5]. Skunk adenovirus-1, a related pathogen, has been identified in gray foxes co-infected with canine distemper virus and Listeria monocytogenes [58], further illustrating the complex multi-host ecology of adenoviruses in wildlife.

CAdV-2, while primarily a respiratory pathogen of dogs, has also demonstrated a broad host range. In Korea, CAdV-2 was detected in wild raccoon dogs, with the isolated strains showing high genetic similarity to the vaccine strain Toronto A26/61, suggesting transmission from vaccinated domestic dogs to wildlife [10]. Serological surveys in Korea detected anti-CAV-2 antibodies in 51.3% of raccoon dogs, 85% of cattle, 48.6% of sows, 35% of horses, and 2.8% of cats [53], indicating that CAdV-2 or antigenically related viruses can infect a wide array of mammalian species. The detection of CAdV-2 antibodies in cattle and horses is particularly noteworthy, as it raises questions about the role of these species as incidental hosts or potential reservoirs. Experimental studies have demonstrated that CAdV-2 can efficiently infect cells of swine, canine, and human origin, including HeLa cells, though replication capacity declines after serial passage in human cells [3]. This finding has significant implications for the use of CAdV-2 as a gene therapy vector and for understanding potential zoonotic risks, although no evidence of natural human infection exists.

Transmission Dynamics and Risk Factors

The transmission of CAdV occurs through multiple routes, with distinct patterns for each serotype that reflect their differing tissue tropisms. CAdV-1 is primarily transmitted via the fecal-oral route, with the virus shed in feces, urine, and saliva of infected animals [11, 40]. The virus exhibits remarkable environmental stability, surviving for weeks to months at room temperature and resisting inactivation by many common disinfectants, facilitating indirect transmission through contaminated fomites, food bowls, bedding, and the hands of handlers. A critical epidemiological feature of CAdV-1 is the establishment of persistent infection in renal tubular epithelium, leading to prolonged urinary shedding for months to years following clinical recovery [11, 40]. This carrier state is of paramount importance for maintaining viral circulation within populations, as apparently healthy animals can serve as unsuspected sources of infection. In red foxes in the United Kingdom, CAdV-1 was detected in urine samples from three animals with inapparent infections, confirming that subclinically infected wildlife can actively shed virus into the environment [40].

CAdV-2, in contrast, is predominantly transmitted via the respiratory route through aerosolized droplets and direct contact with infected respiratory secretions [11, 35]. The virus replicates in the epithelium of the upper respiratory tract, causing ciliostasis and epithelial necrosis that predispose animals to secondary bacterial infections, a hallmark of CIRD. CAdV-2 is also shed in feces, and fecal-oral transmission may occur, particularly in kennel environments where sanitation is compromised [11]. The virus is highly contagious in crowded settings such as shelters, boarding kennels, and breeding facilities, where the basic reproductive number (R₀) can exceed 1, leading to rapid outbreaks [35].

Several risk factors have been robustly associated with CAdV infection and seropositivity. Age is a consistently identified determinant: in wolves from Asturias, Spain, 95% of CAdV-1-positive animals were younger than 2 years, with only one of 46 adults testing positive [18]. Similarly, in domestic dogs, the highest incidence of CAdV-1 infection occurs in puppies aged 0–3 months [28], and seropositivity for CAV-2 increases with age in Korean dogs [19]. Vaccination status is the most critical modifiable risk factor. Unvaccinated dogs are at dramatically higher risk: in India, 83.33% of CAdV-1-positive dogs were unvaccinated [28], and in Portugal, a fatal CAdV-1 infection occurred in a 56-day-old unvaccinated puppy [57]. In Ecuador, dogs that had previously received veterinary care had significantly greater odds of having antibodies against CAV [26]. Sex may also play a role, as male dogs in Ecuador had significantly greater odds of CAV seropositivity [26], though this finding is not universal across studies [19]. Breed susceptibility has been noted, with Labrador Retrievers overrepresented among CAdV-1-positive dogs in India [28], and purebred dogs showing higher rates of CAdV co-infection with CPV-2 compared to mixed-breed dogs in Italy [13]. Geographic location and population density are also influential: in wolves, CAdV-1 prevalence was highest in the western area of Asturias [18], and dogs from urban areas in Germany were more likely to respond to CAV-2 vaccination [29], likely reflecting differential exposure pressures.

Co-infection and Syndemic Interactions

A defining epidemiological feature of CAdV infections is their frequent occurrence as part of polymicrobial disease complexes, particularly in the context of canine infectious respiratory disease (CIRD) and canine viral enteritis. CAdV-2 is a classic component of the CIRD complex, often co-infecting with canine parainfluenza virus, Bordetella bronchiseptica, canine respiratory coronavirus (CRCoV), and other pathogens, leading to exacerbated clinical signs and prolonged disease [4, 35, 55]. In a study of 1688 clinical samples from China, co-infections involving CAdV-2 with CCoV, CRCoV, and canine norovirus were documented [4]. Similarly, in dogs with parvoviral enteritis in Italy, 29.5% of CPV-2-positive dogs were co-infected with other viruses, including CAdV-2 (20%), CAdV-1 (2.1%), and canine circovirus (7.4%) [13]. A retrospective immunohistochemical study of 15 puppies that died suddenly revealed that all animals had intralesional antigens of canine distemper virus (CDV), with concomitant infections by CAdV-1 (100%), CAdV-2 (100%), CPV-2 (100%), and Neospora caninum (100%) in a subset of animals, demonstrating the extreme complexity of polymicrobial disease in vulnerable populations [52]. The clinical significance of these co-infections is variable; while some studies suggest that co-infections do not result in more severe disease [13], others indicate that the cumulative pathogenic burden can overwhelm host defenses, particularly in young, unvaccinated animals.

Wildlife Reservoirs and Spillover Events

Wild canids and other carnivores serve as critical reservoirs for CAdV, maintaining viral circulation in ecosystems where domestic dog vaccination is incomplete or absent. The role of red foxes as a reservoir for CAdV-1 has been extensively documented. In the United Kingdom, 18.8% of red foxes had inapparent CAdV-1 infections detected by nested PCR, with viral DNA identified in liver, kidney, spleen, brain, and lung tissues, and 64.4% were seropositive [40]. In a multicentric European study, CAdV-1 DNA was detected in 22% of kidney samples from red foxes in the UK, Italy, and Germany, with no associated pathological changes, confirming that foxes can be asymptomatic shedders [31]. In Norway, seroprevalence against CAdV in Arctic foxes increased from 25–40% in the late 1990s to 68% in 2002–2003, and in red foxes from 31–67% to 80% over a similar period, indicating enzootic circulation in these populations [36]. Wolves also play a significant role: in Asturias, Spain, 14% of 149 free-ranging wolves had CAdV-1 DNA in spleen samples, with prevalence highest in juveniles and in the western region where three fatal brown bear ICH cases occurred [18]. This spatial and temporal correlation strongly suggests that wolves act as sentinels and likely reservoirs for CAdV-1 spillover into sympatric brown bear populations, a finding of major conservation concern [18, 33]. In northern Canada, CAdV-1 was detected in 1% of 303 gray wolves, with sequences highly identical to reference strains [8]. The detection of CAdV-1 in an oral swab from an Italian wolf, with genetic characterization revealing strict relationship to viruses from dogs, wolves, and red foxes, further supports cross-species transmission [56].

Molecular Epidemiology and Evolutionary Dynamics

The molecular epidemiology of CAdV has been profoundly illuminated by genomic sequencing and phylogenetic analyses, revealing distinct geographic lineages, evidence of recombination, and ongoing viral evolution. Complete genome sequencing of CAdV-2 strains from Turkey (OQ596341) demonstrated nucleotide similarity rates exceeding 99.04% with other global CAdV-2 genomes, yet identified several unique mutations, including H34Y in E1A, P55A in E1B 55K, and E250K in E3 ORFA, the latter reported for the first time in Turkey [1]. This study also detected two recombination events, highlighting the role of genetic exchange in shaping viral diversity. In China, analysis of 19 CAdV-2 strains from central China revealed that the Fiber gene harbored the most variant sites, with only 79.0–80.5% nucleotide identity with the vaccine strain CLL, and pairwise sequence comparisons of the Hexon gene identified a novel genotype [27]. Protein modeling predicted that amino acid mutations in the fiber protein head region could alter surface structure, potentially affecting receptor binding and antigenicity [27]. In India, a unique signature insertion of a “G” nucleotide at position 1077 in the E3 gene of CAdV-2 isolates leads to a frameshift and extension of the E3 protein by 11 amino acids, defining a novel phylogenetic group (Group III) distinct from the previously recognized America-Europe and Asia subgroups [21, 23]. This insertion has been consistently identified in Indian isolates, suggesting it may confer a selective advantage or reflect founder effects in the Indian dog population [21, 23, 25]. Phylogenetic analysis of CAdV-1 from a fatal case in India showed close similarity with the Australasian-Europe cluster and high identity (99.78% nucleotide) with a UK fox isolate, with a unique substitution (W126S) in the E3 protein [7]. In Italy, CAdV-1 sequences from dogs and wolves showed divergence from previously described Italian strains and closer relation with older international strains, suggesting genetic heterogeneity and possible reintroduction events [9]. The CAdV-1 strain isolated from mink in China (F1301) had a Fiber gene located on a separate phylogenetic branch, closely related to strains from Norwegian Arctic fox and red fox, and the 100K protein showed evidence of selective pressure and recombination

Genomic Diversity and Evolutionary Dynamics of Canine Adenovirus

The genomic landscape of canine adenovirus (CAdV) is a tapestry woven from ancient evolutionary lineages, contemporary mutational events, and the selective pressures exerted by host immune systems and anthropogenic interventions such as vaccination. As a member of the genus Mastadenovirus within the family Adenoviridae, CAdV exists as two distinct serotypes, CAdV-1, the etiological agent of infectious canine hepatitis (ICH), and CAdV-2, primarily associated with infectious tracheobronchitis (ITB) and implicated in canine infectious respiratory disease complex [11, 12]. While sharing a common genomic architecture, comprising a linear double-stranded DNA genome of approximately 30–31 kb, these two serotypes exhibit profound differences in pathogenicity, tissue tropism, and genetic composition that underpin their distinct clinical manifestations. The genomic diversity observed across global CAdV isolates is not merely a static catalog of sequence variations; it is a dynamic record of viral adaptation, host-jumping events, recombination, and the emergence of novel genotypes with potential implications for vaccine efficacy, diagnostic accuracy, and our fundamental understanding of adenovirus evolution.

