Salmonid Alphavirus

Overview, Taxonomy, and Genetic Diversity of Salmonid Alphavirus

Overview

Salmonid alphavirus (SAV) is the etiological agent of pancreas disease (PD) in Atlantic salmon (Salmo salar) and sleeping disease (SD) in rainbow trout (Oncorhynchus mykiss), two of the most economically devastating viral diseases affecting European salmonid aquaculture [1, 20]. The virus is a member of the family Togaviridae, genus Alphavirus, and is classified as a non-zoonotic pathogen with no known human health implications, though its impact on food security and animal welfare is profound [6, 7]. The World Organisation for Animal Health (WOAH) lists SAV infection as a notifiable disease, and the European Union’s Animal Health Law has assessed its eligibility for Union intervention, concluding that while it does not meet criteria for Category A (eradication), it remains uncertain whether it meets thresholds for Categories B through E, reflecting the complexity of its epidemiology and control [6].

SAV is an enveloped, single-stranded, positive-sense RNA virus with a genome of approximately 11.7 kb [20, 29]. The genomic organization is typical of alphaviruses: a 5′ cap, nonstructural proteins (nsP1–nsP4) encoded in the first two-thirds of the genome, followed by a subgenomic promoter driving expression of the structural proteins, capsid (C), envelope glycoproteins E3, E2, 6K, and E1, from a 26S subgenomic RNA [9, 13]. Unlike terrestrial alphaviruses such as chikungunya virus, SAV is transmitted horizontally through water without an arthropod vector, a feature that has profound implications for its evolutionary trajectory and host adaptation [13, 18]. The virus is shed into the surrounding seawater by infected fish, and transmission occurs via the waterborne route, with viral loads as low as 7 TCID₅₀ L⁻¹ of seawater sufficient to establish infection in naïve populations [24, 25]. This direct horizontal transmission, combined with the high density of fish in aquaculture settings, drives rapid spread and sustained outbreaks [4, 26].

The clinical and pathological hallmarks of PD include necrosis of the exocrine pancreas, cardiomyopathy, and skeletal myopathy, leading to reduced growth, poor feed conversion, increased mortality, and significant economic losses [2, 21]. In Atlantic salmon, PD outbreaks can result in growth reductions of up to 0.7 kg per fish and increases in feed conversion ratio of 0.07 points, with mortality rates varying widely depending on viral dose, host susceptibility, and environmental conditions [21, 24]. The disease is most severe in the marine phase of production, though SAV1 has been documented causing clinical PD in freshwater Atlantic salmon in Scotland, expanding the known risk period for the disease [23]. In rainbow trout, SAV2 causes SD, characterized by lethargy, abnormal swimming behavior, and extensive necrosis of red skeletal muscle, with the virus exhibiting a particular tropism for muscle satellite cells [27].

Taxonomy and Phylogenetic Classification

SAV is classified within the genus Alphavirus, family Togaviridae, and is currently divided into six recognized genotypes (SAV1–SAV6) based on phylogenetic analysis of the E2 envelope glycoprotein and nsP3 nonstructural protein genes [1, 20]. The genotypes exhibit distinct geographic distributions and host preferences. SAV1 and SAV4–SAV6 are primarily associated with PD in Atlantic salmon in Ireland and Scotland [20, 22]. SAV2 exists in two epidemiological forms: a freshwater variant (SAV2 FW) causing SD in rainbow trout across continental Europe, and a marine variant (SAV2 MAR) causing PD in Atlantic salmon in Scotland and, since 2010, in Norway [10, 20]. SAV3 is the dominant genotype in Norwegian Atlantic salmon aquaculture, responsible for the majority of PD outbreaks in that country [1, 2, 12]. SAV5 and SAV6 have been isolated less frequently, with SAV6 notably recovered from wild-caught ballan wrasse (Labrus bergylta) in Ireland, representing the first isolation of this subtype from a non-salmonid fish and highlighting the potential for reservoir hosts in the marine environment [14, 22].

The taxonomic boundaries between genotypes are not absolute, and recent genomic evidence supports the existence of a seventh genotype, provisionally designated SAV7. Nanopore whole-genome sequencing of a unique SAV strain isolated from ballan wrasse, combined with partitioned phylogenetic analysis that accounts for variation in evolutionary rates across individual genes and codon positions, revealed branch lengths and substitution patterns that justify the designation of a new genotype [14]. This finding underscores the dynamic nature of SAV taxonomy and the importance of comprehensive genomic surveillance. Furthermore, the detection of SAV in non-salmonid species, including ballan wrasse, dab (Limanda limanda), and other flatfish, has prompted suggestions that the species name be amended to the more inclusive “piscine alphavirus” to reflect its broader host range [14, 19].

Genetic Diversity and Evolutionary Dynamics

The genetic diversity of SAV is driven by the error-prone nature of its RNA-dependent RNA polymerase (RdRp), which generates a quasispecies cloud of closely related but genetically distinct variants within individual hosts [5, 19]. This intrahost diversity is a critical feature of SAV biology, as it provides the raw material for rapid adaptation to host immune pressures, antiviral interventions, and environmental changes. Nanopore sequencing of SAV3 genomes during experimental infections of Atlantic salmon and brown trout (Salmo trutta) has revealed a multitude of single nucleotide variants (SNVs) and deletions distributed across the entire genome, with specific variants observed in individual fish at different time points post-challenge [5]. Notably, two relatively frequent SNVs were detected in multiple fish within the same experimental group, while other minor variants increased in frequency specifically in brown trout, suggesting host-specific selective pressures [5]. Nonmetric multidimensional scaling analysis of these data indicated that SAV3 genomes isolated from brown trout late in infection exhibited greater genetic variation than those from Atlantic salmon, implying that host species can shape the evolutionary trajectory of the virus [5].

At the population level, SAV exhibits substantial genetic diversity both within and between genotypes. Genome-wide target-enriched viral sequencing of farmed and wild fish populations in Scotland and Ireland has uncovered extensive “hidden” diversity, including mixed-subtype infections in individual fish and in pooled samples [19]. Co-infections involving SAV2 and SAV3, or SAV1 and SAV5, have been documented in both farmed Atlantic salmon and wild flatfish, indicating that multiple subtypes can circulate simultaneously within the same geographic region and even within the same host [12, 19]. The pooling of tissue samples from multiple animals, a common practice in diagnostic surveillance, was shown to underestimate this diversity, masking the presence of minor variants and limiting the power to detect the introduction of novel genotypes [19]. These findings have profound implications for disease control, as the coexistence of multiple subtypes within a farm or region increases the potential for recombination and the emergence of novel strains with altered virulence or antigenic properties.

