Piscine Myocarditis Virus: CMS Reference

Overview and Taxonomy of Piscine Myocarditis Virus (PMCV) and Cardiomyopathy Syndrome (CMS)

Historical Context and Emergence of Cardiomyopathy Syndrome

Cardiomyopathy syndrome (CMS) represents one of the most economically devastating cardiac diseases afflicting Atlantic salmon (Salmo salar L.) aquaculture, with profound implications for fish welfare, production sustainability, and global seafood security. The disease was first recognized in farmed Atlantic salmon in Norway in 1985, emerging as a clinical entity characterized by severe myocardial inflammation and necrosis that predominantly affected large, market-sized fish nearing slaughter [19]. This initial description heralded a growing awareness of a previously unrecognized pathological condition that would subsequently be documented across all major salmon-producing regions of the North Atlantic. Following its Norwegian identification, CMS was confirmed in farmed salmon in the Faroe Islands, Scotland, and Ireland, and has also been described in wild Atlantic salmon populations in Norway [19]. The temporal and geographic expansion of CMS closely paralleled the intensification and global proliferation of Atlantic salmon aquaculture, a pattern that has been corroborated by recent phylodynamic analyses indicating that the causative agent likely emerged concurrently with the global expansion of aquaculture operations [6]. The World Organisation for Animal Health (WOAH) recognizes CMS as a significant disease affecting aquatic animal health, and the Food and Agriculture Organization (FAO) has documented its impact on aquaculture productivity and economic viability in affected regions.

The seminal discovery that CMS is a transmissible disease was made in 2009, fundamentally altering the scientific understanding of its etiology [19]. This breakthrough was followed by the detection and initial characterization of piscine myocarditis virus (PMCV) in 2010 and 2011, providing the definitive link between a specific viral pathogen and the characteristic cardiac pathology observed in CMS [19]. These discoveries catalyzed a sustained research effort that has progressively elucidated the molecular virology, epidemiology, and pathogenesis of PMCV, though substantial gaps in knowledge persist, particularly regarding the virus’s complete life cycle and ecological reservoirs [4, 19].

Taxonomic Classification and Virion Structure

PMCV is a non-enveloped, double-stranded RNA (dsRNA) virus whose genomic architecture and structural features place it within a taxonomic framework that has undergone considerable refinement since its initial characterization. The virus possesses a bipartite genome comprising three open reading frames (ORFs), designated ORF1, ORF2, and ORF3, which are arranged in a configuration that shares structural similarities with members of the family Totiviridae [3, 16, 19]. This taxonomic affiliation is based on genomic organization, sequence homology, and predicted protein structural motifs rather than on traditional serological or morphological criteria, as PMCV has resisted all attempts at propagation in established cell culture systems [4, 5, 19]. The inability to cultivate PMCV in vitro has profoundly constrained virological investigation, necessitating reliance on molecular detection methods, experimental infection models using tissue homogenates, and advanced sequencing technologies to characterize the virus.

Phylogenetic analyses consistently position PMCV within the order Ghabrivirales, a taxonomic grouping that encompasses dsRNA viruses infecting diverse hosts including fungi, protozoa, and arthropods [3, 7]. The virus’s genomic organization, with ORF1 encoding a putative capsid protein, ORF2 encoding an RNA-dependent RNA polymerase (RdRp), and ORF3 encoding a protein of uncertain function, is characteristic of totivirus-like genomes [3, 19]. The predicted capsid protein, which has been demonstrated to form virus-like particles (VLPs) when expressed recombinantly in plant systems, represents the primary structural component of the virion and a logical target for vaccine development [5]. However, the precise three-dimensional architecture of the native PMCV virion remains unresolved due to the absence of purified virus preparations suitable for cryo-electron microscopy or X-ray crystallography.

Genetic Diversity and Phylogeography

Contemporary genomic analyses have revealed that PMCV exhibits remarkably low genetic diversity relative to many other RNA viruses, a feature that has both epidemiological and evolutionary implications. Comprehensive whole-genome sequencing studies encompassing isolates from Norway, the Faroe Islands, Ireland, and Scotland have demonstrated that PMCV genomes form distinct, country-specific phylogenetic clusters [1, 3, 6]. This national-level genetic structure suggests that viral spread is primarily driven by within-country transmission dynamics, punctuated by relatively infrequent inter-country introductions that establish persistent, locally circulating lineages [6]. The most recent common ancestor (MRCA) of contemporary PMCV strains has been estimated to date to approximately the 1970s–1990s, coinciding with the period of rapid expansion of Atlantic salmon aquaculture [6].

Analysis of PMCV genomes from the Faroe Islands has provided particularly illuminating insights into viral introduction and maintenance dynamics. Dahl et al. [1] sequenced 48 novel Faroese PMCV genomes and demonstrated that these isolates form a homogeneous, monophyletic cluster distinct from Norwegian and Irish strains. This genetic homogeneity, confirmed by principal component analyses showing no spatiotemporal clustering or association with roe or smolt origin, strongly suggests a single introduction event into the Faroe Islands, most likely from Norway, where broodfish are known to be infected [1]. Strikingly, despite a steadily increasing import of Norwegian roe, the data indicate no continuous reintroduction of persisting Norwegian PMCV strains to Faroese farmed salmon, implying that the original introduced lineage has become self-sustaining within the Faroese aquaculture system [1]. An intriguing outlier was identified in a genome derived from a returning wild salmon, which differed substantially from all farmed isolates and formed an outgroup, raising questions about wild fish reservoirs and potential bidirectional virus exchange between wild and farmed populations [1, 2].

In Norway, the epicenter of PMCV research and the country with the highest CMS burden, genomic studies have revealed additional complexity within the otherwise conserved viral genome. Amono et al. [3] added 34 complete genome sequences and 202 novel ORF3 sequences from Norwegian CMS cases to the existing dataset, confirming the high degree of sequence conservation while simultaneously documenting the presence of multiple PMCV variants within single CMS outbreaks. This intra-outbreak diversity suggests that individual fish may harbor heterogeneous viral populations, potentially reflecting ongoing mutation, selection, and possibly co-infection with multiple viral strains [3]. Temporal analyses indicated increasing sequence diversity in the population over time, consistent with continuous evolutionary pressure and the accumulation of mutations, particularly in ORF3, which exhibits a comparatively lower degree of selective constraint than ORF1 and ORF2 [1, 3]. All three ORFs show evidence of purifying selection, but the relaxation of selective pressure on ORF3 may indicate that this gene product is less critical for fundamental viral functions and may be more tolerant of sequence variation [1].

Irish PMCV isolates have been characterized as largely homogenous, with little genetic diversity observed across samples collected from various marine sites around Ireland [12]. However, consistent amino acid variations unique to Irish strains were identified in both ORF1 and ORF3 when compared with Norwegian sequences, and phylogenetic evidence suggests that PMCV was introduced into Ireland in two separate waves, both originating from the southern part of the virus’s range in Norway [12]. More recent genomic analyses comparing Irish and Faroese isolates have further revealed that some Irish PMCV strains may have origins in wild fish populations, underscoring the potential role of wild salmonids in viral maintenance and dissemination [2]. Notably, over three-quarters of the Irish PMCV strains sequenced originated from fish without clinical signs of CMS, raising questions about viral attenuation, host immunity, or environmental factors that modulate disease expression [12].

The most geographically comprehensive phylodynamic study to date, encompassing PMCV isolates from Scotland and Norway, has provided a high-resolution view of viral dispersal dynamics [6]. Zhao et al. [6] generated one of the largest genomic datasets for any aquatic viral pathogen and demonstrated distinctive national clustering of virus genomes, reflecting intranational spread interspersed with multiple introductions of distinct PMCV lineages into Scotland. Statistical modeling highlighted the importance of geographic proximity and at-sea farm density in driving local transmission, while identifying boat connectivity as a plausible mechanism facilitating long-distance dispersal events [6]. These findings have direct implications for biosecurity planning, suggesting that interventions targeting the most highly connected farms and reducing connectivity through managed movements could substantially mitigate the risk of PMCV spread [6, 9].

ORF-Specific Evolution and Functional Implications

The three ORFs of PMCV exhibit distinct evolutionary trajectories and likely perform specialized functions in the viral life cycle, though the precise roles of their protein products remain incompletely defined. ORF1 encodes the putative major capsid protein, which is the most highly conserved of the three ORFs and under strong purifying selection [1, 3]. This conservation likely reflects structural and functional constraints imposed by the capsid’s role in protecting the dsRNA genome, mediating host cell attachment and entry, and potentially evading host immune recognition. The capsid protein has been successfully expressed in Nicotiana benthamiana plants, where it self-assembles into VLPs that morphologically resemble native virions, providing a critical tool for structural and immunological studies [5].

ORF2 encodes the RNA-dependent RNA polymerase (RdRp), the enzymatic machinery responsible for viral genome replication and transcription. The RdRp is similarly highly conserved, as expected for a polymerase with stringent catalytic requirements [1, 3]. Mutations in the RdRp could have profound effects on replication fidelity, viral fitness, and the capacity to generate genetic diversity, though no naturally occurring polymerase variants with altered functional properties have been definitively characterized.

ORF3 is the most enigmatic of the three ORFs and exhibits the greatest sequence variability, with a comparatively relaxed selective constraint relative to ORF1 and ORF2 [1, 3]. The function of the ORF3 protein remains unknown, but its greater tolerance for amino acid substitution suggests that it may play an accessory or modulatory role rather than being essential for core replicative functions. Intriguingly, transcriptional analyses have revealed that the production of full-length transcripts is regulated, with a reduction in the relative number of ORF3-containing transcripts at high transcription rates, suggesting a potential regulatory mechanism that may influence viral gene expression and pathogenesis [4].

Relationship Between Viral Genetics and Pathogenesis

A central question in PMCV research concerns the relationship between viral genetic variation and the severity of cardiac pathology. Despite extensive investigation, no clear association has been established between specific PMCV variants and the severity of heart lesions in either experimental or field settings [1, 3]. Comprehensive analyses of Faroese PMCV genomes found no virulence-determining amino acid substitutions or motifs that correlated with CMS case status, leading to the conclusion that the observed pathology is not attributable to particular viral genotypes [1]. Similarly, Amono et al. [3] examined Norwegian isolates and found no relation between genetic variants and the severity of heart pathology, despite identifying specific non-synonymous and synonymous mutations that might impact protein function and gene expression efficiency.

These findings suggest that host factors, environmental conditions, and husbandry practices may be more significant determinants of disease outcome than intrinsic viral virulence. The identification of quantitative trait loci (QTLs) associated with CMS resistance on chromosomes 12, 23, and 27 provides compelling evidence for a substantial host genetic component in disease susceptibility [18, 20]. Boison et al. [20] identified two chromosomal regions containing single nucleotide polymorphisms associated with CMS resistance, which together explained approximately half of the total genetic variance for the trait. The QTL on chromosome 27 harbors genes with functional roles affecting viral resistance, including magi1, pi4kb, bnip2, and ha1f, while the QTL on chromosome 23 lies in proximity to delta-5 fatty acyl desaturase and fatty acid desaturase 2 genes, linking genetic resistance to polyunsaturated fatty acid metabolism [18, 20]. The estimated SNP-based heritability for viral load and survival phenotypes ranges from 0.12 to 0.55, indicating that genetic improvement through selective breeding represents a viable strategy for mitigating CMS impact [18, 20].

The Role of Defective Viral Genomes and Transcriptional Dynamics

A novel and potentially significant finding in PMCV biology is the presence of defective viral genomes (DVGs), a phenomenon previously unreported for viruses of the order Ghabrivirales [3]. DVGs are truncated viral genomes that arise during replication through polymerase errors or recombination events and can interfere with full-length virus replication, modulate host immune responses, and influence pathogenesis. The identification of DVGs in PMCV-infected tissues suggests that the virus may be subject to additional layers of replication regulation and that DVG-mediated interference could contribute to the variability in disease outcomes observed within infected populations [3].

Transcriptional profiling of PMCV-infected hearts has revealed a striking disconnect between viral genomic RNA load and the extent of pathology. In situ hybridization studies demonstrate intense staining of PMCV RNA in myocardial cells of the spongiform layer of the heart ventricle, with almost no staining in the compact layer, indicating a highly compartmentalized pattern of viral replication [4]. However, the high ratio of viral mRNA to genomic dsRNA in heart tissue suggests active transcription but limited production of new viral particles, leading to the hypothesis that the histopathological changes in CMS are primarily caused by viral mRNA and corresponding viral proteins rather than by the physical accumulation of virions [4]. This observation has profound implications for understanding CMS pathogenesis, as it shifts the focus from direct cytopathic effects of viral assembly to the immunopathological consequences of host recognition of viral transcripts and proteins.

