Canine Parvovirus Variants: CPV-2a, CPV-2b, and CPV-2c

Overview and Taxonomy of Canine Parvovirus Variants: CPV-2a, CPV-2b, and CPV-2c

The emergence of canine parvovirus type 2 (CPV-2) in the late 1970s represented a watershed moment in veterinary virology, precipitating a global pandemic of acute hemorrhagic gastroenteritis and myocarditis in domestic and wild canids [6, 11, 26]. The virus, a member of the species Protoparvovirus carnivoran 1 within the genus Protoparvovirus of the family Parvoviridae, is a small, non-enveloped, single-stranded DNA virus characterized by a remarkably high mutation rate that approximates that of many RNA viruses [7, 19]. This intrinsic genetic plasticity, driven by a fast-evolving DNA polymerase, has been the primary engine behind the rapid and successive emergence of three distinct antigenic variants, CPV-2a, CPV-2b, and CPV-2c, which have completely supplanted the original CPV-2 strain in the global canine population [7, 11, 26]. Understanding the taxonomy, molecular basis, and evolutionary trajectory of these variants is not merely an academic exercise; it is fundamental to comprehending the shifting landscape of vaccine efficacy, diagnostic accuracy, and epidemiological surveillance.

The Molecular Foundation of Variant Classification: The VP2 Capsid Protein

The taxonomic differentiation of CPV-2 variants rests almost exclusively on the analysis of the major capsid protein, VP2, which constitutes approximately 90% of the viral capsid and is the primary determinant of antigenicity, host range, and receptor binding [2, 4, 8, 31]. The VP2 gene, a 1755-base-pair open reading frame encoding 584 amino acids, is the most variable region of the CPV genome and the target of neutralizing antibodies [4, 14, 31]. The canonical classification system, established through decades of molecular epidemiological work, defines the three variants based on specific non-synonymous substitutions at a single critical residue, amino acid 426, within the VP2 protein [2, 8, 26, 27].

• CPV-2a: Characterized by the presence of asparagine (Asn) at residue 426 [2, 8, 27].
• CPV-2b: Defined by the substitution of asparagine with aspartic acid (Asp) at residue 426 [2, 8, 27].
• CPV-2c: Distinguished by the replacement of aspartic acid with glutamic acid (Glu) at residue 426 [2, 8, 27, 29].

This single amino acid change at position 426, located within a major neutralizing epitope on the three-dimensional structure of the capsid, has profound implications for antigenic escape and host-virus interaction [13, 30]. Indeed, cross-antigenic evaluations using serum neutralization and hemagglutination inhibition assays have demonstrated clear antigenic differences among the variants, with CPV-2c exhibiting a particularly unique antigenic profile. Sera from animals immunized with original CPV-2, CPV-2a, or CPV-2b often show poor recognition of CPV-2c, raising significant concerns about the breadth of cross-protection afforded by vaccines based on older genotypes [30]. This antigenic drift is further compounded by a constellation of additional mutations in the VP2 gene that have accumulated over time, giving rise to "new" or "Asian-like" variants that further complicate the taxonomic landscape [4, 8, 9, 17].

The Emergence and Global Succession of Antigenic Variants

The evolutionary timeline of CPV-2 is a compelling narrative of viral emergence and displacement. The original CPV-2, first identified in the 1970s, was rapidly replaced within two years by CPV-2a, a variant that displayed an expanded host range and enhanced pathogenicity [3, 26]. This was followed by the emergence of CPV-2b approximately 14 years later, and then CPV-2c was first detected in Italy in 2000, marking a significant turning point in global CPV epidemiology [3, 26, 29]. Since its initial detection, CPV-2c has demonstrated a remarkable capacity for global dissemination, progressively replacing CPV-2a and CPV-2b as the dominant variant in numerous regions across Asia, South America, North America, and Africa [6, 9, 15, 21, 22, 24].

This global shift is not a uniform process but is characterized by complex regional dynamics. For instance, in China, a comprehensive analysis of VP2 sequences from 1978 to 2022 revealed that CPV-2c has become the dominant genotype, with a particularly aggressive Asian lineage bearing hallmark mutations such as A5G and Q370R [6, 15, 21]. Similarly, in Vietnam, a 2016-2018 survey found that CPV-2c constituted a staggering 96.54% of all isolates, completely eclipsing CPV-2a and CPV-2b [24]. In contrast, the epidemiological picture in South America is more heterogeneous. While CPV-2c is highly prevalent in Argentina and parts of Brazil, Uruguay has witnessed a complete replacement of CPV-2c by CPV-2a since 2014, illustrating the fluid and unpredictable nature of variant circulation [12]. In Europe, a longitudinal study in Italy from 1994 to 2017 showed that CPV-2a was initially predominant, but CPV-2c has since surged, with the Asian lineage of CPV-2c progressively replacing the European lineage [7, 9]. This dynamic replacement underscores the ongoing selective pressures acting on the virus, likely driven by host immunity, vaccination strategies, and the inherent fitness of emerging strains.

Beyond Residue 426: The Rise of "New" and "Asian-Like" Variants

The classification of CPV-2 variants has become increasingly nuanced as whole-genome and full-length VP2 sequencing has become routine. It is now clear that the simple tripartite classification based solely on residue 426 is insufficient to capture the full genetic diversity of circulating strains [8, 20]. A suite of additional amino acid substitutions in the VP2 protein, particularly at residues 87, 101, 267, 297, 300, 305, 324, 370, and 440, have been identified and are used to define "new" CPV-2a, "new" CPV-2b, and "Asian-like" variants [2, 4, 8, 10, 15, 17, 21].

These mutations are not silent; they are frequently located in regions of the capsid that are critical for receptor binding and immune recognition. For example, the substitutions F267Y, Y324I, and Q370R are now considered signature mutations of the Asian CPV-2c lineage, which has been detected with increasing frequency in Europe, Africa, and the Americas [9, 17, 18, 21]. The Q370R mutation, in particular, has been identified as a site under positive selection pressure, suggesting it confers a significant fitness advantage, potentially by altering the electrostatic surface of the capsid and enhancing binding to the transferrin receptor [10, 21]. Similarly, the S297A and A300G substitutions are common in new CPV-2a and new CPV-2b variants and have been linked to altered antigenicity and host range [2, 4, 5, 8]. The emergence of these "new" variants, which often carry a constellation of mutations that distinguish them from the prototype strains first identified in the 1980s and 1990s, has led to the recognition of distinct phylogenetic clades. For instance, the "Asia I" clade encompasses many of the recent CPV-2a and CPV-2c strains from Asia, Africa, and South America, while the "Europe I" clade contains older CPV-2c strains from Europe and the Americas [8, 12, 20]. This phylogenetic structure highlights the role of geographic isolation and long-distance viral exchange in shaping the global diversity of CPV-2.

Co-Infection and Recombination: Drivers of Genetic Diversity

The taxonomic complexity of CPV-2 is further amplified by the phenomena of co-infection and recombination. Deep-sequencing analyses have revealed that individual dogs can be simultaneously infected with multiple CPV-2 variants, creating a "melting pot" for genetic exchange [25]. This co-infection provides the necessary conditions for homologous recombination, a process that can generate chimeric viruses with novel combinations of genetic material. A landmark study utilizing deep sequencing identified a recombinant strain arising from an inter-genotypic recombination event between CPV-2c and CPV-2a strains at the VP1/VP2 gene boundary [25]. Such recombination events can accelerate viral evolution, potentially allowing the virus to rapidly acquire advantageous mutations from different lineages and escape host immune pressures. The detection of recombinant strains, while still relatively rare, underscores the importance of continuous genomic surveillance to track the emergence of novel viral forms that may not fit neatly into the existing taxonomic framework.

Implications for Global Health and Surveillance

The World Organisation for Animal Health (WOAH) recognizes CPV-2 as a significant pathogen of canids, and the continuous evolution of its variants poses a persistent challenge to disease control. The taxonomic framework outlined here is not static; it is a living system that must be continually updated as new genetic data emerge. The shift from a simple classification based on a single amino acid to a more complex, phylogenetically-informed taxonomy reflects the reality of a rapidly evolving virus. The dominance of CPV-2c, particularly the Asian lineage, in many parts of the world has direct implications for vaccine efficacy. While many current vaccines, including those based on the original CPV-2 or CPV-2b, have demonstrated cross-protection in laboratory settings, the emergence of antigenically distinct variants raises legitimate concerns about breakthrough infections in vaccinated animals [1, 4, 15, 30]. Indeed, numerous studies have reported CPV-2 infection in vaccinated dogs, highlighting the potential for antigenic drift to erode vaccine-induced immunity [2, 3, 7, 16, 23, 28]. Therefore, a deep understanding of the taxonomy and evolutionary trajectory of CPV-2a, CPV-2b, and CPV-2c is indispensable for designing next-generation vaccines, optimizing diagnostic assays, and implementing effective global surveillance strategies to mitigate the impact of this devastating pathogen.

Molecular Pathogenesis: Genetic Determinants of CPV-2 Variants (VP2 Gene Mutations and Structural Implications)

The molecular pathogenesis of canine parvovirus type 2 (CPV-2) is fundamentally governed by the genetic architecture of its major capsid protein, VP2. This 64–65 kDa structural protein, encoded by the VP2 gene spanning approximately 1,755 nucleotides, constitutes the primary determinant of antigenicity, host range, receptor binding, and tissue tropism [2, 11, 31]. The emergence and global dissemination of CPV-2 variants, CPV-2a, CPV-2b, and CPV-2c, represent a paradigm of viral evolution driven by cumulative point mutations within the VP2 coding sequence, particularly at residue 426, which defines the canonical antigenic classification [2, 4, 8]. However, a comprehensive examination of the VP2 gene across temporally and geographically diverse isolates reveals a far more intricate mutational landscape, with multiple residues under positive selection pressure that collectively modulate capsid structure, receptor interactions, and immune evasion.

The VP2 Protein as the Molecular Epicenter of Variant Emergence

VP2 is the sole component of the CPV-2 icosahedral capsid, assembled as 60 copies that form a T=1 symmetry shell approximately 260 Å in diameter [11, 26]. The capsid surface is characterized by three major structural features: the three-fold spike, the two-fold depression, and the five-fold cylinder, which collectively define the topographical landscape for host receptor engagement and antibody recognition [26]. The three-fold spike, formed by loops emanating from the VP2 β-barrel core, constitutes the most variable region of the capsid and contains the primary neutralizing epitopes targeted by the host humoral immune response [42]. Consequently, mutations in these exposed loop regions, particularly those involving residues 93, 300, 323, 324, 426, and 440, have profound implications for viral fitness, transmissibility, and vaccine breakthrough potential [4, 6, 8, 15].

The evolutionary trajectory from the original CPV-2 to the contemporary antigenic variants began within approximately two years of the virus's initial emergence in the late 1970s, when CPV-2a appeared, followed by CPV-2b in the 1980s and CPV-2c in 2000 [3, 26, 29]. This rapid succession of variants, occurring over a period of unprecedented evolutionary acceleration for a single-stranded DNA virus, was driven by specific amino acid substitutions that altered capsid surface chemistry without compromising the structural integrity required for viral assembly and infectivity [20, 26]. The VP2 gene of CPV-2 evolves at a rate approaching that of RNA viruses, approximately 10⁻⁴ to 10⁻⁵ substitutions per site per year, enabling the virus to explore sequence space and adapt to immunological pressures imposed by widespread vaccination [7, 19, 38].

The Canonical Determinant: Residue 426 and Antigenic Classification

The defining genetic determinant that distinguishes the three major CPV-2 variants resides at amino acid position 426 of the VP2 protein, located within a prominent loop region of the three-fold spike that constitutes a major antigenic epitope [2, 8, 26, 27, 33]. In the original CPV-2 and CPV-2a, residue 426 is occupied by asparagine (Asn, N); in CPV-2b, this residue is substituted by aspartic acid (Asp, D); and in CPV-2c, the substitution involves glutamic acid (Glu, E) [2, 8, 13, 26, 35, 40]. The functional consequences of these substitutions extend beyond simple antigenic reclassification. Aspartic acid and glutamic acid are both negatively charged amino acids, whereas asparagine is neutral and polar. The introduction of a negative charge at position 426 in CPV-2b and CPV-2c alters the electrostatic potential of the capsid surface, potentially affecting interactions with host cell receptors and neutralizing antibodies [8, 26, 30].

Critically, the N426E substitution characteristic of CPV-2c confers a unique antigenic profile that has been demonstrated to reduce recognition by antibodies raised against CPV-2, CPV-2a, and CPV-2b [30]. Early neutralization studies using monoclonal antibodies revealed that CPV-2c isolates lacked reactivity with several antibodies that effectively neutralized earlier variants, indicating that residue 426 sits within a conformational epitope critical for antibody-mediated neutralization [42]. Cross-neutralization assays further demonstrated that sera from animals immunized with CPV-2, CPV-2a, or CPV-2b exhibited significantly lower neutralizing titers against CPV-2c compared to the homologous viruses [30]. This antigenic distinctiveness has raised concerns regarding the efficacy of vaccines based on original CPV-2 or CPV-2b against emerging CPV-2c field strains, although recent studies have demonstrated that certain modern vaccines, such as Vanguard C4, can induce cross-neutralizing antibodies against all three variants [1].