Genomic Architecture and Key Regions of Diversification

The CAdV genome is organized into early (E1A, E1B, E2, E3, E4) and late (L1–L5) transcription units, encoding proteins essential for viral replication, host cell manipulation, virion assembly, and immune evasion. The late region encodes the major structural proteins, hexon, penton base, and fiber, which are the primary targets of the host neutralizing antibody response and thus are under significant evolutionary pressure [11]. Among these, the hexon protein is the most abundant capsid component and the principal determinant of serotype specificity. Comparative genomic analyses have revealed that the hexon gene, while relatively conserved across CAdV-2 isolates (sharing >97.4% nucleotide identity among strains from central China), harbors sufficient variability to allow for the delineation of novel genotypes. Specifically, pairwise sequence comparisons of the hexon from Chinese isolates CH-JS-1901 and CH-HN-1801 against the Indian strain India2006 revealed a novel genotype, underscoring the ongoing diversification of this critical antigenic target [27]. More strikingly, the fiber gene, which mediates primary attachment to host cell receptors, exhibits markedly higher variability. Among CAdV-2 strains circulating in central China, fiber gene sequences shared only 79.0–80.5% nucleotide identity with the vaccine strain CLL, a degree of divergence that is astonishing for a virus with a proofreading DNA polymerase [27]. This hypervariability is concentrated in the fiber knob domain, the globular head region responsible for receptor engagement, where amino acid mutations are predicted to induce substantial structural alterations on the virion surface [27]. These changes may directly impact viral tropism, immune escape, and the efficacy of existing vaccines.

The E3 region, a non-essential early gene cassette involved in modulating host immune responses, represents another major hotspot for genomic diversification. Perhaps the most compelling evidence for ongoing CAdV-2 evolution comes from Indian isolates, where a unique frameshift mutation, an insertion of a single guanine nucleotide at position 1077 of the E3 gene, has been consistently identified [21, 23]. This insertion results in a premature termination codon being bypassed, leading to an elongated polypeptide chain with an additional eleven amino acids at the C-terminal end compared to global reference strains [21, 23]. This signature insertion has been proposed to define a novel, third genogroup of CAdV-2, distinct from the previously established American-European and Asian subgroups [23]. The functional consequences of this E3 extension are not yet fully elucidated, but given the role of E3 proteins in inhibiting apoptosis and downregulating MHC class I expression, such a modification could alter the virus’s ability to persist within the host or evade cellular immune responses [21]. Furthermore, the spatial and temporal stability of this insertion across multiple Indian isolates over several years suggests it is not a transient artifact but rather a fixed, adaptive trait in this geographic lineage [23, 25].

Recombination, Mutation Rates, and the Emergence of Novel Variants

Despite the general fidelity of adenoviral DNA replication, CAdV genomes are not static. Whole-genome sequencing of a Turkish CAdV-2 strain (OQ596341), the first complete genome from Europe, revealed two recombination events and a suite of novel point mutations distributed across multiple genes [1]. These mutations included H34Y in E1A; P55A in E1B 55K; D13N and D202N in IVa2; K679R, V934I, and K989N in the DNA polymerase (Pol) gene; E205K in the preterminal protein (pTP); T455A in pIIIa; A310V in protein V; G151R in the protease; E268K in the 100K scaffolding protein; G66S and G141S in the 33K protein; and a cluster of changes in the E3 ORFA gene (T14A, E250K, D287N, I293T) [1]. Notably, the E250K mutation in the E3 ORFA gene was reported for the first time in Turkey, highlighting the value of geographically diverse genomic surveillance [1]. The recombination events identified in the Turkish strain are particularly significant. Recombination between co-infecting adenovirus strains can generate chimeric genomes with novel combinations of alleles, potentially allowing the virus to rapidly acquire advantageous traits such as expanded host range or resistance to neutralizing antibodies. The presence of recombination in a DNA virus with a proofreading polymerase challenges the conventional view that these viruses evolve slowly; rather, CAdV appears to possess a dynamic capacity for genetic innovation through both point mutation and reassortment of genomic segments.

The accumulation of amino acid substitutions in functionally critical proteins, such as the DNA polymerase and terminal protein, raises questions about their impact on viral fitness and replication fidelity. While the direct phenotypic effects of individual mutations require experimental validation, their presence in geographically and temporally distinct isolates suggests they are not merely random noise but may reflect adaptive evolution in response to host or environmental pressures [1]. The 100K protein, a multifunctional scaffold that shuttles hexon trimers into the nucleus and suppresses host protein synthesis, has been the subject of targeted bioinformatic analyses in CAdV-1 strains from minks. This protein exhibits evidence of selective pressure and recombination, possesses one glycosylation site, 48 phosphorylation sites, 60 dominant B-cell epitopes, and nine MHC-I and MHC-II binding peptides [5]. Such a rich repertoire of immunogenic and post-translational modification sites suggests that the 100K protein is a nexus of host-virus interaction, and its variation may influence viral pathogenesis and the host immune response.

Evolutionary Dynamics, Phylogeography, and Host-Tropism Shifts

The evolutionary history of CAdV is inextricably linked to its canid hosts, but the virus has demonstrated a remarkable capacity for cross-species transmission. The two serotypes are believed to have diverged from a common ancestor, with CAdV-1 representing the ancestral, more virulent form that causes systemic disease, and CAdV-2 representing a lineage that adapted to replicate primarily in the respiratory tract [11]. This divergence is reflected not only in clinical presentation but also in genomic signatures: CAdV-1 and CAdV-2 differ in their hemagglutination patterns, coding proteins, and genome sequences, with the latter being sufficiently distinct to be used as a live attenuated vaccine against the former [11]. The widespread use of CAdV-2-based modified live vaccines has, paradoxically, introduced its own evolutionary pressures. Evidence from Korea demonstrates that CAdV-2 strains isolated from wild raccoon dogs (Nyctereutes procyonoides) exhibit extremely high nucleotide similarity (>99%) to the vaccine strain Toronto A26/61, suggesting that vaccine-derived viruses can spill over into wildlife populations [10]. This finding raises critical questions about the safety of live adenovirus vaccines in ecosystems where wild canids are present; the shedding of vaccine strains by vaccinated domestic dogs may lead to unintended infection, and potentially recombination with wild-type strains, in naive wildlife hosts.

Phylogeographic analyses have revealed a complex pattern of CAdV-2 distribution, with at least two major clades: an American-European subgroup and an Asian subgroup, the latter of which includes isolates from China and Turkey and is characterized by distinct amino acid differences in the E3 gene product [27, 35]. Within the Asian subgroup, further substructuring is evident, with Indian strains carrying the E3 frameshift insertion forming a separate cluster that may represent an emerging genotype [23, 25]. Conversely, one Indian strain (ABT/MVC/CAV2/001) from Chennai grouped with the America-Europe subgroup, defying simple geographic assignment and suggesting multiple introductions or long-distance dispersal events [25]. This genetic heterogeneity within a single country underscores the dynamic nature of CAdV evolution and the importance of sustained molecular surveillance.

The evolutionary dynamics of CAdV-1 are equally compelling, particularly in light of its ability to infect a broad range of non-canid hosts. CAdV-1 has been identified as a cause of fatal hepatitis in European brown bears (Ursus arctos arctos) in the Cantabrian Mountains of Spain, where it poses a significant threat to this endangered population [33]. Genetic characterization of CAdV-1 from wolves and bears in this region has shown that the viruses circulating in sympatric carnivores are closely related, suggesting a shared viral pool maintained through trophic interactions or environmental contamination [18, 56]. In red foxes (Vulpes vulpes) across Europe, CAdV-1 is endemic, with seroprevalence rates as high as 64.4% in the United Kingdom and viral DNA detected in 22% of kidney samples from apparently healthy animals in a multicentric European study [31, 40]. These foxes serve as asymptomatic shedders, excreting virus in urine and likely contaminating the environment for months or years, thereby acting as a persistent reservoir for infection of more vulnerable species [31, 40]. The detection of CAdV-1 in arctic foxes (Vulpes lagopus) and red foxes in Norway, with seroprevalence increasing to 80% in some seasons, further demonstrates the virus’s ability to establish itself in extreme environments and maintain transmission cycles even at low host densities [36]. Critically, the observation that CAdV-1 can infect minks (Neovison vison), a species phylogenetically distant from canids, underscores the virus’s potential for host-range expansion [5]. The fiber protein of the mink CAdV-1 strain was found to cluster in a separate phylogenetic branch, closely related to strains from Norwegian Arctic fox and red fox, but distinct from those circulating in domestic dogs, suggesting that the virus may have entered the mink population through spillover from wild foxes and is now undergoing independent evolution within this novel host [5].

Selection Pressures and Adaptive Evolution

The evolutionary trajectory of CAdV is shaped by a complex interplay of positive, negative, and diversifying selection. Bioinformatic analyses of the CAdV-1 fiber protein have revealed evidence of both negative (purifying) selection, which maintains the structural and functional integrity of the receptor-binding domain, and recombination, which can generate novel mosaic alleles [14]. Negative selection predominates in core structural genes like the hexon, where critical epitopes are often conserved to maintain viral fitness. However, the identification of conserved linear neutralizing epitopes (e.g., 634RIKQRETPAL643 and 736PESYKDRMYS745 within the hexon) that are invariant across all known CAdV isolates highlights the existence of functionally constrained regions that are indispensable for viral entry and assembly [22]. These conserved epitopes are attractive targets for vaccine design and universal diagnostic assays, as they are unlikely to be lost under immune pressure.

Conversely, certain genomic regions, particularly the fiber knob and the E3 cassette, appear to be under positive or diversifying selection, as evidenced by the high non-synonymous to synonymous substitution ratios observed in these loci [27]. The accumulation of mutations in the fiber knob is predicted to alter surface topology and electrostatic charge, potentially shifting receptor usage away from the classical coxsackievirus and adenovirus receptor (CAR) towards alternative, yet unidentified, molecules [17, 27]. Indeed, work on CAdV-2 transduction patterns in canine and human cell lines has demonstrated that CAR expression does not correlate with infectivity; CAdV-2 can efficiently transduce cells with low or minimal CAR, and it infects cell types that human adenovirus type 5 (HAdV-5) cannot [17]. This suggests that CAdV-2 utilizes one or more novel receptors, the identification of which would be transformative for both virology and vector development. The ability of CAdV-2 vectors to transduce human neurons with high efficiency and undergo retrograde axonal transport, properties that have made it a premier tool for neuroscience research, is likely a consequence of its unique receptor interactions [15, 20, 24, 34, 43, 44, 50].

The selective pressures encountered by CAdV are not limited to the host immune system. The use of CAdV-2 as a vaccine vector against heterologous pathogens, such as rabies virus, foot-and-mouth disease virus, and Toxoplasma gondii, introduces additional constraints [37, 47, 62]. Engineered replication-defective vectors, deleted in the E1 region and propagated in complementing cell lines, represent a dead-end for the virus but must retain the ability to efficiently transduce target cells and express transgenes without generating a strong anti-vector immune response that would preclude repeated administration [16, 45]. The low seroprevalence of anti-CAdV-2 antibodies in human populations, a key advantage over HAdV-5 for clinical gene therapy, may itself be a consequence of the virus’s evolutionary history, having adapted to canine hosts without extensive prior exposure in humans [16, 24].