Transmission Pathways and Epidemiological Drivers

The epidemiology of PMCV transmission has been elucidated through a combination of molecular epidemiology, network modeling, and longitudinal surveillance studies. Vertical transmission from infected broodstock to progeny has been demonstrated, representing a critical pathway for viral introduction into hatcheries and subsequent dissemination to sea sites [11]. Jensen et al. [11] detected PMCV in eggs, larvae, fingerlings, and presmolt originating from broodstock with high viral prevalence (98% in hearts, 69% in roe, 59% in milt), establishing that the virus can be carried through early life stages and into the freshwater production phase. This finding provides farmers with options for risk mitigation, including screening of broodstock and selective use of PMCV-negative or low-virus individuals for spawning [11].

Horizontal transmission during the seawater phase drives the majority of between-farm spread and is strongly influenced by connectivity patterns within the aquaculture network. Data-driven network modeling in Ireland demonstrated that local spread accounted for only a small proportion of new infections, while the movement of subclinically infected fish was most important for explaining countrywide dissemination [9]. The model estimated that after assumed initial introduction in 2009, it took approximately five years for between-farm prevalence to reach 100% in late 2014, with older fish being most affected [9]. Targeted interventions focusing on the most connected farms, those with the highest number of outward fish movements, proved highly effective in silico, reducing peak between-farm prevalence from 100% to below 20% when applied proactively [9].

Longitudinal surveillance in Norway has confirmed that PMCV is more widespread than previously recognized, with the virus detected at all 12 study sites and in all sampled cages [10]. Initial infection occurred between one and seven months post-sea transfer, and the median time from infection to clinical CMS outbreak was 6.5 months [10]. Sites that developed clinical CMS had higher viral titers and higher prevalence compared to sites that did not develop disease, and the virus persisted until slaughter at most sites [10]. These findings underscore the importance of monitoring viral load as a potential tool for predicting impending outbreaks, though such screening must be integrated with health status observations and risk factor assessment [10].

Disease Expression and Host–Pathogen Interactions

The clinical manifestation of CMS is highly variable, ranging from subclinical infections with no detectable pathology to catastrophic outbreaks with mortality exceeding 20% in affected cages [10, 19]. This variability is influenced by viral load, host genetics, age, stress, nutrition, and likely other environmental factors. Experimental infection models have been essential for dissecting host–pathogen interactions, though they consistently fail to reproduce the high mortality observed in severe field outbreaks, highlighting the importance of additional co-factors in natural disease expression [5, 14]. The lack of a robust experimental challenge model that fully recapitulates field mortality represents a significant impediment to vaccine development and therapeutic testing [5].

Transcriptomic profiling of infected hearts has provided a detailed picture of the host immune response to PMCV infection and its relationship to disease outcome. The earliest responses involve systemic induction of antiviral and interferon-dependent genes, detectable as early as two weeks post-infection, which subsequently level off during the progression of infection [16, 17]. This initial innate response is followed by a biphasic activation of B cell and MHC antigen presentation pathways, peaking at the time of maximal clinical pathology [16]. A distinct cardiac activation of complement at six weeks post-infection suggests complement-dependent activation of humoral antibody responses, while the peak of cardiac pathology and viral load coincides with cardiac-specific upregulation of T cell response genes and splenic induction of complement genes [16].

Comparison of high responder (HR) fish, characterized by persistent inflammation and necrosis, with low responder (LR) fish that recover without significant pathology has revealed key immunological determinants of disease outcome [17]. Both groups mount equivalent early antiviral and innate immune responses, but differences emerge in the kinetics and magnitude of subsequent adaptive responses. LR fish exhibit earlier activation of NK cell-mediated cytotoxicity and NOD-like receptor signaling pathways, suggesting that more rapid engagement of cytotoxic mechanisms may facilitate viral clearance before pathology becomes established [17]. In contrast, HR fish show a stronger and sustained expression of adaptive immunity genes in heart tissue at late stages, likely reflecting increased lymphocyte infiltration and a persistent, ultimately pathological immune response [17]. This sustained immunopathology occurs despite decreasing viral transcription, indicating that the host immune response itself becomes a major driver of cardiac damage [15, 17].

A detailed comparison of immune responses in different cardiac compartments has revealed that the atrium exhibits higher expression of proinflammatory, anti-inflammatory, and cytotoxic T cell marker genes than the ventricle, correlating with more severe disease progression in the atrial compartment [8]. This compartmentalization of the immune response suggests that regional differences in virus replication, immune cell recruitment, or tissue-specific susceptibility contribute to the heterogeneous distribution of lesions observed in CMS hearts [8]. The presence of IFN-γ and IFNb-producing cells has been confirmed in the ventricles of HR fish but not LR fish, demonstrating that sustained interferon signaling is a hallmark of severe disease and likely contributes to the immunopathological cascade [15].

Nutritional and Environmental Modulation of Disease

The recognition that host nutrition can profoundly influence the outcome of PMCV infection has opened new avenues for disease management. Functional feeds enriched with eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) have been shown to reduce hepatic steatosis associated with CMS, modulate cardiac immune responses, and attenuate the severity of heart lesions in experimental infection models [13]. Fish fed functional feeds with reduced lipid content and increased EPA levels exhibited a milder and delayed inflammatory response and less severe heart lesions at both early and later stages after infection [13]. These effects are mediated through changes in membrane phospholipid composition

Molecular Pathogenesis of PMCV: Genomic Architecture, ORF Functions, and Virulence Determinants

The molecular pathogenesis of piscine myocarditis virus (PMCV) is a paradigm of how a structurally simple double-stranded RNA (dsRNA) virus can induce complex, often severe, immunopathological disease in a teleost host. Despite over a decade of intensive research since its initial characterization, PMCV remains refractory to in vitro propagation, a limitation that has profoundly shaped our understanding of its molecular biology [4, 5, 19]. Consequently, insights into its genomic architecture, the functional roles of its open reading frames (ORFs), and the determinants of its virulence have been derived almost exclusively from in vivo studies, comparative genomics, and transcriptomic profiling of infected Atlantic salmon (Salmo salar L.). This section synthesizes the current state of knowledge, emphasizing the interplay between a highly conserved viral genome, a tightly regulated transcriptional program, and a host immune response that is both the primary driver of pathology and the key to viral clearance.

Genomic Architecture and Phylogenetic Conservation

PMCV possesses a non-segmented, linear dsRNA genome of approximately 7.3 kilobases (kb), a feature that places it within the order Ghabrivirales, with structural and phylogenetic affinities to the family Totiviridae [7, 16, 17, 19]. The genome is organized into three major open reading frames (ORF1, ORF2, and ORF3), a configuration that is remarkably stable across geographic isolates. Whole-genome sequencing efforts, accelerated by the development of amplicon-based and high-throughput methods, have now generated hundreds of PMCV genomes from Norway, the Faroe Islands, Ireland, and Scotland [1-3, 6, 12]. A consistent finding across all these studies is the exceptionally high degree of nucleotide sequence conservation. Phylogenetic analyses reveal a clear clustering by country of origin, with Faroese genomes forming a homogenous monophyletic cluster distinct from Norwegian and Irish lineages [1, 3, 6]. This pattern strongly suggests that PMCV was introduced into the Faroe Islands from Norway, likely via infected broodstock or eggs, and has since evolved in relative isolation without evidence of continuous reintroduction [1]. Similarly, Irish PMCV strains appear to have originated from southern Norway in at least two introduction events, with some lineages potentially linked to wild fish reservoirs [2, 12].

The high genomic conservation is not merely a reflection of a recent common ancestor. Phylodynamic analyses estimate that the most recent common ancestor of contemporary PMCV strains emerged in the 1970s–1990s, coinciding with the global expansion of intensive salmon aquaculture [6]. Despite decades of circulation, the virus has maintained a low genetic diversity, with all three ORFs exhibiting signatures of strong purifying (negative) selection [1, 3]. This indicates that most non-synonymous mutations are deleterious and are rapidly purged from the population, likely because they disrupt essential protein functions. However, a critical nuance emerges when comparing the three ORFs: ORF3 consistently displays a comparatively lower degree of selective constraint than ORF1 or ORF2 [1, 3]. This suggests that ORF3 may tolerate a greater degree of amino acid variation, potentially reflecting its role in host interaction or immune evasion, where some level of plasticity could be advantageous. Indeed, specific non-synonymous and synonymous mutations within ORF3 have been identified that may impact protein function and gene expression efficiency, although no clear association with clinical pathology has been established [3, 12].

Functional Dissection of Open Reading Frames (ORFs)

The functional annotation of the three PMCV ORFs has been inferred primarily through sequence homology, structural predictions, and experimental expression studies, given the absence of a permissive cell line for live virus propagation.

ORF1 encodes the putative major capsid protein (CP). This is the most abundant structural protein and the primary target for host immune recognition. Expression of recombinant ORF1 in plant systems (Nicotiana benthamiana) results in the spontaneous assembly of virus-like particles (VLPs), confirming its intrinsic capacity for capsid formation [5]. The CP is therefore the logical candidate antigen for vaccine development, and VLP-based vaccines have been shown to trigger innate immunity and induce a modest, albeit non-significant, reduction in viral replication in experimental challenge models [5]. The high degree of conservation in ORF1 across all geographic isolates [1, 3] is consistent with its structural role; even minor alterations in the capsid architecture could compromise virion stability or assembly.

ORF2 encodes the RNA-dependent RNA polymerase (RdRp), the enzymatic machinery responsible for viral genome replication and transcription. As the most conserved protein across dsRNA viruses, the ORF2 sequence is the primary target for phylogenetic classification and evolutionary studies [6, 19]. The strong purifying selection acting on ORF2 [1] reflects the functional constraints of an enzyme that must maintain precise catalytic activity for the virus to replicate. No specific virulence determinants have been mapped to ORF2, but its essential role in replication makes it an indirect determinant of viral load and, consequently, of disease severity.

ORF3 is the most enigmatic and arguably the most intriguing component of the PMCV genome. It encodes a small protein of unknown function, with no significant sequence homology to any characterized viral or cellular proteins. Several lines of evidence point to a critical, possibly regulatory, role in pathogenesis. First, as noted, ORF3 is under comparatively relaxed selective pressure, with a higher frequency of non-synonymous substitutions observed in field isolates [1, 3, 12]. Second, the transcription of ORF3 is not constitutive. A landmark study using in situ hybridization (ISH) and quantitative PCR on naturally infected salmon revealed that the production of full-length viral transcripts is tightly regulated, with a specific reduction in the relative number of ORF3-containing transcripts at high transcription rates [4]. This suggests a sophisticated transcriptional control mechanism, possibly involving subgenomic RNA production or alternative splicing, that modulates the expression of ORF3 in response to the intracellular environment or viral load. Third, the ratio of viral mRNA to genomic dsRNA in infected hearts is very high, indicating active transcription but limited production of new viral particles [4]. This observation implies that the histopathological changes characteristic of CMS are driven primarily by the accumulation of viral mRNA and its cognate proteins, including the ORF3 product, rather than by the cytopathic effect of virion assembly. The ORF3 protein may therefore act as a pathogenicity factor, directly contributing to cardiomyocyte dysfunction or modulating the host immune response.

Virulence Determinants and the Role of Defective Viral Genomes

A central question in PMCV research is whether specific genetic determinants of virulence exist. Despite extensive genomic surveillance, no single amino acid substitution or conserved motif has been consistently associated with the severity of CMS pathology [1, 3]. In both Norwegian and Faroese cohorts, high viral load correlates strongly with histopathological damage, but the viral variants themselves do not segregate with disease outcome [3, 10, 16]. This suggests that virulence is not a fixed property of a particular PMCV genotype but is instead an emergent property of the host–virus interaction, modulated by host genetics, immune status, and environmental stressors.

However, a paradigm-shifting discovery has been the identification of defective viral genomes (DVGs) in PMCV populations [3]. DVGs are truncated, non-infectious viral genomes that arise from errors in replication and can interfere with the replication of full-length helper virus. Their presence is a novel finding for viruses of the order Ghabrivirales and has profound implications for pathogenesis. DVGs can act as potent agonists of the innate immune system, particularly the interferon response, and their accumulation has been linked to both viral persistence and disease attenuation in other RNA virus systems. In the context of PMCV, the detection of DVGs in clinical CMS cases suggests that the within-host viral population is not a homogeneous swarm but a dynamic mixture of full-length and defective genomes [3]. The interplay between these populations could influence the trajectory of infection, potentially explaining the wide variation in disease outcomes observed within a single outbreak.