Expanding the Mutational Repertoire: Beyond Residue 426

While residue 426 provides the taxonomic framework for CPV-2 classification, comprehensive full-length VP2 sequencing from global surveillance efforts has identified numerous additional amino acid substitutions that contribute to the molecular diversity and adaptive capacity of circulating strains. These mutations cluster within discrete structural and functional domains of VP2, each with distinct implications for viral biology.

The GH Loop and the "New CPV-2a/2b" Signature Mutations: One of the most consequential sets of mutations involves residues 297 (alanine to serine, A297S or alanine to glycine, S297A) and 300 (alanine to glycine, A300G or glycine to alanine, G300A), located within the GH loop of VP2, a structurally flexible region that constitutes a major antigenic site [2, 5, 8, 10, 15, 36]. These substitutions, often accompanied by mutations at residue 305 (tyrosine to aspartic acid, Y305D), define the "new CPV-2a" and "new CPV-2b" variants that have largely replaced the original CPV-2a and CPV-2b in many regions worldwide [5, 15, 22, 26, 36]. The S297A substitution is particularly prevalent, having been detected in CPV-2a, CPV-2b, and CPV-2c isolates from South America, Europe, Asia, and Africa [2, 4, 5, 8, 10, 36]. Structural modeling indicates that residue 297 sits at the apex of the three-fold spike, and the serine-to-alanine change alters the hydrogen bonding capacity of this loop, potentially modifying the conformation of adjacent antigenic epitopes and affecting antibody binding [15, 19].

The Emerging Asian Clade and the A5G/Q370R Mutations: A particularly noteworthy evolutionary development is the emergence and global expansion of an "Asian lineage" of CPV-2c characterized by two distinctive mutations: alanine to glycine at residue 5 (A5G) and glutamine to arginine at residue 370 (Q370R) [6, 9, 10, 15, 17, 21, 24, 32, 37]. These mutations, first documented in Chinese isolates in the early 2010s, have since been detected in CPV-2c strains circulating in Italy, Romania, Nigeria, Egypt, Thailand, Korea, and Vietnam, indicating a progressive global diffusion of this lineage [9, 17, 18, 21, 24, 37]. Residue 5 is located in the N-terminal region of VP2, which is thought to be involved in nuclear translocation and capsid assembly, while residue 370 resides on the capsid surface near the three-fold spike [10]. The Q370R substitution introduces a positively charged arginine at a position that is highly conserved across carnivore protoparvoviruses, suggesting that this mutation may confer a selective advantage, potentially through enhanced receptor binding or immune evasion [10, 21]. Importantly, selection pressure analyses have identified both residues 5 and 370 as being under positive selection and parallel evolution, indicating that these mutations are adaptive rather than neutral [21].

Mutations at Residues 267, 324, and 440: Modulating Antigenicity and Receptor Binding: Several additional residues have emerged as critical determinants of VP2 structure and function. The F267Y (phenylalanine to tyrosine) substitution is nearly ubiquitous in contemporary CPV-2a, CPV-2b, and CPV-2c isolates and is considered a canonical marker of the "new" variants [10, 15, 17, 18]. Residue 267 is located within the VP2 β-barrel core, proximal to the sialic acid binding pocket that mediates hemagglutination [10, 15]. The tyrosine substitution introduces a hydroxyl group that may alter the hydrogen bonding network within this pocket, potentially affecting the virus's ability to bind sialic acid receptors on erythrocytes and host cells [15].

The Y324I (tyrosine to isoleucine) substitution represents another hallmark of emerging CPV-2c variants of Asian origin [10, 15, 17, 18, 34]. This mutation is located in a loop region adjacent to the three-fold spike and has been shown to be under positive selection in CPV-2c populations [4]. Structural modeling predicts that the replacement of a bulky aromatic tyrosine with a smaller, hydrophobic isoleucine at this position alters the local surface topology, potentially exposing or masking adjacent antigenic determinants [15]. Intriguingly, a novel CPV-2a mutant with a Y324L (tyrosine to leucine) substitution was identified in Italy, representing the first evidence of this mutation in the CPV-2a background and underscoring the ongoing evolutionary experimentation within the VP2 gene [19].

Residue 440, located near the three-fold spike, exhibits variability among CPV-2c strains, with the T440A (threonine to alanine) or A440T (alanine to threonine) substitutions being commonly reported [4, 15, 18]. This residue is under positive selection in CPV-2c populations and has been implicated in antigenic drift, as antibodies raised against earlier CPV-2c strains may show reduced binding to isolates carrying the 440 mutation [4].

Structural Implications of VP2 Mutations: From Atomic to Functional Consequences

The cumulative effect of VP2 mutations is not merely additive; rather, mutations at spatially distant residues can act synergistically to remodel the capsid surface, alter its biophysical properties, and shift its antigenic profile. The three-fold spike, formed by interdigitating loops from three VP2 monomers, presents a large, contiguous surface area for antibody recognition [42]. Mutations at residues 93, 297, 300, 305, 324, and 426, all of which are surface-exposed on or adjacent to the three-fold spike, can collectively reshape this immunodominant region, enabling the virus to escape neutralizing antibodies while preserving the structural framework required for receptor engagement [4, 8, 15, 42].

The sialic acid binding site, located in a depression near the two-fold axis of symmetry, is another hotspot for functionally relevant mutations. Residues 93 (lysine to asparagine, K93N) and 323 (asparagine to aspartic acid, N323D) within this pocket have been implicated in modulating host range, particularly the ability of CPV-2 variants to infect feline cells [10, 26]. The original CPV-2 could not replicate efficiently in cats, whereas CPV-2a, CPV-2b, and CPV-2c have acquired mutations that expand their host range to include felids, a shift that has significant ecological and epidemiological implications for cross-species transmission [26, 41].

Recent structural modeling studies using CPV-2c strains from East China have provided atomic-level predictions of how specific mutations alter the tertiary structure of VP2. The L87M, T101I, Y267F, A297S, G300A, Y305D, I324Y, Q370R, N426E, A440T, and I447M substitutions were predicted to induce localized conformational changes that could alter the electrostatic surface potential, hydrogen bonding patterns, and loop flexibility [15]. The I447M mutation, first reported in Chinese CPV-2c strains, is of particular interest because residue 447 is located within a conserved structural element that may influence capsid stability and uncoating dynamics [15, 21].

Selection Pressures and Evolutionary Dynamics Shaping the VP2 Gene

Despite the overall high conservation of the VP2 gene, which is maintained under predominantly negative (purifying) selection that eliminates deleterious mutations, specific codons are subject to positive selection driven by host immune pressure [4, 21, 32]. Selection pressure analyses conducted on global CPV-2 datasets have identified residues 5, 297, 324, 370, 426, and 440 as positively selected sites, consistent with their location in antigenic epitopes or receptor-binding domains [4, 21]. The identification of parallel evolution at residues 5, 370, and 426, where identical amino acid changes have arisen independently in different geographic lineages, provides compelling evidence that these mutations confer a fitness advantage and are not simply stochastic [21].

The evolutionary dynamics of CPV-2 VP2 are further complicated by co-infection and recombination events. Deep-sequencing analyses of canine fecal samples from Uruguay revealed that co-infection with distinct CPV-2 variants is not uncommon, and in one instance, inter-genotypic recombination between CPV-2c and CPV-2a strains was detected at the VP1/VP2 gene boundary [25]. This recombination event generated a chimeric virus carrying the VP2 capsid from one parent and the non-structural proteins from another, potentially combining advantageous traits from both lineages [25]. Although recombination is infrequent in parvoviruses compared to substitution mutations, its occurrence adds another dimension to the genetic plasticity of CPV-2 and the potential for rapid phenotypic change.

Implications for Vaccine Design and Molecular Surveillance

The detailed characterization of VP2 mutations and their structural consequences has direct implications for veterinary vaccinology and public health. The original CPV-2 and CPV-2b strains used in many commercial vaccines are increasingly divergent from the contemporary field strains, particularly the Asian-lineage CPV-2c variants that now dominate in many regions [1, 15, 24, 26, 37]. The accumulation of mutations in the GH loop and three-fold spike, the very regions that constitute the immunodominant epitopes targeted by vaccine-induced antibodies, raises the theoretical risk of vaccine escape, although cross-protection studies have shown that current vaccines can still induce neutralizing antibodies against heterologous variants [1]. However, the emergence of CPV-2c with unique antigenic features, combined with the high prevalence of this variant in vaccinated dog populations, underscores the need for continuous molecular surveillance and periodic reassessment of vaccine strain composition [4, 15, 26, 39].

The molecular pathogenesis of CPV-2 variants is thus a dynamic interplay between viral genetic determinants and host selective forces, encoded in the ever-evolving VP2 gene. Each amino acid substitution carries the potential to alter capsid structure, receptor specificity, antigenicity, and ultimately, the clinical and epidemiological profile of the virus. The ongoing global surveillance of VP2 sequences, coupled with structural biology and functional assays, remains essential for predicting future evolutionary trajectories and informing evidence-based disease control strategies.

Epidemiology of CPV-2 Variants: Global and Regional Prevalence Patterns

The epidemiology of canine parvovirus type 2 (CPV-2) represents a dynamic and rapidly evolving landscape, characterized by the progressive replacement of the original CPV-2 strain by its antigenic variants, CPV-2a, CPV-2b, and CPV-2c, and their subsequent, ongoing diversification into distinct phylogenetic lineages with differential geographic distributions [1, 6, 26]. Since the emergence of CPV-2 in the late 1970s, the virus has demonstrated a remarkable capacity for genetic and antigenic change, driven by a single-stranded DNA genome that, paradoxically, evolves at rates approaching those of RNA viruses [7, 26]. This evolutionary trajectory has resulted in a complex global pattern wherein the three primary variants co-circulate with varying frequencies, and newer sub-variants or mutants (e.g., "new CPV-2a," "new CPV-2b," and "Asian-like" CPV-2c) are continuously emerging and displacing older strains [6, 15, 19]. Understanding these intricate prevalence patterns is critical for assessing vaccine efficacy, predicting future outbreaks, and implementing effective control measures, as the World Organisation for Animal Health (WOAH) recognizes CPV-2 as a globally significant pathogen of canids [26, 51].

Global Distribution and Temporal Shifts

The triumvirate of CPV-2 variants, 2a, 2b, and 2c, has achieved a truly worldwide distribution, yet their relative proportions are neither static nor uniform [1, 11, 49]. A comprehensive analysis of all available VP2 sequences from the NCBI database, spanning from 1978 to 2022, revealed a profound global trend: CPV-2c is progressively replacing CPV-2a as the dominant variant across multiple continents, including Asia, South America, North America, and Africa [6]. This shift is not a recent phenomenon but has been unfolding over the past two decades. The original CPV-2 was rapidly supplanted by CPV-2a in the early 1980s, which was then partially overtaken by CPV-2b in the 1990s, and finally, CPV-2c emerged in 2000 in Italy and has been expanding its geographical and ecological niche ever since [1, 26, 29]. The driving forces behind this succession are likely multifaceted, involving subtle antigenic differences that allow for partial immune escape in vaccinated populations, differential fitness and transmissibility, and stochastic founder effects during viral introductions into new regions [4, 30].

Bayesian coalescent analyses of global whole-genome sequences have estimated that the time to the most recent common ancestor (tMRCA) for the major circulating clades dates back to the 1980s and 1990s, originating in North America before disseminating worldwide [20]. Importantly, these analyses have demonstrated that the phylogenetic classification of CPV-2 strains does not always align neatly with the traditional antigenic typing based solely on residue 426 of the VP2 protein. Instead, the global phylogeny comprises multiple well-defined clades, some of which contain a mix of antigenic variants, indicating complex evolutionary histories involving recombination and convergent evolution [20, 25]. For instance, a whole-genome phylogenetic tree revealed two main clades distributed worldwide, while a more detailed VP2 gene tree showed four distinct clades, including one comprising sequences exclusive to Brazil [20]. This suggests that local viral populations are not merely passive recipients of globally circulating strains but are active sites of viral evolution and diversification [12, 20].