Clinical and Epidemiological Implications of Genomic Diversity

The genomic diversity of CAdV has direct and profound implications for disease pathogenesis, diagnosis, and control. The emergence of neurotropic CAdV-2 strains, capable of causing neurological signs in addition to respiratory disease, represents a worrying phenotypic shift [21]. An Indian CAdV-2 isolate from a vaccinated dog presenting with neurological and respiratory symptoms was successfully propagated in MDCK cells, and phylogenetic analysis of its partial E3 region placed it in a separate clade distinct from established subgroups [21]. The unique E3 frameshift mutation present in this and other Indian isolates may contribute to altered pathogenesis by modulating the virus’s ability to evade apoptosis or innate immune responses within the central nervous system [21]. Similarly, CAdV-1 has been documented to cause neurological signs in puppies, including circling, ataxia, and obtundation, with characteristic vasculitis and intranuclear inclusions in the brain [51]. The ability of CAdV-1 to induce PANoptosis, a multifaceted programmed cell death pathway involving pyroptosis, apoptosis, and necroptosis, in canine splenocytes via activation of the AIM2 inflammasome represents a newly discovered mechanism of immunopathogenesis that may explain the severe inflammatory responses observed in ICH [48].

From a diagnostic standpoint, the genetic variability within the E3 gene has necessitated the development of molecular assays that can accurately distinguish between CAdV-1 and CAdV-2. SYBR Green real-time PCR with melting curve analysis offers a rapid, economical method for typing, exploiting sequence differences in the amplified product [42]. Multiplex and quadruplex RT-qPCR assays, targeting conserved regions such as the hexon gene, are now available for simultaneous detection of CAdV-2 alongside other canine respiratory and enteric pathogens, enabling the elucidation of co-infection patterns [4]. The high frequency of co-infections, CAdV-2 has been detected alongside canine parvovirus, canine coronavirus, canine distemper virus, and canine circovirus in clinical samples, highlights the importance of understanding how genomic diversity within CAdV influences interspecies and inter-viral interactions [13, 19, 28, 46, 52, 59].

The impact of genomic diversity on vaccinology cannot be overstated. The currently used modified live CAdV-2 vaccines provide cross-protection against CAdV-1, but the emergence of novel CAdV-2 genotypes, such as those harboring the E3 frameshift or the fiber knob mutations seen in Chinese isolates, raises legitimate concerns about vaccine breakthrough [11, 27]. Sero-surveillance studies from South Korea indicate that while overall seropositivity for CAV-2 is high (77.8%), the protection rate, defined as the proportion of dogs with neutralizing antibody titers considered protective, was only 44.8%, the lowest among the core vaccine pathogens evaluated [19]. This discrepancy between exposure and protection may reflect antigenic drift in the circulating virus population, waning immunity, or both. The development of inactivated CAdV-1 vaccines, such as those derived from the F1301 fox strain, offers an alternative safety profile for use in wildlife and could circumvent some of the risks associated with live vaccines, including reversion to virulence or shedding [60, 61]. Furthermore, the combination of inactivated CAdV-2 vaccines with adjuvants like Cabopol has been shown to induce robust neutralizing antibody responses in dogs, representing a potential improvement over existing formulations [61].

In conclusion, the genomic diversity of canine adenovirus is a product of deep evolutionary divergence between serotypes, ongoing mutational drift, recombination, and adaptive evolution in response to host immune pressure and cross-species transmission. The virus exists not as a monolithic entity but as a constellation of genetically distinct lineages, some of which are expanding their host range, altering their tissue tropism, or evading vaccine-induced immunity. Sustained genomic surveillance, coupled with functional characterization of emerging mutations, is essential to anticipate future shifts in CAdV epidemiology and to ensure that diagnostic and prophylactic tools remain effective. The study of CAdV evolution not only informs the management of this important canine pathogen but also provides a valuable model for understanding the broader principles of DNA virus adaptation and emergence.

Clinical Manifestations and Pathophysiology of Canine Adenovirus Infections

Canine adenoviruses (CAdVs), comprising the two distinct serotypes CAdV-1 and CAdV-2, produce a spectrum of clinical diseases that range from acute, systemic, and often fatal infections to localized, self-limiting respiratory or enteric conditions. The clinical outcome is governed by a complex interplay between viral serotype, host immune status, age, route of inoculation, and the presence of concurrent infections. Understanding the pathophysiological mechanisms underpinning these distinct clinical syndromes is essential for accurate diagnosis, effective therapeutic intervention, and the development of robust vaccination strategies. The two serotypes, while antigenically cross-reactive, exhibit profoundly different tropisms and pathogenic mechanisms, a divergence encoded within their genomes and manifested through differential receptor utilization, tissue-specific replication kinetics, and distinct capacities for immune modulation [11, 14].

Pathophysiology of Infectious Canine Hepatitis (CAdV-1)

Systemic Dissemination and Target Organ Injury. CAdV-1 is a pantropic virus that, following oronasal inoculation, undergoes initial replication in the tonsils and pharyngeal lymphoid tissue. A subsequent cell-associated viremia disseminates the virus to parenchymal organs, with a particular predilection for the liver, kidneys, vascular endothelium, and the central nervous system [7, 9, 51]. The hallmark of CAdV-1 infection is a profound necrotizing hepatitis and vasculitis. Within the liver, viral replication within hepatocytes and Kupffer cells induces multifocal to massive hepatocellular necrosis, accompanied by intranuclear inclusion bodies (Cowdry type A) that are pathognomonic for the infection [33, 57]. The liver becomes friable, pale, and mottled, and grossly exhibits a characteristic nutmeg appearance. Hepatic injury leads to a dramatic rise in serum hepatic enzymes (ALT, AST) and, in severe cases, to fulminant hepatic failure and coagulopathy. A distinctive pathophysiological feature is the development of a generalized vasculitis, which manifests as widespread petechial and ecchymotic hemorrhages across serosal surfaces, subcutaneous tissues, and internal organs [57]. This vascular damage is a direct consequence of viral replication within and subsequent lysis of endothelial cells, leading to increased vascular permeability, edema, and hemorrhagic diatheses.

Renal and Ocular Pathophysiology. The kidney is another primary target organ. Viral replication within glomerular endothelial cells and tubular epithelium leads to a focal to diffuse interstitial nephritis, which can progress to renal failure in chronic or recovered cases. A unique and clinically striking manifestation of CAdV-1 infection is corneal edema, commonly known as "blue eye." This condition arises from the deposition of virus-antibody complexes within the corneal endothelium, leading to endothelial cell damage and subsequent corneal edema. This phenomenon is frequently observed during the recovery phase of the disease or following vaccination with modified-live CAdV-1 vaccines, occurring 7–21 days post-infection or vaccination [6, 52]. Histopathologically, "blue eye" is characterized by corneal epithelial degeneration and edema without a significant inflammatory infiltrate, underscoring its immune-mediated nature [52]. The epidemiological link between wild-type CAdV-1 infection and keratouveitis continues to be documented even in populations vaccinated with CAdV-2 vaccines, with rising convalescent titers confirming recent infection in a subset of affected juvenile dogs [6].

Splenic and Lymphoid Pathophysiology and PANoptosis. A critical and recently elucidated aspect of CAdV-1 pathophysiology is its profound impact on the immune system. CAdV-1 induces a severe lymphocytolysis and splenic necrosis. Groundbreaking research has demonstrated that CAdV-1 triggers a unique form of programmed cell death termed PANoptosis, a coordinated, multifaceted inflammatory cell death pathway involving pyroptosis, apoptosis, and necroptosis, within canine splenocytes. This process is initiated by the recognition of viral double-stranded DNA (dsDNA) by the absent in melanoma 2 (AIM2) inflammasome [48]. Activation of the AIM2 inflammasome in infected T and B cells within the spleen, tonsils, and lymph nodes leads to the concurrent activation of caspase-1 (pyroptosis), caspase-3/7 (apoptosis), and RIPK3/MLKL (necroptosis), resulting in an explosive release of pro-inflammatory cytokines (IL-1β, IL-18) and cellular debris [48]. This massive and dysregulated inflammatory response, rather than direct viral cytopathology alone, is now understood to be a primary driver of the systemic inflammation, coagulopathy, and multi-organ failure observed in acute ICH.

Neurological Pathophysiology. While less common, CAdV-1 can invade the central nervous system, causing a non-suppurative meningoencephalitis and vasculitis. The virus gains access to the brain either via infected endothelial cells or through the blood-brain barrier. Neurological signs, including ataxia, circling, obtundation, vocalization, and seizures, are a grave prognostic indicator and are most frequently reported in young, unvaccinated puppies [38, 51]. Histopathological examination reveals vasculitis, perivascular cuffing, and neuronal necrosis, with intranuclear inclusion bodies occasionally identified in endothelial cells and neurons [51]. The pathophysiological basis for the neurotropism of CAdV-1 is not fully understood but appears to be more pronounced in certain CAdV-1 lineages, such as those causing fox encephalitis [5, 30, 60]. The virus’s Fiber protein, particularly its knob domain, is a critical determinant of cell tropism and host range, with sequence variations influencing the affinity for the coxsackievirus and adenovirus receptor (CAR) and potentially dictating the capacity to infect neuronal cell types [14, 30].

Pathophysiology of Infectious Tracheobronchitis (CAdV-2)

Respiratory Epithelial Tropism and Mucosal Infection. CAdV-2 is primarily an epitheliotropic virus that targets the respiratory tract. Following oronasal exposure, the virus replicates in the ciliated epithelial cells of the nasal mucosa, trachea, bronchi, and bronchioles. This replication induces a characteristic cytopathic effect, leading to ciliostasis, epithelial cell necrosis, and sloughing, which disrupts the mucociliary escalator and compromises the airway's primary defense mechanism [11, 35]. The resulting pathology is a catarrhal to purulent tracheobronchitis, characterized by cough, serous to mucopurulent nasal discharge, and conjunctivitis. The disease is typically self-limiting in immunocompetent adult dogs but can be severe or fatal in young puppies or in animals stressed by overcrowding or poor ventilation, such as those in shelters [35]. CAdV-2 is a core component of the canine infectious respiratory disease complex (CIRD) and frequently participates in polymicrobial infections with viruses like canine distemper virus, canine parainfluenza virus, and bacteria such as Bordetella bronchiseptica, which synergistically exacerbate clinical signs and disease severity [4, 52, 55].

Gastrointestinal and Occasional Neurological Involvement. Unlike the systemic nature of CAdV-1, CAdV-2 infection is largely confined to the respiratory tract. However, a subset of studies has robustly documented the presence of CAdV-2 in the feces of dogs with gastroenteritis, and the virus is frequently detected in cases of canine viral enteritis, often alongside canine parvovirus (CPV-2) [4, 13, 54]. The pathophysiological basis for this enteric involvement is likely the ingestion of the virus and subsequent replication within the intestinal epithelium, though the enteric phase is usually less severe than the respiratory component. Notably, CAdV-2 has also been associated with rare cases of neurological disease in vaccinated dogs. A unique Indian isolate of CAdV-2 was identified in a dog presenting with both respiratory and neurological signs, including seizures and ataxia, where other common neurotropic pathogens were ruled out [21]. This finding suggests that specific CAdV-2 strains may possess the capacity for neuroinvasion, challenging the long-held dogma that CAdV-2 infection is exclusively non-neurological. The genetic basis for this neurotropism may lie in specific mutations within the E3 gene, which encodes proteins involved in immune evasion and modulation of host cell signaling. The unique frameshift mutation in the E3 gene of Indian CAdV-2 isolates, resulting in an extended protein with eleven additional amino acids, may alter the protein's function and influence viral fitness or tropism [21, 23].