Host-Pathogen Interface and Immunopathological Drivers of Pathogenesis

The molecular pathogenesis of CMS is ultimately defined by the host’s response to the virus. Transcriptomic and proteomic studies have painted a detailed picture of a biphasic immune response that transitions from an early, systemic antiviral state to a late, localized, and often destructive adaptive immune reaction in the heart [15-17].

Early infection (2–4 weeks post-infection, wpi) is characterized by a strong and systemic induction of interferon (IFN)-dependent genes and antiviral effectors such as Mx [7, 8, 16]. This response is initially effective at controlling viral replication, and both high-responder (HR) and low-responder (LR) fish show comparable upregulation of these innate pathways [17]. However, a critical divergence occurs around 6–9 wpi. In LR fish, which ultimately clear the virus and recover without significant pathology, there is an earlier activation of natural killer (NK) cell-mediated cytotoxicity and NOD-like receptor signaling pathways [17]. This suggests that a swift and targeted innate cytotoxic response is sufficient to eliminate infected cells before the virus can establish a foothold.

In contrast, HR fish, which develop severe myocarditis and sustained high viral loads, exhibit a delayed but ultimately overwhelming adaptive immune response. By 9–12 wpi, the hearts of HR fish are infiltrated by large numbers of T cells, particularly CD8+ cytotoxic T lymphocytes, and IFN-γ-secreting cells [15, 17]. RNAscope ISH has confirmed the presence of IFN-γ and IFNb-positive cells in the ventricular myocardium of HR fish, but not in LR fish [15]. This robust T-cell response is associated with a massive upregulation of chemokines, pro-inflammatory cytokines, and genes involved in antigen presentation [15, 16]. While this response is theoretically aimed at viral clearance, it comes at a high cost: the collateral damage to cardiomyocytes, driven by perforin/granzyme-mediated cytotoxicity and inflammatory cytokine release, is the direct cause of the myocardial necrosis and fibrosis that define CMS [15-17].

Critically, this immunopathology persists even as viral transcription begins to decline [15]. The sustained presence of IFN-γ and IFNb-secreting cells in the heart indicates that the immune response becomes self-perpetuating, driven by persistent antigenic stimulation from residual viral RNA or by the release of damage-associated molecular patterns (DAMPs) from dying cells. This phenomenon explains why CMS can progress to severe, often fatal, outcomes even when viral loads are decreasing. The host’s own immune system, rather than the virus itself, is the primary effector of cardiac damage.

The genetic basis for this differential host response is beginning to be elucidated. Genome-wide association studies (GWAS) have identified several quantitative trait loci (QTLs) associated with resistance to CMS. A major QTL on chromosome 27, near genes involved in viral resistance (magi1, pi4kb, bnip2, hla-f), explains a substantial portion of the genetic variance [20]. A second QTL on chromosome 12 has also been confirmed [20]. More recently, a novel QTL on chromosome 23, in proximity to delta-5 fatty acyl desaturase and fatty acid desaturase

Epidemiology of PMCV: Transmission Pathways, Spatiotemporal Clustering, and Reintroduction Dynamics in Farmed Atlantic Salmon

Introduction to PMCV Epidemiology and Global Distribution

Piscine myocarditis virus (PMCV), the etiological agent of cardiomyopathy syndrome (CMS), represents one of the most economically consequential viral pathogens confronting Atlantic salmon (Salmo salar L.) aquaculture across the North Atlantic basin [6, 19]. Since its initial recognition in Norwegian farmed salmon in 1985, CMS has emerged as a persistent threat to production, with approximately 100 outbreaks reported annually in Norway alone and substantial economic losses estimated at up to €9 million per year [5, 13]. The disease has subsequently been documented in all major salmon-producing regions, including the Faroe Islands, Scotland, and Ireland, where the first recorded outbreak occurred in 2012 [9, 12, 19]. The epidemiological complexity of PMCV transmission is compounded by the virus’s resistance to propagation in conventional cell culture systems, which has historically impeded detailed virological characterization and the development of effective prophylactic interventions [4, 5]. The World Organisation for Animal Health (WOAH) recognizes CMS as a significant transboundary disease of aquatic animals, underscoring its importance for international trade and biosecurity protocols. Understanding the transmission pathways, spatiotemporal clustering patterns, and reintroduction dynamics of PMCV is therefore paramount for designing evidence-based control strategies that can mitigate the substantial welfare and economic burdens imposed by this pathogen.

Transmission Pathways: Horizontal, Vertical, and Environmental Vectors

The transmission ecology of PMCV is multifaceted, involving both direct horizontal transmission among farmed cohorts and compelling evidence for vertical transmission from infected broodstock to progeny. Horizontal transmission is widely considered the primary mechanism of viral spread within and between marine production sites, facilitated by the high density of fish in sea cages and the continuous shedding of virus from infected individuals [9, 10]. Longitudinal epidemiological studies conducted in Norway between 2015 and 2018 revealed that PMCV was detected at all 12 monitored sites and in every sampled cage, with initial infection occurring between 1 and 7 months post-sea transfer [10]. This ubiquitous presence underscores the efficiency of horizontal transmission once the virus is introduced into a marine environment. The median time from initial infection to clinical outbreak of CMS was 6.5 months, with sites experiencing clinical disease exhibiting significantly higher viral titers and prevalence compared to subclinically infected sites [10]. Importantly, the virus persisted until slaughter in 11 of 12 sites, indicating that once established, PMCV becomes endemic within a population and is not readily cleared [10].

The role of subclinically infected fish in PMCV transmission cannot be overstated. Data-driven network modeling of the Irish salmon farming industry demonstrated that the movement of subclinically infected fish was the most important factor explaining the observed countrywide spread of PMCV, with local spread accounting for only a small proportion of new infections [9]. The model indicated that after an assumed introduction in 2009, it took only 5 years for between-farm prevalence to reach 100% by late 2014, with older fish being disproportionately affected [9]. This finding aligns with phylogenetic evidence from Ireland suggesting that PMCV may have been introduced in two distinct waves originating from southern Norway, with over three-quarters of sequenced strains coming from fish not exhibiting clinical signs of CMS [12]. The existence of a substantial subclinical reservoir presents a formidable challenge for surveillance and control, as apparently healthy fish can serve as silent vectors for viral dissemination across extensive geographic distances.

Perhaps the most significant advance in understanding PMCV transmission has been the demonstration of vertical transmission from infected broodstock to their offspring. Jensen et al. (2019) provided compelling evidence for this pathway, detecting PMCV in eggs, larvae, fingerlings, and presmolts originating from broodstock with a 98% prevalence of infection in heart tissue, 69% in roe, and 59% in milt [11]. The virus was detected in all progeny stages up to and including the 40 g stage, and was also present in presmolt samples examined for tissue tropism [11]. This finding has profound implications for hatchery management and biosecurity, as it suggests that vertical transmission may represent a mechanism for viral persistence across generations and for introduction of PMCV into naive populations through the movement of infected eggs or juveniles. The detection of PMCV in post-smolts shortly after sea transfer further supports the hypothesis of carry-over from freshwater hatcheries [10, 11]. The Faroese experience is particularly instructive in this regard: an outbreak of CMS in the Faroe Islands was linked to the importation of eggs from a CMS-endemic area, and subsequent genomic analysis confirmed that PMCV was introduced into the Faroe Islands from Norway, where brood fish are known to be infected [1, 11].

Environmental transmission pathways, while less well-characterized, warrant consideration. The detection of PMCV RNA in all organs of infected fish, with the highest loads in heart, kidney, and spleen, suggests that viral shedding into the aquatic environment through feces, urine, and sloughed epithelial cells is plausible [4]. The high ratio of viral mRNA to genomic dsRNA observed in heart tissue indicates active transcription but limited production of new viral particles, which may influence the efficiency of environmental transmission [4]. The role of alternative hosts or environmental reservoirs remains an open question. Efforts to identify fungal hosts have been inconclusive, with fungal sequences found inconsistently in salmon tissue samples [4]. The possibility that PMCV may persist in marine invertebrates or other fish species has not been rigorously excluded and represents a critical knowledge gap for comprehensive risk assessment.

Spatiotemporal Clustering and Phylogeographic Structure

The genomic epidemiology of PMCV has revealed a striking pattern of national clustering, with viral genomes forming distinct monophyletic groups corresponding to their country of origin [1, 3, 6]. This phylogeographic structure is evident across the North Atlantic, with Norwegian, Faroese, Irish, and Scottish isolates each occupying separate clades in phylogenetic reconstructions [1, 3, 6]. The most comprehensive phylodynamic analysis to date, encompassing one of the largest genomic datasets for any aquatic viral pathogen, estimated that PMCV likely emerged in farmed salmon concurrent with the global expansion of aquaculture, with a most recent common ancestor dating to approximately the 1970s–1990s [6]. This timeline coincides with the intensification of salmon farming and the increased movement of live fish and ova across international borders.

Within Norway, the epicenter of PMCV diversity and the likely source population for other regions, phylogenetic analyses have revealed a more complex picture. While overall PMCV sequences are highly conserved, single CMS outbreaks frequently harbor multiple viral variants, and temporal data indicate increasing sequence diversity in the population over time [3]. This intra-outbreak diversity suggests that multiple viral lineages can coexist within a single farm, potentially reflecting multiple introductions or the generation of de novo diversity through mutation and recombination. The presence of defective viral genomes, a novel finding for viruses of the order Ghabrivirales, may further contribute to the evolutionary dynamics and epidemiological behavior of PMCV [3]. The identification of specific non-synonymous and synonymous mutations that may impact protein function and gene expression efficiency highlights the ongoing adaptive evolution of the virus within its host population [3].

The Faroese archipelago presents a particularly instructive case study in spatiotemporal clustering and the consequences of founder effects. Genomic analysis of 48 novel Faroese PMCV genomes, representing broad spatiotemporal coverage, revealed a remarkably homogenous monophyletic cluster compared to Norwegian and Irish isolates [1]. Principal component analysis demonstrated no spatiotemporal clustering of genotypes, nor any clustering based on roe or smolt origin [1]. This homogeneity is consistent with a single introduction event, likely from Norway, followed by local circulation without continuous reintroduction of novel strains. Notably, one genome obtained from a returning wild salmon differed considerably from all farmed isolates and formed an outgroup, suggesting that wild salmon may harbor genetically distinct PMCV lineages that are not currently circulating in the farmed population [1]. This finding raises important questions about the role of wild fish as reservoirs or sentinels for PMCV diversity.

In Ireland, phylogenetic analysis of PMCV open reading frames (ORFs) 1 and 3 revealed a largely homogenous viral population with limited genetic diversity [12]. However, several amino acid positions within both ORF1 and ORF3 showed consistent variations unique to Irish strains when compared with Norwegian sequences [12]. The phylogeny suggested two waves of introduction from southern Norway, with subsequent local evolution [12]. More recent whole-genome sequencing has revealed that over 80% of the genetic diversity of PMCV lies outside the commonly sequenced ORFs, underscoring the importance of complete genome sequencing for accurate phylogeographic inference [2]. Comparison of Irish and Faroese sequences further suggests that some strains in Ireland may originate from wild fish, adding another layer of complexity to the transmission dynamics [2].

Scotland exhibits a distinct epidemiological pattern, with phylodynamic analyses identifying multiple introductions of distinct PMCV lineages into the country [6]. This pattern of repeated introduction contrasts with the single introduction event observed in the Faroe Islands and suggests that Scotland may be subject to ongoing viral incursions, likely from Norway or other endemic regions [6]. The factors facilitating these long-distance dispersal events are not fully understood, but the possible role of boat connectivity, including well-boat transport of live fish and movement of equipment, has been highlighted as a mechanism for long-distance viral dissemination [6].

Reintroduction Dynamics and the Role of Anthropogenic Vectors

The reintroduction dynamics of PMCV are intimately linked to anthropogenic activities, particularly the international trade in salmonid ova and the movement of live fish. The Faroese experience provides the clearest example of the risks associated with egg importation. Despite a steadily increasing import of Norwegian roe following the reemergence of CMS in 2013, genomic surveillance has demonstrated no continuous reintroduction of persisting PMCV strains to Faroese farmed salmon [1]. This finding suggests that while the initial introduction was likely via infected eggs, subsequent outbreaks have been driven by local circulation of the established founder strain rather than repeated importation of novel variants [1]. However, the detection of PMCV in 69% of roe samples from infected broodstock indicates that the potential for reintroduction through this route remains substantial [11].