The emergence of SAV2 in Norway provides a compelling case study in the introduction and spread of a novel genotype. Full-length genome sequencing of SAV2 strains sampled between 2006 and 2012 in Norway and Scotland revealed that all Norwegian marine SAV2 strains share a recent last common ancestor with Scottish marine SAV2, supporting the hypothesis of a single introduction event, likely via well-boat traffic, sometime between 2006 and 2010 [10]. Following its introduction, SAV2 split into two distinct clades, designated SAV2a and SAV2b, around 2013 [12]. These clades exhibit non-overlapping geographic distributions in Norway: SAV2a is predominantly found in southern regions, while SAV2b is more common in northern latitudes, though both clades have been detected at the boundary of Møre og Romsdal and Trøndelag [12]. Genomic epidemiology has further demonstrated recent long-distance transmission events, including the spread of SAV2a from Møre og Romsdal to Rogaland in 2019–2020, and a northward jump of SAV2b of approximately 100 km within Trøndelag [12]. These data illustrate the power of genomic epidemiology to reconstruct transmission networks and inform targeted control measures.

Structural and Functional Correlates of Genetic Diversity

The genetic diversity of SAV is not merely a static catalogue of sequence variants; it has direct functional consequences for viral fitness, virulence, and host interactions. In silico reconstruction of the SAV virion using AlphaFold and structural bioinformatics has revealed a distinctive, exposed α-helical feature in the E2 envelope protein that is unique among alphaviruses [9]. This structural element is predicted to be a key determinant of host cell attachment and entry, and its sequence diversity across SAV genotypes correlates with differences in virulence and host species adaptation [9]. Mutations in the N-linked glycosylation sites of E2 have been shown to attenuate the virus in cell culture, reducing cytopathic effect and infectious particle production, while mutation of the E1 glycosylation site completely abrogates viral replication [16]. These findings highlight the critical role of envelope glycoprotein structure in modulating viral fitness and suggest that naturally occurring genetic variation in these regions could influence the outcome of infection.

At the molecular level, SAV nonstructural protein 2 (nsP2) plays a central role in immune evasion by blocking the RIG-I signaling cascade and inhibiting type I interferon (IFN) induction [13]. The C-terminal domain of nsP2, which contains two nuclear localization sequences (NLS), is both necessary and sufficient for this inhibitory activity, and mutation of the NLS is deleterious to the virus [13]. The ability of nsP2 to suppress the host innate immune response is a key determinant of viral virulence, and genetic variation in this region could modulate the balance between viral replication and host clearance. Similarly, SAV nonstructural protein 2 has been identified as a key activator of the NF-κB signaling pathway, upregulating TLR3, TLR7, and TLR8, and inducing inflammatory cytokine expression [8]. This dual role, suppressing IFN induction while promoting inflammation, reflects the complex interplay between viral factors and host immune responses, and genetic diversity in nsP2 may contribute to the observed variation in disease severity across SAV genotypes and isolates.

Implications for Disease Control and Future Directions

The genetic diversity of SAV poses significant challenges for disease control. Current vaccines, including oil-adjuvanted inactivated whole-virus vaccines and DNA vaccines, are designed to protect against specific genotypes, and their efficacy may be compromised by the emergence of antigenically distinct variants [4, 15]. DNA vaccines have shown promise in reducing viral shedding and transmission, and in achieving herd immunity under experimental conditions, but their long-term effectiveness in the face of ongoing viral evolution remains to be determined [4, 26]. Live attenuated vaccines, generated by site-directed mutagenesis of virulence determinants such as the E2 glycosylation site and the capsid NLS, have demonstrated protective efficacy in experimental trials, but concerns about genetic stability, reversion to virulence, and environmental shedding of genetically modified viruses necessitate careful evaluation [1, 11].

The discovery of extensive intrahost and interhost genetic diversity, including mixed-subtype infections and the presence of SAV in non-salmonid reservoir hosts, underscores the need for routine genome-wide surveillance to capture the true diversity of circulating strains [19]. The application of nanopore sequencing and other high-throughput technologies to field samples, combined with genomic epidemiology, offers a powerful tool for tracking the emergence and spread of novel variants and for informing the design of next-generation vaccines and control strategies [5, 12]. As the aquaculture industry continues to expand, understanding the evolutionary dynamics of SAV and the genetic basis of host resistance, including the identification of quantitative trait loci on chromosome 3 associated with PD resistance [3, 17, 28], will be essential for developing sustainable approaches to disease management.

Molecular Pathogenesis: Virulence Factors and Host Interaction

Salmonid alphavirus (SAV), the etiological agent of pancreas disease (PD) and sleeping disease (SD), represents a significant threat to global salmonid aquaculture, with substantial economic and animal welfare implications recognized by the World Organisation for Animal Health (WOAH) [6, 7, 20]. The molecular pathogenesis of SAV is a complex, multifactorial process governed by a sophisticated arsenal of virulence factors that orchestrate host cell entry, subvert innate immune defenses, manipulate cellular signaling cascades, and ultimately cause the characteristic tissue pathology. Understanding these intricate molecular interactions is fundamental to developing rational control strategies, including next-generation vaccines and antiviral therapeutics.

Molecular Architecture and Host Cell Entry

Virion Structure and Glycoprotein Organization

The SAV virion, like other alphaviruses, is an enveloped particle with an icosahedral nucleocapsid housing a single-stranded, positive-sense RNA genome. However, recent in silico reconstructions leveraging AlphaFold have revealed distinctive molecular features that set SAV apart from terrestrial arthropod-borne alphaviruses [9]. The structural modeling predicted an exposed and distinctive α-helical feature in the E2 envelope protein, a domain that may govern host species adaptation and virulence [9]. This structural peculiarity likely reflects the unique evolutionary trajectory of SAV as an aquatic alphavirus that has dispensed with the requirement for an arthropod vector, transmitting directly through water [13, 20]. The E2 glycoprotein is the primary determinant of receptor binding and host cell tropism, and its N-glycosylation status is a critical virulence determinant. Mutagenesis studies targeting the N-glycosylation consensus motif in E2 (specifically E2319Q and E2319A substitutions) resulted in attenuated viruses that produced reduced cytopathic effects (CPE) and lower titers of infectious particles in cell culture, while mutations in the E1 glycoprotein N-glycosylation site proved lethal, completely abrogating viral replication [16]. This differential requirement for glycosylation underscores the essential role of E1 in membrane fusion and the more modulatory role of E2 glycosylation in viral fitness and pathogenesis.

Receptor Engagement and Cell Tropism

SAV exhibits broad tissue tropism, though the heart, pancreas, and skeletal muscle are the primary targets for pathology [29, 30]. Detailed in situ hybridization studies using RNAscope® technology have demonstrated SAV3 RNA in a wide array of tissues during acute infection, with the heart ventricle showing the most extensive signal [30]. Notably, the virus displayed a specific affinity for adipocyte components, being detected in various adipose tissues associated with internal organs [30]. This tropism for adipocytes may have profound implications for pathogenesis, as adipose tissue could serve as a viral reservoir and contribute to the metabolic disturbances observed during PD. The pseudobranch also emerged as a site of persistent, low-level infection, suggesting a pivotal role in SAV pathogenesis, possibly as an initial entry portal or a site of viral persistence that seeds systemic infection [30]. At the ultrastructural level, cytolytic changes consistent with virus-induced apoptosis were observed in the heart, while other tissues such as the gill and head kidney displayed evidence of viral morphogenesis without overt cytopathology, indicating that tissue-specific factors govern the outcome of infection [29].