Pathophysiological Mechanisms of CAdV-2 Attenuation and Replication Deficiency. The fundamental pathophysiological difference between the two serotypes is encoded in their genomes. While CAdV-1 is a highly virulent, systemic pathogen, CAdV-2 is naturally attenuated, causing a localized, milder disease. This intrinsic attenuation makes CAdV-2 the serotype of choice for modified-live vaccines, as it confers cross-protective immunity against CAdV-1 without inducing the severe systemic and ocular side effects associated with earlier whole-virus or CAdV-1-based vaccines [11, 29, 63]. The genetic basis for this differential pathogenicity is complex, involving variations in the E3 region and structural protein genes, which modulate host immune responses and tissue tropism [11, 27].

In contemporary gene therapy and vaccine vectorology, CAdV-2 vectors are rendered replication-defective through deletion of the E1 gene region. This genetic engineering ensures that the vector can infect target cells and deliver a therapeutic transgene but cannot produce infectious progeny, thereby eliminating the risk of vaccine-induced disease or viral shedding. This E1-deleted platform, propagated in specialized complementing cell lines like AD-293, represents the safest strategy for using CAdV-2 as a gene delivery vehicle in both veterinary and human medicine [16, 24, 34, 43]. In this context, the vector’s natural tropism for neurons and its ability to undergo efficient retrograde axonal transport are exploited for neuroscience research, specifically for circuit mapping and chemogenetic modulation, including in non-human primates [15, 20, 44, 50].

Co-infections and Synergistic Pathophysiology

A major clinically relevant factor in CAdV infections is the high incidence of co-infections with other canine pathogens. Both serotypes are frequently detected as part of polymicrobial infections. CAdV-1 is commonly found alongside canine parvovirus (CPV-2), canine distemper virus (CDV), and Neospora caninum, forming a "pathogen load" that can overwhelm a young host's immune system and dramatically accelerate disease progression and mortality [13, 28, 52]. In a large retrospective study, concomitant triple and quadruple infections with CDV, CPV-2, CAdV-1, and N. caninum were documented in a cohort of puppies that died suddenly, underscoring the synergistic lethal potential of these co-infections [52]. Similarly, CAdV-2 is a near-constant companion in cases of CIRD, where its presence potentiates the pathogenicity of B. bronchiseptica and other respiratory viruses [35]. The pathophysiological synergy is likely multifaceted: viral-induced damage to the respiratory or gastrointestinal epithelium provides a portal of entry for secondary bacterial invaders, while viral immune evasion strategies (such as those encoded by the CAdV E3 region) suppress antiviral and antibacterial host defenses, creating a permissive environment for superinfection [13, 52]. This high rate of co-infection has driven the development of complex, multiplex diagnostic panels, such as quadruplex RT-qPCR assays, capable of simultaneously detecting CAdV alongside CRCoV, CCoV, and norovirus, reflecting the clinical reality that canine viral disease is rarely monotypic [4].

Diagnostic Approaches for Canine Adenovirus Detection

The accurate and timely detection of canine adenovirus (CAdV), encompassing both CAdV-1 (the etiological agent of infectious canine hepatitis) and CAdV-2 (associated with infectious tracheobronchitis and canine infectious respiratory disease complex), is paramount for effective clinical management, epidemiological surveillance, and the implementation of control strategies. The diagnostic landscape has evolved dramatically, moving from classical virological and serological methods to a sophisticated armamentarium of molecular, point-of-care, and high-throughput genomic technologies. This evolution reflects the pressing need to address the limitations of traditional approaches, which are often time-consuming, require specialized laboratory infrastructure and personnel, and may lack the sensitivity to detect subclinical infections or discriminate between closely related serotypes [2, 11]. The following analysis provides a comprehensive, authoritative dissection of the contemporary diagnostic toolkit, evaluating each modality’s biological underpinnings, analytical performance, and practical applicability in both clinical and research settings, with a particular emphasis on differentiating CAdV-1 from CAdV-2.

Classical Virological and Serological Methodologies

Historically, the gold standard for CAdV diagnosis rested upon virus isolation (VI) in permissive cell lines, most notably Madin-Darby Canine Kidney (MDCK) cells and Vero cells [3, 10, 14]. The characteristic cytopathic effect (CPE), described as a "bunch of grapes" morphology, provides a presumptive identification, which is then confirmed via immunofluorescence assays (IFA) using serotype-specific monoclonal antibodies or virus neutralization tests (VNT) [21, 69]. The isolation of CAdV-2 strains from clinical samples, such as the APQA1601 and CAV-HN45 strains, remains a critical step for characterizing circulating field isolates, assessing their growth kinetics, and generating viral stocks for vaccine development and gene therapy vector production [3, 69]. However, VI is inherently labor-intensive, requires several days to weeks for observable CPE, and is dependent on the viability of the virus in the sample, making it impractical for rapid clinical decision-making.

Serological diagnostics, which detect host antibodies, provide a retrospective view of infection or vaccination status. The VNT is considered the reference standard for measuring neutralizing antibody titers, quantifying the biological function of antibodies in preventing viral entry. It has been extensively used to assess vaccine-induced immunity and seroprevalence in canine populations, with studies demonstrating a strong correlation between VNT results and protection [19, 29, 53, 65]. For instance, serosurveys in Korean dogs revealed CAV-2 VNT seropositivity rates of 77.8%, with significant variations observed by age and vaccination history [19]. Hemagglutination inhibition (HI) assays, which exploit the ability of CAdV to agglutinate guinea pig or human erythrocytes, offer an alternative serological platform, though this property is more pronounced for CAdV-1 than CAdV-2 [11, 68].

Enzyme-linked immunosorbent assays (ELISAs) have been developed to circumvent the time and complexity of VNT, enabling high-throughput serological screening. Indirect ELISAs (I-ELISAs) using column-purified viral antigens have shown high sensitivity (97.0-98.6%) and specificity (74.2-86.4%) compared to VNT for both CAdV-1 and CAdV-2, and are particularly valuable for large-scale epidemiological studies [64, 66, 67]. A significant advancement is the development of a sandwich ELISA for CAdV-1 antigen detection in foxes, utilizing monoclonal antibodies for capture and HRP-labeled polyclonal antibodies for detection, which demonstrated a 92.86% coincidence rate with RT-PCR, thereby offering a rapid tool for mass screening in resource-limited settings [64]. Point-of-care (POC) dot-blot ELISA assays have also been validated, providing clinicians with a rapid (minutes) and reliable method for assessing protective antibody levels against CAdV, with reported sensitivity and specificity exceeding 96% and 87%, respectively, allowing for evidence-based booster vaccination decisions [26, 65]. However, serology cannot differentiate between antibodies elicited by vaccination and those from natural infection, nor can it detect active, early-stage infections before seroconversion, a critical window for therapeutic intervention.

Molecular Detection: Polymerase Chain Reaction and Its Modern Variants

The advent of polymerase chain reaction (PCR) has revolutionized CAdV diagnostics, offering unparalleled sensitivity, specificity, and speed. Conventional PCR assays, often targeting the E3, hexon, or fiber genes, have been instrumental in detecting CAdV DNA in a wide array of clinical specimens, including whole blood, serum, feces, urine, nasal swabs, and tissue homogenates [40, 42]. These assays are capable of distinguishing CAdV-1 from CAdV-2 through the use of serotype-specific primers or subsequent restriction fragment length polymorphism (RFLP) analysis, a crucial capability given the overlapping clinical presentations and the widespread use of CAdV-2-based vaccines that can confound diagnosis [11, 21, 25].

Real-time quantitative PCR (qPCR) has emerged as the contemporary workhorse for molecular detection, providing not only qualitative yes/no answers but also precise quantification of viral load. This quantitative capacity is invaluable for monitoring disease progression, assessing therapeutic efficacy, and studying viral pathogenesis. The development of a SYBR Green real-time PCR assay with melting curve analysis is a particularly elegant solution for simultaneous detection and differentiation of CAdV-1 and CAdV-2. This method exploits the distinct melting temperatures (Tm) of the amplicons generated from each serotype, allowing for a single-tube, closed-system assay that minimizes contamination risk and obviates the need for post-PCR processing, such as gel electrophoresis [42]. The assay has demonstrated high sensitivity (detecting as few as 10-100 copies of viral DNA) and reproducibility, making it a simple, economical, and robust option for diagnostic laboratories [42].

The latest frontier in PCR-based diagnostics is represented by multiplex and pan-pathogen panels. Given the high frequency of co-infections in canine respiratory and enteric disease complexes, the ability to simultaneously screen for multiple pathogens in a single reaction is a significant clinical advantage. A recently developed quadruplex RT-qPCR for the detection of canine coronavirus, canine respiratory coronavirus, CAdV-2, and canine norovirus exemplifies this trend. Targeting the conserved hexon gene of CAdV-2, this assay achieved a limit of detection (LOD) of 1.0 × 10² copies/reaction with excellent specificity, sensitivity, and repeatability (intra-assay variability 0.19–1.31%) [4]. Such multiplex panels dramatically reduce the time to diagnosis, conserve precious sample material, and provide a comprehensive etiological profile that is essential for appropriate case management and outbreak control.

Isothermal Amplification and CRISPR-Based Point-of-Care Technologies

To address the logistical constraints of PCR, which requires thermal cyclers and skilled technicians, isothermal amplification methods have been developed, bringing molecular diagnostics closer to the patient. Recombinase polymerase amplification (RPA) is a prominent example, operating at a constant low temperature (e.g., 39°C) and providing results in under 15 minutes. A real-time RPA assay for CAdV-2, targeting the hexon gene, demonstrated a detection limit of 214 copies/μL and a 97.7% coincidence rate with qPCR when tested on field samples, with no cross-reactivity to other major canine viruses [55]. This technology is particularly suited for deployment in field settings, shelters, and low-resource veterinary clinics.

The most groundbreaking innovation in POC diagnostics, however, is the integration of CRISPR-Cas systems with isothermal amplification. The SHERLOCK (Specific High-sensitivity Enzymatic Reporter UnLOCKing) platform, utilizing Cas13a, has been adapted for CAdV-2 detection. This method, combining HUDSON (Heating Unextracted Diagnostic Samples to Obliterate Nucleases) for rapid nucleic acid extraction, RAA (recombinase-aided amplification), and CRISPR-Cas13a collateral cleavage activity, allows for naked-eye visual readout on a lateral flow strip. The assay achieves a sensitivity of 10² copies/μL (approximately 750 copies per reaction), exhibits no cross-reactivity with canine parvovirus, distemper virus, or coronavirus, and demonstrates a 95% concordance with conventional PCR on clinical samples [2]. The entire process, from sample to result, can be completed in approximately 90 minutes using only basic equipment (e.g., a heat block), representing a paradigm shift for field-based surveillance and outbreak management [2].