In Norway, the detection of PMCV at all 12 monitored sites in a longitudinal study, despite varying clinical outcomes, indicates that the virus is far more widespread than previously recognized [10]. This ubiquity suggests that reintroduction may be less of a concern than the management of endemic infection. However, the identification of multiple PMCV variants within single outbreaks and the occasional grouping of sequences from different cases indicate that viral movement between farms does occur, likely facilitated by the movement of subclinically infected fish or contaminated equipment [3]. The importance of geographic proximity and at-sea farm density as drivers of viral spread has been quantitatively demonstrated, with network models highlighting the role of farm-to-farm linkages through outward fish movements in sustaining transmission [6, 9].

The potential for PMCV reintroduction through wild salmon populations represents an underexplored but potentially significant pathway. The discovery of a genetically distinct PMCV genome in a returning wild salmon in the Faroe Islands, which formed an outgroup to all farmed isolates, suggests that wild fish may harbor viral lineages that are not currently circulating in aquaculture [1]. Whether these wild strains represent ancient lineages that have been maintained in the wild population or recent introductions from an unknown source remains unclear. The detection of PMCV in wild Atlantic salmon in Norway further supports the possibility of a wild reservoir [19]. The implications for disease management are profound: if wild salmon can serve as a source of viral reintroduction to farmed populations, then biosecurity measures must extend beyond the control of anthropogenic movements to include consideration of interactions with wild fish.

The role of boat connectivity in facilitating long-distance dispersal events has been increasingly recognized as a critical factor in PMCV epidemiology [6]. Well-boats, which are used to transport live fish between sites, can serve as mechanical vectors for viral transmission if not properly disinfected. The movement of equipment, personnel, and processing waste between farms may also contribute to viral spread. The relative importance of these different anthropogenic vectors likely varies by region and production system, but their cumulative effect is to create a highly connected network that facilitates rapid viral dissemination once introduced.

Genetic Diversity, Selection Pressures, and Implications for Transmission

The genomic landscape of PMCV is characterized by high conservation, with all three open reading frames (ORF1, ORF2, and ORF3) exhibiting signs of purifying selection [1, 3]. This low genetic diversity is consistent with a relatively recent emergence and a stable host-virus relationship. However, ORF3 displays a comparatively lower degree of selective constraint than ORF1 and ORF2, suggesting that it may be under less stringent functional constraints or may be evolving in response to host immune pressure [1]. The identification of specific non-synonymous mutations that might impact protein function and gene expression efficiency indicates that adaptive evolution is occurring, even within the context of overall genomic stability [3].

The absence of virulence-determining amino acid substitutions in Faroese PMCV genomes, with no association found between CMS cases and specific amino acid substitutions or motifs, suggests that virulence may be more strongly influenced by host factors and environmental conditions than by viral genotype [1]. This finding is consistent with the observation that single CMS outbreaks can display multiple PMCV variants, and that no clear relation has been found between viral variants and the severity of heart pathology [3]. The hypothesis that multiple bottlenecks and changing infection dynamics in the host population, with transfer to naive individuals over time, represent a continuous selection pressure on viral populations provides a framework for understanding the observed patterns of genetic diversity [3].

The detection of defective viral genomes (DVGs) in PMCV represents a novel finding for viruses of the order Ghabrivirales and may have significant implications for transmission dynamics [3]. DVGs are truncated viral genomes that can interfere with the replication of full-length viruses and modulate the host immune response. The presence of DVGs in PMCV infections could influence viral fitness, transmission efficiency, and the severity of clinical disease. Further research is needed to determine the prevalence of DVGs in different epidemiological contexts and their impact on PMCV transmission and pathogenesis.

Host Factors Influencing Transmission and Disease Outcome

Individual host susceptibility plays a critical role in determining the outcome of PMCV infection and, by extension, the transmission dynamics within a population. Genetic studies have identified quantitative trait loci (QTLs) on chromosomes 12, 23, and 27 that are associated with resistance to CMS [18, 20]. The QTL on chromosome 23 is particularly noteworthy, as it is in proximity to delta-5 fatty acyl desaturase and fatty acid desaturase 2 genes, both of which play a role in the production of polyunsaturated fatty acids [18]. This genetic link between fatty acid metabolism and viral resistance provides a mechanistic basis for the observed effects of dietary lipid composition on disease outcome [13, 14]. The estimated SNP-based heritability for resistance to PMCV ranges from 0.12 to 0.55, indicating that genetic improvement through selective breeding is feasible [18, 20].

The host immune response to PMCV infection is characterized by a complex interplay between innate and adaptive immunity, with the outcome of infection determined by the timing and magnitude of these responses. Transcriptomic studies have revealed that fish with severe disease outcomes (high responders) exhibit a sustained and exaggerated adaptive immune response in the heart, characterized by increased expression of chemokines, antiviral response molecules, and Th1 pro-inflammatory genes [15, 17]. In contrast, fish that recover (low responders) show earlier activation of NK cell-mediated cytotoxicity and NOD-like receptor signaling pathways, followed by a controlled adaptive immune response and activation of genes involved in cardiac energy metabolism [17]. The presence of IFN-γ and IFNb-secreting cells in the hearts of high responders but not low responders suggests that persistent antiviral responses and sustained immunopathology contribute to severe disease, even as viral transcription declines [15].

The implications of these host factors for transmission dynamics are significant. Fish with high viral loads and severe pathology are likely to shed more virus into the environment, increasing the force of infection for cohabiting individuals. Conversely, fish that mount an effective early immune response and clear the infection may contribute less to onward transmission. The observation that PMCV can persist in lymphoid tissues, including the head kidney and spleen, with no signs of clearance, suggests that these tissues may serve as reservoirs for long-term viral persistence and potential reactivation [8]. This persistence could contribute to the maintenance of infection within a population even after clinical disease has resolved.

Conclusions and Future Directions for Epidemiological Research

The epidemiology of PMCV in farmed Atlantic salmon is characterized by a complex interplay of horizontal and vertical transmission pathways, distinct phylogeographic structure, and dynamic reintroduction patterns driven by anthropogenic activities. The virus is far more widespread than previously recognized, with subclinical infections serving as a silent reservoir for dissemination across extensive geographic distances. The demonstration of vertical transmission from infected broodstock to progeny has profound implications for hatchery management and the international trade in salmonid ova. The distinct national clustering of PMCV genomes reflects a combination of founder effects, local evolution, and sporadic long-distance dispersal events, with the Faroe Islands representing a clear example of a single introduction followed by endemic circulation.

The identification of genetic markers associated with resistance to CMS offers promise for selective breeding programs, while the observed effects of dietary lipid composition on disease outcome suggest that nutritional interventions may provide a complementary approach to disease management. The development of effective vaccines remains a priority, but the inability to propagate PMCV in cell culture has hindered progress [5]. The recent establishment of experimental challenge models and the application of plant-produced virus-like particle vaccines represent important steps forward [5].

Future epidemiological research should focus on several key areas: (1) the role of wild fish populations as reservoirs and sources of reintroduction; (2) the contribution of environmental transmission pathways, including the potential for persistence in marine invertebrates; (3) the impact of defective viral genomes on transmission dynamics and virulence; (4) the development of non-lethal diagnostic tools for early detection of infection; and (5) the integration of genomic surveillance with network modeling to predict and prevent viral spread. The application of whole-genome sequencing to large-scale epidemiological studies, as demonstrated by recent work in Norway, the Faroe Islands, Ireland, and Scotland, provides the foundation for a comprehensive understanding of PMCV transmission dynamics [1-3, 6]. As the global salmon farming industry continues to expand, the insights gained from these studies will be essential for developing evidence-based biosecurity strategies that can mitigate the impact of this economically significant pathogen.

Diagnostics and Whole-Genome Sequencing of PMCV: Amplicon-Based Methods, MinION vs. Illumina Comparison, and Field Sample Application

The advancement of molecular diagnostics for piscine myocarditis virus (PMCV) has been historically constrained by the intractable nature of this double-stranded RNA (dsRNA) virus, which, akin to members of the Totiviridae family, has consistently resisted propagation in conventional cell culture systems [4, 5, 19]. This fundamental limitation has necessitated a paradigm shift toward direct molecular detection and genomic characterization from clinical specimens, placing diagnostics and whole-genome sequencing (WGS) at the epicenter of PMCV research and cardiomyopathy syndrome (CMS) surveillance. The development of robust, field-deployable sequencing methodologies has proven indispensable for elucidating viral transmission pathways, evolutionary dynamics, and population-level genetic diversity, which are critical for informing biosecurity strategies in an industry where CMS causes substantial economic losses and compromises fish welfare across the North Atlantic [1, 6, 9, 19]. The following sections provide an exhaustive examination of the contemporary diagnostic and sequencing landscape for PMCV, with a specific focus on amplicon-based WGS approaches, a critical comparative analysis of MinION versus Illumina sequencing platforms, and the practical application of these technologies to field samples.

Amplicon-Based Whole-Genome Sequencing: Overcoming the Cell Culture Bottleneck

The inability to isolate PMCV in vitro [4, 5] has driven the innovation of amplicon-based WGS protocols that bypass the need for viral propagation by directly amplifying the viral genome from host tissue RNA. The foundational work by Dahl et al. (2025) [1] established a rapid and cost-effective amplicon-based method that has become a cornerstone for large-scale genomic surveillance. This approach was specifically designed to generate complete PMCV genomes directly from total RNA extracted from heart tissue homogenates, a strategy that leverages the high viral loads typically present in the myocardium of infected fish [1, 4]. The method involves a multiplexed PCR amplification of contiguous, overlapping fragments that tile across the entire ~8.5 kb dsRNA genome, which encodes three open reading frames (ORF1, ORF2, and ORF3) [1, 3]. The success of this targeted amplification hinges on the use of highly conserved primer-binding sites, identified through alignment of the initial PMCV reference genomes. The resultant amplicons are then subjected to high-throughput sequencing, primarily on the Illumina platform, providing sufficient depth to resolve the viral quasi-species present within a single host [1, 3].

Critically, this amplicon-based strategy has revealed that over 80% of the total genetic diversity of PMCV resides outside the commonly analyzed ORF1 and ORF3 regions [2]. This finding underscores a profound limitation of earlier diagnostic approaches that relied solely on partial gene sequencing (e.g., ORF1 fragments) for molecular epidemiology [12]. The non-coding regions and the ORF2 (encoding the putative capsid protein) harbor a substantial reservoir of genetic variation that is essential for robust phylogenetic inference and transmission network reconstruction [2, 6]. The Dahl method [1] has been successfully applied to generate 48 novel PMCV genomes from Faroese farmed salmon, demonstrating its efficacy on a wide range of sample qualities and viral loads typical of field submissions. Subsequent large-scale studies have built upon this foundation, generating extensive genomic datasets (e.g., 34 novel genomes from Norway as per Amono et al., 2024 [3]) that collectively have transformed our understanding of PMCV phylogeography [6].

A Comparative Analysis of Sequencing Platforms: MinION vs. Illumina

The choice of sequencing platform is a pivotal decision in any genomic surveillance program, balancing considerations of accuracy, throughput, cost, portability, and turnaround time. For PMCV, a direct comparison of the Oxford Nanopore MinION (long-read) and the Illumina MiSeq (short-read) platforms has been rigorously conducted by Tighe et al. (2025) [2], providing critical empirical data for researchers. This study [2] generated a full PMCV genome from an Irish field isolate using both platforms, enabling a head-to-head assessment of sequence fidelity and utility for downstream analyses.

The results demonstrated that the MinION-derived consensus genome shared 99.59% identity with the Illumina-derived genome [2]. At face value, this high degree of concordance suggests that MinION sequencing is a viable tool for applications such as rapid pathogen identification, confirmation of PMCV presence in new geographic regions, and addressing deep evolutionary questions (e.g., establishing the monophyly of country-level clades [1, 6]). The MinION’s advantages, real-time data generation, low capital cost, and extreme portability, make it an attractive option for near-source diagnostics in remote aquaculture settings or in scenarios where rapid outbreak response is paramount [2]. Furthermore, its long-read capability can potentially resolve complex genomic structures, such as the defective viral genomes (DVGs) recently discovered in PMCV, which are hypothesized to impact replication dynamics [3].