Nonstructural Proteins as Master Regulators of Host Immunity

nsP2: A Multifunctional Antagonist of the Innate Immune Response

The nonstructural protein 2 (nsP2) of SAV has emerged as a central virulence factor, functioning as a potent antagonist of the host type I interferon (IFN) system. The SAV nsP2 effectively blocks the induction of type I IFN downstream of the transcription factor IRF3, thereby crippling the RIG-I-like receptor (RLR) signaling cascade, the first line of antiviral defense [13]. Critically, this inhibition is independent of the protease activity carried by nsP2. Instead, the C-terminal domain of nsP2, which contains two nuclear localization sequences (NLS), is both necessary and sufficient to block IFN induction [13]. The localization of nsP2 within the host cell nucleus is a prerequisite for its inhibitory function, and mutation of these NLS sequences is deleterious to the virus, highlighting the essential nature of this immune evasion strategy [13]. The precise mechanism involves a punctate distribution of nsP2 within discrete areas of chromatin, suggesting that it interferes with the transcriptional machinery required for IFN-β gene expression, rather than directly sequestering IRF3 [13].

Concurrently, nsP2 also activates the NF-κB signaling pathway, a master regulator of inflammation [8]. SAV, primarily through nsP2, upregulates Toll-like receptors TLR3, TLR7, and TLR8, subsequently activating the adaptor molecules MyD88 and TRAF6, leading to IκB degradation, p65 phosphorylation, and nuclear translocation [8]. This activation drives the expression of downstream inflammatory cytokines, which contributes to the hallmark inflammatory lesions of PD, including pancreatic necrosis, cardiomyopathy, and skeletal myopathy [8]. Intriguingly, nsP2 also upregulates mitochondrial antiviral signaling protein (MAVS) and can promote the expression of IFNa1 and the antiviral protein Mx, suggesting a delicate balancing act where the virus simultaneously induces and subverts the antiviral response to establish a productive infection [8]. The pro-inflammatory NF-κB activation may be a double-edged sword, driving tissue damage while also potentially contributing to virus clearance, a trade-off that is particularly relevant in the context of live-attenuated vaccine design [1].

nsP2 Mediated Disruption of the JAK-STAT Signaling Axis

Beyond the RLR and NF-κB pathways, SAV nsP2 orchestrates a broader assault on the host cellular signaling network, specifically targeting the Janus kinase/signal transducer and activator of transcription (JAK-STAT) pathway. Transcriptome analyses of SAV3-infected macrophage/dendritic-like TO-cells revealed that the virus downregulates several critical components of the JAK-STAT pathway, including type I and II receptor genes, Jak2, Tyk2, Stat3, and Stat5 [37]. This wholesale suppression effectively blocks the transcription of antiviral genes that depend on STAT signaling, creating a cellular environment permissive for viral replication [37]. In addition, SAV3 infection was shown to downregulate suppressor of cytokine signaling 3 (SOCS3), a negative regulator of cytokine signaling, while SOCS1 and SOCS3 were upregulated in IFN-treated control cells [37]. This selective inhibition of SOCS3 suggests a sophisticated viral strategy to fine-tune the host response. The combined effect of suppressing the JAK-STAT cascade while simultaneously activating NF-κB demonstrates a highly coordinated viral manipulation of host signaling to favor replication while driving immunopathology.

Structural Proteins and Virulence Determinants

The Capsid Protein and Nuclear-Cytosolic Trafficking

The capsid protein of SAV, beyond its structural role in encapsidating the genomic RNA, harbors a nuclear localization signal (NLS) that is a critical determinant of virulence. The NLS within the capsid protein is responsible for suppressing cellular nuclear-cytosolic trafficking, a mechanism that likely contributes to immune evasion by disrupting the transport of transcription factors and other signaling molecules [1, 11]. Live attenuated virus clones generated by site-directed mutagenesis of this capsid NLS, particularly when combined with mutations in the E2 glycoprotein, exhibited profound attenuation in vivo [1, 11]. These double-mutant clones were characterized by rapid viral clearance from the host, minimal clinical symptoms, and greatly reduced transmission to cohabitant fish [1, 11]. The attenuation phenotype demonstrated that the capsid NLS is essential for full virulence, likely by enabling the virus to manipulate the host cell’s transcriptional and signaling environment. This finding has direct implications for vaccine development, as such mutants represent promising, albeit genetically modified, live attenuated vaccine candidates [1].

The Envelope E2 Protein: Glycosylation, Attachment, and Immunogenicity

The E2 glycoprotein is the primary target of neutralizing antibodies, and its sequence diversity underpins the antigenic variation observed among SAV subtypes [20]. Mutations targeting the N-glycosylation site in E2 (E2319Q or E2319A) resulted in attenuation in vitro, characterized by reduced CPE and lower infectious particle production [16]. In vivo, the E2 mutant clones retained some residual virulence but induced potent protective immunity against subsequent challenge with wild-type SAV3, suggesting that the attenuation conferred by these mutations does not completely abrogate immunogenicity [1]. This balance between attenuation and immunogenicity is a critical consideration for live vaccine design. The E2 protein also contains the primary attachment site for host cell receptors. Disruption of this attachment, as seen in the E2 mutant clones, leads to reduced viral entry and spread, contributing to the attenuated phenotype [1]. The structural modeling by Biacchesi et al. [9] further suggests that the α-helical feature in E2 may be a key determinant of host range and virulence, potentially mediating interactions with host-specific receptors or factors that influence replication efficiency.

Host Genetic Determinants of Resistance and Pathogenesis

The gig1 Locus: A Major QTL for PD Resistance

Host genetics play a substantial role in determining the outcome of SAV infection. Multiple studies have independently mapped a major quantitative trait locus (QTL) for PD resistance to chromosome Ssa03 in Atlantic salmon, explaining a significant proportion of the heritable variation in resistance [3, 17, 28]. This QTL was validated across different populations and life stages, confirming a common underlying genetic mechanism for resistance [28]. The most likely causal candidates within this QTL are three tandemly duplicated copies of the gig1-like gene, a fish-specific antiviral effector [3, 17]. The SNP within the 3′ untranslated region (UTR) of gig1 showed the highest association with viral load in a genome-wide association study (GWAS) [17]. Expression analyses revealed significant differences in gig1-like transcript levels among resistant and susceptible fish, and allele-specific expression studies identified SNPs with putative functional impact on gene regulation [3]. The gig1 gene product is thought to function as an interferon-stimulated gene (ISG) with direct antiviral activity, potentially restricting SAV replication. The presence of multiple copies and regulatory variation suggests that selective breeding programs could exploit this locus to enhance PD resistance [3, 28].