Regional Prevalence Patterns: A Continent-by-Continent Analysis

Asia: The Epicenter of CPV-2c Dominance and Novel Mutations

Asia, particularly East and Southeast Asia, has emerged as a crucial epicenter for CPV-2 evolution and the rapid expansion of CPV-2c, often carrying a distinct set of amino acid substitutions that define an "Asian lineage" [6, 22, 32]. In China, the shift towards CPV-2c dominance has been dramatic and well-documented. In Henan Province, between 2020 and 2021, CPV-2c constituted a staggering 91.84% of 98 sequenced positive samples, with new CPV-2a accounting for only the remaining 8.16% [21]. Similarly, in East China from 2018 to 2020, CPV-2c represented 77.19% of isolates, while new CPV-2a and new CPV-2b were detected at much lower frequencies [22]. A long-term temporal study in Shanghai from 2016 to 2025 confirmed this complete genotype replacement: isolates from 2016 to 2020 were predominantly new CPV-2a, but CPV-2c became the absolute dominant genotype from 2021 onwards [15]. This pattern is echoed across the region. In Vietnam, an extensive survey conducted from November 2016 to February 2018 across three regions (northern, central, and southern) found CPV-2c in 96.54% (251/260) of positive samples, with CPV-2a making up the remainder and CPV-2b being completely absent [24]. In Thailand, a retrospective analysis from 2003 to 2019 showed that CPV-2a and CPV-2b were the only variants detected until 2010; CPV-2c emerged in 2014 and by 2019 had become the major variant, completely replacing the earlier types [34]. Taiwan also reported a similar scenario, with CPV-2a and CPV-2b being displaced by CPV-2c, which accounted for 54.6% of 88 isolates from 2014 to 2016, representing the first detection of this variant on the island [48].

Critically, these Asian CPV-2c strains are not identical to the prototype CPV-2c first identified in Europe. They are characterized by a specific set of amino acid changes in the VP2 protein, most notably A5G and Q370R, and often also include F267Y, Y324I, and N426E [6, 15, 17, 18, 21, 32]. These mutations are under positive selection pressure and are considered markers of the rapidly expanding "Asian lineage" [6, 15, 21, 32]. The emergence of this lineage is believed to be driven by unique gradual point mutations in key VP2 sites within China, rather than by multiple introductions from outside [32]. This concept is supported by phylogenetic analyses showing that Chinese CPV-2c isolates, including the representative strain CPV-SH1516, form a monophyletic cluster distinct from European and American strains [22, 32]. Furthermore, the detection of Asian-like CPV-2c strains in dogs in Romania and Italy, where they were associated with acute gastroenteritis, indicates that this lineage is progressively expanding its global reach, likely through the movement of infected animals [9, 17]. The biological consequences of these mutations are significant; structural modeling suggests that they may alter the tertiary structure of the VP2 protein, potentially affecting antigenicity and receptor recognition, which raises concerns about the efficacy of vaccines based on older CPV-2 or CPV-2b genotypes [15].

South America: A Dynamic Interplay of Variants and Phylogenetic Diversity

South America presents a more heterogeneous epidemiological picture, with different countries and even regions within countries harboring distinct variant profiles. Ecuador provides a striking example of this intracontinental variability. An early study in Quito from 2018 reported a prevalence of 57.1% for CPV-2a, 34.3% for CPV-2c, and only 8.5% for CPV-2b [2]. However, a more recent study covering 2022-2023 in Ecuador found that CPV-2b had become the predominant genotype, accounting for 84.54% of positive samples, a dramatic shift in just a few years [5]. This highlights the rapid and unpredictable nature of variant fluctuations at a local level. In contrast, Colombia has reported the co-circulation of CPV-2a and CPV-2b, with CPV-2c not being detected in one major study from Antioquia, where 93.1% of amplified samples belonged to a new CPV-2a variant carrying the Ala514Ser mutation, characteristic of a South American I clade [8]. The Colombian CPV-2b strains, meanwhile, showed amino acid changes (Phe267Tyr, Tyr324Ile, Thr440Ala) that related them to the Asia-I clade, suggesting multiple and geographically distant origins for the variants circulating in the country [8].

In the Southern Cone, a comparative genomic study of samples from Argentina and Uruguay revealed contrasting evolutionary patterns. In Argentina, samples collected from 2008 to 2018 were predominantly CPV-2c [12]. In Uruguay, however, a complete variant replacement occurred, with CPV-2a variants entirely displacing CPV-2c after 2014 [12]. This study also showed that the CPV-2c strains from Argentina and Uruguay clustered within the Europe I phylogroup, while the CPV-2a strains from Uruguay formed a distinct Asia I group, further emphasizing the complex global admixture [12]. In Brazil, the historical trend shows a clear progression. In Rio de Janeiro, samples from 1995 to 2003 were all "new CPV-2a," followed by a mixed period of CPV-2a and CPV-2b from 2004 to 2006, after which CPV-2b became dominant [33, 45]. Importantly, a single CPV-2c sample was detected in 2008, marking its first identification in the state [33, 45]. A more recent study analyzing VP2 sequences from Brazilian cats also detected a mix of CPV-2a (28%) and CPV-2b (36%), alongside feline panleukopenia virus [41]. The detection of the Y324L mutation in Brazilian CPV strains from cats, a mutation not previously reported in canine strains, suggests ongoing adaptation and host-range expansion [41]. The presence of a CPV clade exclusive to Brazil in global phylogenies further underscores the role of local, endemic evolution in this region [20].

Europe: Historical Shifts and the Influx of the Asian Lineage

Europe, where the prototype CPV-2c was first identified in Italy in 2000, has a long history of CPV-2 variant surveillance [26, 29]. The epidemiological picture is one of continuous fluctuation, with temporal shifts in dominance and the recent introduction of novel lineages. In Italy, a 23-year study (1994-2017) analyzed 123 samples and found that all three variants circulated, but with notable regional and temporal differences. CPV-2a was the prominent genotype overall, followed by CPV-2c and then CPV-2b [7]. However, CPV-2a was characterized by the highest genetic variability, while CPV-2c showed notable stability with a predominant amino acid profile [7]. Interestingly, CPV-2b re-emerged in the later years of the study (2015-2017) with a new and distinctive amino acid profile [7]. In Sicily, a more recent survey from 2019 to 2022 documented a rapid and significant shift: the frequency of CPV-2a and CPV-2b decreased, while CPV-2c increased dramatically to represent 69% of the 215 samples [9]. Most strikingly, the European lineage of CPV-2c was progressively replaced by the Asian lineage (carrying A5G, Q370R), which had become dominant [9]. This finding in a geographically isolated island location demonstrates the potent ability of the Asian CPV-2c lineage to outcompete and replace established endemic strains.

In Spain, historical data from central Spain (2003-2014) showed that CPV-2c was the predominant variant (42.9% of sequenced samples), followed by CPV-2a (31.0%) and then CPV-2b (9.5%), with the Cornell vaccine strain and FPV also detected [44]. This early dominance of CPV-2c is consistent with its initial spread from Italy [29]. In other European countries, the variant picture is more varied. In Turkey, a study in the southeast Anatolia region found CPV-2b to be the most prevalent (16/18 samples), with only a single detection each of CPV-2a and CPV-2c [36]. Conversely, a more recent study in the Black Sea region of Turkey found that all six isolates characterized by whole-genome sequencing were classified as CPV-2a variants [28]. This highlights how even within a single country, regional differences can be pronounced. In Romania, ten CPV-positive samples from dogs with acute gastroenteritis were all identified as CPV-2c of Asian origin (5Gly, 267Tyr, 324Ile, 370Arg), suggesting a recent introduction and expansion of this lineage into Eastern Europe [17]. A study in Northern Italy from 2017 to 2023 using qPCR with high-resolution melting (HRM) analysis also reported a significant proportion (45.4%) of CPV-2 carrying Asian-like residues, further confirming the widespread penetration of this lineage into Europe [47].

Africa: Under-sampled but Revealing a Dominant CPV-2c Presence

Despite being a continent with vast canine populations and limited surveillance, the available data from Africa consistently point towards a high prevalence of CPV-2c, often of the emerging Asian lineage. In Nigeria, a 2018 study on 59 selected positive samples from a larger pool of 320 rectal swabs revealed a striking prevalence of 91.5% for CPV-2c and only 8.5% for CPV-2a [37]. This represented a significant shift from earlier 2010 data from Nigeria, and the VP2 gene sequences showed a close connection with CPV strains of Asian origin [37]. An earlier, smaller study from 2013-2014 in Nigeria had detected only CPV-2a in six sequenced samples, illustrating a rapid epidemiological shift over just a few years [16]. In Morocco, a study from 2011 to 2015 found a more balanced distribution, with CPV-2b at 47.7% and CPV-2c at 43.3%, along with a small proportion of CPV-2a (1.1%) and co-infections (4.4%) [50]. This suggests that the replacement by CPV-2c may not have been as complete in North Africa at that time as it was in West Africa. In Egypt, a study covering seven governorates from 2019 to 2021 confirmed the co-circulation of all three variants, but with a notable temporal shift: CPV-2b was not detected in 2020 and 2021, while CPV-2c frequency increased significantly in 2021 [10]. The Egyptian strains also clustered with Asian strains in phylogenetic analyses and carried the characteristic A5G and Q370R mutations, aligning them with the global Asian lineage [10]. In Gabon, Central Africa, the first molecular characterization of CPV-2 variants identified both CPV-2a and CPV-2c in samples from young, vaccinated dogs, with CPV-2a predominating [3]. This detection of CPV-2c in vaccinated animals raises concerns for the region, similar to findings in South America and Asia [2, 3, 7, 15]. Overall, the African data, while still sparse, aligns with the global trend of CPV-2c expansion, particularly of the Asian lineage, and underscores the urgent need for more comprehensive epidemiological surveys across the continent [3, 37].

The Indian Subcontinent: A Persistent Stronghold of CPV-2a

In contrast to many other regions, the Indian subcontinent appears to maintain a different epidemiological profile, with CPV-2a remaining the most prevalent variant. A large-scale review of CPV-2 outbreaks in India from 2010 to 2023, analyzing VP2 gene sequences from the NCBI database, found a seroprevalence of 45% for CPV-2a, followed by 36% for CPV-2c and 12% for CPV-2b [46]. This is supported by individual molecular studies. In Hyderabad, a 2024 study using PCR-RFLP detected CPV-2a as the most common type (in 17 of 67 positive samples that were not CPV-2b), with CPV-2b also being highly prevalent (50/100 samples) and CPV-2c being absent [13]. Another study from the same year in Jaipur reported a high number of CPV-2a (147) and CPV-2b (141) positives among 166 total cases, with CPV-2c (115) also being common but less frequent [43]. Earlier work in Himachal Pradesh (2016) detected only CPV-2b in affected dogs, showing that local and temporal variation occurs even within a single country [23]. The first detection of CPV-2c in India was reported in 2006 in a small study of 16 positive samples, confirming its early presence but not its dominance [40]. This persistent high prevalence of CPV-2a in India, contrasting with the CPV-2c dominance in much of Asia, suggests that regional differences in host population immunity, vaccination practices, and viral fitness may be shaping distinct evolutionary trajectories.

Emerging Variants, Co-Infections, and the Role of Vaccination

The epidemiological picture is further complicated by the continuous emergence of new mutants and the frequent occurrence of co-infections. In Italy, a novel CPV-2a mutant with a VP2 Tyr324Leu mutation was first reported, representing a previously undocumented escape mutant [19]. Similarly, a novel I447M mutation was identified in several CPV-2c isolates from Shanghai, China, and in five CPV-2c strains from Henan Province, China, highlighting the ongoing generation of genetic diversity [15, 21]. The occurrence of co-infections with multiple CPV-2 variants is not rare and has been documented in several studies, including in Morocco where co-infections with CPV-2b and CPV-2c accounted for 4.4% of cases [50]. Deep-sequencing analysis of a co-infected sample from Uruguay revealed the presence of both CPV-2c and CPV-2a strains, as well as a major recombinant strain (86.7% of the viral population) that arose from inter-genotypic recombination between the two variants within the VP1/VP2 gene boundary [25]. This demonstrates that co-infection provides a mechanism for generating novel recombinant viruses with potentially altered antigenic and pathogenic properties [25].

A recurrent and concerning finding across multiple epidemiological studies is the detection of CPV-2 variants in dogs that have been previously vaccinated [2, 3, 7, 16, 23, 28, 49]. In the Italian study from 1994 to 2017, 32.5% of the sick dogs had been vaccinated, and no statistical association was found between the CPV-2 variant and vaccination status, indicating that vaccine breakthroughs occur with all variants [7]. In Gabon, the study was triggered by clinical cases among vaccinated dogs, and the CPV-2c variant was detected in these animals [3]. Similarly, in Ecuador, CPV-2 was affecting even vaccinated puppies [2]. While the primary cause of vaccination failure is often interference by maternally derived antibodies (MDA), there is accumulating evidence that the antigenic differences between vaccine strains (often original CPV-2 or CPV-2b) and circulating field variants (especially the newly emerging CPV-2c and its Asian lineage) may contribute to reduced vaccine efficacy [4, 15, 26, 30]. Serological studies have demonstrated antigenic differences between the variants, with CPV-2c exhibiting a unique antigenic pattern and being poorly recognized by sera from animals immunized with CPV-2, CPV-2a, or CPV-2b [30]. This has led a growing number of researchers to call for an update to current CPV vaccines to better match the currently dominant field variants, particularly the widespread CPV-2c [15, 22, 26, 29].