However, the 0.41% residual discrepancy between the two platforms is not negligible for fine-scale molecular epidemiology. Tighe et al. (2025) [2] explicitly caution that the error rate of the MinION platform, particularly for homopolymer regions, renders it “insufficient for accurately tracking viral transmission pathways.” When tracking the spread of a highly genetically homogeneous virus like PMCV, where transmission links may be defined by a single nucleotide polymorphism (SNP) or a short indel, the error model of nanopore sequencing introduces unacceptable ambiguity [2, 3]. In contrast, the Illumina platform, with its significantly lower per-base error rate (~0.1% vs. ~5-15% raw read accuracy for MinION), provides the high-fidelity consensus sequences necessary for robust phylodynamic analyses that discern between alternative hypotheses of viral dispersal (e.g., single introduction vs. continuous reintroduction) [1, 2, 6]. Therefore, a practical and cost-effective workflow typically involves using the MinION for initial screening and real-time monitoring, followed by Illumina sequencing of select positive cases for high-resolution phylogenetic and population genetic analyses [2]. This hybrid approach maximizes the strengths of both technologies while mitigating their individual weaknesses.

Application to Field Samples: From Tissue Homogenates to Evolutionary Insights

The successful application of WGS to field samples has been the linchpin of recent advances in PMCV epidemiology. The optimized amplicon-based protocols have proven robust across a variety of sample matrices, though heart tissue remains the gold standard due to the extraordinarily high viral loads in the myocardium, particularly in the spongiform layer of the ventricle, where PMCV RNA is most intensely detected via in situ hybridization (ISH) [4, 10]. However, studies have also demonstrated the utility of kidney and spleen tissues for diagnostic detection, as PMCV is present in all organs during active infection, albeit at lower concentrations [4, 8, 10]. This pan-tropic distribution has important implications for sampling strategies, particularly for non-lethal monitoring where gill or blood biopsies offer a humane alternative [7, 21, 22].

The raw material for WGS is typically total RNA extracted from homogenized tissue. The quality and quantity of input RNA are critical; the presence of host-derived ribosomal RNA can interfere with virus-specific amplification [1]. However, the amplicon-based approach is more tolerant of degraded or mixed RNA than metagenomic sequencing, as the targeted PCR step selectively enriches for viral sequences of interest [1, 2]. The application of these methods to field samples has yielded transformative insights into the evolution and spread of PMCV. Whole-genome sequencing of 48 Faroese genomes revealed a remarkably homogenous, monophyletic cluster, suggesting a single, stable introduction of the virus from Norway, with no evidence of continuous reintroduction despite a steady import of Norwegian roe [1]. This contrasts with the pattern observed in Ireland, where phylogenetic data suggest the possibility of multiple introductions, perhaps even from wild fish reservoirs, adding a layer of complexity to the epidemiological picture [2, 12].

Furthermore, the depth of data provided by WGS has enabled the detection of intra-host viral diversity. Single CMS outbreaks are now known to harbor multiple distinct PMCV variants, and temporal sampling has revealed increasing sequence diversity over time within populations, likely driven by continuous selection pressure as the virus is transmitted to naïve individuals in successive production cycles [3]. The identification of specific non-synonymous mutations in ORF1 and ORF3, while not yet definitively linked to virulence, provides a mechanistic basis for investigating potential adaptation to the host [3, 19]. The recent discovery of DVGs in PMCV, a first for the order Ghabrivirales, was only possible through deep sequencing of field isolates and opens a new avenue for understanding viral persistence and the host immune response [3]. Finally, the large-scale phylodynamic analysis by Zhao et al. (2026) [6], which sequenced hundreds of PMCV genomes from Norway and Scotland, used field sample data to estimate the virus’s emergence date (1970s–1990s, concurrent with the expansion of industrial aquaculture) and to quantify the roles of geographic proximity, farm density, and vessel connectivity in driving long-distance dispersal events.

In summary, the diagnostic landscape for PMCV has evolved from reliance on clinical signs and histopathology to a sophisticated, genomics-driven discipline. Amplicon-based WGS, particularly when executed on the Illumina platform, provides the resolution needed to disentangle the complex transmission networks of this economically devastating virus. The complementary use of MinION sequencing for rapid, field-based diagnostics offers a powerful tool for real-time monitoring. The application of these methods to field samples has not only confirmed the country-level genetic structure of PMCV but has also revealed the hidden complexity of intra-host dynamics and the potential role of novel genomic elements like DVGs. This genomic infrastructure is essential for informing targeted biosecurity measures, as advocated by the WOAH and FAO for the sustainable management of aquatic animal health, and for underpinning future strategies in selective breeding for resistance and vaccine development [6, 9, 18, 20].

Genetic Diversity and Evolutionary Trajectories of PMCV: Purifying Selection, Homogeneity, and Outgroup Analysis

The genomic architecture of piscine myocarditis virus (PMCV) presents a compelling paradox within the landscape of RNA virus evolution. While many RNA viruses exhibit high mutation rates and extensive genetic diversity driven by error-prone polymerases, PMCV demonstrates a remarkable degree of genomic conservation, particularly within its open reading frames (ORFs). This apparent stability, however, belies a complex evolutionary narrative shaped by purifying selection, distinct population bottlenecks, and the emergence of divergent lineages that challenge our understanding of the virus's transmission ecology. A comprehensive analysis of the global PMCV sequence dataset, now significantly expanded through recent genomic surveillance efforts, reveals that the virus is subject to strong selective constraints while simultaneously exhibiting population-level homogeneity, interrupted by rare but phylogenetically significant outgroup lineages that provide crucial insights into its evolutionary origins and dispersal history.

Purifying Selection Across Open Reading Frames

A defining feature of PMCV evolution is the pervasive influence of purifying (negative) selection across all three ORFs, a finding consistently reported across multiple independent studies and geographical regions. Genome-wide analyses conducted by Dahl et al. [1] on Faroese PMCV genomes, representing a comprehensive spatiotemporal dataset, demonstrated unequivocal signatures of purifying selection acting on ORF1, ORF2, and ORF3. This indicates that most non-synonymous mutations are deleterious and are promptly removed from the viral population, a pattern typical of viruses with highly constrained functional requirements. This selective regime is likely driven by the dual pressures of maintaining essential protein functions, including the viral RNA-dependent RNA polymerase (RdRp) encoded by ORF1 and the structural capsid proteins, and navigating the host immune environment. Critically, however, Dahl et al. [1] identified a comparatively lower degree of selective constraint on ORF3 relative to ORF1 and ORF2. This differential selection pressure suggests a more permissive evolutionary landscape for the protein encoded by ORF3, which may function as a non-structural or accessory protein, potentially involved in host immune modulation or other auxiliary roles where minor structural variations are more tolerable.

This pattern of ORF3 relaxation is corroborated by the larger-scale Norwegian study by Amono et al. [3], which added 34 complete genomes and 202 novel ORF3 sequences to the public repository. Their phylogenetic analyses confirmed that while the overall genomic landscape is highly conserved, ORF3 exhibits a higher proportion of variable sites, contributing disproportionately to the observed genetic diversity. Importantly, Amono et al. [3] identified specific non-synonymous and synonymous mutations within ORF3 that may impact protein function and gene expression efficiency, respectively. The identification of synonymous mutations with potential functional consequences introduces a layer of complexity, suggesting that even silent changes can influence codon usage bias, mRNA stability, or translational kinetics, thereby shaping viral fitness. The presence of defective viral genomes (DVGs), a novel finding for the order Ghabrivirales, further complicates the evolutionary dynamics [3]. DVGs are truncated or rearranged genomes that arise during replication and can modulate the host immune response, potentially exerting additional selective pressures on the intact viral population by competing for replication resources or triggering antiviral pathways.

The biological basis for this purifying selection is multifaceted and deeply intertwined with the virus's replication strategy and transmission cycle. As a double-stranded RNA (dsRNA) virus with structural similarities to the Totiviridae family [16, 17, 19], PMCV likely possesses a relatively stable replication complex that, while not proofreading-proficient in the classical sense of large DNA viruses, may still constrain mutational tolerance. More critically, the virus's ecology within the Atlantic salmon host imposes severe bottlenecks during transmission. The findings of Nyman et al. [4] are particularly instructive: their in situ hybridization studies revealed intense staining of PMCV RNA in the spongiform layer of the heart ventricle, with almost no staining in the compact layer, and only sporadic detection in the kidneys. This highly localized replication pattern, coupled with the observation that high viral loads in the heart were due to extensive mRNA transcription rather than abundant production of new viral particles, suggests a replication strategy that prioritizes transcriptional activity over particle assembly. The high ratio of viral mRNA to genomic dsRNA indicates active transcription but limited production of infectious virions [4]. Such a strategy would inherently reduce the effective population size of actively replicating genomes, intensifying the stochastic effects of genetic drift and purifying selection, as most newly arising mutations would be eliminated before they could be packaged and transmitted. Furthermore, the observation in the same study that the production of full-length transcripts is regulated, with a reduction in the relative number of ORF3-containing transcripts at high transcription rates [4], suggests a dynamic post-transcriptional regulatory mechanism that could differentially affect the evolutionary trajectory of ORF3 compared to the other ORFs.

Population Homogeneity and Geographic Structuring

The overwhelming signal emerging from genomic surveillance across the North Atlantic is the pronounced homogeneity of PMCV populations, particularly within national or regional cohorts. This homogeneity is most strikingly exemplified by the Faroese PMCV isolates. Dahl et al. [1] demonstrated that all Faroese genomes, derived from a broad spatiotemporal range of farmed Atlantic salmon, formed a tight, monophyletic cluster when compared to Norwegian and Irish sequences. Principal component analyses (PCA) further substantiated this finding, revealing no spatiotemporal clustering of genotypes based on geographic location within the Faroe Islands, nor any clustering based on roe or smolt origin [1]. This lack of internal structure indicates that the Faroese PMCV population was founded by a single or very limited introduction event and has since undergone predominantly clonal expansion with minimal ongoing genetic differentiation. The implication is profound: despite a steadily increasing import of Norwegian roe, from which PMCV is known to be transmitted vertically [11], there has been no continuous reintroduction of persisting, genetically distinct PMCV strains into Faroese farmed salmon [1]. This suggests either that the imported roe is free of the virus, that the viral strains present in imported material are rapidly outcompeted by the resident founder strain, or that immigration events are so rare and the population so well-adapted that they leave no detectable phylogenetic footprint.

This pattern of national clustering is not unique to the Faroe Islands. Zhao et al. [6], in one of the most comprehensive phylodynamic analyses ever conducted for an aquatic viral pathogen, generated a large genomic dataset from Scotland and Norway and demonstrated distinctive national clustering of PMCV genomes. Their analyses revealed that the global PMCV population likely emerged in farmed salmon concurrent with the expansion of aquaculture, with a most recent common ancestor (MRCA) estimated around the 1970s–1990s [6]. This relatively recent emergence is consistent with the first recognition of CMS in Norway in 1985 [19]. The phylodynamic data indicate that virus spread is predominantly intranational, characterized by dissemination within a country's farm network, interspersed with multiple introductions of distinct PMCV lineages into Scotland [6]. The drivers of this intranational spread are critically linked to farm connectivity: geographic proximity, high at-sea farm density, and boat-mediated connectivity all facilitate long-distance dispersal events [6]. This network-driven transmission, as also modeled by Yatabe et al. [9] for Ireland, explains how a genetically homogenous virus can spread rapidly across a region yet maintain its overall sequence conservation. The virus is effectively moving through a well-connected host population, encountering frequent population bottlenecks as it jumps between farms, which further purges genetic variation.

The Irish PMCV population also exhibits substantial homogeneity, although with important nuances. Tighe et al. [12] sequenced ORF1 and ORF3 from Irish farmed salmon and found that PMCV was largely homogenous, mirroring the Faroese pattern. However, they identified several consistent amino acid substitutions unique to Irish strains when compared with Norwegian sequences [12]. This suggests that following its introduction into Ireland, likely from southern Norway [12], the virus underwent a period of independent evolution, accumulating a small number of fixed, population-specific mutations. This pattern aligns with the concept of a founder effect followed by local adaptation or genetic drift. More strikingly, Tighe et al. [2], using whole-genome sequencing, revealed that over 80% of the total genetic diversity of PMCV in Irish strains lies outside the commonly sequenced ORFs. This finding underscores the critical importance of whole-genome sequencing for understanding the true evolutionary landscape of PMCV. The non-coding regions or intergenic sequences may harbor regulatory elements or RNA structural motifs under different selective pressures, and their exclusion from routine surveillance could lead to a significant underestimation of viral diversity.