Major Histocompatibility Complex and Immunoglobulin Loci

Beyond the gig1 QTL, additional genomic regions contribute to resistance. The immunoglobulin heavy chain locus (IGH), mapped to a region containing multiple significant SNPs, suggests that antibody diversity and the ability to mount a robust humoral response are critical for controlling infection [17]. Furthermore, QTLs on other chromosomes, such as Ssa07, have been linked to infection-specific responses, suggesting that different genetic pathways may be activated depending on the route of exposure or the stage of infection [3]. Heritability estimates for PD resistance range from moderate to high (0.15–0.21 for viral load at 4 weeks post-infection, and 0.4–0.5 for survival) [17, 28], indicating that selective breeding is a viable strategy for disease management.

Cellular and Molecular Mechanisms of Tissue Pathology

Apoptosis and Cytolysis in Target Organs

The characteristic pathology of PD, necrosis of the exocrine pancreas, cardiomyopathy, and skeletal myopathy, is driven by a combination of direct viral cytolysis and host-mediated immunopathology. Ultrastructural studies have confirmed cytolytic changes in the heart, with signs of viral morphogenesis and apoptosis-mediated cell death, consistent with in vitro observations [29]. In the pancreas, viral replication leads to the destruction of acinar cells, resulting in a loss of digestive enzyme production and subsequent malabsorption and growth impairment [2, 21, 38]. The selective precipitation reaction (SPR) assay, which detects leaked muscle proteins such as enolase and aldolase in the serum of infected fish, provides a biochemical correlate of this tissue damage, with the magnitude of the SPR signal correlating with histopathological scores in the pancreas, heart, and muscle [38].

The Interplay of Inflammation and Tissue Damage

The activation of NF-κB by nsP2 is a primary driver of the inflammatory response that characterizes PD [8]. While inflammation is a necessary defense mechanism, the uncontrolled or prolonged upregulation of pro-inflammatory cytokines can exacerbate tissue damage. The expression of microRNAs (miRNAs) is dynamically regulated during SAV infection and likely plays a crucial role in modulating the balance between protective immunity and immunopathology [34]. For instance, the miRNA-21 family showed decreased expression early during infection, while other miRNAs increased in the later phase, potentially acting as a brake on inflammation [34]. Putative targets of these differentially expressed miRNAs include key immune regulators such as IRF3, IRF7, and IRF4, suggesting a complex post-transcriptional regulatory network that fine-tunes the antiviral response [34]. Disruption of this miRNA-mediated homeostasis could contribute to the excessive inflammation seen in severe cases of PD.

Host-Environment Interactions Influencing Pathogenesis

The Role of Smoltification and Seawater Adaptation

The physiological state of the host at the time of exposure profoundly influences the outcome of SAV infection. Atlantic salmon undergoing smoltification and the transition to seawater are particularly vulnerable. Post-smolts challenged only two weeks after seawater transfer exhibited higher viral loads, more severe histopathological lesions, and a greater magnitude of viremia compared to fish adapted for nine weeks [25, 36]. This increased susceptibility correlates with a distinct transcriptional profile. Salmon newly transferred to seawater showed a higher and more prolonged upregulation of the anti-inflammatory cytokine IL-10, which may suppress the development of an effective antiviral response [36]. In contrast, fish adapted for longer to seawater mounted a more robust humoral and cell-mediated immune response, characterized by earlier production of neutralizing antibodies, stronger expression of cellular response genes such as CD40 and MHCII, and a more rapid resolution of infection [36]. The inability of newly transferred fish to mount a timely and effective adaptive response likely creates a window of extreme vulnerability to SAV, which aligns with the epidemiological observation that PD outbreaks typically occur after seawater transfer [25, 36].

Temperature-Dependent Viral Fitness

Environmental temperature is a critical abiotic factor that modulates both viral replication kinetics and the host immune response. Laboratory studies using SAV3 in culture medium and seawater demonstrated that viral survival and infectivity are inversely related to temperature [18]. At 16°C, SAV3 shed from infected fish was completely inactivated within three weeks, whereas at 12°C, inactivation took four weeks [18]. The minimal infectious dose required for horizontal transmission was estimated to be 448 TCID50 per liter of seawater in an in vivo bath challenge model, whereas as few as 24 TCID50 could infect cells in vitro [18, 24]. These data have direct bearing on epidemiological modeling, suggesting that outbreaks may be more protracted at lower water temperatures due to extended environmental persistence of the virus.

The Impact of Co-infections: Sea Lice and Bacteria

SAV infection rarely occurs in isolation, and co-infections with other pathogens can dramatically alter disease pathogenesis. Infestation with the sea louse Lepeophtheirus salmonis has been shown to compromise the ability of peripheral blood monocytic cells (PBMCs) to control SAV replication ex vivo [33]. PBMCs from lice-infested fish exhibited significantly higher viral titers following SAV infection, which coincided with an inability to upregulate key antiviral genes such as IFIT5, IRF9, and Mx [33]. This finding provides a direct cellular mechanism for the field observation that sea lice infestations predispose salmon to more severe viral diseases, an interaction of profound importance in integrated pest management strategies. Furthermore, SAV infection itself can disrupt the host microbiome, leading to skin dysbiosis characterized by a decrease in protective Proteobacteria (e.g., Oleispira sp.) and an increase in opportunistic bacteria such as Flavobacteriaceae and Tenacibaculum sp., thereby rendering the host more susceptible to secondary bacterial infections [32]. Diploid salmon infected with SAV3 also showed greater histopathological signs of epitheliocystis caused by Candidatus Branchiomonas compared to controls, illustrating the complex interplay between viruses and bacteria in the gill microbiome [31]. The presence of Acholeplasma laidlawii in cell cultures has even been shown to enhance the cytopathic effect of SAV2, suggesting that bacterial co-factors can potentiate viral replication and pathology [35].

Genetic Variation and Quasispecies Dynamics

Intra-Host Genetic Diversity and Immune Escape

As an RNA virus with a high mutation rate, SAV exists as a quasispecies, a swarm of genetically diverse but related variants within a single host. Nanopore sequencing has unveiled a multitude of single nucleotide variants (SNVs) and deletions distributed across the SAV3 genome during experimental infections of Atlantic salmon and brown trout [5]. While many SNVs were transient and fish-specific, two relatively major SNVs were observed in multiple fish, and minor SNVs showed evidence of positive selection in brown trout, suggesting host-specific adaptive pressures [5]. The genetic diversity of SAV in farmed and wild fish populations is even more extensive than previously appreciated. A genome-wide target-enriched viral sequencing study revealed compelling evidence of mixed subtype infections (e.g., SAV2 and SAV3) in individual fish, including wild flatfish such as dab [19]. This intra-host diversity has profound implications for pathogenesis, as it provides the raw material for immune escape, the emergence of more virulent variants, and the potential for recombination. The presence of co-infecting subtypes in the same host could facilitate the exchange of genetic material, accelerating viral evolution [12, 19].