Diagnostic Approaches for CPV-2 Variants: From Immunochromatography to Molecular Characterization

The accurate and timely diagnosis of canine parvovirus type 2 (CPV-2) infection, coupled with the precise identification of circulating antigenic variants, is a cornerstone of effective clinical management, epidemiological surveillance, and the development of robust vaccination strategies. The continuous genetic and antigenic drift of CPV-2, which has led to the global predominance of variants CPV-2a, CPV-2b, and CPV-2c, and their subsequent sub-lineages, presents a formidable diagnostic challenge [6, 9, 26]. Traditional diagnostic modalities, while still valuable in resource-limited settings, often lack the sensitivity and specificity required to differentiate between variants or to detect low viral loads, particularly in the later stages of disease or in vaccinated animals that may shed virus at lower titers [54, 56]. This section provides a comprehensive, mechanistic analysis of the diagnostic armamentarium available for CPV-2, ranging from rapid point-of-care immunochromatographic tests to advanced molecular characterization techniques, with a specific focus on the biological and epidemiological contexts that dictate their appropriate application.

Point-of-Care Diagnostics: Immunochromatography and Its Limitations

The rapid immunochromatographic test (RICT), commonly referred to as a lateral flow assay, remains the most widely deployed frontline diagnostic tool for CPV-2 in veterinary practice worldwide. Its appeal lies in its simplicity, speed (typically 5–15 minutes), and ease of use without the need for specialized laboratory equipment [54, 56]. These tests operate on the principle of antigen capture, where viral particles present in fecal samples bind to monoclonal antibodies conjugated to colloidal gold or other labels, which then migrate across a nitrocellulose membrane to be captured by a second set of immobilized antibodies, forming a visible line. Commercially available CPV-specific RICTs are designed to detect the original CPV-2 and its major variants, and they demonstrate generally acceptable sensitivity for acute clinical cases with high viral shedding [2, 53, 54].

However, the reliability of immunochromatography is profoundly influenced by the antigenic heterogeneity of circulating CPV-2 strains. A critical evaluation of six commercial RICT kits revealed that while most kits produced strong positive results for CPV-2a, CPV-2b, and CPV-2c, two kits failed to generate a positive signal for the original CPV-2 or CPV-2b at a titer of 10⁵ FAID₅₀/mL [54]. This finding underscores a fundamental biological limitation: the monoclonal antibodies employed in these kits are raised against specific epitopes on the VP2 capsid protein. As CPV-2 undergoes antigenic drift, particularly at key residues such as 426 (which defines the variant) and other immunodominant sites like 297, 300, and 324, the binding affinity of these capture and detection antibodies can be sufficiently compromised to yield false-negative results [4, 30, 54]. This is not merely a theoretical concern; the emergence of "Asian-like" CPV-2c strains carrying mutations such as A5G and Q370R, which are now globally disseminated, may further challenge the recognition profile of older-generation RICTs [6, 17, 21]. Furthermore, the sensitivity of RICTs is demonstrably lower than that of molecular methods. In a study from Egypt, only 45% of clinically suspected cases were confirmed by immunochromatography, whereas PCR identified 86% of the same cohort as positive [56]. This discrepancy highlights the risk of underdiagnosis when relying solely on RICTs, particularly in convalescent animals or those with mild or peracute disease where fecal viral loads may be near the limit of detection. Despite these limitations, the cross-reactivity of CPV-specific RICTs with feline panleukopenia virus (FPV) has been exploited for off-label use in cats, with recent studies demonstrating near-perfect diagnostic agreement (Cohen’s κ = 0.919–1.000) between canine- and feline-specific RICTs for the diagnosis of feline panleukopenia, offering a cost-effective alternative in multi-species practice [53].

Hemagglutination, Virus Isolation, and Serological Assays

Prior to the widespread adoption of molecular diagnostics, hemagglutination (HA) and hemagglutination inhibition (HI) assays were the mainstays of CPV-2 diagnosis and typing. CPV-2 possesses the ability to agglutinate porcine or feline erythrocytes, a property mediated by the VP2 protein’s interaction with sialic acid receptors [31, 33]. The HA test is straightforward and inexpensive, but its sensitivity is poor compared to PCR, and it is prone to interference from non-specific agglutinins in fecal samples. Furthermore, HA cannot differentiate between CPV-2 variants, as the hemagglutinating phenotype is largely conserved across all antigenic types [26]. Virus isolation in cell culture, typically using Madin-Darby canine kidney (MDCK) cells or feline cell lines like CRFK, remains the gold standard for definitive diagnosis and for obtaining live virus for further characterization [23, 28, 52]. However, it is labor-intensive, time-consuming (requiring multiple passages to observe cytopathic effects), and requires a BSL-2 facility. The success of isolation is also highly dependent on sample quality and the presence of viable virus, which is rapidly inactivated by environmental conditions [23, 52].

Serological assays, such as HI and serum neutralization (SN), are primarily employed for assessing post-vaccination immunity and for seroepidemiological surveys, rather than for acute diagnosis [1, 30]. The HI test measures antibodies that block hemagglutination, while SN quantifies neutralizing antibodies. Cross-neutralization studies have been instrumental in evaluating vaccine efficacy. For instance, the Vanguard C4 vaccine was shown to induce cross-neutralizing antibodies against Australian isolates of CPV-2a, CPV-2b, and CPV-2c, demonstrating a broad protective potential [1]. However, a seminal study by Cavalli et al. (2007) using SN revealed clear antigenic differences among the variants. Notably, CPV-2c exhibited a unique antigenic pattern; it was poorly recognized by sera from dogs immunized with CPV-2, CPV-2a, or CPV-2b [30]. Conversely, animals immunized with CPV-2c had higher neutralizing titers against CPV-2b than against the homologous CPV-2c virus [30]. This asymmetric cross-reactivity has profound implications for vaccine design and suggests that serological data based on a single variant may not accurately reflect population-level immunity. Monoclonal antibody (MAb)-based typing, using panels of MAbs like those described by Nakamura et al. (2003), has historically been used to distinguish antigenic variants [42]. These MAbs target specific antigenic sites (e.g., site A and site B) on the VP2 protein, and the differential reactivity patterns can differentiate CPV-2a, -2b, and -2c. However, the generation and maintenance of such MAb panels is technically demanding, and their utility is being superseded by genotyping methods.

Conventional and End-Point Polymerase Chain Reaction (PCR)

The introduction of polymerase chain reaction (PCR) revolutionized the diagnosis of CPV-2, offering orders of magnitude greater sensitivity than antigen detection methods [11, 31, 56]. Conventional PCR assays typically target a conserved region of the VP2 gene, enabling the detection of all CPV-2 variants and even FPV in a single reaction [23, 57]. The high sensitivity of PCR is critical for detecting the virus in samples with low viral loads, such as those from vaccinated dogs that may still shed a small amount of virus, or from dogs in the late stages of infection [7, 16, 57]. In a study from India, nested PCR increased the detection rate from 50% (by conventional PCR) to 89% among suspected cases, highlighting the value of this approach for maximizing diagnostic yield [57].

Genotyping of CPV-2 variants by conventional PCR is achieved through several strategies. The most straightforward approach involves amplifying the region encompassing residue 426 of the VP2 gene, followed by restriction fragment length polymorphism (RFLP) analysis. Using the MboII restriction enzyme, for example, CPV-2c can be differentiated from CPV-2a and CPV-2b based on a specific digestion pattern [13, 35, 36, 39, 40]. This PCR-RFLP method has been widely employed for epidemiological surveillance [13, 35, 36, 40]. Alternatively, variant-specific primer sets can be used in a multiplex PCR format. Kumar and Nandi (2010) used a combination of primers to first differentiate CPV-2a from CPV-2b, and then a secondary PCR to detect CPV-2c [35]. Similarly, Purushotham et al. (2024) used CPV-2ab primers to generate a 681 bp amplicon, followed by CPV-2b-specific primers to differentiate CPV-2b from CPV-2a, and finally an MboII digestion to confirm CPV-2c [13]. While effective, these multi-step conventional PCR approaches are time-consuming and increase the risk of cross-contamination, which has driven the development of more sophisticated real-time PCR methods.

Real-Time PCR and High-Resolution Melting (HRM) Analysis

Real-time quantitative PCR (qPCR) represents a significant advancement, providing not only qualitative detection but also quantification of viral DNA, which can be correlated with disease severity and prognosis [5, 27, 47]. The use of hydrolysis probes (e.g., TaqMan) or intercalating dyes (e.g., SYBR Green) allows for continuous monitoring of amplification, eliminating the need for post-PCR processing and reducing turnaround time. For variant differentiation, several elegant qPCR strategies have been developed.

Minor Groove Binder (MGB) Probe Technology: This approach uses short, allele-specific MGB probes that have very high melting temperature (Tm) discrimination, allowing them to distinguish single-nucleotide polymorphisms (SNPs) such as the A4062G and T4064A changes that define CPV-2a, -2b, and -2c at residue 426 [39, 50]. In a study from Morocco, a TaqMan-based real-time PCR using MGB probes successfully characterized the vast majority of 91 CPV-positive samples, identifying CPV-2a, -2b, and -2c, as well as co-infections [50]. However, the authors noted that three samples could not be typed by the MGB probe assay due to a G4068A mutation, which altered the probe-binding site [50]. This serves as a critical reminder that the very genetic plasticity of CPV-2 can undermine assays that rely on a single, static target sequence, necessitating periodic redesign of probes to match evolving field strains.

Multiplex TaqMan Real-Time PCR: To overcome the limitations of single-target assays, multiplex systems have been developed that can differentiate all four major antigenic types (CPV-2, CPV-2a, CPV-2b, and CPV-2c) in two separate reaction tubes. Sun et al. (2018) designed a set of primers and probes targeting positions 305 and 426 of the VP2 gene [27]. The first reaction (targeting residue 305) differentiates original CPV-2 from the variants, while the second reaction (targeting residue 426) distinguishes CPV-2a/2 from CPV-2b and CPV-2c. This assay demonstrated 100% agreement with DNA sequencing and a detection limit of 10–100 genome copies/μL, highlighting its high sensitivity and specificity [27]. Such assays are now considered the standard for high-throughput diagnostic laboratories.

High-Resolution Melting (HRM) Analysis: HRM is a post-PCR technique that analyzes the melting behavior of amplicons. After amplification in the presence of a saturating DNA dye, the PCR product is slowly denatured, and the change in fluorescence is monitored. The resulting melting curve is a characteristic signature determined by the amplicon’s GC content, length, and sequence, enabling the differentiation of variants without the need for expensive, sequence-specific probes [47, 55]. Balboni et al. (2026) developed two qPCR-HRM assays: one to differentiate FPV, original CPV-2, and the CPV-2 antigenic variants, and a second to identify "Asian-like" CPV strains carrying characteristic mutations (e.g., A5G, Q370R) [47]. These assays successfully classified all 33 clinical samples tested, demonstrating their reliability and utility for both diagnostic and epidemiological purposes [47]. A particularly elegant application of this concept is the duplex fluorescence melting curve analysis (FMCA) developed by Liu et al. (2019) [55]. This assay uses two TaqMan probes in a single reaction tube. The probes are designed to have varying levels of mismatch with the different target sequences. For example, a FAM-labeled probe perfectly matches CPV-2a but has a 1-bp mismatch with CPV-2b and a 2-bp mismatch with CPV-2c, resulting in distinct melting temperatures (Tm). This allows for the unambiguous genotyping of CPV-2, CPV-2a, CPV-2b, CPV-2c, and two vaccine strains (CPVpf and CPVint) in a single tube, with a lower detection limit of 1–10 copies per reaction [55]. Furthermore, the FMCA technique can directly identify co-infections, which is a significant advantage over sequencing-based methods that may mask minor variants [25, 55].

Sequencing and Next-Generation Sequencing (NGS): The Gold Standard for Molecular Characterization

While PCR-based genotyping is effective for identifying known variants, only nucleotide sequencing can provide the definitive, high-resolution characterization required to detect novel mutations, track evolutionary trajectories, and identify recombination events. Sanger sequencing of the full-length VP2 gene (1,755 bp) is the most common approach for comprehensive molecular analysis [2, 8, 10, 15, 20, 34]. This method has been instrumental in identifying the specific amino acid substitutions that define the CPV-2a, -2b, and -2c variants (most critically at residue 426), as well as numerous other mutations that contribute to antigenic drift and host range expansion. For instance, full-length VP2 sequencing of Ecuadorian strains revealed substitutions at residues 87, 101, 139, 219, 297, 300, 305, 322, 324, 375, 386, 426, 440, and 514 when compared to the original CPV-2 [2]. Similarly, a study in Chile identified mutations in regions of high antigenicity, such as CPV-2b (297 and 324) and CPV-2c (440), and reported novel mutations (e.g., CPV-2c: 188, 322, 379, 427, and 463) that may have implications for immune evasion [4].