Advanced Genomic Characterization and Serotyping

While PCR and isothermal methods are excellent for detection, they provide limited information on the genetic identity and evolutionary trajectory of the virus. For comprehensive characterization, whole-genome sequencing (WGS) and next-generation sequencing (NGS) are indispensable. NGS has enabled the complete genome characterization of CAdV-2 strains from diverse geographical origins, such as Türkiye, Korea, and India, revealing novel mutations, recombination events, and evolutionary dynamics that would remain cryptic with partial gene sequencing [1, 10, 23]. For example, WGS of a Turkish CAdV-2 strain identified unique non-synonymous mutations across multiple genes (E1A, E1B, IVa2, Pol, pTP, pIIIa, V, protease, 100K, 33K, E3), including the E250K mutation in the E3 ORFA gene, which has implications for viral tropism and immune evasion [1]. Similarly, WGS of CAdV-2 from Korean raccoon dogs revealed a high similarity to the vaccine strain Toronto A26/61, suggesting potential vaccine-derived virus circulation in wildlife [10].

Bioinformatic analysis of these genomes allows for sophisticated phylogenetic clustering, which has redefined our understanding of CAdV-2 global diversity. Phylogenetic analyses of the E3, fiber, hexon, and penton genes have consistently revealed a split into distinct genogroups: an "America-Europe" cluster and an "Asian" (or "China-Turkey") cluster [27, 35]. The fiber gene, in particular, harbors the highest degree of variability, with amino acid changes in the knob domain potentially affecting receptor binding and cell tropism [27]. Furthermore, a unique frameshift mutation in the E3 gene (an insertion of a "G" nucleotide at position 1077) has been identified as a signature of Indian CAdV-2 isolates, resulting in an elongated protein that may have functional consequences for host immune modulation [21, 23]. These findings underscore the necessity of ongoing genomic surveillance to track the emergence of novel variants and to inform the selection of appropriate vaccine strains.

Differentiating CAdV-1 from CAdV-2: A Critical Diagnostic Challenge

The clinical distinction between CAdV-1 and CAdV-2 infections can be ambiguous, as both serotypes can present with respiratory signs, gastroenteritis, and systemic involvement, including, rarely, neurological signs [11, 21, 51]. This diagnostic challenge is compounded by the widespread use of modified-live CAdV-2 vaccines, which are protective against CAdV-1 but can themselves cause mild respiratory signs and be shed in feces, leading to false-positive PCR results if not properly interpreted [52]. Therefore, a definitive diagnosis relies on molecular differentiation.

Several strategies exist for serotype discrimination. The SYBR Green real-time PCR with melting curve analysis provides a robust and rapid solution, as the amplicon from CAdV-1 has a distinct Tm from that of CAdV-2 [42]. Conventional multiplex PCR assays, designed to amplify different-sized fragments from the E3 or fiber genes of each serotype, are also widely used [11, 25]. Sequence analysis of the E3 gene or the full hexon gene provides the highest resolution and is essential for confirming the serotype of novel or atypical strains [7, 21]. It is crucial to note that the E3 gene is not present in all non-replicating adenoviral vectors (which are E1-deleted), so diagnostic assays targeting alternative regions are necessary for those applications [16]. Histopathological examination, while not a primary diagnostic tool, can provide supportive evidence: CAdV-1 characteristically produces large basophilic intranuclear inclusion bodies in hepatocytes, Kupffer cells, and vascular endothelial cells, whereas CAdV-2 inclusions are typically found in bronchial and alveolar epithelial cells [7, 51, 52, 57]. Immunohistochemistry (IHC) using serotype-specific monoclonal antibodies can confirm the presence of viral antigen in tissue sections and has been instrumental in retrospective studies of archived specimens, including those from fatal cases in wolves and bears [33, 52, 57].

Choosing the Appropriate Diagnostic Modality: An Algorithmic Approach

The selection of a specific diagnostic test for CAdV must be guided by the clinical context, the objective of testing, and the available resources. For a clinically ill dog with acute respiratory or hepatic disease, a rapid and sensitive molecular test is paramount. In a well-equipped laboratory, the SYBR Green real-time PCR assay offers a rapid (under 2 hours), quantitative, and serotype-discriminating result [42]. For a field setting or shelter with limited infrastructure, the CRISPR/SHERLOCK lateral flow assay provides a visual result within 90 minutes with minimal equipment [2]. For epidemiological surveys or herd-level immunity assessments, serological tests such as I-ELISA or the POC dot-blot ELISA are appropriate for their high throughput and low cost [64-66]. For vaccine efficacy trials or investigations into viral pathogenesis, virus isolation combined with WGS is essential to fully characterize the viral strain and its genetic determinants [1, 10, 69].

It is imperative that all diagnostic testing be performed in conjunction with a thorough clinical history, including vaccination status, age, and exposure history, as the interpretation of test results, especially serology and PCR, can be profoundly affected by recent vaccination. The World Organization for Animal Health (WOAH) recommends molecular confirmation of clinical cases, and serosurveillance should be a cornerstone of national control programs. The diagnostic armamentarium for CAdV is now exceptionally diverse, moving from the slow and laborious classical methods to a suite of rapid, sensitive, specific, and accessible tools. The challenge for the practitioner and the epidemiologist lies not in a lack of options, but in the judicious selection and interpretation of the most appropriate test for the specific diagnostic question at hand, leveraging the molecular and genomic insights that have so fundamentally reshaped our understanding of this important canine pathogen.

Vaccine Development and Therapeutic Strategies for Canine Adenovirus

The development of effective vaccines and therapeutic interventions against canine adenovirus (CAdV) represents a critical frontier in veterinary medicine, encompassing both prophylactic immunization against infectious canine hepatitis (ICH) and infectious tracheobronchitis (ITB), as well as the repurposing of CAdV vectors for gene therapy, oncolytic virotherapy, and the construction of vectored vaccines against other pathogens. The dual nature of CAdV, as both a pathogen of significant concern and a powerful molecular tool, necessitates a comprehensive understanding of its biology, antigenic structure, and host interactions to inform rational vaccine design and therapeutic deployment. The global framework established by the World Organisation for Animal Health (WOAH) and national veterinary authorities underscores the importance of robust, safe, and enduring immunity against CAdV, particularly given its documented circulation in both domestic and wild canid populations [8, 18, 33] and the emergence of novel genetic variants that may challenge existing vaccine-induced protection [1, 11, 27].

Conventional Vaccination Strategies: Modified Live and Inactivated Platforms

The cornerstone of CAdV prophylaxis has long been the modified live virus (MLV) vaccine, predominantly utilizing the CAdV-2 serotype. This approach is predicated on the robust cross-protective immunity engendered by CAdV-2 against the more virulent CAdV-1, which causes ICH, as well as homologous protection against CAdV-2-associated respiratory disease [11, 29]. The immunological basis of this cross-protection lies in the shared antigenic epitopes, particularly within the hexon protein, which elicits potent neutralizing antibody responses capable of neutralizing both serotypes [22]. Studies have consistently demonstrated that a single dose of MLV CAdV-2 vaccine can induce a durable neutralizing antibody response, with seropositivity rates in vaccinated populations often exceeding 85-95% [29]. The duration of this immunity is a subject of active investigation; while current guidelines from the American Animal Hospital Association (AAHA) recommend re-vaccination at three-year intervals, evidence suggests that protective antibody titers can persist for five years or longer in a substantial proportion of dogs [63]. Bergmann et al. [29] demonstrated that 87% of dogs possessed pre-vaccination anti-CAdV antibodies even when the last vaccination was administered more than three years prior, and only 6% of dogs exhibited a vaccination response upon re-vaccination, indicating that booster intervals may be extended safely in many patients. However, despite the overall efficacy of MLV vaccines, concerns persist regarding their safety profile. The CAdV-2 vaccine strain is attenuated, but reports of vaccine-associated disease, particularly in non-target species such as fennec foxes, where co-infection with vaccine-strain canine distemper virus led to fatal outcomes [70], highlight the potential for residual virulence. Furthermore, the risk of reversion to virulence, though theoretically low for a DNA virus with a stable genome, remains a consideration [11]. The potential for CAdV-2 vaccine strains to shed in the environment and infect naive wild canids has been raised, as evidenced by the identification of CAdV-2 strains in Korean raccoon dogs exhibiting high genetic similarity to the Toronto A26/61 vaccine strain, suggesting spillover from vaccinated domestic dogs [10]. This phenomenon underscores the need for careful risk assessment, particularly in regions where domestic and wild canid populations overlap.

In response to the safety concerns associated with MLV platforms, significant efforts have been directed toward the development of inactivated (killed) vaccines. These vaccines offer an inherently safer profile, as they cannot replicate in the host or revert to virulence, making them particularly attractive for vaccinating immunocompromised animals, pregnant bitches, or wildlife where the consequences of vaccine-induced disease could be catastrophic. Fu et al. [60] developed an inactivated CAdV-1 vaccine for foxes utilizing the F1301 strain, demonstrating that it effectively protected silver foxes against virulent CAdV-1 challenge with an improved safety profile compared to the commercial MLV vaccine. Similarly, Yang et al. [61] constructed an inactivated CAdV-2 vaccine candidate from the Korean APQA1701-40P strain, which had been passaged 40 times in MDCK cells and inactivated with formaldehyde. This vaccine, adjuvanted with Cabopol, induced a significantly higher virus-neutralizing antibody titer in dogs compared to unvaccinated controls, confirming its immunogenicity [61]. A critical advantage of inactivated platforms is the ability to incorporate novel adjuvants that can shape the quality and magnitude of the immune response. Broutin et al. [47] demonstrated that the water-in-oil-in-water adjuvant Montanide™ ISA 201 VG significantly increased rabies antibody titers when co-administered with a non-replicating CAV-2-vectored rabies vaccine, without affecting the anti-CAV-2 vector response. Transcriptomic analysis revealed upregulation of RIG-I, TLR, NLR, and IFN signaling pathways, indicating that the adjuvant engaged innate immune sensors to potentiate adaptive immunity [47]. This adjuvantation strategy could be directly applied to inactivated CAdV vaccines to enhance their potency and potentially reduce the required antigen dose, making them more economical for large-scale vaccination campaigns. The Carnican-5R vaccine, a combined vaccine developed in Russia, incorporates inactivated components against canine distemper, parvovirus, coronavirus, adenovirus, and rabies. Testing in dogs demonstrated that double administration at a 21-day interval induced a 5.36-fold increase in antibody titers against CAdV-2, with protective immunity lasting at least 12 months [59]. This multi-valent, inactivated approach aligns with current trends in veterinary vaccinology, aiming to minimize injections while maximizing safety and coverage.

Next-Generation Vectored Vaccines: Replication-Defective and Replication-Competent Adenovirus Platforms

The inherent immunogenicity of adenovirus vectors, combined with their large cloning capacity (up to 30-36 kb for helper-dependent vectors), renders them exceptionally versatile platforms for the development of multivalent vaccines against canine pathogens [16, 43]. Canine adenovirus type 2 (CAV-2) vectors are particularly attractive due to their low seroprevalence in human populations, their ability to infect a broad range of cell types including respiratory epithelial cells and neurons, and their capacity for retrograde axonal transport, enabling targeted delivery of transgenes to the central nervous system [16, 24]. The standard approach for generating replication-defective CAV-2 vectors involves deletion of the E1 region, which is essential for viral replication. These vectors are propagated in complementing cell lines (e.g., AD-293 or E1-transformed MDCK cells) that provide the E1 protein in trans [16]. The deletion of E1 not only renders the vector non-replicative in normal cells, enhancing safety, but also provides space for the insertion of foreign transgenes.