Emerging Genotypes and Host Range

The classification of SAV has expanded from six to potentially seven genotypes, with a unique strain isolated from wild-caught ballan wrasse (Labrus bergylta) proposed as SAV7 [14]. This isolate, along with others from wrasse, demonstrates that SAV can infect non-salmonid fish species, raising concerns about the role of wild fish as reservoirs for the virus [14, 22]. The isolation of SAV6 from clinically healthy ballan wrasse suggests that these fish may act as asymptomatic carriers, potentially facilitating the spread of the virus into salmon farms where they are deployed as cleaner fish for sea lice control [22]. This expanded host range adds another layer of complexity to the molecular pathogenesis of SAV, as the virus must adapt to replicate in phylogenetically distant fish species, which likely involves selection for specific virulence determinants encoded by the E2 glycoprotein and other host-interaction domains [9, 14].

Epidemiology of Pancreas Disease in Salmonid Aquaculture

Pancreas disease (PD), caused by infection with salmonid alphavirus (SAV), represents one of the most economically significant viral diseases affecting European salmonid aquaculture, with profound implications for fish welfare, production efficiency, and international trade. The epidemiology of PD is a complex tapestry woven from viral genetic diversity, host population dynamics, environmental drivers, and anthropogenic factors such as farm management practices and vaccine deployment. A comprehensive understanding of these epidemiological patterns is essential for designing effective control strategies, informing risk assessments, and guiding policy decisions at both national and international levels, as recognized by the World Organisation for Animal Health (WOAH) and the European Food Safety Authority (EFSA), which have assessed SAV for listing under the Animal Health Law [6].

Global Distribution and Genotypic Biogeography

The epidemiological landscape of SAV is characterized by the spatial segregation and co-circulation of six distinct genotypes (SAV1–SAV6), each with a defined, albeit evolving, geographic range and host preference [7, 20, 41]. SAV1 and SAV4–SAV6 are predominantly associated with PD outbreaks in Atlantic salmon (Salmo salar) farmed in the marine environments of Ireland and Scotland, though notable exceptions exist. A landmark study confirmed the first clinical case of SAV1 infection in freshwater Atlantic salmon in Scotland, demonstrating that the virus can cause pathology consistent with PD outside the typical marine production phase and challenging the long-held assumption that PD is exclusively a disease of seawater-adapted fish [23]. SAV3 is the dominant genotype in Norway, the world’s largest producer of Atlantic salmon, and is responsible for the majority of PD outbreaks in that country, affecting both salmon and rainbow trout (Oncorhynchus mykiss) [12, 20, 30, 39, 42]. SAV2 exists in two distinct epidemiological forms: a freshwater variant (SAV2 FW), the etiological agent of sleeping disease (SD) primarily in rainbow trout across continental Europe, and a marine variant (SAV2 M) that causes PD in Atlantic salmon in Scotland and, more recently, in Norway [10, 20]. The emergence of SAV2 in Norway around 2010, hypothesized to be a single introduction from Scotland via well-boat traffic, has fundamentally altered the Norwegian PD epizootic, establishing a new endemic zone and complicating legislative control measures designed to limit the spread of SAV3 [10, 12]. Subsequent genomic epidemiological analyses have revealed that, following its introduction, the Norwegian marine SAV2 lineage diverged into two distinct clades (SAV2a and SAV2b) around 2013, with SAV2a predominantly circulating in southern regions and SAV2b in more northern latitudes, and with evidence of long-distance transmission events linking distant production regions [12]. Furthermore, the discovery of SAV6 in wild-caught ballan wrasse (Labrus bergylta) in Ireland, and the isolation of SAV7 from the same species, expands the known host range and highlights the role of non-salmonid fish, particularly those used as cleaner fish for sea lice control, as potential reservoirs or vectors for viral transmission, a finding with significant implications for biosecurity management [14, 22].

Host Range, Species Susceptibility, and the Role of Wild Fish

While the primary clinical and economic impact of PD is borne by intensively farmed Atlantic salmon and rainbow trout, the host range of SAV is broader than previously appreciated. Experimental infections have confirmed that Arctic char (Salvelinus alpinus) are susceptible to SAV2 FW, developing clinical signs of sleeping disease, and SAV infection has been documented in farmed Arctic char in Austria, extending the geographic and host range of the virus [47]. The detection of SAV RNA and the isolation of infectious virus from wild fish populations is a critical area of epidemiological inquiry. Comprehensive surveillance of migrating Atlantic salmon post-smolts in Norwegian fjord systems, conducted over two years, failed to detect SAV in any of the 651 fish sampled, suggesting that migrating smolts are at low risk of infection from farm-derived virus during their seaward migration [39]. Similarly, a study of returning adult Atlantic salmon caught at sea in northern Norway found no evidence of SAV infection, despite a high prevalence of piscine orthoreovirus (PRV) in the same population, although a significant proportion (10%) of the sampled fish were identified as farm escapees, which may act as a bridge for pathogen introduction into wild stocks [44]. In stark contrast, the repeated detection of SAV in wild flatfish, such as dab (Limanda limanda), in Scottish and Irish waters, and the confirmation of mixed-subtype co-infections in individual wild fish, provides compelling evidence that wild marine fish can sustain SAV infection and may constitute a true reservoir, capable of seeding outbreaks in naïve farmed populations [19]. This finding is further supported by the isolation of SAV6 from asymptomatic wild ballan wrasse, a species that shares a farm environment as cleaner fish [22]. The implications are profound: the aquatic environment is not merely a medium for passive viral transport but may harbor a dynamic and genetically diverse viral pool circulating within wild fish communities, making eradication of SAV from an endemic region an exceptionally challenging, if not impossible, goal.

Transmission Dynamics: Horizontal Spread, Viral Shedding, and Environmental Persistence

SAV is transmitted horizontally through the water column, with horizontal transmission from infected to naïve cohabitating fish being the primary route of spread both within and between farms [4, 24]. The virus is shed into the water primarily via the feces and urine of infected fish, and the magnitude of shedding is directly correlated with the viral dose to which the population is exposed [24]. Experimental bath challenge models have demonstrated that SAV3 is highly infectious, with a dose as low as 7 TCID₅₀ L⁻¹ of seawater sufficient to establish infection in a population, and that the prevalence of viremia and the peak shedding rate are dose-dependent [24]. Peak shedding rates can reach 2.4 × 10⁴ TCID₅₀ L⁻¹ of seawater h⁻¹ kg⁻¹, representing a substantial infectious burden being released into the surrounding environment [24]. The presence of SAV in tank water can be detected by RT-qPCR as early as 7 days post-challenge, preceding its detection in fish tissues by several days, and a significant positive correlation exists between viral RNA concentration in water and in fish mid-kidney samples, validating the use of water sampling as an early-warning surveillance tool [40, 42]. Field validation of this approach in Norwegian marine farms demonstrated that SAV could be detected in seawater samples an average of several weeks before it was detected in monthly fish samples, offering a faster, less invasive, and more cost-effective method for monitoring farm-level infection status [42]. The survival of SAV in the aquatic environment is temperature-dependent; infectivity decays more rapidly at higher temperatures, with complete inactivation of shed virus occurring within three weeks at 16°C and four weeks at 12°C, as measured by TCID₅₀, although RT-qPCR signals may persist longer due to detection of non-infectious RNA [18]. A critical epidemiological threshold was identified: while cell culture could be infected with as little as 24 TCID₅₀ L⁻¹ of seawater, successful transmission to naïve post-smolts required a starting concentration above 448 TCID₅₀ L⁻¹, indicating that a minimum infectious dose must be exceeded for horizontal transmission to occur in a field setting [18].