The power of sequencing is magnified when applied to the complete viral genome (approximately 5.2 kb). Whole-genome sequencing provides data on both the structural (VP1, VP2) and non-structural (NS1, NS2) proteins, offering a holistic view of viral evolution. Phylogenetic analysis of whole-genome sequences has revealed a complex global structure, with well-defined clades that do not always correlate with the simple

Immunological Cross-Protection and Vaccine Efficacy Against CPV-2 Variants

The relentless evolution of canine parvovirus type 2 (CPV-2) has presented a profound and persistent challenge to the cornerstone of infectious disease control in companion animals: prophylactic vaccination. Since the replacement of the original CPV-2 by the antigenic variants CPV-2a, CPV-2b, and CPV-2c, the central question for veterinary clinicians and immunologists has shifted from whether vaccines work to how well they work against genetically and antigenically distinct field strains. This section provides an exhaustive analysis of the mechanisms of immunological cross-protection, the empirical evidence for vaccine efficacy against heterologous CPV-2 variants, and the emerging threats posed by antigenic drift and the progressive global dominance of the CPV-2c Asian lineage.

The Molecular Basis of Antigenic Variation and Cross-Neutralization

The foundation of any discussion on cross-protection rests on the molecular architecture of the CPV-2 capsid, specifically the major structural protein VP2. The canonical classification of CPV-2 into variants a, b, and c is defined by a single, critical amino acid substitution at residue 426 of VP2: asparagine (Asn) in CPV-2a, aspartic acid (Asp) in CPV-2b, and glutamic acid (Glu) in CPV-2c [1, 2, 26]. This residue lies within a major antigenic epitope, and its alteration profoundly influences the virus's interaction with neutralizing antibodies. Early antigenic cartography, utilizing panels of monoclonal antibodies (MAbs) and polyclonal sera, unequivocally demonstrated clear antigenic differences among the variants. Critically, the variant CPV-2c exhibited a unique antigenic pattern, being poorly recognized by sera from animals immunized with original CPV-2, CPV-2a, or CPV-2b [30]. This seminal finding raised immediate and justifiable alarm regarding the potential for CPV-2c to escape vaccine-induced immunity.

However, the immunological landscape is far more complex than a single epitope. The VP2 protein possesses multiple linear and conformational B-cell epitopes, as well as T-cell epitopes within VP2 and the non-structural proteins NS1 and NS2. An exhaustive immunoinformatics analysis identified conserved B-cell, cytotoxic T lymphocyte (CTL), and helper T lymphocyte (HTL) epitopes across CPV-2 variants, demonstrating that despite extensive variation in neutralizing antibody targets, highly conserved, immunodominant regions exist [14]. This diversity of epitopes is the mechanistic basis for cross-protection. When a dog is vaccinated with a modified-live virus (MLV) vaccine based on the original CPV-2 or CPV-2b, the immune system is presented with a vast array of antigens. While the antibody response specific to the residue 426 epitope may be suboptimal against a heterologous variant, the response directed against more conserved epitopes can still provide robust protection. This polyclonal response, encompassing both humoral and cell-mediated arms, is why vaccines developed decades ago continue to provide clinical protection against modern field strains.

Empirical Evidence of Cross-Protection from Vaccine Challenge Studies

The most robust evidence for cross-protection comes from controlled laboratory challenge studies. A pivotal evaluation of the Vanguard® C4 vaccine, which contains the original CPV-2 strain, demonstrated its ability to induce cross-neutralizing antibodies against Australian isolates of CPV-2a, CPV-2b, and CPV-2c [1]. Virus neutralization (VN) assays, the gold standard for assessing functional antibody, clearly showed that sera from vaccinated dogs could effectively neutralize all three heterologous variants. This study is particularly compelling as it used contemporary field isolates (2019-2020), confirming that the vaccine-induced immunity is not just a laboratory artifact but is relevant to circulating viruses. This cross-neutralization capacity is the immunological correlate of protection, explaining the long-standing success of CPV-2-based vaccines even as new variants emerged.

Further supporting this paradigm, a study evaluating a canine-derived chimeric monoclonal antibody (MAb 11D9) derived from a mouse MAb (10H4) showed exceptionally high hemagglutination inhibition (HI) and VN titers against new CPV-2a, new CPV-2b, and CPV-2c variants [58]. The neutralization titers against the CPV-2c variant were only marginally lower than those against the 2a and 2b variants (1:10,615.7 vs. 1:11,046.5), indicating that a potent, single B-cell epitope can be broadly neutralizing. Moreover, therapeutic administration of this MAb in beagles infected with new CPV-2a resulted in a high therapeutic effect, proving that the cross-neutralizing capacity translates directly to in vivo protection [58]. These findings collectively affirm that the critical epitopes conferring protection against severe disease are largely conserved across the variants, a concept that underpins the current global vaccination strategy.

The Paradox of Vaccine Failure: Maternally Derived Antibody and Immune Gaps

Despite robust evidence for cross-protection, an undeniable epidemiological reality persists: CPV-2 infection is frequently diagnosed in vaccinated dogs. Numerous molecular surveys have recorded CPV-2 in animals with a history of vaccination. For example, a study in Ecuador found CPV-2 affecting vaccinated puppies [2]. Research in Gabon was explicitly triggered by clinical cases among vaccinated dogs, leading to the characterization of CPV-2a and CPV-2c in that population [3]. Similarly, a long-term Italian study from 1994 to 2017 reported that 32.5% of CPV-2-positive dogs had been vaccinated, with no statistical association between vaccination status and the specific infecting variant [7]. In Nigeria, 60% of CPV-2-positive dogs were documented as vaccinated [16]. These recurring observations from diverse geographic regions, India, Turkey, Spain, and China, demand a nuanced explanation that moves beyond simple assertions of vaccine failure [7, 15, 23, 28, 29, 36].

The primary and most widely accepted cause of these "vaccine breaks" is interference from maternally derived antibodies (MDA). High titers of passively acquired antibodies, while protective against natural infection, can neutralize the vaccine virus before it can replicate and stimulate a robust active immune response. This leaves a critical "window of susceptibility" between the waning of MDA and the development of active immunity. The epidemiology of CPV-2, with its highest incidence in puppies under six months of age, aligns perfectly with this mechanism. As a seminal review on the subject notes, "the primary cause of failure of CPV vaccination is interference by maternally derived immunity" [26]. Therefore, infection in vaccinated dogs is frequently a failure of the vaccination protocol (timing, number of doses) rather than a failure of the vaccine itself.

However, this does not preclude the possibility of genuine antigenic escape. The concept of "antigenic drift" in CPV-2 is gaining traction. Studies analyzing selection pressures on the VP2 gene have identified sites under positive selection, notably at residues 297, 324, 426, and 440, which are located within or near major antigenic regions [4, 21]. An investigation in Chile specifically documented mutations in CPV-2b (at residues 297 and 324) and CPV-2c (at residue 440) that are associated with evasion of the immune response via antigenic drift [4]. These mutations are not random; they are driven by immune pressure from the vaccinated population. A comprehensive analysis of CPV-2 in Henan, China, identified 17 positive selection sites and 10 parallel evolution sites in VP2, with residues 5, 370, and 426 being under both positive selection and parallel evolution, a hallmark of convergent adaptive evolution [21]. This molecular evolution suggests that while current vaccines provide a broad umbrella of protection, the virus is continuously probing for "weak spots," and the emergence of drift variants is a real, albeit slow, phenomenon.

The Progressive Dominance of CPV-2c: A Test of Vaccine Durability

The global epidemiological landscape is characterized by a dramatic shift: the replacement of CPV-2a and CPV-2b by the CPV-2c variant. This trend is not a localized event but a global replacement underway across Asia, South America, North America, and Africa [6]. In China, studies report that CPV-2c has become the dominant genotype, accounting for 77.19% to 91.84% of infections in different provinces [15, 21, 22]. In Vietnam, an astonishing 96.54% of isolates were CPV-2c [24]. Similar patterns are seen in Italy, where CPV-2c has progressively replaced both CPV-2a and CPV-2b, and notably, the European CPV-2c lineage is itself being replaced by the Asian CPV-2c lineage [9]. This "Asian lineage" of CPV-2c carries hallmark mutations including A5G and Q370R, which are now spreading throughout Europe, Africa, and the Middle East [6, 9, 17, 21].

The critical question is whether this global replacement is driven by vaccine-induced selective pressure. The unique antigenic profile of CPV-2c, as demonstrated in the early MAb and polyclonal sera studies [30], suggests that it may have a selective advantage in a population with high vaccine coverage. The CPV-2c variant is poorly neutralized by antibodies raised against the original CPV-2 and CPV-2b strains [30]. While current vaccines still prevent clinical disease in the vast majority of vaccinated dogs, the ability of CPV-2c to replicate to higher titers and transmit more efficiently in a partially immune population could explain its rapid ascendancy. Indeed, the vaccine strain (often CPV-2 or CPV-2b) is rarely detected, while field variants, predominantly CPV-2c, are overwhelmingly present [23, 47]. This suggests that the vaccine creates a bottleneck that the CPV-2c variant is better able to traverse.

Future Horizons: The Imperative for Vaccine Update and DIVA Capabilities

The accumulation of genetic and antigenic data is moving the veterinary community toward a consensus that current vaccines, while still effective, may eventually require updating. The repeated isolation of CPV-2c from vaccinated dogs and the identification of novel mutations (e.g., I447M in China, Q370R globally) that may alter VP2 tertiary structure and receptor binding [15, 21] are flags that cannot be ignored. Structural modeling of the VP2 protein from Shanghai isolates indicates that mutations at key residues like Q370R and I447M may alter the protein's antigenic surface [15]. A study in Mongolia even identified novel mutations Pro580Thr and Tyr584His in CPV-2c strains, further expanding the genetic diversity of this variant [52].

To address these challenges, the next generation of CPV-2 vaccines will likely need to incorporate contemporary field strains, particularly the circulating CPV-2c Asian lineage. Such an update would ensure that the antigenic match between vaccine and field virus is optimized, potentially reducing the viral load and shedding in vaccinated animals and thereby diminishing transmission. Furthermore, there is a compelling need for the development of DIVA (Differentiating Infected from Vaccinated Animals) vaccines. Currently, serological tests cannot distinguish a naturally infected dog from a vaccinated one. A rationally designed, multi-epitope adenovirus-vectored vaccine incorporating conserved T-cell epitopes and a DIVA marker from a non-CPV source (e.g., Capripoxvirus) has been proposed, offering a path toward both improved safety and enhanced epidemiological monitoring [14]. The global virological community, guided by organizations like the World Organisation for Animal Health (WOAH), must prioritize the continuous molecular surveillance of CPV-2 evolution to inform vaccine strain selection, just as is done for influenza virus in humans. The era of a single, static vaccine strain for a rapidly evolving pathogen must be carefully re-evaluated to ensure the long-term effectiveness of our most powerful tool against this devastating disease.

Clinical Manifestations and Disease Severity Across CPV-2 Variants

The clinical presentation of canine parvovirus type 2 (CPV-2) infection in domestic dogs represents a spectrum of disease severity that is influenced by a complex interplay of viral genetic determinants, host factors, and environmental variables. While the classic syndrome of acute hemorrhagic gastroenteritis, immunosuppression, and myocarditis has been well-characterized since the virus’s emergence in the late 1970s, the subsequent evolution of three major antigenic variants, CPV-2a, CPV-2b, and CPV-2c, has prompted considerable investigation into whether these genetic distinctions translate into meaningful differences in clinical expression and outcome. The available evidence, drawn from both retrospective epidemiological studies and prospective clinical analyses, suggests that while all three variants can induce the full spectrum of parvoviral disease, there exist subtle yet significant variations in the pathophysiological impact, hematological derangements, and mortality risk associated with each variant.

At the most fundamental level, the hallmark clinical signs of CPV-2 infection remain consistent across variants: affected dogs typically present with a history of lethargy, anorexia, pyrexia, vomiting, and profuse, often hemorrhagic, diarrhea [11, 26]. The disease targets rapidly dividing cell populations, particularly in the intestinal crypt epithelium, bone marrow, lymphoid tissues, and, in very young puppies, the myocardium. The resulting villous atrophy and crypt necrosis lead to malabsorption, protein-losing enteropathy, and disruption of the intestinal barrier, while concurrent lymphopenia and neutropenia predispose the patient to secondary bacterial sepsis and systemic inflammatory response syndrome [31, 43]. However, emerging data from comparative clinical studies indicate that the magnitude of these derangements may vary according to the infecting variant.