The utility of CAV-2 vectored vaccines has been demonstrated across a spectrum of canine pathogens. A seminal study by Vleeschauwer et al. [37] evaluated a CAV-2 vector expressing the P1-3C cassette of foot-and-mouth disease virus (FMDV) in guinea pigs. The Cav-P1/3C R° vaccine elicited a strong humoral immune response and conferred protection against challenge with a heterologous FMDV strain, comparable to that afforded by a high-potency conventional FMD vaccine. This proof-of-concept study highlights the potential of CAV-2 vectors as marker vaccines that can differentiate infected from vaccinated animals (DIVA), a critical attribute for FMD control programs [37]. In the context of toxoplasmosis, Li et al. [62] constructed a CAV-2 vector expressing the Toxoplasma gondii ROP18 antigen (CAV-2-ROP18). Intramuscular immunization of mice elicited robust humoral and cellular immune responses, characterized by a mixed IgG1/IgG2a profile, significant production of IFN-γ and IL-2, and activation of CD4+ and CD8+ T cells. This response translated into a 40% survival rate against acute challenge with the virulent RH strain and a significant reduction in brain cyst burden in the chronic infection model [62]. The ability of CAV-2 vectors to induce strong Th1-biased responses is particularly relevant for intracellular pathogens, where cellular immunity is paramount for protection.

The canine adenovirus vector platform is not limited to canine-specific pathogens. Hu et al. [39] developed a replication-defective human adenovirus type 5 (Ad5) vector co-expressing canine distemper virus hemagglutinin (H) and canine parvovirus viral protein 2 (VP2). This Ad5-(VP2+H) vaccine elicited potent and durable antigen-specific immune responses in mice and dogs, and exhibited a superior safety profile compared to a commercial live-attenuated vaccine, eliminating the risk of virulence reversion [39]. Similarly, a bivalent Ad5 vector expressing rabies virus glycoprotein and CDV-H (rAd5-G-H) conferred 100% protection against lethal RABV and CDV challenge in mice and foxes, demonstrating the potential of adenovirus-vectored vaccines for wildlife vaccination [74]. The development of replication-competent adenovirus vectors represents another frontier, particularly for oral vaccination of wildlife, where injectable vaccines are impractical. Du et al. [71] explored a mouse adenovirus type 1 (MAV-1) vector expressing CDV-H as an oral vaccine in mice. A single oral immunization elicited a neutralizing antibody response, and a second oral dose effectively boosted immunity, even in the presence of pre-existing vector immunity. This study validates the concept of replicating AdV-based oral vaccines for wildlife, though the challenge remains to develop species-specific vectors that can replicate efficiently in target carnivores without causing disease [71].

Therapeutic Strategies: Oncology, Gene Therapy, and Antiviral Interventions

Beyond prophylactic vaccination, canine adenovirus vectors and oncolytic adenoviruses are being intensively investigated as therapeutic modalities for canine cancers. Oncolytic virotherapy harnesses the ability of viruses to selectively replicate in and lyse tumor cells, while simultaneously inducing antitumor immune responses. Several oncolytic adenovirus platforms have been evaluated in canine patients. Hashimoto et al. [49] assessed the telomerase-specific oncolytic human adenoviruses OBP-301 and OBP-302 against canine breast cancer cell lines. Both viruses exhibited cytopathic effects in vitro, and intratumoral administration significantly suppressed the growth of subcutaneous tumors in a murine xenograft model. Notably, combination therapy with the CDK4/6 inhibitor palbociclib enhanced the antitumor activity of OBP-301, suggesting that cell cycle modulation can augment oncolytic virus replication [49]. A separate study by Martín-Carrasco et al. [41] evaluated the safety and efficacy of ICOCAV15, an oncolytic canine adenovirus, in eight dogs with spontaneous carcinomas/adenocarcinomas. Intratumoral administration was well-tolerated, with no clinically relevant adverse effects. Two dogs exhibited a partial response, and six achieved stable disease, with median survival times exceeding those historically reported for chemotherapy. Immunohistochemical analysis revealed increased infiltration of CD8+, CD3+, and CD20+ lymphocytes in the tumor microenvironment following treatment, indicative of an induced antitumor immune response [41]. A significant limitation of CAdV-based oncolytics, however, is the potential for pre-existing neutralizing antibodies in vaccinated dogs to attenuate the therapeutic effect. To circumvent this, Matsugo et al. [73] explored a bat adenovirus (BtAdV) as an alternative oncolytic platform. The BtAdV Mm32 strain replicated efficiently in canine tumor cell lines and exhibited no serological cross-reactivity with CAdV. A recombinant Mm32 construct with a tumor-specific canine telomerase reverse transcriptase promoter (Mm32-E1Ap+cTERTp) showed superior growth and cytotoxicity in canine tumor cells compared to normal cells, positioning it as a promising candidate for canine cancer therapy in patients with pre-existing anti-CAdV immunity [73].

Gene therapy applications of CAV-2 vectors are equally promising, particularly for the treatment of monogenic diseases affecting the central nervous system and other tissues. The ability of CAV-2 to undergo retrograde axonal transport and transduce neurons with high efficiency makes it an ideal vector for addressing neurological disorders [24, 43]. Stevens et al. [20] optimized parameters for transducing the locus coeruleus (LC) in rats using a CAV-2 vector expressing an excitatory DREADD (hM3Dq). By carefully titrating the viral particle number and injection volume, they achieved highly specific and efficient transduction of LC noradrenergic neurons (up to 87% of LC neurons) with minimal off-target expression and no signs of toxicity [20]. This level of precision is essential for chemogenetic modulation of defined neural circuits in preclinical models of epilepsy, depression, and other neurological conditions. Furthermore, CAV-2 vectors have been successfully employed for transgene expression in non-human primate motoneurons following intramuscular injection [34] and in cholinergic interneurons in the monkey striatum using a cell-type-specific promoter [50]. These studies pave the way for translational applications in human and veterinary neurology. In the context of metabolic diseases, helper-dependent (HD) CAV-2 vectors have been used to deliver the human β-glucuronidase (GUSB) cDNA into the corneal stroma of dogs with mucopolysaccharidosis VII (MPS VII). HD-RIGIE injection led to efficient transduction of keratocytes, restoration of β-glu activity, and reduction of glycosaminoglycan accumulation, demonstrating the feasibility of CAV-2-mediated gene therapy for corneal clouding [45]. This approach could be extended to other inherited corneal dystrophies in dogs.

Antiviral therapeutic strategies targeting CAdV directly are less well-developed but are beginning to emerge. The use of recombinant adenoviruses expressing canine interferon lambda 3 (Ad-caIFNλ3) has shown promise as a broad-spectrum antiviral. Kim et al. [72] demonstrated that Ad-caIFNλ3 transduction suppressed replication of canine coronavirus, canine parvovirus, and canine distemper virus in A72 and MDCK cell lines without cytotoxicity, suggesting a potential therapeutic role in treating mixed viral infections that frequently complicate CAdV disease [13, 52]. The induction of type III interferons via a viral vector could provide a rapid, non-specific antiviral defense that is particularly useful in outbreak scenarios.

Challenges and Future Directions: Genetic Drift, Vaccine Matching, and the One Health Imperative

The continuous evolution of CAdV poses a significant challenge to vaccine efficacy. Genomic surveillance has revealed an increasing diversity among circulating strains, with mutations accumulating in critical immunogenic regions. Temizkan and Temizkan [1] identified 16 novel amino acid substitutions in a Turkish CAdV-2 strain, including mutations in the E1A, E1B 55K, IVa2, Pol, pTP, pIIIa, V, protease, 100K, 33K, and E3 ORFA genes. Of particular concern is the E250K mutation in the E3 ORFA gene, which was reported for the first time in Türkiye [1]. The E3 region encodes proteins involved in immune evasion, and mutations in this region could potentially alter viral pathogenesis or the host's ability to recognize and neutralize the virus. Ji et al. [27] conducted a comprehensive analysis of CAdV-2 strains circulating in central China between 2017 and 2019. Their phylogenetic analysis of the fiber, hexon, and penton genes revealed that the fiber gene harbored the most variant sites, with only 79.0–80.5% nucleotide identity between Chinese field strains and the CLL vaccine strain. This degree of divergence raises the possibility that vaccine-induced immune responses may be suboptimal against contemporary field strains, a phenomenon reminiscent of antigenic drift seen in influenza and other RNA viruses. The identification of a novel genotype within the hexon gene of Chinese isolates further underscores the need for ongoing antigenic surveillance and periodic vaccine strain updates [27]. Similarly, the discovery of a unique frameshift mutation in the E3 gene of Indian CAdV-2 isolates, resulting in an 11-amino acid extension of the E3 protein, has led to the classification of a novel genetic group (Group III) of CAdV-2 [21, 23]. While the functional impact of this mutation on viral tropism or immune evasion remains to be fully elucidated, it may have implications for the design of E3-deleted vectors and the effectiveness of existing vaccines. The documented circulation of CAdV-1 and CAdV-2 across a wide range of hosts, including wolves, foxes, raccoon dogs, and even brown bears [8, 9, 18, 31, 33, 36, 38, 40, 56], highlights the importance of a One Health approach. Vaccination of domestic dogs is not only for their own protection but also serves to reduce the reservoir of virus that can spill over into vulnerable wildlife populations. The detection of CAdV-1 in free-ranging brown bears in Spain, causing fatal infectious canine hepatitis [33], and the high seroprevalence (up to 68%) in Arctic foxes in Norway [36] demonstrate the ecological impact of this pathogen. A concerted global effort, coordinated by WOAH and national veterinary services, is required to standardize vaccine quality, monitor emerging strains, and implement vaccination strategies that safeguard both companion animals and biodiversity.