Risk Factors: Physiological State, Smoltification, Ploidy, and Co-infections

The susceptibility of Atlantic salmon to SAV is not uniform across all life stages or physiological conditions. A seminal finding is that post-smolts are significantly more susceptible to SAV3 infection when challenged two weeks after seawater transfer compared to nine weeks post-transfer [25, 36]. Fish adapted to seawater for the longer period developed a more robust humoral and cell-mediated immune response, including the production of neutralizing antibodies, which were entirely absent in the group infected shortly after transfer [36]. This suggests that the physiological stress and immunological immaturity associated with the smoltification window creates a period of heightened vulnerability, which may explain why PD outbreaks in Norway are typically detected after sea transfer [36]. The anti-inflammatory cytokine IL-10 was upregulated to a greater degree in fish infected early after transfer, potentially reflecting an attempt to control excessive inflammation but perhaps also contributing to a less effective antiviral state [36]. The ploidy of the host also influences infection dynamics. While both diploid and triploid Atlantic salmon are susceptible to SAV, experimental challenges with SAV3 using a low-dose bath model showed that triploid post-smolts accumulated prevalence more slowly than their diploid counterparts, reaching 100% prevalence later in the infection course [46, 48]. In SAV1-challenged fry, triploids exhibited lower viral RNA copy numbers in heart and liver and less severe histopathological lesions in the pancreas and myocardium compared to diploids, suggesting a potentially lower susceptibility or reduced viral replication capacity in triploid fish [45]. The underlying mechanism remains unclear but may be related to differences in cell metabolism or innate immune signaling [45]. Co-infections with other pathogens are a hallmark of the multifactorial disease ecology in aquaculture. Sea lice (Lepeophtheirus salmonis) infestation, a ubiquitous problem, has been shown to directly compromise the ability of host immune cells to control SAV replication. Peripheral blood mononuclear cells (PBMCs) isolated from lice-infested fish supported significantly higher SAV replication ex vivo compared to PBMCs from uninfested controls, concomitant with a failure to upregulate key antiviral genes such as IFIT5, IRF9, and Mx [33]. This provides a mechanistic link between parasite burden and heightened viral disease risk. Furthermore, SAV3 infection itself can induce dysbiosis of the skin microbiota, characterized by a loss of protective Proteobacteria (e.g., Oleispira sp.) and an overgrowth of opportunistic pathobionts such as Flavobacteriaceae and Tenacibaculum sp., potentially rendering the host more susceptible to secondary bacterial infections and further complicating the clinical picture [32]. The interaction between SAV and other viruses is also evident; co-infections of SAV2a and SAV3 have been documented in individual farmed fish in Norway, and the presence of Candidatus Branchiomonas in the gill microbiome was more pronounced in diploid salmon infected with SAV3 compared to uninfected controls, suggesting that viral infection can exacerbate existing bacterial conditions [12, 31].

Genetic and Genomic Epidemiology: Host Resistance and Viral Evolution

The host genetic architecture plays a substantial role in determining resistance to PD, with moderate-to-high heritability estimates (h² ~0.4–0.5) for survival after SAV challenge, providing a strong foundation for selective breeding programs [28]. A major quantitative trait locus (QTL) on chromosome Ssa03 has been consistently and independently validated across multiple populations and life stages (fry and post-smolts), explaining a significant proportion of the genetic variation in resistance [3, 17, 28]. Fine-mapping and transcriptomic analyses have identified three tandemly duplicated gig1-like genes as the most likely causal candidates within this QTL region, with specific single nucleotide polymorphisms (SNPs) in the 3' untranslated region of gig1 showing the strongest association with viral load [3, 17]. Gig1 is a fish-specific antiviral effector, and allele-specific expression analysis suggests that functional variation in these genes is a key determinant of resistance [3]. A second QTL on chromosome Ssa07 has also been linked to infection-specific response, indicating that different genetic pathways may govern resistance depending on the route or timing of exposure [3]. On the viral side, SAV, as an RNA virus, exhibits inherently high mutation rates, and this genetic plasticity has measurable epidemiological consequences. Intra-host variation is extensive, with Nanopore sequencing revealing a multitude of single nucleotide variants (SNVs) and deletions distributed across the SAV3 genome during experimental infections of both Atlantic salmon and brown trout, with greater genetic diversity observed in brown trout late in infection [5]. This capacity for rapid evolution was demonstrated by the emergence of two distinct SAV2 clades in Norway within just a few years of its introduction [12]. Furthermore, genome-wide target-enriched sequencing of field isolates has uncovered extensive "hidden" diversity, including the circulation of multiple SAV subtypes on the same farm and mixed-subtype co-infections in individual fish, both farmed and wild [19]. The pooling of samples, a common diagnostic practice, was shown to significantly underestimate this diversity, masking the true complexity of circulating viral populations and potentially hindering the detection of emergent genotypes with altered virulence or antigenicity [19]. The identification of novel SAV3 lineages that diverged from previously characterized strains more than 25 years ago underscores the long-term, cryptic evolution of this virus in the marine environment [12]. These genomic tools are now being deployed for real-time tracking of transmission networks, providing a powerful epidemiological lens through which to investigate the origins and spread of outbreaks and to inform targeted control measures [12].