A landmark prospective investigation by Mandawat et al. (2025) provided some of the most detailed comparative clinico-physiological and hematological data across all three CPV-2 variants in naturally infected dogs in India [43]. This study, which analyzed 52 cases each of CPV-2a and CPV-2b and 44 cases of CPV-2c, documented significantly elevated heart and respiratory rates in dogs infected with CPV-2b and CPV-2c compared to those infected with CPV-2a, suggesting a more pronounced systemic inflammatory response or greater cardiovascular compromise with the former two variants [43]. Furthermore, while all infected groups exhibited reductions in hemoglobin, packed cell volume, and total erythrocyte count, reflecting the hemorrhagic component of the disease, the CPV-2b and CPV-2c groups demonstrated the highest neutrophil counts and the lowest lymphocyte percentages [43]. This finding is particularly noteworthy, as the degree of lymphopenia is a well-established prognostic indicator in canine parvovirus infection; a more profound lymphodepletion may correlate with greater immunosuppression and a higher risk of secondary complications. The study also reported elevated eosinophil and basophil counts across all infected groups, a less commonly appreciated finding that may reflect a Th2-skewed immune response or concurrent parasitic co-infections [43].

Biochemical analyses from the same investigation revealed variant-specific differences in organ function markers. Serum total protein and albumin were significantly reduced across all infected dogs, consistent with protein-losing enteropathy, but dogs infected with CPV-2c exhibited the lowest values, indicating potentially more severe intestinal protein loss or hepatic synthetic dysfunction [43]. Hepatic enzyme elevations, specifically aspartate aminotransferase (AST) and alkaline phosphatase (ALP), were observed in all groups, confirming that hepatic damage is a component of CPV-2 infection regardless of variant; however, the severity of elevation varied. Perhaps most strikingly, dogs infected with CPV-2c demonstrated decreased serum creatinine levels, a finding that the authors interpreted as suggestive of possible renal impairment or altered renal perfusion in this group, potentially indicating a more severe systemic disease process involving the kidneys [43]. These data collectively suggest that CPV-2c may be associated with a more profound multisystemic insult than the other variants, though the underlying mechanisms remain to be fully elucidated.

Historical data from Spain provide additional context regarding the clinical severity of CPV-2c infection. In an outbreak investigation by Decaro et al. (2006), CPV-2c was identified as the causative agent of fatal hemorrhagic enteritis in basset hound puppies, occurring in association with co-infection by canine coronavirus types I and II [29]. The authors noted that the CPV-2c variant had initially been detected in Italy and subsequently in Vietnam, and its emergence in Spain was associated with severe clinical outcomes, raising concerns about its potential pan-European spread [29]. More recently, a large retrospective and prospective analysis of CPV-2 cases in central Spain spanning 2003 to 2014 found that CPV-2c was the predominant variant among hospitalized cases (42.9%), followed by CPV-2a (31.0%), although the study did not identify a statistically significant association between antigenic variant and in-hospital mortality [44]. Instead, the strongest predictors of mortality were small breed size (dogs weighing <15 kg had 2.74 times higher odds of death) and the presence of gastrointestinal signs in combination with neurological and/or respiratory signs (odds ratio 9.14) [44]. This finding underscores the critical importance of host factors and co-morbidities in determining clinical outcome, independent of the infecting variant.

Nevertheless, the antigenic properties of CPV-2c merit particular attention in any discussion of disease severity. Cross-antigenic evaluations using serum neutralization assays have demonstrated that CPV-2c exhibits a unique antigenic profile, being poorly recognized by sera from animals immunized with CPV-2, CPV-2a, or CPV-2b [30]. This finding has direct implications for vaccine-mediated protection and, consequently, for the clinical severity of breakthrough infections. Cavalli et al. (2007) reported that while animals immunized with CPV-2c developed high neutralizing titers to the homologous virus and to CPV-2b, the converse was not observed, animals vaccinated with earlier variants showed reduced neutralization of CPV-2c [30]. This antigenic disparity may contribute to the increasingly frequent reports of CPV-2c causing clinically significant disease in vaccinated dogs, as documented in studies from Gabon [3], Ecuador [2, 5], Italy [7, 9, 39], Vietnam [24], and China [15, 21, 22, 32]. The detection of CPV-2c in young, vaccinated dogs in Libreville, Gabon, for instance, highlights the potential for this variant to overcome vaccine-induced immunity, leading to clinical disease in populations that would otherwise be expected to be protected [3].

The molecular basis for these clinical differences likely resides in specific amino acid substitutions in the VP2 capsid protein that influence both antigenicity and receptor binding. The characteristic residue 426 substitution (asparagine in CPV-2a, aspartic acid in CPV-2b, glutamic acid in CPV-2c) is situated in a major neutralizing epitope and directly impacts the virus’s interaction with the host immune system [26, 42]. Additionally, recent CPV-2c strains, particularly those of Asian origin, have accumulated further mutations at residues 5 (Ala→Gly), 267 (Phe→Tyr), 324 (Tyr→Ile), and 370 (Gln→Arg) [6, 17, 18, 21]. The rapid global spread of these Asian-lineage CPV-2c strains, which have been documented replacing earlier variants in multiple continents [6, 15, 24, 34], suggests that these mutations may confer a fitness advantage, potentially through enhanced transmission, altered receptor tropism, or immune evasion. The A5G and Q370R mutations, in particular, have been identified as being under positive selection pressure and associated with parallel evolution, indicating their functional importance [21]. Structural modeling has further suggested that mutations at these key residues can alter the tertiary structure of the VP2 protein, potentially affecting antigenicity and receptor recognition [15].

The clinical significance of co-infections with multiple CPV-2 variants is an emerging area of concern. Co-infections with CPV-2b and CPV-2c have been documented in Morocco [50], and deep-sequencing analyses have revealed the presence of mixed infections and even inter-genotypic recombination between CPV-2c and CPV-2a within individual animals [25]. The implications of such co-infections for disease severity are not yet fully understood, but the potential for recombination to generate novel variants with altered pathogenic properties represents a significant evolutionary threat. The detection of a recombinant strain arising from CPV-2c and CPV-2a at the VP1/VP2 gene boundary in Uruguay highlights the capacity of this virus to generate genetic diversity even within a single host [25].

It is also critical to recognize that the clinical severity of CPV-2 infection is modulated by factors that transcend the infecting variant. Vaccination status, maternally derived antibody interference, breed predisposition, age, nutritional status, and the presence of concurrent parasitic or viral infections all contribute to the final clinical picture [26, 38, 49]. The high proportion of vaccinated dogs found to be infected with CPV-2 variants in multiple studies, 32.5% in Italy [7], 60% in Nigeria [16], and frequent reports from Ecuador [5] and Gabon [3], does not necessarily indicate vaccine failure but may instead reflect inadequate vaccine timing relative to maternal antibody waning, incomplete vaccination series, or overwhelming exposure pressure. Nonetheless, the consistent detection of CPV-2c in vaccinated populations, coupled with its unique antigenic profile, warrants continued vigilance and consideration of updated vaccine formulations [15, 39].

The hematological hallmark of CPV-2 infection, leukopenia, particularly neutropenia and lymphopenia, remains a critical determinant of outcome. The degree of leukopenia correlates with the extent of lymphoid and bone marrow destruction and is a strong predictor of mortality. In the comparative Indian study, while all variants induced lymphopenia, the most profound reductions were observed in CPV-2b and CPV-2c cases, and these same groups also demonstrated the most significant alterations in cardiac and respiratory parameters [43]. This constellation of findings suggests a more aggressive disease course characterized by greater immunosuppression and systemic inflammation. The elevated neutrophil counts in these groups, despite concurrent lymphopenia, may reflect a robust innate immune response or, alternatively, a stress response secondary to more severe tissue damage or endotoxemia.

From a pathophysiological standpoint, the intestinal tropism of CPV-2 leads to a predictable cascade of events: crypt epithelial cell destruction results in villous atrophy, loss of absorptive surface area, and compromised barrier function. This permits translocation of enteric bacteria and their toxins into the systemic circulation, triggering a systemic inflammatory response that can progress to sepsis, disseminated intravascular coagulation, and multi-organ failure. The more profound hypoproteinemia and hypoalbuminemia observed in CPV-2c-infected dogs in the Indian study may indicate more extensive intestinal damage or a greater degree of protein-losing enteropathy in this group [43]. Similarly, the elevated hepatic enzymes across all groups, but particularly in CPV-2c, suggest that hepatic injury, whether from hypoxic damage, direct viral cytopathology, or systemic inflammation, is a component of the disease that may be exacerbated with this variant.

The cardiac manifestations of CPV-2 infection, while most commonly associated with myocarditis in very young puppies from the original CPV-2 strain, have been reported with variants as well. The elevated heart rates documented in CPV-2b and CPV-2c infections in the Indian study may reflect tachycardia secondary to fever, dehydration, pain, or systemic inflammation rather than primary myocarditis [43]. However, the possibility of subclinical myocardial involvement in older dogs infected with emerging variants cannot be entirely discounted and warrants further investigation.

The epidemiological landscape of CPV-2 variants provides essential context for understanding current patterns of disease severity. Global surveillance data indicate that CPV-2c is progressively replacing CPV-2a and CPV-2b in many regions, including Asia [6, 15, 21, 24, 34], South America [9, 12], Africa [10, 37], and Europe [7, 9, 38]. In China, a complete genotype shift has been documented, with CPV-2c becoming the dominant variant after 2021, and isolates from Shanghai and Henan provinces showing 91-100% CPV-2c prevalence [15, 21]. In Italy, CPV-2c has been reported to constitute up to 69% of circulating strains, with the Asian lineage progressively replacing the European lineage [9]. In Africa, CPV-2c prevalence reached 91.5% in Nigerian isolates [37] and was also predominant in Egyptian studies [10]. This global replacement pattern suggests that CPV-2c may possess a selective advantage over earlier variants, whether through enhanced transmissibility, broader host range, or increased capacity to infect partially immune populations. The clinical consequence of this shift is that the variant potentially associated with more severe hematological and biochemical derangements is becoming the most common cause of canine parvovirus disease worldwide.

In summary, while all three CPV-2 variants, 2a, 2b, and 2c, are capable of causing the full clinical spectrum of parvoviral enteritis, accumulating evidence points toward CPV-2c as potentially inducing a more profound systemic illness, characterized by more severe lymphopenia, greater protein-losing enteropathy, and more marked alterations in hepatic and possibly renal function markers. The unique antigenic profile of CPV-2c, which renders it poorly neutralized by sera raised against earlier variants, may contribute to breakthrough infections in vaccinated animals, further amplifying its clinical impact. The ongoing global replacement of CPV-2a and CPV-2b by CPV-2c, particularly the emerging Asian lineage carrying additional mutations at residues 5, 267, 324, and 370, represents a significant shift in the epidemiological landscape that demands continued clinical and molecular surveillance.

Evolutionary Dynamics and Emergence of CPV-2 Variants

The evolutionary trajectory of canine parvovirus type 2 (CPV-2) represents one of the most compelling examples of rapid viral adaptation and emergence in a novel host species. Since its initial recognition as an emerging pathogen of dogs in the late 1970s, CPV-2 has undergone a series of profound genetic and antigenic shifts that have fundamentally altered its epizootiology, host range, and clinical impact [11, 26]. The emergence of CPV-2 is itself a story of cross-species transmission, likely originating from feline panleukopenia virus (FPV) or a related carnivore parvovirus, followed by a remarkable acceleration of evolutionary change that continues to this day [20, 26]. Understanding the evolutionary dynamics that have driven the emergence and global dissemination of CPV-2a, CPV-2b, and CPV-2c is not merely an academic exercise; it is essential for predicting future viral trajectories, optimizing vaccine strategies, and informing global surveillance efforts as recommended by the World Organisation for Animal Health (WOAH) for emerging infectious diseases of companion animals.

The Molecular Basis of Emergence: Mutational Hotspots in the VP2 Capsid

The evolutionary engine of CPV-2 is centered on the VP2 capsid protein, the primary determinant of antigenicity, host range, and receptor binding [11, 31]. CPV-2, like other parvoviruses, possesses a single-stranded DNA genome with a high substitution rate that approaches that of many RNA viruses, a paradoxical feature attributed to the rapid replication kinetics and the error-prone nature of host cell DNA polymerases operating under conditions of cellular stress [19, 25]. The critical mutational hotspot that defines the three major antigenic variants is residue 426 of the VP2 protein. The original CPV-2 and CPV-2a both encode asparagine (Asn) at this position, while CPV-2b features aspartic acid (Asp), and CPV-2c features glutamic acid (Glu) [2, 8, 13]. This single amino acid substitution at a key neutralizing epitope has profound consequences for antigenic structure, antibody recognition, and viral fitness [30, 42].

However, the evolutionary narrative extends far beyond residue 426. Comprehensive sequencing studies have revealed a constellation of additional mutations that accumulate over time, defining distinct sub-lineages and driving antigenic drift. The transition from the original CPV-2 to the "new" CPV-2a and CPV-2b variants involved a series of characteristic substitutions including Ser297Ala, Ala300Gly, and Tyr324Ile, which are now considered hallmarks of contemporary circulating strains [10, 15, 21]. These mutations are not randomly distributed; they cluster in specific surface loops of the VP2 capsid that correspond to major antigenic sites recognized by neutralizing antibodies [4, 42]. The selective pressure exerted by the host immune system, whether from natural infection or vaccination, drives the fixation of mutations that allow the virus to escape antibody neutralization while maintaining or enhancing receptor binding affinity for the canine transferrin receptor (TfR) [4, 30].