The biological mechanisms underlying this homogeneity extend beyond simple founder effects. The strong purifying selection described previously acts as a powerful homogenizing force, preventing the accumulation of neutral or slightly deleterious mutations that would otherwise generate diversity. Additionally, the host immune response may impose a homogenizing effect by selecting for a narrow range of antigenic variants. The detailed immune profiling of CMS by Timmerhaus et al. [16, 17] and Sun et al. [15] reveals that severe outcomes are associated with a robust, persistent adaptive immune response, including cytotoxic T cells and interferon-gamma (IFN-γ) secreting cells. This T-cell-mediated pressure could select for viral variants that escape immune recognition, but the apparent lack of such immune-escape mutations in the population [3] suggests that either the virus has evolved a highly effective strategy to avoid immune detection (e.g., by modulating replication or hiding within cells), or that the capsid structure is so constrained that any mutation that alters an epitope is also catastrophically deleterious. The observation that specific non-synonymous mutations in ORF1 and ORF3 were not associated with the severity of heart pathology [3] further supports the idea that virulence in CMS is not determined by particular PMCV genotypes but is instead a complex interplay of host genetics, environmental stressors, and viral load dynamics.

Outgroup Analysis and Evolutionary Origins

Amidst the sea of genomic homogeneity, the identification of phylogenetically distinct outgroup genomes provides critical windows into the evolutionary history and potential origins of PMCV. The most striking example comes from the Faroe Islands, where a single genome obtained from a returning wild Atlantic salmon differed considerably from all the farmed Faroese isolates, forming a clear outgroup in phylogenetic analyses [1]. This singular finding is of paramount importance. It suggests that wild Atlantic salmon harbor PMCV lineages that are genetically distinct from those circulating in the intensively sampled farmed populations. This wild reservoir may represent an ancestral or divergent lineage that has been maintained in the wild fish population, potentially for a longer evolutionary timescale, and has not yet become established in aquaculture environments. Alternatively, this outgroup genome could represent a rare introduction from a geographically distant wild population, such as from the North American stock of Atlantic salmon, with which European salmon rarely interact.

The existence of wild outgroup lineages is further supported by findings from Ireland. Tighe et al. [2] reported that comparing Irish and Faroese genomes revealed that some strains of PMCV in Ireland may actually originate from wild fish. This is a critical observation that challenges the paradigm of aquaculture as the sole reservoir and amplifier of PMCV. It raises the possibility of bidirectional movement of PMCV between farmed and wild compartments, or that wild salmon serve as a persistent, low-level source of viral diversity that occasionally spills over into farmed populations. The epidemiological study by Svendsen et al. [10], which detected PMCV at all 12 monitored Norwegian farm sites and in all sampled cages, underscores the ubiquity of the virus in the farming environment. However, the discovery of a divergent wild genome in the Faroe Islands [1] suggests that the virus's natural ecology extends beyond the farm net-pen.

The phylodynamic analyses by Zhao et al. [6] provide a temporal and ancestral context for these outgroup relationships. Their estimation of the PMCV MRCA in the 1970s–1990s places the virus's emergence squarely within the period of global aquaculture expansion. This timeline is consistent with the hypothesis that the intensification of salmon farming created a new ecological niche for the virus, allowing it to amplify and spread from an ancestral wild reservoir. The outgroup sequences, therefore, may represent glimpses of this pre-amplification diversity. The finding that the Faroese farmed population is monophyletic and distinct from the wild outgroup [1] suggests that the founder introduction into the Faroe Islands was not from a wild source but rather from a farmed source, likely from Norway, where brood fish are known to be infected [11]. The detection of PMCV in roe and milt of broodstock, and its vertical transmission to progeny [11], provides a direct mechanism for this introduction. Imported Norwegian roe could have carried the virus, establishing the founder population in the Faroe Islands, which then evolved in isolation, diverging from its Norwegian ancestors and forming the homogenous cluster observed today.

The outgroup analysis also sheds light on the relative roles of vertical and horizontal transmission. The lack of continuous reintroduction from imported roe to the Faroe Islands [1], despite high prevalence in Norwegian broodstock [11], suggests that the founder strain may have a competitive advantage, or that the numbers of particles in imported eggs are insufficient to initiate a new infection in the face of competition from the established strain. The data-driven network model from Yatabe et al. [9] for Ireland demonstrated that the spread of PMCV is primarily driven by horizontal movement of subclinically infected smolts, not by vertical transmission. Together, these findings paint a picture where the initial introduction of PMCV into a new region may occur via infected eggs or broodstock, but subsequent within-region dissemination is dominated by horizontal fish movements. The outgroup genomes from wild salmon [1, 2] add a further dimension, suggesting that occasional spillback from wild to farmed populations or independent evolution in wild hosts can generate the phylogenetic distinctiveness observed.

The evolutionary trajectory of PMCV is, therefore, a story of strong constraint punctuated by rare, high-impact events. The virus's genome is subject to intense purifying selection, particularly in ORF1 and ORF2, which encode essential replicative and structural proteins. This constraint, combined with population bottlenecks during transmission between farms and within hosts, results in the striking national-level homogeneity observed in the Faroe Islands, Ireland, Norway, and Scotland. The differential relaxation of selection on ORF3 provides a locus for some adaptive potential, but this is tightly regulated and does not translate into widespread antigenic or virulence diversity. The outgroup genomes, particularly the divergent wild salmon isolate from the Faroe Islands, serve as crucial evolutionary archives, hinting at a broader genetic pool that exists outside the intensively farmed environment. These outgroups are not mere phylogenetic curiosities; they are essential for understanding the virus's deep evolutionary history, its potential for future change, and the true extent of its host range. The implication for biosecurity and regulatory frameworks, such as those of the World Organisation for Animal Health (WOAH) and the Food and Agriculture Organization (FAO), is clear: surveillance and control strategies must not only focus on the inter-farm movement of fish but must also account for the potential role of wild salmonid populations as reservoirs of genetic diversity and occasional sources of novel viral introductions into aquaculture systems. The evidence from these 25 studies collectively indicates that the genetic diversity and evolutionary trajectory of PMCV are a delicate balance between powerful homogenizing forces and rare, phylogenetically informative outliers that together define the virus's past and constrain its future.

Host-Virus Interactions and Disease Progression: CMS Pathology, Immune Response, and Brood Fish Infection

Cardiomyopathy syndrome (CMS), caused by the double-stranded RNA virus piscine myocarditis virus (PMCV), represents one of the most significant infectious cardiac diseases affecting Atlantic salmon (Salmo salar) aquaculture globally [6, 10, 19]. The disease was first recognized in Norwegian farms in 1985 and has since emerged as a major cause of morbidity and mortality in production cycles across Norway, the Faroe Islands, Scotland, and Ireland [2, 6, 19]. The host-virus interplay underlying CMS pathogenesis is extraordinarily complex, involving precise viral tropism for cardiac tissues, a robust yet often immunopathogenic host response, dynamic viral replication strategies, and a growing understanding of vertical transmission via brood fish [1, 4, 11]. This section provides an exhaustive examination of the biological mechanisms governing PMCV infection and CMS progression, synthesizing the most recent genomic, transcriptomic, proteomic, and epidemiological advances to delineate the intricate dialogue between virus and host.

Cardiac Pathology: Viral Tropism and Histopathological Progression

The defining hallmark of CMS is a severe, often fatal myocarditis that primarily targets the heart ventricle, with differential involvement of its constituent layers [4, 8]. Histopathological examination consistently reveals extensive inflammation, myocyte degeneration, and necrosis concentrated within the spongiform (inner) layer of the ventricular myocardium [4, 19]. In situ hybridization (ISH) studies have demonstrated intense staining of PMCV RNA almost exclusively within myocardial cells of the spongiform layer, with negligible signal detected in the compact (outer) layer, indicating a highly selective tropism [4]. This compartmentalization is functionally significant; the spongiform layer is responsible for generating the majority of ventricular contractile force in salmonids, and its destruction directly precipitates the circulatory collapse and sudden mortality characteristic of acute CMS outbreaks [19].

The temporal progression of cardiac pathology has been meticulously characterized in experimental challenge models and field studies. Following infection, histopathological lesions typically become detectable by 4–6 weeks post-infection (wpi), reaching peak severity between 8–9 wpi before entering a phase of potential recovery in surviving fish [16, 17]. The pathological changes evolve through distinct stages: an early phase marked by mild myocyte degeneration and inflammatory cell infiltration, followed by a peak phase characterized by extensive myocyte necrosis, multifocal loss of striated muscle architecture, and profound infiltration by mononuclear leukocytes [16, 17]. At the ultrastructural level, myofibrillar lysis, mitochondrial swelling, and the accumulation of cellular debris are prominent [19]. Crucially, high viral loads quantified by RT-qPCR correlate strongly with histopathology scores, strongly suggesting that direct cytopathic effects of PMCV replication are a primary driver of myocardial damage [10, 16]. However, this is not the entire story; as will be discussed, the host immune response contributes significantly to tissue injury.

Immune Response Dynamics: From Antiviral Defense to Immunopathogenesis

The host immune response to PMCV is a double-edged sword. While ultimately necessary for viral clearance, the activation of both innate and adaptive immune pathways in the heart is intimately linked to the severity of tissue damage [15, 17]. The response unfolds in a biphasic, temporally regulated manner, as elucidated by transcriptome profiling and high-throughput qPCR analyses [16, 17].

Innate and Interferon-Mediated Responses

The earliest detectable host response is a robust, systemic induction of antiviral and interferon (IFN)-dependent genes, observable as early as 2 wpi [7, 16]. Type I IFN responses, including the upregulation of Mx protein, are activated across infected tissues, with Mx transcript levels showing significant elevation in both heart and blood of infected fish [7, 8]. However, the initial type I IFN response does not always exhibit a clear, consistent pattern throughout the infection, and its kinetics can vary between individuals [8]. This early innate activation is critical for controlling initial viral replication, but its persistence can be detrimental.

Remarkably, work by Sun et al. (2024) demonstrated that fish with severe disease outcomes (High Responders; HR) continue to display elevated transcription of IFNb at 9 and 12 wpi, despite declining viral loads [15]. RNAscope in situ hybridization confirmed the presence of high quantities of IFN-γ and IFNb-secreting cells within the heart ventricles of HR fish, which were absent in low responders (LR) [15]. This indicates that a sustained, localized IFN response within the myocardium, potentially driven by infiltrating immune cells, contributes to persistent inflammation and immunopathology. The presence of these IFN-secreting cells suggests that the adaptive immune system, particularly Th1-type responses, is actively driving a chronic inflammatory state that exacerbates cardiac injury even after viral replication has been curtailed [15].

Adaptive Immunity: T Cell Infiltration and Biphasic Activation

A hallmark of severe CMS is the massive infiltration of T lymphocytes into the cardiac interstitium. Transcriptomic analyses have revealed a biphasic activation of adaptive immune pathways [16]. An initial wave of B cell and MHC class II antigen presentation gene expression is observed, which is followed by a pronounced and sustained upregulation of T cell response genes that coincides with the peak of cardiac pathology at 8–9 wpi [16]. Genes associated with cytotoxic T cell (CD8α) activity, as well as chemokines that recruit these cells to the heart, are highly upregulated in infected fish [7, 15]. Indeed, a strong positive correlation between viral load and cell-mediated immune markers, including CD8α and granzyme-like molecules, has been confirmed [8].

A critical insight comes from comparing High Responders (HR) and Low Responders (LR) to infection [17]. At the late stage of infection, HR fish exhibit a broad and sustained activation of genes involved in adaptive immunity, particularly those associated with Th1 pro-inflammatory and cytotoxic T cell responses [15, 17]. This prolonged activation is thought to directly cause the extensive myocardial necrosis observed in HR fish, a classic case of immunopathology. In stark contrast, LR fish, which ultimately recover, show a significantly reduced transcription of adaptive immunity genes and instead activate pathways involved in cardiac energy metabolism and tissue repair [17]. This suggests that a controlled, self-limiting adaptive immune response, coupled with an efficient metabolic shift towards repair, is the key to a favorable outcome.

Compartmentalized Differences in Cardiac Immune Response

Recent work has highlighted that the two heart compartments, atrium and ventricle, mount distinct immune responses to PMCV. Malachowski et al. (2025) demonstrated that the atrium consistently exhibits higher expression of pro-inflammatory (e.g., IL-1β, TNF-α), anti-inflammatory (e.g., IL-10), and cytotoxic T cell marker genes compared to the ventricle [8]. This increased inflammatory tone in the atrium is correlated with higher viral load and more severe disease progression in this chamber [8]. Conversely, the spleen and head kidney, primary lymphoid organs, showed no such compartmentalized differences, indicating that the unique immunological environment of the heart is pivotal in shaping disease outcome [8].