Vaccination, Herd Immunity, and Shedding Dynamics

Vaccination is the cornerstone of PD control, and epidemiological studies have provided critical insights into how different vaccine platforms influence infection dynamics at the population level. While oil-adjuvanted, inactivated whole-virus vaccines are widely used and reduce mortality and clinical severity, they do not fully prevent infection or transmission [4, 43]. In contrast, a DNA vaccine (DNA-PD) has demonstrated the capacity to significantly curb the spread of SAV. In a controlled transmission study, DNA-PD vaccinated fish that were subsequently challenged with SAV2 did not transmit the infection to cohabiting DNA-PD vaccinated fish, a finding that signifies the potential for achieving herd immunity if a sufficiently high proportion of the population is vaccinated with this platform [4]. This effect was attributed to significantly lower viremia levels, faster clearance of viral RNA from heart tissue, and a marked reduction in viral shedding into the water, as measured by RT-droplet digital PCR [4, 15]. In contrast, oil-adjuvanted vaccinated fish shed virus and infected cohabiting oil-vaccinated fish at levels comparable to naïve controls [4]. The DNA vaccine also induced significantly higher neutralizing antibody titers and reduced the prevalence and severity of histopathological lesions in target organs, including the pancreas, heart, and muscle [15, 43]. The importance of vaccination in reducing environmental viral load was further underscored by a study demonstrating that PD vaccination significantly reduces viral shedding from infected individuals, thereby lowering the infection pressure on neighboring farms, a recognized key risk factor for PD outbreaks in endemic regions [26]. Live attenuated vaccines, developed through targeted mutagenesis of the E2 glycoprotein and/or the capsid protein nuclear localization signal, have shown promise in experimental settings by providing robust protection against subsequent challenge, but they highlight a critical trade-off: greater attenuation results in lower immunogenicity and weaker protection, while strains with residual virulence provide stronger protection but also cause significant reductions in weight gain and can be transmitted to naïve cohabitants, raising concerns about environmental safety and reversion to virulence [1, 11]. The development of a live attenuated vaccine would require a delicate balance between safety and efficacy, and its use would necessitate careful risk assessment regarding the shedding of genetically modified virus [1].

Clinical Signs, Pathophysiology, and Biomarkers of SAV Infection

The clinical manifestation of salmonid alphavirus (SAV) infection represents a complex interplay between viral virulence factors, host genetic susceptibility, environmental stressors, and the ontogenetic stage of the fish. As a notifiable pathogen under the World Organisation for Animal Health (WOAH) guidelines, SAV infection presents with a spectrum of clinical outcomes ranging from subclinical infection to catastrophic mortality, with the hallmark pathological triad of exocrine pancreatic necrosis, cardiomyopathy, and skeletal myopathy defining the disease. Understanding the nuanced clinical signs, the underlying pathophysiological mechanisms, and the utility of biomarkers is essential for both clinical management and epidemiological surveillance in Atlantic salmon (Salmo salar) and rainbow trout (Oncorhynchus mykiss) aquaculture.

Clinical Signs and Disease Progression

The clinical presentation of SAV infection varies considerably depending on the viral genotype, host species, age, water temperature, and the presence of co-infections. In Atlantic salmon, pancreas disease (PD) typically manifests during the marine phase, although freshwater outbreaks have been documented [23]. The incubation period ranges from 7 to 14 days post-exposure under experimental conditions, with clinical signs becoming apparent as viremia peaks and target organ pathology develops [24]. Early clinical signs are often nonspecific and include anorexia, lethargy, and darkened pigmentation. Hoel et al. [2] documented an 89% reduction in appetite at 4 weeks post-challenge in a controlled trial, with weekly mortality peaking at 4.1% in week 5. This profound anorexia is a critical clinical feature, as it directly contributes to the substantial growth impairment that represents the primary economic impact of PD [21].

As the disease progresses, affected fish exhibit more characteristic signs. In rainbow trout infected with SAV subtype 2 (the causative agent of sleeping disease), the most pathognomonic clinical sign is the eponymous "sleeping" behavior, where fish lie on their sides at the bottom of tanks, exhibiting abnormal swimming patterns due to severe necrosis and atrophy of red skeletal muscle [27]. This behavioral manifestation is directly attributable to the virus's tropism for muscle satellite cells, as demonstrated by Biacchesi et al. [27], who established that SAV2 specifically infects these progenitor cells, leading to impaired muscle regeneration and persistent atrophy. In Atlantic salmon, while the sleeping posture is less common, affected fish often show reduced swimming activity, loss of equilibrium, and increased respiratory effort, reflecting the combined effects of cardiac dysfunction and muscle damage.

Mortality patterns in PD are highly variable. Field outbreaks can result in cumulative mortality ranging from negligible levels to over 50%, depending on the virulence of the circulating strain and management conditions. However, as Røsæg et al. [21] emphasized in their retrospective cohort study of SAV2-infected Atlantic salmon, the most significant economic consequence is often not acute mortality but rather the chronic reduction in growth and feed conversion efficiency. In a scenario with fixed slaughter time, the estimated impact corresponded to a growth reduction of 0.7 kg and a 0.07-point increase in feed conversion ratio [21]. This growth impairment is a direct consequence of the destruction of exocrine pancreatic tissue, which eliminates the primary source of digestive enzymes, leading to maldigestion and malnutrition. Braaen et al. [1] further demonstrated that even attenuated SAV clones, which cause minimal clinical symptoms, still induce significant reductions in weight gain, underscoring the metabolic cost of viral infection even in the absence of overt disease.

Pathophysiology: Cellular and Molecular Mechanisms

The pathophysiology of SAV infection is rooted in the virus's capacity to hijack host cellular machinery, evade innate immune responses, and induce direct cytolytic damage in target tissues. At the cellular level, SAV exhibits a broad tropism, with in situ hybridization studies by Tartor et al. [30] detecting viral RNA in heart ventricle, pancreas, skeletal muscle, adipose tissue, and notably, the pseudobranch. The heart ventricle shows the most extensive infection during the acute phase, consistent with the severe cardiomyopathy that characterizes PD [30]. Ultrastructural studies by Herath et al. [29] revealed that cytolytic changes are prominent in the heart, with evidence of membrane-dependent morphogenesis and apoptosis-mediated cell death, while other tissues such as gill and head kidney may harbor replicating virus without exhibiting degenerative lesions.

The molecular pathogenesis of SAV is orchestrated by both structural and nonstructural proteins. The nonstructural protein nsP2 emerges as a master regulator of host immune evasion. Jami et al. [13] demonstrated that nsP2 effectively blocks the induction of type I interferon (IFN) by acting downstream of IRF3, the transcription factor responsible for activating the IFN promoter. This inhibition is dependent on the nuclear localization of nsP2, which is mediated by two nuclear localization sequences (NLS) in its C-terminal domain. The C-terminal domain alone is sufficient to block IFN induction, and mutation of the NLS is deleterious to the virus, highlighting the critical importance of this immune evasion strategy for viral fitness [13]. Complementing this, Gao et al. [8] showed that nsP2 simultaneously activates the NF-κB signaling pathway by upregulating TLR3, 7, and 8, leading to IκB degradation, p65 phosphorylation and transnucleation, and subsequent expression of pro-inflammatory cytokines. This dual role of nsP2, suppressing the antiviral IFN response while promoting inflammatory signaling, creates a microenvironment that facilitates viral replication while contributing to the immunopathology observed in target tissues.