Selective Pressures and the Emergence of Antigenic Variants

The evolutionary dynamics of CPV-2 are shaped by a complex interplay of selective forces, including host immunity, receptor specificity, and transmission efficiency. Analyses of selection pressure acting on the VP2 gene have consistently demonstrated that the gene is predominantly under purifying (negative) selection, which removes deleterious mutations and maintains the structural integrity of the capsid [4, 32]. However, specific codons are subject to positive (diversifying) selection, indicating that adaptive evolution is occurring at functionally critical sites. In a comprehensive study of Chilean CPV-2 isolates, positive selection was detected at residues 324, 426, and 440 in CPV-2c, and at residues 297 and 324 in CPV-2b, all of which are located within or adjacent to known antigenic epitopes [4]. Similarly, a large-scale genomic analysis of Chinese CPV-2 strains identified 17 positive selection sites and 10 parallel evolution sites in the VP2 protein, with residues 5, 370, and 426 being under both positive selection and parallel evolution, suggesting convergent adaptation across geographically distinct viral populations [21].

The emergence of CPV-2c in Italy around 2000 represented a major evolutionary event [26, 29]. This variant, defined by the Glu426 substitution, exhibited a unique antigenic profile that distinguished it from both CPV-2a and CPV-2b. Cross-neutralization studies demonstrated that sera from dogs immunized with CPV-2, CPV-2a, or CPV-2b showed significantly reduced neutralizing activity against CPV-2c, while sera from CPV-2c-immunized animals paradoxically showed higher titers against CPV-2b than against the homologous virus [30]. This asymmetric antigenic relationship suggests that CPV-2c may have emerged as a consequence of immune pressure from populations with high vaccine coverage, selecting for a variant that could partially escape vaccine-induced immunity while retaining the ability to infect and transmit among vaccinated hosts [26, 30]. The rapid global spread of CPV-2c following its initial detection underscores the fitness advantage conferred by this antigenic shift [6, 26].

Global Phylodynamics and Lineage Replacement

The evolutionary history of CPV-2 is characterized by successive waves of lineage replacement, where newly emerged variants progressively displace their predecessors on a global scale. The original CPV-2 was completely replaced by CPV-2a and CPV-2b within a few years of its emergence, and these variants themselves have been undergoing continuous replacement by newer sub-lineages [7, 20]. Bayesian coalescent analyses of global CPV-2 sequences have estimated that the time to the most recent common ancestor (tMRCA) for the major circulating clades dates to the 1980s and 1990s, with a major expansion in viral population size occurring at the end of the 1980s and a secondary increase between 2000 and 2004 [20]. These temporal dynamics correlate with the emergence and global dissemination of the antigenic variants.

The most recent and dramatic shift in global CPV-2 phylodynamics is the emergence and rapid expansion of an Asian lineage of CPV-2c, which is now displacing both CPV-2a and the European lineage of CPV-2c in multiple continents [6, 9, 15]. This Asian CPV-2c lineage is characterized by a distinctive set of amino acid substitutions, including Ala5Gly, Phe267Tyr, Tyr324Ile, and Gln370Arg, which are rarely found in European or American CPV-2c strains [6, 17, 18, 21]. The Ala5Gly and Gln370Arg mutations are particularly noteworthy, as they have been identified as sites under positive selection and parallel evolution, suggesting they confer a significant fitness advantage [21]. Structural modeling indicates that these mutations may alter the tertiary structure of the VP2 protein, potentially affecting both antigenicity and receptor recognition [15].

The displacement of older variants by the Asian CPV-2c lineage has been documented across diverse geographic regions. In China, CPV-2c has become the dominant genotype, accounting for over 90% of isolates in several provinces, with a complete replacement of New CPV-2a occurring between 2016 and 2021 [15, 21, 22]. In Vietnam, CPV-2c constituted 96.5% of isolates collected between 2016 and 2018 [24]. In Thailand, CPV-2c emerged in 2014 and had become the major variant by 2019, replacing CPV-2a and CPV-2b [34]. Critically, this Asian lineage is not confined to Asia. It has been detected in Italy, where it is progressively replacing the European CPV-2c lineage [9], in Romania [17], in Nigeria [37], and in Egypt [10]. The detection of Asian CPV-2c in Sicily, southern Italy, with evidence of local transmission and establishment, demonstrates the capacity of this lineage for long-distance dispersal and successful colonization of new geographic regions [9]. The global spread of this lineage represents an ongoing evolutionary event with significant implications for vaccine efficacy and disease control.

Recombination and Co-Infection as Drivers of Genetic Diversity

While the accumulation of point mutations is the primary mechanism of CPV-2 evolution, recombination between co-infecting variants represents an additional and potentially underappreciated source of genetic diversity. Co-infection of individual dogs with multiple CPV-2 variants has been documented in several studies, providing the necessary conditions for recombination to occur [25, 50]. In Morocco, co-infections with CPV-2b and CPV-2c were detected in 4.4% of positive samples [50]. More significantly, deep-sequencing analysis of a naturally infected dog in Uruguay revealed the presence of a major recombinant strain (86.7% of the viral population) arising from inter-genotypic recombination between CPV-2c and CPV-2a strains within the VP1/VP2 gene boundary [25]. This finding demonstrates that recombination is not merely a theoretical possibility but an active evolutionary process that can generate novel viral genomes with unique combinations of parental alleles.

The detection of recombination in CPV-2 has important implications for understanding viral emergence. Recombination can accelerate the rate of adaptive evolution by bringing together beneficial mutations that have arisen independently in different lineages, potentially creating viruses with enhanced fitness, altered host range, or novel antigenic properties [25]. The fact that the recombination breakpoint in the Uruguayan isolate occurred at the VP1/VP2 gene boundary, a region critical for capsid structure and function, suggests that recombination may play a role in generating antigenic diversity [25]. As surveillance efforts intensify and sequencing technologies improve, it is likely that additional recombinant CPV-2 strains will be identified, further complicating the evolutionary landscape of this virus.

Host Range Evolution and Spillover Events

The evolutionary dynamics of CPV-2 are inextricably linked to changes in host range. The original CPV-2 emerged as a host range variant of FPV, acquiring the ability to infect and replicate in dogs while losing some capacity for replication in cats [26, 41]. The subsequent antigenic variants, CPV-2a, CPV-2b, and CPV-2c, have further expanded the host range, regaining the ability to infect and cause disease in cats [26, 41]. This host range expansion is mediated by specific amino acid substitutions in the VP2 capsid that alter the interaction with the host TfR, allowing the virus to utilize receptors from multiple carnivore species [11, 31]. The detection of CPV-2a and CPV-2b in domestic cats in Brazil, with evidence of genetic changes including the Y324L mutation not previously reported in canine strains, suggests ongoing adaptation to the feline host [41].

The evolutionary implications of this expanded host range are profound. The ability to infect multiple host species creates additional ecological niches for viral persistence and transmission, potentially increasing the overall viral population size and the opportunities for adaptive evolution [26]. Spillover events from domestic dogs to wild carnivores have been documented, raising concerns about the impact of CPV-2 on endangered wildlife populations [51]. The World Organisation for Animal Health (WOAH) recognizes the importance of monitoring CPV-2 in both domestic and wild animal populations, as the virus poses a threat to biodiversity and conservation efforts. The continued evolution of CPV-2 in multiple host species, each with potentially different selective pressures, may drive the emergence of variants with altered pathogenicity, transmissibility, or antigenicity, underscoring the need for integrated surveillance across domestic and wild carnivore populations.

References

[1] Packianathan R, Hodge A, Wright J, Lavidis L, Ameiss K, Yip HY, et al.. Cross-Neutralization of Vanguard C4 Vaccine Against Australian Isolates of Canine Parvovirus Variants CPV-2a, CPV-2b, and CPV-2c. Viral immunology. 2022. DOI: https://doi.org/10.1089/vim.2022.0027

[2] Torre DL, Mafla E, Puga B, Erazo L, Astolfi-Ferreira C, Ferreira AP. Molecular characterization of canine parvovirus variants (CPV-2a, CPV-2b, and CPV-2c) based on the VP2 gene in affected domestic dogs in Ecuador. Veterinary World. 2018. DOI: https://doi.org/10.14202/vetworld.2018.480-487

[3] Maganga G, Labouba I, Ikapi SZM, Nkili-Meyong AA, Dikongo AMN, Boundenga L, et al.. Molecular Characterization of Canine Parvovirus Variants CPV-2a and CPV-2c, Associated with Vaccinated Dogs at Libreville, Gabon. Viruses. 2023. DOI: https://doi.org/10.3390/v15051169

[4] Alexis V, Sonia V, Daniela S, Miguel G, Timothy H, Valentina F, et al.. Molecular Analysis of Full-Length VP2 of Canine Parvovirus Reveals Antigenic Drift in CPV-2b and CPV-2c Variants in Central Chile. Animals. 2021. DOI: https://doi.org/10.3390/ani11082387

[5] Loor-Giler A, Santander-Parra S, Castillo-Reyes S, Campos M, Mena-Pérez R, Prado-Chiriboga S, et al.. Characterization, Quantification, and Molecular Identification of Co-Infection of Canine Parvovirus (CPV-2) Variants in Dogs Affected by Gastroenteritis in Ecuador During 2022–2023. Veterinary Sciences. 2025. DOI: https://doi.org/10.3390/vetsci12010046

[6] Hao X, Li Y, Xiao X, Chen B, Zhou P, Li S. The Changes in Canine Parvovirus Variants over the Years. International Journal of Molecular Sciences. 2022. DOI: https://doi.org/10.3390/ijms231911540

[7] Battilani M, Modugno F, Mira F, Purpari G, Bella SD, Guercio A, et al.. Molecular epidemiology of canine parvovirus type 2 in Italy from 1994 to 2017: recurrence of the CPV-2b variant. BMC Veterinary Research. 2019. DOI: https://doi.org/10.1186/s12917-019-2096-1

[8] Giraldo-Ramirez S, Rendon-Marin S, Ruíz-Sáenz J. Phylogenetic, Evolutionary and Structural Analysis of Canine Parvovirus (CPV-2) Antigenic Variants Circulating in Colombia. Viruses. 2020. DOI: https://doi.org/10.3390/v12050500

[9] Mira F, Schiró G, Franzo G, Canuti M, Purpari G, Giudice E, et al.. Molecular epidemiology of canine parvovirus type 2 in Sicily, southern Italy: A geographical island, an epidemiological continuum. Heliyon. 2024. DOI: https://doi.org/10.1016/j.heliyon.2024.e26561

[10] Adly M, Elgaml MA, Khalek AFA, Saeed OS, Shalaby M, Amer H. Molecular characterization of full-length VP2 gene of canine parvovirus type 2 strains circulating in Egypt 2019-2021.. Comparative Immunology, Microbiology & Infectious Diseases. 2024. DOI: https://doi.org/10.1016/j.cimid.2024.102190

[11] Nguyen L, Danh U. OVERVIEW OF THE CANINE PARVOVIRUS TYPE 2: ETIOLOGY, PATHOGENESIS, DIAGNOSIS, TREATMENT, AND PREVENTION. TRA VINH UNIVERSITY JOURNAL OF SCIENCE. 2024. DOI: https://doi.org/10.35382/tvujs.14.2.2024.3908

[12] Grecco S, Condon E, Bucafusco D, Bratanich A, Panzera Y, Pérez R. Comparative genomics of canine parvovirus in South America: Diversification patterns in local populations.. Infection, Genetics and Evolution. 2024. DOI: https://doi.org/10.1016/j.meegid.2024.105633

[13] Purushotham P, P K, K D. CHARACTERIZATION OF CANINE PARVOVIRUS-2, CIRCULATING STRAINS IN HYDERABAD BY PCR AND RESTRICTION FRAGMENT LENGTH POLYMORPHISM - PCR FROM CLINICAL CASES IN DOGS. Asian Journal of Microbiology, Biotechnology & Environmental Sciences. 2024. DOI: https://doi.org/10.53550/ajmbes.2024.v26i01.006

[14] Sharma S, Sharma P, Mandar RP, Kumar R, Gaur A, Gurjar M, et al.. Design of Multi‐Epitope DIVA‐Compatible Vaccine Candidate Against Canine Parvovirus 2. Journal of Molecular Recognition. 2026. DOI: https://doi.org/10.1002/jmr.70036

[15] Xia Q, Liu J, Gui Y, Xia L, Cao C, Chen B, et al.. Molecular Characteristics and Genetic Diversity of Canine Parvovirus in Shanghai, China, from 2016 to 2025. Microorganisms. 2026. DOI: https://doi.org/10.3390/microorganisms14040761