The Paradox of Viral Replication and Persistence

PMCV exhibits a replication strategy that is essential to understanding its pathogenesis. The virus is a double-stranded RNA (dsRNA) virus, and its lifecycle relies on a robust transcription of viral mRNA. Nyman et al. (2024) performed a landmark study showing that in naturally infected salmon hearts, the ratio of viral mRNA to viral genomic dsRNA is very high, indicating active transcription but surprisingly limited production of new viral particles [4]. They concluded that the histopathological changes in the heart are likely driven by the accumulation of viral mRNA and its translation products (viral proteins), rather than by the formation of numerous progeny virions [4]. This helps explain why tissue damage can be so severe even when attempts to propagate the virus in cell culture have consistently failed [5].

Furthermore, PMCV can persist in its host. In field studies, Svendsen et al. (2019) detected PMCV in 11 of 12 farm sites until the point of slaughter, with persistence lasting for months after initial infection [10]. The virus is first detected in lymphoid tissues (head kidney) as early as 1 wpi before spreading to the heart [8]. While some evidence points to clearance of the virus from the ventricle in a subset of fish by 16 wpi, a low but persistent viral load is frequently maintained in other tissues, including the head kidney and spleen [8]. This long-term persistence in lymphoid tissues suggests that these organs may serve as a viral reservoir, enabling recrudescence of infection under stress or maintaining a constant, low-level antigenic stimulation that drives the chronic immunopathology seen in severe cases [8]. Adding another layer of complexity, the presence of defective viral genomes (DVGs) was a novel finding for the order Ghabrivirales [3]; DVGs can modulate the host interferon response and influence viral replication dynamics, potentially affecting disease outcome.

Brood Fish Infection and Vertical Transmission: A Critical Route of Introduction

A seminal discovery in PMCV epidemiology is the high prevalence of infection in brood fish and the strong evidence for vertical transmission. Jensen et al. (2019) demonstrated that in commercial broodstock populations, the prevalence of PMCV in the heart can be as high as 98%, with the virus also detected in 69% of roe (eggs) and 59% of milt (sperm) samples [11]. Crucially, they detected PMCV in all stages of progeny, from eggs and larvae to fingerlings and pre-smolts, confirming that the virus is transferred vertically from infected parents to their offspring [11]. This finding has profound implications for the industry. It explains the otherwise puzzling emergence of CMS in post-smolts shortly after sea-transfer, as observed in Norwegian and Faroese farms [10, 11].

Phylodynamic studies have further corroborated the role of brood fish infection in the international spread of PMCV. Dahl et al. (2025) showed that the Faroese PMCV genomes form a homogenous monophyletic cluster, likely originating from a single introduction event from Norway [1]. This introduction was linked to the importation of infected roe from Norway, as no continuous reintroduction of new strains was observed despite continued egg import [1]. Similarly, analyses of Irish PMCV strains suggest the virus was introduced from the southern range of Norway via imported brood fish [9, 12]. Network modeling by Yatabe et al. (2020) further demonstrated that the movement of subclinically infected fish between sites is the most important factor explaining the rapid, countrywide spread of PMCV in Ireland [9]. The detection of PMCV in wild returning salmon also suggests that wild fish may act as a potential source, although the primary driver remains farmed brood fish [1, 2]. This vertical transmission pathway provides farmers with potential control points, such as screening broodstock and using only PMCV-negative eggs, but its scale and economic implications represent a major ongoing challenge [9, 11].

Genetic Determinants of Host Susceptibility

The variation in disease outcome, from subclinical infection to fatal myocarditis, has a strong genetic basis in the host. Genome-wide association studies (GWAS) have identified several quantitative trait loci (QTL) associated with resistance to CMS. Boison et al. (2019) initially identified significant QTLs on chromosomes SSA27 and SSA12, which together explain approximately half of the total genetic variance for resistance [20]. More recently, Moghadam and Røsaeg (2023) described a new major QTL on chromosome 23, which in an experimental challenge explained up to 46% of the genetic variation in viral load [18]. This QTL is located in proximity to genes crucial for polyunsaturated fatty acid (PUFA) metabolism, specifically delta-5 fatty acyl desaturase and fatty acid desaturase 2 [18]. This genetic link between lipid metabolism and resistance is particularly compelling, as it directly aligns with experimental evidence showing that dietary enrichment with eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) can reduce heart pathology and viral load in CMS outbreaks [13, 14]. The genomic heritability estimates for resistance to CMS range from 0.12 to 0.55, clearly demonstrating that selective breeding is a viable and effective strategy for mitigating the impact of this devastating disease [18, 20]. These findings, combined with the identification of serum biomarkers like fibrinogen, creatine kinase, and lactate dehydrogenase that reflect cardiac damage, offer powerful tools for both genetic selection and non-lethal diagnostics [21-23].

Prevention and Control Strategies for PMCV in Aquaculture: Biosecurity, Roe Import Monitoring, and Surveillance Implications

The prevention and control of piscine myocarditis virus (PMCV) in Atlantic salmon aquaculture necessitates a paradigm shift from reactive outbreak management to a comprehensive, multi-layered biosecurity framework. Unlike many viral pathogens for which efficacious vaccines or antiviral therapeutics exist, PMCV currently eludes such direct interventions; no commercially available vaccine has demonstrated consistent, broad protection in field conditions, and the virus resists propagation in conventional cell culture systems, severely hampering classical vaccine development pathways [5, 19]. Consequently, the cornerstone of PMCV control rests upon a triad of interdependent strategies: (1) rigorous biosecurity protocols that interrupt transmission pathways at local, regional, and international scales; (2) stringent monitoring and risk-based management of roe and smolt imports to prevent the introduction of novel viral lineages; and (3) the deployment of sophisticated surveillance systems capable of early detection, risk stratification, and informed decision-making. The urgency of these measures is underscored by the pathogen’s substantial economic footprint, annual losses in Norway alone are estimated at up to €9 million, coupled with its capacity for cryptic circulation within populations prior to the onset of clinical cardiomyopathy syndrome (CMS) [10, 13, 19]. Furthermore, the World Organisation for Animal Health (WOAH) recognizes the significance of emerging aquatic viral diseases, and while PMCV is not currently listed, its transnational spread and high morbidity in farmed salmon align with criteria for enhanced international reporting and coordinated biosecurity governance.

The Biosecurity Imperative: Disentangling Transmission Pathways

Effective biosecurity cannot be implemented without a granular understanding of how PMCV moves through, and persists within, the aquaculture production network. A seminal phylodynamic analysis by Zhao et al. [6], which generated one of the largest genomic datasets for any aquatic viral pathogen, provides critical insights: the virus likely emerged in farmed salmon concurrent with the global expansion of aquaculture during the 1970s–1990s, and its contemporary dispersal is driven overwhelmingly by anthropogenic factors. Geographic proximity between marine farms and the density of farms within a given region emerged as the strongest predictors of viral spread, implicating local hydrographic connectivity and farm-to-farm contact as dominant forces [6]. Notably, the same study identified boat connectivity as a plausible vector for long-distance dispersal events, suggesting that fomites, including well-boats, transport vessels, and shared equipment, play a previously underappreciated role in viral dissemination. These findings have profound biosecurity implications: interventions limited to individual farm-level hygiene are insufficient if the broader seascape connectivity among farms is not addressed.

The Faroese experience offers a cautionary yet instructive example. Dahl et al. [1] performed whole-genome sequencing on 48 PMCV genomes from Faroese farmed salmon and demonstrated a striking genetic homogeneity, forming a monophyletic cluster distinct from Norwegian and Irish strains. Critically, despite a steadily increasing import of Norwegian roe, the authors found no evidence of continuous reintroduction of persisting PMCV strains; the virus appears to have undergone a single introduction event, likely from Norway via infected broodstock, and then established a self-sustaining endemic cycle within the Faroe Islands [1]. This observation carries a dual biosecurity message. On one hand, it suggests that strategic, high-quality biosecurity at the point of roe importation can prevent the repeated seeding of new viral variants into a naïve region. On the other hand, it starkly illustrates that once a genetically homogenous viral population becomes established locally, it can persist and cause recurrent disease outbreaks without external reintroduction, meaning that biosecurity must encompass both import controls and internal “firebreaks” to limit within-country spread. The Irish epidemiological trajectory corroborates this concern. Yatabe et al. [9] used data-driven network modeling to demonstrate that following the initial introduction of PMCV into Ireland in approximately 2009, the between-farm prevalence reached 100% within five years, driven predominantly by the movement of subclinically infected fish rather than local waterborne spread. Their models showed that local spread was only a minor contributor to new infections, whereas the movement of live fish, often from a relatively small number of highly connected “hub” farms, was the primary engine of countrywide dissemination [9]. This finding reframes biosecurity not merely as a hygiene protocol but as a logistics and supply-chain management challenge.

Roe Import Monitoring: A Critical Checkpoint Against Viral Invasion

The detection of PMCV in broodstock tissues, with one study reporting 98% prevalence in hearts, 69% in roe, and 59% in milt from PMCV-positive broodstock, alongside the demonstration of vertical transmission to progeny up to the 40-gram stage, establishes the importation of infected roe as a high-risk pathway for viral introduction into naïve regions [11]. This vertical transmission pathway is biologically plausible given PMCV’s broad tissue tropism, including infection of the reproductive tissues [4, 11]. However, the genomic data complicate a simplistic narrative. Despite the high prevalence in broodstock and the known importation of Norwegian roe into the Faroe Islands, Dahl et al. [1] could find no evidence that these imports resulted in the establishment of new, persistent viral lineages; all Faroese sequences clustered tightly together irrespective of roe origin, suggesting that either the imported roe carried viral loads below a transmission threshold or that the resident Faroese strain outcompeted any introduced variants. Similarly, Amono et al. [3] and Tighe et al. [2] found that PMCV sequences form distinct clusters by country of origin, with limited evidence for ongoing mixing between national populations, despite the global trade in salmonid germplasm. These observations should not be interpreted as a justification for lax import controls. Rather, they underscore a nuanced biosecurity reality: the risk of roe-mediated introduction may be stochastic, contingent upon viral load in the roe, the health status of the broodstock, and the presence of competing viral strains at the destination. A precautionary principle is therefore warranted. The implementation of mandatory, PCR-based screening of broodstock for PMCV prior to roe collection, with the culling or segregation of high-titer individuals, could dramatically reduce the probability of vertical transmission [11]. Furthermore, the use of disinfectant protocols for egg surfaces, while not proven against intracellular virus within oocytes, remains a prudent adjunctive measure.

The monitoring of roe imports should be coupled with genomic surveillance to detect cryptic introductions. Tighe et al. [2] and Amono et al. [3] both demonstrated that whole-genome sequencing, rather than sequencing of open reading frames (ORFs) alone, captures >80% of the virus’s genetic diversity, making it essential for tracking transmission pathways. The routine inclusion of MinION-based sequencing at import inspection points could provide rapid, field-deployable genotyping to identify whether a viral strain in a new outbreak is a recent import or an endemic variant, thereby triggering appropriate biosecurity responses [2]. However, a cautionary note from Tighe et al. [2] must be heeded: MinION-generated genomes, while useful for phylogenetics, showed only 99.59% identity to Illumina-derived genomes, insufficient accuracy for fine-scale transmission tracing. Therefore, confirmatory short-read sequencing should complement rapid sequencing platforms when the resolution of transmission chains is critical for epidemiological investigations.

Surveillance Implications: From Passive Detection to Proactive Risk Anticipation

Traditional surveillance for CMS has relied upon the observation of clinical signs, lethargy, anemia, ascites, and mortality, followed by histopathological confirmation and RT-qPCR. However, an exhaustive body of evidence demonstrates that PMCV infection precedes clinical CMS by a considerable latency period. Svendsen et al. [10] conducted a longitudinal epidemiological study across 12 Norwegian farm sites and found that initial PMCV infection occurred between 1 and 7 months post-sea transfer, with a median interval of 6.5 months from first infection to clinical outbreak. Crucially, PMCV was detected at all 12 sites, yet only six developed overt CMS, indicating that subclinical infection is the rule rather than the exception [10]. This has profound surveillance implications: monitoring based solely on mortality events is akin to driving by looking only in the rearview mirror. Proactive surveillance programs must incorporate regular, scheduled RT-qPCR testing of sentinel populations, ideally at the cage level, beginning early in the sea phase. The detection of rising viral loads, even in the absence of pathology, should trigger heightened monitoring and management interventions.