References

[1] Temizkan SS, Temizkan MC. Complete Genome Characterization of Canine Adenovirus From Türkiye With Next‐Generation Sequencing. Veterinary Medicine and Science. 2025. DOI: https://doi.org/10.1002/vms3.70163

[2] Liu B, Li Y, Yang Z, Wu J, Jiang Y, Zhao L, et al.. A rapid and visual detection for canine Adenovirus-2 using CRISPR-Cas13a-based SHERLOCK technology.. Journal of Microbiological Methods. 2025. DOI: https://doi.org/10.1016/j.mimet.2025.107314

[3] Wang C, Wang D, Yin G, Wang Y, Lu M, Zhang Z, et al.. Isolation and characterization of canine adenovirus type 2 (CAV-HN45) and its selective infection of human cervical cancer cells with preliminary oncolytic potential. Frontiers in Veterinary Science. 2025. DOI: https://doi.org/10.3389/fvets.2025.1692395

[4] Shi K, Shi Y, Shi Y, Long F, Yin Y, Pan Y, et al.. Establishment of a Quadruplex RT-qPCR for the Detection of Canine Coronavirus, Canine Respiratory Coronavirus, Canine Adenovirus Type 2, and Canine Norovirus. Pathogens. 2024. DOI: https://doi.org/10.3390/pathogens13121054

[5] Hou J, Xu J, Wang B, Zhang H, Yin B, Li G, et al.. First identification of canine adenovirus 1 in mink and bioinformatics analysis of its 100 K protein. Frontiers in Microbiology. 2023. DOI: https://doi.org/10.3389/fmicb.2023.1245581

[6] Joyce H, Oliver J, Fricker GV, Weir W, Fleming L. Keratouveitis in juvenile dogs and its presumed association with canine adenovirus infection.. Veterinary Ophthalmology. 2023. DOI: https://doi.org/10.1111/vop.13085

[7] Syamily S, Rajasekhar R, Anoopraj R, Ravishankar C, Jishnu HP. Detection and molecular characterisation of canine adenovirus type 1 from a fatal case of infectious canine hepatitis from India. Veterinary Record Case Reports. 2023. DOI: https://doi.org/10.1002/vrc2.698

[8] Canuti M, Fry K, Cluff HD, Mira F, Fenton H, Lang A. Co-circulation of five species of dog parvoviruses and canine adenovirus type 1 among gray wolves (Canis lupus) in northern Canada.. Transboundary and Emerging Diseases. 2022. DOI: https://doi.org/10.1111/tbed.14474

[9] Mira F, Puleio R, Schiró G, Condorelli L, Bella SD, Chiaramonte G, et al.. Study on the Canine Adenovirus Type 1 (CAdV-1) Infection in Domestic Dogs in Southern Italy. Pathogens. 2022. DOI: https://doi.org/10.3390/pathogens11111254

[10] Kim Y, Lee S, Kim Y, Na E, Park J, Oem J. Genetic Characteristics of Canine Adenovirus Type 2 Detected in Wild Raccoon Dogs (Nyctereutes procyonoides) in Korea (2017–2020). Veterinary Sciences. 2022. DOI: https://doi.org/10.3390/vetsci9110591

[11] Zhu Y, Xu J, Lian S, Zhang R, Hou J, Wang M, et al.. Difference Analysis Between Canine Adenovirus Types 1 And 2. Frontiers in Cellular and Infection Microbiology. 2022. DOI: https://doi.org/10.3389/fcimb.2022.854876

[12] . canine adenovirus. CABI Compendium. 2022. DOI: https://doi.org/10.1079/cabicompendium.96011

[13] Balboni A, Terrusi A, Urbani L, Troìa R, Stefanelli SAM, Giunti M, et al.. Canine circovirus and Canine adenovirus type 1 and 2 in dogs with parvoviral enteritis. Veterinary research communications. 2021. DOI: https://doi.org/10.1007/s11259-021-09850-y

[14] Wang B, Wang M, Zhang H, Xu J, Hou J, Zhu Y. Canine Adenovirus 1 Isolation Bioinformatics Analysis of the Fiber. Frontiers in Cellular and Infection Microbiology. 2022. DOI: https://doi.org/10.3389/fcimb.2022.879360

[15] Kabanova A, Cavani E, Logothetis N, Eschenko O. Feasibility of Canine Adenovirus Type 2 (CAV2) Based Vector for the Locus Coeruleus Optogenetic Activation in Non-Transgenic Rats: Implications for Functional Studies. Brain Science. 2022. DOI: https://doi.org/10.3390/brainsci12070904

[16] Omara D, Ndekezi C, Mugaba S, Nakyanzi A, Bukenya H, Bwogi J, et al.. Practical guidelines for producing non-replicating canine adenovirus vectors. PLoS ONE. 2026. DOI: https://doi.org/10.1371/journal.pone.0341642

[17] Hogans MD, Kretzschmar WP, Higgins TA, Chakrabortty A, Nance RL, Smith BF, et al.. Characterization of Canine Adenovirus Type 2 Virus Infection Pattern in Canine and Human Cell Lines. Advances in Virology. 2022. DOI: https://doi.org/10.1155/2022/3658970

[18] Oleaga Á, Balseiro A, Espí A, Royo L. Wolf (Canis lupus) as canine adenovirus type 1 (CAdV-1) sentinel for the endangered Cantabrian brown bear (Ursus arctos arctos).. Transboundary and Emerging Diseases. 2021. DOI: https://doi.org/10.1111/tbed.14010

[19] Yang D, Kim H, Lee H, Cheong Y, Hyun L, Kim M, et al.. Incidence of canine viral diseases and prevalence of virus neutralization antibodies of canine distemper virus, adenovirus type 2, parvovirus, and parainfluenza virus type 5 in Korean dogs. Korean Journal of Veterinary Research. 2024. DOI: https://doi.org/10.14405/kjvr.20230045

[20] Stevens L, Larsen L, Lysebettens WV, Carrette E, Boon P, Raedt R, et al.. Optimized Parameters for Transducing the Locus Coeruleus Using Canine Adenovirus Type 2 (CAV2) Vector in Rats for Chemogenetic Modulation Research. Frontiers in Neuroscience. 2021. DOI: https://doi.org/10.3389/fnins.2021.663337

[21] Chander V, Sharma G, Bhatt M, Nandi S, Mahajan S, Singh M, et al.. Isolation and genetic characterization of canine adenovirus type 2 from a domestic dog showing neurological symptoms. Brazilian Journal of Microbiology. 2021. DOI: https://doi.org/10.1007/s42770-021-00540-0

[22] Wang S, Wang C, Ren X, Xue W, He H, Zhu Y, et al.. Identification of Two Novel Linear Neutralizing Epitopes within the Hexon Protein of Canine Adenovirus Using Monoclonal Antibodies. Vaccines. 2021. DOI: https://doi.org/10.3390/vaccines9020135

[23] Ramidi A, Ganji V, Buddala B, Yella NR, Manthani GP, Putty K. E3 Gene-Based Genetic Characterization of Canine Adenovirus-2 Isolated from Cases of Canine Gastroenteritis in India Revealed a Novel Group of the Virus. Intervirology. 2020. DOI: https://doi.org/10.1159/000507329

[24] Lavoie A, Liu B. Canine Adenovirus 2: A Natural Choice for Brain Circuit Dissection. Frontiers in Molecular Neuroscience. 2020. DOI: https://doi.org/10.3389/fnmol.2020.00009

[25] Raja P, Sachin V, Parthiban M, Janaki PA. Molecular characterization of canine adenovirus type 2 in dogs from India. VirusDisease. 2021. DOI: https://doi.org/10.1007/s13337-021-00690-7

[26] DiGangi BA, Dingman P, Grijalva CJ, Belyeu M, Tucker S, Isaza R. Prevalence and risk factors for the presence of serum antibodies against canine distemper, canine parvovirus, and canine adenovirus in communities in mainland Ecuador.. Veterinary Immunology and Immunopathology. 2019. DOI: https://doi.org/10.1016/j.vetimm.2019.109933

[27] Ji J, Li W, Hu W, Xu X, Kan Y, Yao L, et al.. Novel Genotype Definition and the First Epidemiological Investigation of Canine Adenovirus Type 2 in Dogs in Central China. Frontiers in Veterinary Science. 2020. DOI: https://doi.org/10.3389/fvets.2020.00534

[28] Chethan G, Singh M, Chander V, Akhilesh, Singh D, Rajesh J, et al.. Occurrence of Canine parvovirus-2 and Canine adenovirus-1 Infections in Dogs: A Hospital Based Study. Indian Journal of Animal Research. 2020. DOI: https://doi.org/10.18805/IJAR.B-3936

[29] Bergmann M, Freisl M, Zablotski Y, Speck S, Truyen U, Hartmann K. Antibody Response to Canine Adenovirus-2 Virus Vaccination in Healthy Adult Dogs. Viruses. 2020. DOI: https://doi.org/10.3390/v12101198

[30] Zhu Y, Sun J, Yan M, Lian S, Hu B, Lv S, et al.. The biological characteristics of the canine adenovirus type 1 from fox and the transcriptome analysis of the infected MDCK cell. Cell Biology International. 2020. DOI: https://doi.org/10.1002/cbin.11537

[31] Verin R, Forzan M, Schulze C, Rocchigiani G, Balboni A, Poli A, et al.. Multicentric Molecular and Pathologic Study On Canine Adenovirus Type 1 in Red Foxes (Vulpes vulpes) in Three European Countries.. Journal of Wildlife Diseases. 2019. DOI: https://doi.org/10.7589/2018-12-295

[32] Wang S, Wen Y, An T, Duan G, Sun M, Ge J, et al.. Development of an Immunochromatographic Strip for Rapid Detection of Canine Adenovirus. Frontiers in Microbiology. 2019. DOI: https://doi.org/10.3389/fmicb.2019.02882

[33] Marín JFG, Royo L, Oleaga Á, Gayo E, Alarcia O, Pinto D, et al.. Canine adenovirus type 1 (CAdV‐1) in free‐ranging European brown bear (Ursus arctos arctos): A threat for Cantabrian population?. Transboundary and Emerging Diseases. 2018. DOI: https://doi.org/10.1111/tbed.13013

[34] Bohlen MO, El-Nahal HG, Sommer M. Transduction of Craniofacial Motoneurons Following Intramuscular Injections of Canine Adenovirus Type-2 (CAV-2) in Rhesus Macaques. Frontiers in Neuroanatomy. 2019. DOI: https://doi.org/10.3389/fnana.2019.00084

[35] Timurkan M, Aydın H, Alkan F. Detection and molecular characterization of canine adenovirus type 2 (CAV-2) in dogs with respiratory tract symptoms in shelters in Turkey. Veterinarski arhiv (Tisak). 2018. DOI: https://doi.org/10.24099/VET.ARHIV.0052

[36] Tryland M, Balboni A, Killengreen S, Mørk T, Nielsen O, Yoccoz N, et al.. A screening for canine distemper virus, canine adenovirus and carnivore protoparvoviruses in Arctic foxes (Vulpes lagopus) and red foxes (Vulpes vulpes) from Arctic and sub-Arctic regions of Norway. Polar Research. 2018. DOI: https://doi.org/10.1080/17518369.2018.1498678

[37] Vleeschauwer ADD, Zhou X, Lefebvre D, Garnier A, Watier F, Pignon C, et al.. A canine adenovirus type 2 vaccine vector confers protection against foot-and-mouth disease in guinea pigs.. Vaccine. 2018. DOI: https://doi.org/10.1016/j.vaccine.2018.02.074

[38] Pizzurro F, Marcacci M, Zaccaria G, Orsini M, Cito F, Rosamilia A, et al.. Genome Sequence of Canine Adenovirus Type 1 Isolated from a Wolf (Canis lupus) in Southern Italy. Genome Announcements. 2017. DOI: https://doi.org/10.1128/genomeA.00225-17

[39] Hu Y, Han X, Zhu S, Zhang Y, Peng Y, Zhang C, et al.. A replication-defective bivalent adenovirus-vectored vaccine provides robust and durable protection against both canine distemper virus and canine parvovirus.. Veterinary Microbiology. 2026. DOI: https://doi.org/10.1016/j.vetmic.2026.110933

[40] Walker D, Fee S, Hartley G, Learmount J, O'Hagan M, Meredith A, et al.. Serological and molecular epidemiology of canine adenovirus type 1 in red foxes (Vulpes vulpes) in the United Kingdom. Scientific Reports. 2016. DOI: https://doi.org/10.1038/srep36051