[16] Apaa TT, Daly J, Tarlinton R. Canine parvovirus (CPV-2) variants circulating in Nigerian dogs. Veterinary Record Open. 2016. DOI: https://doi.org/10.1136/vetreco-2016-000198

[17] Balboni A, Niculae M, Vito SD, Urbani L, Terrusi A, Mureșan C, et al.. The detection of canine parvovirus type 2c of Asian origin in dogs in Romania evidenced its progressive worldwide diffusion. BMC Veterinary Research. 2021. DOI: https://doi.org/10.1186/s12917-021-02918-6

[18] Moon B, Jang J, Kim S, Kim Y, Lee H, So B, et al.. Genetic characterization of canine parvovirus type 2c from domestic dogs in Korea. Transboundary and Emerging Diseases. 2020. DOI: https://doi.org/10.1111/tbed.13501

[19] Mira F, Dowgier G, Purpari G, Vicari D, Bella SD, Macaluso G, et al.. Molecular typing of a novel canine parvovirus type 2a mutant circulating in Italy. Infection, Genetics and Evolution. 2018. DOI: https://doi.org/10.1016/j.meegid.2018.03.010

[20] Santana WdO, Silveira VP, Wolf J, Kipper D, Echeverrigaray S, Canal C, et al.. Molecular phylogenetic assessment of the canine parvovirus 2 worldwide and analysis of the genetic diversity and temporal spreading in Brazil.. Infection, Genetics and Evolution. 2022. DOI: https://doi.org/10.1016/j.meegid.2022.105225

[21] Fu P, He D, Cheng X, Niu X, Wang C, Fu Y, et al.. Prevalence and Characteristics of Canine Parvovirus Type 2 in Henan Province, China. Microbiology spectrum. 2022. DOI: https://doi.org/10.1128/spectrum.01856-22

[22] Chen Y, Wang J, Bi Z, Tan Y, Lv L, Zhao H, et al.. Molecular epidemiology and genetic evolution of canine parvovirus in East China, during 2018-2020.. Infection, Genetics and Evolution. 2021. DOI: https://doi.org/10.1016/j.meegid.2021.104780

[23] Sharma S, Dhar P, Thakur A, Sharma V, Sharma M. First detection of canine parvovirus type 2b from diarrheic dogs in Himachal Pradesh. Veterinary World. 2016. DOI: https://doi.org/10.14202/vetworld.2016.964-969

[24] Hoang M, Lin W, Le V, Nga BT, Chiou M, Lin C. Molecular epidemiology of canine parvovirus type 2 in Vietnam from November 2016 to February 2018. Virology Journal. 2019. DOI: https://doi.org/10.1186/s12985-019-1159-z

[25] Pérez R, Calleros L, Marandino A, Sarute N, Iraola G, Grecco S, et al.. Phylogenetic and Genome-Wide Deep-Sequencing Analyses of Canine Parvovirus Reveal Co-Infection with Field Variants and Emergence of a Recent Recombinant Strain. PLoS ONE. 2014. DOI: https://doi.org/10.1371/journal.pone.0111779

[26] Decaro N, Buonavoglia C. Canine parvovirus, A review of epidemiological and diagnostic aspects, with emphasis on type 2c. Veterinary Microbiology. 2011. DOI: https://doi.org/10.1016/j.vetmic.2011.09.007

[27] Sun Y, Cheng Y, Lin P, Yi L, Tong M, Cao Z, et al.. A multiplex TaqMan real-time PCR for detection and differentiation of four antigenic types of canine parvovirus in China. Molecular and Cellular Probes. 2018. DOI: https://doi.org/10.1016/j.mcp.2018.02.004

[28] Kuruçay H, Gumusova S. Virological investigation on the presence of CPV-2 in the Middle Black Sea Region of Turkey. Polish journal of veterinary sciences. 2026. DOI: https://doi.org/10.24425/pjvs.2026.158496

[29] Decaro N, Martella V, Desario C, Bellacicco A, Camero M, Manna L, et al.. First Detection of Canine Parvovirus Type 2c in Pups with Haemorrhagic Enteritis in Spain. Journal of veterinary medicine. B, Infectious diseases and veterinary public health. 2006. DOI: https://doi.org/10.1111/j.1439-0450.2006.00974.x

[30] Cavalli A, Martella V, Desario C, Camero M, Bellacicco A, Palo PD, et al.. Evaluation of the Antigenic Relationships among Canine Parvovirus Type 2 Variants. Clinical and Vaccine Immunology. 2007. DOI: https://doi.org/10.1128/CVI.00444-07

[31] P M, Brar B, K S, Thomas JV, J S, Na I, et al.. CANINE PARVOVIRUS- AN INSIGHT INTO DIAGNOSTIC ASPECT. Journal of Experimental Biology and Agricultural Sciences. 2016. DOI: https://doi.org/10.18006/2016.4(3S).279.290

[32] Li C, Tang J, Chen Z, Niu G, Liu G. A divergent canine parvovirus type 2c (CPV-2c) isolate circulating in China.. Infection, Genetics and Evolution. 2019. DOI: https://doi.org/10.1016/j.meegid.2019.05.004

[33] Castro TX, Costa E, Leite J, Labarthe N, Garcia RC. Partial VP2 sequencing of canine parvovirus (CPV) strains circulating in the state of Rio de Janeiro, Brazil: detection of the new variant CPV-2c.. Brazilian Journal of Microbiology. 2010. DOI: https://doi.org/10.1590/S1517-83822010000400031

[34] Jantafong T, Ruenphet S, Garner H, Ritthipichai K. Tracing the Genetic Evolution of Canine Parvovirus Type 2 (CPV-2) in Thailand. Pathogens. 2022. DOI: https://doi.org/10.3390/pathogens11121460

[35] Kumar M, Nandi S. Molecular typing of canine parvovirus variants by polymerase chain reaction and restriction enzyme analysis.. Transboundary and Emerging Diseases. 2010. DOI: https://doi.org/10.1111/j.1865-1682.2010.01167.x

[36] Polat P, Şahan A, Aksoy G, Timurkan M, Dinçer E. Molecular and restriction fragment length polymorphism analysis of canine parvovirus 2 (CPV-2) in dogs in southeast Anatolia, Turkey. Onderstepoort Journal of Veterinary Research. 2019. DOI: https://doi.org/10.4102/ojvr.v86i1.1734

[37] Ogbu K, Mira F, Purpari G, Nwosuh C, Loria G, Schiró G, et al.. Nearly full‐length genome characterization of canine parvovirus strains circulating in Nigeria. Transboundary and Emerging Diseases. 2019. DOI: https://doi.org/10.1111/tbed.13379

[38] Tucciarone C, Franzo G, Mazzetto E, Legnardi M, Caldin M, Furlanello T, et al.. Molecular insight into Italian canine parvovirus heterogeneity and comparison with the worldwide scenario.. Infection, Genetics and Evolution. 2018. DOI: https://doi.org/10.1016/j.meegid.2018.09.021

[39] Giuseppa P, Francesco M, Santina D, Simona D, Elisabetta G, Annalisa G. Investigation on Canine parvovirus circulation in dogs from Sicily (Italy) by biomolecular assay. Acta Veterinaria. 2018. DOI: https://doi.org/10.2478/acve-2018-0007

[40] Nandi S, Chidri S, Kumar M, Chauhan R. Occurrence of canine parvovirus type 2c in the dogs with haemorrhagic enteritis in India.. Research in Veterinary Science. 2010. DOI: https://doi.org/10.1016/j.rvsc.2009.05.018

[41] Domingues CF, Castro TXd, Lago BVd, Garcia RC. Genetic characterization of the parvovirus full-length VP2 gene in domestic cats in Brazil.. Research in Veterinary Science. 2024. DOI: https://doi.org/10.1016/j.rvsc.2024.105186

[42] Nakamura M, Nakamura K, Miyazawa T, Tohya Y, Mochizuki M, Akashi H. Monoclonal Antibodies That Distinguish Antigenic Variants of Canine Parvovirus. Clinical Diagnostic Laboratory Immunology. 2003. DOI: https://doi.org/10.1128/CDLI.10.6.1085-1089.2003

[43] Mandawat S, Singh R, Mohammed N, Singh Y, Meena DS, Jyoti, et al.. Comparative Pathophysiology of Canine Parvovirus-2 Variants in Naturally Infected Dogs. Asian Journal of Research in Biochemistry. 2025. DOI: https://doi.org/10.9734/ajrb/2025/v15i3395

[44] Penelo S, Simarro I, Fuertes‐Recuero M, Ayllón T, Ortiz-Díez G. Canine parvovirus type 2 antigenic variants and in-hospital mortality in central Spain: Retrospective and prospective data (2003–2014). Open Veterinary Journal. 2026. DOI: https://doi.org/10.5455/ovj.2026.v16.i3.48

[45] Castro TX, Costa E, Leite J, Labarthe N, Garcia RC. Partial VP2 sequencing of canine parvovirus (CPV) strains circulating in the state of Rio de Janeiro, Brazil: detection of the new variant CPV-2c. Brazilian Journal of Microbiology. 2010. DOI: https://doi.org/10.1590/S1517-838220100004000031

[46] Jyothi VP, Bhaskaran MS, Gundi VA. Epidemiology, molecular prevalence and prevention on canine parvovirus in India: A review. Bioinformation. 2024. DOI: https://doi.org/10.6026/973206300200536

[47] Balboni A, Facile V, Urbani L, Magliocca M, Gallina L, Vigato M, et al.. Feline panleukopenia virus and canine parvovirus: development of two qPCR assays with high-resolution melting analysis and molecular epidemiology in dogs and cats from Northen Italy in 2017–2023. Frontiers in Veterinary Science. 2026. DOI: https://doi.org/10.3389/fvets.2026.1789452

[48] Chiang S, Wu H, Chiou M, Chang M, Lin C. Identification of a novel canine parvovirus type 2c in Taiwan. Virology Journal. 2016. DOI: https://doi.org/10.1186/s12985-016-0620-5

[49] Galkina TS, Karaulov A. Canine parvovirus enteritis: epidemic situation analysis and perspectives. Veterinary Science Today. 2021. DOI: https://doi.org/10.29326/2304-196x-2020-4-35-283-289

[50] Amrani N, Desario C, Kadiri A, Cavalli A, Berrada J, Zro K, et al.. Molecular epidemiology of canine parvovirus in Morocco.. Infection, Genetics and Evolution. 2016. DOI: https://doi.org/10.1016/j.meegid.2016.04.005

[51] Jiang F. Bioclimatic and altitudinal variables influence the potential distribution of canine parvovirus type 2 worldwide. Ecology and Evolution. 2018. DOI: https://doi.org/10.1002/ece3.3994

[52] Temuujin U, Tserendorj A, Fujiki J, Sakoda Y, Tseren-Ochir E, Okamatsu M, et al.. The first isolation and identification of canine parvovirus (CPV) type 2c variants during 2016-2018 genetic surveillance of dogs in Mongolia.. Infection, Genetics and Evolution. 2019. DOI: https://doi.org/10.1016/j.meegid.2019.05.006

[53] Safwat MS, Farouk MM, Anwer AZ, Ali ME, Shahen N, Abdelwahed DA, et al.. Perfect diagnostic agreement between canine- and feline-specific parvovirus PoC antigen kits in feline panleukopenia. Veterinary Quarterly. 2025. DOI: https://doi.org/10.1080/01652176.2025.2604234

[54] Hyeon L, Yang D, Kim E, Park Y, Lee H, Hyun B. Evaluation of commercial immunochromatography test kits for diagnosing canine parvovirus. Korean Journal of Veterinary Research. 2023. DOI: https://doi.org/10.14405/kjvr.20230016

[55] Liu Z, Bingga G, Zhang C, Shao J, Shen H, Sun J, et al.. Application of Duplex Fluorescence Melting Curve Analysis (FMCA) to Identify Canine Parvovirus Type 2 Variants. Frontiers in Microbiology. 2019. DOI: https://doi.org/10.3389/fmicb.2019.00419

[56] Salman N, Mahmoud A, El-Sayed EF, Tarabili ME. Detection of Canine Parvovirus in Diarrheic Dogs in Three Egyptian Provinces During 2019-2020. Suez Canal Veterinary Medical Journal. SCVMJ. 2021. DOI: https://doi.org/10.21608/SCVMJ.2021.184760

[57] Vasavi K, Kaur G, Chandra M, Dwivedi P, Kaur N. Identification and Typing of Canine Parvovirus Using Molecular Techniques. Current Journal of Applied Science and Technology. 2021. DOI: https://doi.org/10.9734/cjast/2021/v40i1931457

[58] Zhou L, Wu H, Du M, Song H, Huo N, Chen X, et al.. A canine-derived chimeric antibody with high neutralizing activity against canine parvovirus-2. AMB Express. 2022. DOI: https://doi.org/10.1186/s13568-022-01416-8