The immunological host response further complicates surveillance interpretation. Monte et al. [7] demonstrated that following experimental infection, antiviral genes such as mx, cd8α, and γip were upregulated in heart tissue at terminal time points, but in blood samples, the kinetics of expression varied markedly between individuals. This individual variability means that population-level surveillance using blood-based immune markers must rely on large sample sizes and longitudinal sampling to distinguish early infection from historical exposure. Malachowski et al. [8] added another layer of complexity by showing that PMCV persists in lymphoid tissues (head kidney and spleen) even as the heart begins to clear the virus; at 16 weeks post-infection, three of six fish had cleared virus from the ventricle, yet low-level persistence continued in other tissues. This finding implies that surveillance sampling targeting only heart tissue may underestimate the true prevalence of infection, particularly in recovering or chronically infected populations. A multi-tissue sampling strategy, including head kidney, spleen, and possibly blood, should be standard for prevalence surveys.

The promise of non-lethal surveillance using serum biomarkers is an exciting frontier with direct biosecurity applications. Costa et al. [21, 23, 24] and Riva et al. [22] conducted comprehensive proteomic profiling of serum from CMS-affected fish and identified 27 proteins unique to diseased animals, many of which are orthologous to human cardiac biomarkers, including leakage enzymes (creatine kinase, lactate dehydrogenase) and acute-phase proteins (fibrinogen, haptoglobin). Of particular operational value is the finding that serum fibrinogen, when ratioed to skeletal troponin C, can differentiate CMS from pancreas disease (PD), another important viral cardiomyopathy caused by salmonid alphavirus [22]. The deployment of rapid, field-friendly immunoassays for these biomarkers could enable cage-side screening of large numbers of fish during regular health inspections, identifying cages harboring subclinical disease before viral shedding reaches epidemic proportions. This would represent a transformative advance over current diagnostic paradigms.

Integrating Biosecurity, Genetics, and Health Management

No discussion of PMCV control is complete without acknowledging the role of host genetics as a biosecurity multiplier. Multiple independent genome-wide association studies have identified significant quantitative trait loci (QTLs) for resistance to CMS on chromosomes 12, 23, and 27 [18, 20]. Mogahadam and Røsaeg [18] estimated the SNP-based heritability of viral load at 0.55 in an experimental challenge, with a QTL on chromosome 23, near genes encoding delta-5 fatty acyl desaturase and fatty acid desaturase 2, explaining 46% of the genetic variation. This proximity to fatty acid desaturase genes is particularly intriguing given the mounting evidence that dietary lipid composition modulates the severity of CMS. Martínez-Rubio et al. [13] demonstrated that functional feeds with reduced lipid content and increased eicosapentaenoic acid (EPA) levels led to a milder and delayed inflammatory response and reduced heart lesion severity in experimentally infected salmon. More recently, Rennemo et al. [14] reported that a clinical nutrition intervention enriched with EPA and DHA during a natural CMS outbreak was associated with declining mortality, regression of histopathological changes, and a significant reduction in PMCV RNA load. These findings collectively suggest that biosecurity is not solely a matter of preventing viral entry; it can be enhanced by breeding fish with genetic resistance and by managing their nutritional status to buffer against the immunopathological consequences of inevitable exposure. Marker-assisted selection incorporating the QTLs on chromosomes 12, 23, and 27 should be integrated into breeding programs targeting CMS resistance, and broodstock should be screened for these favorable alleles prior to use [18, 20].

Operationally, the network modeling work of Yatabe et al. [9] provides a clear, actionable framework. Their simulations demonstrated that proactively targeting the top eight most connected farms (based on outward fish movements) for mandatory testing of all outgoing consignments would have prevented the between-farm PMCV prevalence from ever exceeding 20% in Ireland, compared to the uncontrolled simulation where it reached 100%. This approach, effectively a form of “targeted surveillance” informed by centrality metrics, could be adapted to any salmon-producing nation. It requires real-time tracking of fish movements, integration with farm management software, and the legal authority to impose movement restrictions on high-risk shipments. Such a system aligns with the biosecurity principle of “biocontainment at the source” and represents a far more efficient use of diagnostic resources than blanket testing of all farms. The cost of such surveillance is dwarfed by the economic losses from uncontrolled outbreaks.

In sum, the prevention and control of PMCV demands a holistic, intelligence-driven strategy that integrates molecular epidemiology, supply-chain logistics, genetic selection, nutritional intervention, and multi-tissue diagnostic surveillance. The virus’s ability to persist subclinically in lymphoid tissues, transmit vertically via roe, and spread silently through fish movements before clinical disease manifests means that traditional reactive measures will always be several steps behind. A shift toward pre-emptive biosecurity, characterized by genomic surveillance of imports, network-based movement restrictions, and the deployment of non-lethal biomarkers for early detection, offers the most robust pathway to sustainable control.

References

[1] Dahl MM, Petersen PE, Christiansen D. Genomic analysis reveals low genetic diversity and no continuous reintroduction of piscine myocarditis virus in farmed Atlantic salmon in the Faroe Islands. Journal of General Virology. 2025. DOI: https://doi.org/10.1099/jgv.0.002068

[2] Tighe A, Carlsson J, Ruane NM. Comparing Genome Sequencing Methods to Reconstruct the Spread of Piscine Myocarditis Virus in Ireland. Journal of Fish Diseases. 2025. DOI: https://doi.org/10.1111/jfd.14150

[3] Amono R, Markussen T, Singh VK, Lund M, Manji F, Mor SK, et al.. Unraveling the genomic landscape of piscine myocarditis virus: mutation frequencies, viral diversity and evolutionary dynamics in Atlantic salmon. Virus Evolution. 2024. DOI: https://doi.org/10.1093/ve/veae097

[4] Nyman IB, Wessel Ø, Bjørgen H, Alarcón M, Tengs T, Rimstad E. Evaluating Atlantic Salmon (Salmo salar) as a Natural or Alternative Host for Piscine Myocarditis Virus (PMCV) Infection. Pathogens. 2024. DOI: https://doi.org/10.3390/pathogens13090744

[5] Su H, Eerde A, Steen H, Heldal I, Haugslien S, Ørpetveit I, et al.. Establishment of a piscine myocarditis virus (PMCV) challenge model and testing of a plant-produced subunit vaccine candidate against cardiomyopathy syndrome (CMS) in Atlantic salmon Salmo salar. Aquaculture. 2021. DOI: https://doi.org/10.1016/J.AQUACULTURE.2021.736806

[6] Zhao M, Duault H, Fournie G, Solheim MA, Alexandersen S, Matthews C, et al.. Molecular epidemiology and multi-scale drivers of piscine myocarditis virus dispersal in salmon aquaculture. Virus Evolution. 2026. DOI: https://doi.org/10.1093/ve/veag020

[7] Monte M, Urquhart K, Evensen Ø, Secombes C, Collet B. Individual monitoring of immune response in Atlantic salmon Salmo salar following experimental infection with piscine myocarditis virus (PMCV), agent of cardiomyopathy syndrome (CMS).. Developmental and Comparative Immunology. 2019. DOI: https://doi.org/10.1016/j.dci.2019.103406

[8] Malachowski A, Jenberie S, Dalmo R, Swain JK. The infection biology of piscine myocarditis virus in Atlantic salmon: Can a local immune response aid eliminating the virus from the heart.. Fish and Shellfish Immunology. 2025. DOI: https://doi.org/10.1016/j.fsi.2025.110518

[9] Yatabe T, Martínez-López B, Díaz-Cao JM, Geoghegan F, Ruane N, Morrissey T, et al.. Data-Driven Network Modeling as a Framework to Evaluate the Transmission of Piscine Myocarditis Virus (PMCV) in the Irish Farmed Atlantic Salmon Population and the Impact of Different Mitigation Measures. Frontiers in Veterinary Science. 2020. DOI: https://doi.org/10.3389/fvets.2020.00385

[10] Svendsen J, Nylund S, Kristoffersen AB, Takle H, Buhaug JF, Jensen B. Monitoring infection with Piscine myocarditis virus and development of cardiomyopathy syndrome in farmed Atlantic salmon (Salmo salar L.) in Norway. Journal of Fish Diseases. 2019. DOI: https://doi.org/10.1111/jfd.12974

[11] Jensen BB, Nylund S, Svendsen J, Ski PR, Takle H. Indications for a vertical transmission pathway of piscine myocarditis virus in Atlantic salmon (Salmo salar L.). Journal of Fish Diseases. 2019. DOI: https://doi.org/10.1111/jfd.12990

[12] Tighe A, Carlsson J, Morrissey T, Swords F, Ruane N. Genetic diversity of piscine myocarditis virus in Atlantic salmon Salmo salar L. in Ireland.. Journal of Fish Diseases. 2019. DOI: https://doi.org/10.1111/jfd.13018

[13] Martínez-Rubio L, Evensen Ø, Krasnov A, Jørgensen S, Wadsworth S, Ruohonen K, et al.. Effects of functional feeds on the lipid composition, transcriptomic responses and pathology in heart of Atlantic salmon (Salmo salar L.) before and after experimental challenge with Piscine Myocarditis Virus (PMCV). BMC Genomics. 2014. DOI: https://doi.org/10.1186/1471-2164-15-462

[14] Rennemo J, Berge K, Yousaf MN, Eriksen TB, Welde E, Robertsen C, et al.. An Atypical Course of Cardiomyopathy Syndrome (CMS) in Farmed Atlantic Salmon (Salmo salar) Fed a Clinical Nutrition Diet. Microorganisms. 2023. DOI: https://doi.org/10.3390/microorganisms12010026

[15] Sun B, Vindas M, Kavaliauskiene S, Bjørgen H, Koppang E, Wisløff H, et al.. Persistent immune responses in the heart determine the outcome of cardiomyopathy syndrome in Atlantic salmon (Salmo salar).. Fish and Shellfish Immunology. 2024. DOI: https://doi.org/10.1016/j.fsi.2024.109404

[16] Timmerhaus G, Krasnov A, Nilsen P, Alarcón M, Afanasyev S, Rode M, et al.. Transcriptome profiling of immune responses to cardiomyopathy syndrome (CMS) in Atlantic salmon. BMC Genomics. 2011. DOI: https://doi.org/10.1186/1471-2164-12-459

[17] Timmerhaus G, Krasnov A, Takle H, Afanasyev S, Nilsen P, Rode M, et al.. Comparison of Atlantic salmon individuals with different outcomes of cardiomyopathy syndrome (CMS). BMC Genomics. 2012. DOI: https://doi.org/10.1186/1471-2164-13-205

[18] Mogahadam HK, Røsaeg MV. From viral load to survival: Unveiling new genetic links for resistance against PMCV in Atlantic Salmon (Salmo salar).. Journal of Fish Diseases. 2023. DOI: https://doi.org/10.1111/jfd.13847

[19] Garseth Å, Fritsvold C, Svendsen J, Jensen BB, Mikalsen A. Cardiomyopathy syndrome in Atlantic salmon Salmo salar L.: A review of the current state of knowledge.. Journal of Fish Diseases. 2018. DOI: https://doi.org/10.1111/jfd.12735

[20] Boison S, Ding J, Leder E, Gjerde B, Bergtun PH, Norris A, et al.. QTLs Associated with Resistance to Cardiomyopathy Syndrome in Atlantic Salmon. Journal of Heredity. 2019. DOI: https://doi.org/10.1093/jhered/esz042

[21] Costa JZ, Thompson K, Pozo JD, McLean K, Inglis N, Sourd P, et al.. Proteomic Characterisation of Serum Proteins From Atlantic Salmon (Salmo salar L.) With Cardiomyopathy Syndrome (CMS). . 2021. DOI: https://doi.org/10.21203/rs.3.rs-146896/v1

[22] Riva F, Black LE, Thompson KD, Sourd P, Bordeianu A, Chadwick CC, et al.. Serum Biomarkers in Atlantic Salmon for Differential Diagnosis of Cardiomyopathy Syndrome and Pancreas Disease: Proteomic Identification of Serum Fibrinogen to Enhance Troponin Immunoassay as Optimal Diagnostic Approach. Journal of Fish Diseases. 2025. DOI: https://doi.org/10.1111/jfd.14151

[23] Costa JZ, Pozo JD, McLean K, Inglis N, Sourd P, Bordeianu A, et al.. Proteomic characterization of serum proteins from Atlantic salmon (Salmo salar L.) from an outbreak with cardiomyopathy syndrome.. Journal of Fish Diseases. 2021. DOI: https://doi.org/10.1111/jfd.13488

[24] Costa J, Pozo J, Thompson KD, McLean K, Inglis N, Sourd P, et al.. Proteomic characterisation of serum proteins from Atlantic salmon (Salmo salar L.) from an outbreak with Cardiomyopathy Syndrome. . 2021. DOI: https://doi.org/10.21203/rs.3.rs-146896/v3