[41] Martín-Carrasco C, Delgado-Bonet P, Tomeo-Martín B, Pastor J, Riva Cdl, Palau-Concejo P, et al.. Safety and Efficacy of an Oncolytic Adenovirus as an Immunotherapy for Canine Cancer Patients. Veterinary Sciences. 2022. DOI: https://doi.org/10.3390/vetsci9070327

[42] Balboni A, Dondi F, Prosperi S, Battilani M. Development of a SYBR Green real-time PCR assay with melting curve analysis for simultaneous detection and differentiation of canine adenovirus type 1 and type 2.. Journal of Virological Methods. 2015. DOI: https://doi.org/10.1016/j.jviromet.2015.05.009

[43] Simão D, Pinto C, Fernandes P, Peddie C, Piersanti S, Collinson L, et al.. Evaluation of helper-dependent canine adenovirus vectors in a 3D human CNS model. Gene Therapy. 2015. DOI: https://doi.org/10.1038/gt.2015.75

[44] Boender A, Jong JDd, Boekhoudt L, Luijendijk M, Plasse Gvd, Adan R. Combined Use of the Canine Adenovirus-2 and DREADD-Technology to Activate Specific Neural Pathways In Vivo. PLoS ONE. 2014. DOI: https://doi.org/10.1371/journal.pone.0095392

[45] Serratrice N, Cubizolle A, Ibanes S, Mestre-Frances N, Bayó-Puxan N, Creyssels S, et al.. Corrective GUSB transfer to the canine mucopolysaccharidosis VII cornea using a helper-dependent canine adenovirus vector. Journal of Controlled Release. 2014. DOI: https://doi.org/10.1016/j.jconrel.2014.02.022

[46] McRee A, Wilkes R, Dawson J, Parry R, Foggin C, Adams HR, et al.. Serological detection of infection with canine distemper virus, canine parvovirus and canine adenovirus in communal dogs from Zimbabwe.. Journal of the South African Veterinary Association. 2014. DOI: https://doi.org/10.4102/jsava.v85i1.1110

[47] Broutin M, Costa F, Peltier S, Mayé J, Versillé N, Klonjkowski B. An Oil-Based Adjuvant Improves Immune Responses Induced by Canine Adenovirus-Vectored Vaccine in Mice. Viruses. 2023. DOI: https://doi.org/10.3390/v15081664

[48] Li Y, Zhao Y, Wang W, Qi B, Lv Y, Wang S, et al.. Canine adenovirus type 1 induces PANoptosis in canine splenocytes via AIM2 inflammasome activation.. Veterinary Microbiology. 2026. DOI: https://doi.org/10.1016/j.vetmic.2026.110965

[49] Hashimoto N, Tazawa H, Kikuchi S, Kuroda S, Urata Y, Kagawa S, et al.. Abstract 7241: Combination of CDK4/6 inhibitor promotes the antitumor effect of oncolytic adenovirus against canine breast cancer cells. Cancer Research. 2024. DOI: https://doi.org/10.1158/1538-7445.am2024-7241

[50] Martel A, Elseedy H, Lavigne M, Scapula J, Ghestem A, Kremer E, et al.. Targeted Transgene Expression in Cholinergic Interneurons in the Monkey Striatum Using Canine Adenovirus Serotype 2 Vectors. Frontiers in Molecular Neuroscience. 2020. DOI: https://doi.org/10.3389/fnmol.2020.00076

[51] Hornsey SJ, Philibert H, Godson D, Snead ECR. Canine adenovirus type 1 causing neurological signs in a 5-week-old puppy. BMC Veterinary Research. 2019. DOI: https://doi.org/10.1186/s12917-019-2173-5

[52] Headley SA, Oliveira TE, Pereira AHT, Moreira JR, Michelazzo MZ, Pires BG, et al.. Canine morbillivirus (canine distemper virus) with concomitant canine adenovirus, canine parvovirus-2, and Neospora caninum in puppies: a retrospective immunohistochemical study. Scientific Reports. 2018. DOI: https://doi.org/10.1038/s41598-018-31540-0

[53] Yang D, Kim H, Yoon S, Ji M, Cho I. Incidence and Sero-survey of Canine Adenovirus Type 2 in Various Animal Species. Journal of Bacteriology and Virology. 2018. DOI: https://doi.org/10.4167/JBV.2018.48.3.102

[54] Abayli H, Aslan O, Tumer K, Can-Sahna K, Tonbak S. Investigation of canine chaphamaparvovirus, canine bufavirus, and canine adenovirus in dogs with diarrhea: First report of novel canine bufavirus in Turkey. Virus genes. 2023. DOI: https://doi.org/10.1007/s11262-023-01982-4

[55] Xiao L, Zhang M, Tian Z, Ge Y, Zhang T, Yi L, et al.. Development of a Recombinase Polymerase Amplification Fluorescence Assay for the Detection of Canine Adenovirus 2. Frontiers in Veterinary Science. 2021. DOI: https://doi.org/10.3389/fvets.2021.596877

[56] Balboni A, Musto C, Kaehler E, Verin R, Caniglia R, Fabbri E, et al.. Genetic Characterization of Canine Adenovirus Type 1 Detected by Real-Time Polymerase Chain Reaction in an Oral Sample of An Italian Wolf ( Canis lupus).. Journal of Wildlife Diseases. 2019. DOI: https://doi.org/10.7589/2018-08-206

[57] Duarte M, Henriques AM, Lima C, Ochoa C, Mendes F, Monteiro M, et al.. Fatal canine adenovirus type 1 acute infection in a Yorkshire Terrier puppy in Portugal: a case report. Veterinarni Medicina. 2018. DOI: https://doi.org/10.17221/7482-VETMED

[58] Needle D, Marr JL, Park CJ, Andam CP, Wise A, Maes R, et al.. Concurrent Infection of Skunk Adenovirus-1, Listeria monocytogenes, and a Regionally Specific Clade of Canine Distemper Virus in One Gray Fox (Urocyon cinereoargenteus) and Concurrent Listeriosis and Canine Distemper in a Second Gray Fox. Pathogens. 2020. DOI: https://doi.org/10.3390/pathogens9070591

[59] Klimova AA, Komarova AA, Kiselev AM, Galkina TS. Testing of vaccine against canine distemper, parvovirus and coronavirus enteritis, adenovirus infection and dog rabies for its antigenic properties. Veterinary Science Today. 2025. DOI: https://doi.org/10.29326/2304-196x-2025-14-2-179-185

[60] Fu Y, Sun J, Lian S, Deng X, Zhang L, Shao J, et al.. Immunogenicity of an Inactivated Canine Adenovirus Type 1 Vaccine for Foxes. Frontiers in Veterinary Science. 2022. DOI: https://doi.org/10.3389/fvets.2022.678671

[61] Yang D, Kim H, Yoo J, Ji M, Han B, Oh S, et al.. Immunogenicity of a new, inactivated canine adenovirus type 2 vaccine for dogs. Clinical and experimental vaccine research. 2020. DOI: https://doi.org/10.7774/cevr.2020.9.1.40

[62] Li X, Wang X, Xia L, Weng Y, Hernandez JA, Tu L, et al.. Protective efficacy of recombinant canine adenovirus type-2 expressing TgROP18 (CAV-2-ROP18) against acute and chronic Toxoplasma gondii infection in mice. BMC Infectious Diseases. 2015. DOI: https://doi.org/10.1186/s12879-015-0815-1

[63] Gonzalez SE, Gogal R, Meindl AG, Boyer N, Nelson S, Everett SE, et al.. Influence of age and vaccination interval on canine parvovirus, distemper virus, and adenovirus serum antibody titers.. Veterinary Immunology and Immunopathology. 2023. DOI: https://doi.org/10.1016/j.vetimm.2023.110630

[64] Wang B, Xu J, Zhang H, Lian S, Duan Y, Zhang H, et al.. Establishment of ELISA method for canine adenovirus type 1. Frontiers in Veterinary Science. 2024. DOI: https://doi.org/10.3389/fvets.2024.1440124

[65] Egerer A, Schaefer Z, Larson L. A point-of-care dot blot ELISA assay for detection of protective antibody against canine adenovirus, canine parvovirus, and canine distemper virus is diagnostically accurate.. Journal of the American Veterinary Medical Association. 2022. DOI: https://doi.org/10.2460/javma.22.05.0224

[66] Yang D, Kim HH, Lee S, Oh D, Yoo J, Hyun B. Development of indirect ELISA for the detection of canine adenovirus type 2 antibodies in dog sera. Journal of Veterinary Sciences. 2020. DOI: https://doi.org/10.4142/jvs.2020.21.e63

[67] Yang D, Kim H, Lee S, Ji M, Han B, Oh S, et al.. Development of an Indirect ELISA Featuring Plates Coated with Column Chromatographically Purified Canine Adenovirus Type-1 Antigen. Journal of Bacteriology and Virology. 2020. DOI: https://doi.org/10.4167/jbv.2020.50.1.017

[68] Yang D, Kim H, Lee E, Yoo J, Yoon S, Park J, et al.. Recharacterization of the Canine Adenovirus Type 1 Vaccine Strain based on the Biological and Molecular Properties. Journal of Bacteriology and Virology. 2019. DOI: https://doi.org/10.4167/jbv.2019.49.3.124

[69] Yang D, Kim HH, Yoon S, Lee H, Cho I. Isolation and identification of canine adenovirus type 2 from a naturally infected dog in Korea. Korean Journal of Veterinary Research. 2018. DOI: https://doi.org/10.14405/KJVR.2018.58.4.177

[70] Tamukai K, Minami S, Kurihara R, Shimoda H, Mitsui I, Maeda K, et al.. Molecular evidence for vaccine-induced canine distemper virus and canine adenovirus 2 coinfection in a fennec fox. Journal of Veterinary Diagnostic Investigation. 2020. DOI: https://doi.org/10.1177/1040638720934809

[71] Du X, Goffin E, Gillard L, Machiels B, Gillet L. A Single Oral Immunization with Replication-Competent Adenovirus-Vectored Vaccine Induces a Neutralizing Antibody Response in Mice against Canine Distemper Virus. Viruses. 2022. DOI: https://doi.org/10.3390/v14091847

[72] Kim D, Han S, Go H, Kim D, Kim J, Lee J, et al.. Antiviral activity of canine interferon lambda 3 expressed using a recombinant adenovirus against canine coronavirus, canine parvovirus, and canine distemper virus. Veterinary research communications. 2022. DOI: https://doi.org/10.1007/s11259-022-10000-1

[73] Matsugo H, Kitamura-Kobayashi T, Kamiki H, Ishida H, Sekine W, Takenaka-Uema A, et al.. A potential bat adenovirus-based oncolytic virus targeting canine cancers. Scientific Reports. 2021. DOI: https://doi.org/10.1038/s41598-021-96101-4

[74] Yan L, Zhao Z, Xue X, Zheng W, Xu T, Liu L, et al.. A Bivalent Human Adenovirus Type 5 Vaccine Expressing the Rabies Virus Glycoprotein and Canine Distemper Virus Hemagglutinin Protein Confers Protective Immunity in Mice and Foxes. Frontiers in Microbiology. 2020. DOI: https://doi.org/10.3389/fmicb.2020.01070