The mitochondrial antiviral signaling (MAVS) protein serves as a critical intermediary in the RIG-I-like receptor (RLR) signaling cascade. Xu et al. [49] demonstrated that disruption of MAVS in Chinook salmon embryonic cells resulted in significantly lower expression of IRF3, IFNa, and interferon-stimulated genes (ISGs) following SAV3 infection, leading to a 1.5 log10 increase in viral titer compared to wild-type cells. This finding underscores the essential role of the MAVS-dependent pathway in restricting SAV replication. Furthermore, the virus has evolved sophisticated mechanisms to subvert the JAK-STAT signaling pathway. Transcriptome analyses by Xu et al. [37] revealed that SAV3 infection downregulates several JAK-STAT pathway genes, including type I and II receptor genes, Jak2, Tyk2, Stat3, and Stat5, representing a strategic immune evasion tactic to block the transcription of antiviral genes. The suppressor of cytokine signaling 3 (SOCS3) was specifically downregulated in infected cells, further compromising the negative feedback regulation of cytokine signaling [37].

The role of microRNAs (miRNAs) in modulating the host response to SAV infection adds another layer of regulatory complexity. Andreassen et al. [34] identified 20 differentially expressed miRNAs in response to SAV infection, with the majority showing increased expression after viral load had stabilized or was decreasing. Notably, the miRNA-21 family showed decreased expression at early time points post-infection, suggesting a potential role in promoting early inflammatory responses. Target prediction analyses indicated that 17 of these differentially expressed miRNAs could target 24 immune network genes, including IRF3 and IRF7, with the majority of predicted targets promoting inflammatory responses [34]. The temporal expression patterns suggest that miRNAs may function as early promoters or late inhibitors of inflammation, contributing to the delicate balance between viral clearance and immunopathology.

Tissue-Specific Pathology and the Role of Adipose Tissue

The pathological hallmark of PD is the triad of exocrine pancreatic necrosis, cardiomyopathy, and skeletal myopathy. However, recent research has expanded our understanding of SAV tropism beyond these classical targets. Tartor et al. [30] provided compelling evidence for a specific affinity of SAV3 for adipocyte components, detecting viral RNA in adipose tissues associated with internal organs. This finding has significant pathophysiological implications, as adipose tissue is not merely a passive energy storage depot but an active endocrine organ that regulates metabolism and immune function. Infection of adipocytes could contribute to the metabolic dysregulation and growth impairment observed in PD, potentially through altered leptin signaling or adipokine-mediated inflammation.

The pseudobranch, a gill-derived endocrine organ unique to teleosts, has emerged as a previously unrecognized site of SAV persistence. Tartor et al. [30] detected early SAV3 infection in the pseudobranch, followed by persistent low-level infection over the experimental course. The inconsistent immune response to SAV3 in this tissue, characterized by failure to mount an effective antiviral state, likely contributes to viral persistence and may serve as a reservoir for reinfection or transmission. This finding challenges the conventional view that SAV is cleared from most tissues following the acute phase and suggests that certain anatomical niches may harbor the virus for extended periods.

The impact of SAV infection on mucosal microbiomes represents a novel dimension of pathophysiology. Reid et al. [32] demonstrated that SAV3 infection causes significant skin dysbiosis, characterized by decreased abundance of protective Proteobacteria such as Oleispira sp. and increased abundances of opportunistic taxa including Flavobacteriaceae, Streptococcaceae, and Tenacibaculum sp. This viral-induced disruption of the skin microbial community likely renders the host more susceptible to secondary bacterial infections, a clinically relevant observation given the frequent co-occurrence of PD with bacterial diseases such as tenacibaculosis and furunculosis in field settings. Brown et al. [31] further showed that ploidy influences the gill microbiome during SAV infection, with diploid salmon exhibiting greater histopathological signs of epitheliocystis caused by Candidatus Branchiomonas compared to triploids, suggesting complex interactions between host genetics, viral infection, and microbial communities.

Biomarkers of Infection and Disease

The identification and validation of biomarkers for SAV infection have advanced significantly, providing tools for early detection, monitoring disease progression, and assessing physiological status. Traditional diagnostic approaches rely on RT-qPCR detection of viral RNA in heart tissue, which remains the gold standard for confirming infection [51]. However, the temporal dynamics of viral RNA detection must be carefully interpreted. Jansen et al. [51] reported that real-time RT-PCR shows high diagnostic sensitivity (≥0.977) but only moderate diagnostic specificity (0.831), indicating that false positives can occur, particularly in populations with low disease prevalence. This underscores the need for confirmatory testing, such as virus isolation in cell culture, which demonstrated both high diagnostic sensitivity and specificity in their Bayesian latent class analysis [51].

Blood chemistry parameters have proven valuable as biomarkers of tissue damage and physiological disturbance. Hoel et al. [2] conducted a comprehensive analysis of plasma biomarkers in both controlled and field trials, documenting significant increases in creatine kinase (CK), alanine aminotransferase (ALAT), aspartate aminotransferase (ASAT), and lactate dehydrogenase (LDH) following SAV infection. In the controlled trial, CK, ALAT, and ASAT levels increased significantly at 4 weeks post-challenge, peaked at 8 weeks, and while ALAT returned to pre-challenge levels by 12 weeks, CK and ASAT remained elevated, indicating ongoing muscle and cardiac damage [2]. The temporal correlation between these enzyme elevations and clinical signs such as anorexia and mortality supports their utility as objective measures of disease severity. In field trials, these biomarkers peaked concurrently with increasing SAV prevalence, along with stress markers including plasma cortisol and fecal tetrahydrocortisone, despite low mortality (<0.2% weekly) [2].

A particularly innovative biomarker discovery was reported by Braceland et al. [38], who identified a selective precipitation reaction (SPR) in serum from SAV-infected Atlantic salmon. Mixing serum from infected fish with sodium acetate buffer caused a visible precipitation that does not occur with serum from healthy salmon. Proteomic analysis revealed that the precipitate contains muscle proteins (enolase, aldolase) along with serum proteins (serotransferrin, complement C9). The SPR assay, measured as change in optical density at 340 nm, showed a temporal profile that correlated with histopathological scores of pancreas, heart, and muscle damage, peaking at 6 weeks post-infection [38]. This simple, cost-effective test has potential as an on-farm qualitative tool for health assessment, though further validation is needed.

Serological biomarkers of adaptive immunity have also been developed. Teige et al. [50] established a bead-based immunoassay for detecting SAV-specific antibodies using detergent-treated SAV particles as antigens. Increased levels of SAV-specific antibodies were detected after most fish had become negative for viral RNA, indicating that seroconversion occurs during the convalescent phase. This assay offers advantages over traditional virus neutralization tests, including reduced time requirements and suitability for non-lethal testing due to low sample volume requirements [50]. Neutralizing antibody titers have been directly correlated with protection, as Thorarinsson et al. [43] demonstrated that DNA-vaccinated fish with higher neutralizing antibody levels showed significantly lower viremia and reduced transmission to naïve cohabitants.

Environmental Detection and Shedding Dynamics

The detection of SAV in seawater has emerged as a powerful tool for early surveillance and understanding transmission dynamics. Bernhardt et al. [40, 42] developed and validated methods for concentrating SAV from seawater using electronegative membrane filters, achieving recovery rates of 39.5% by RT-ddPCR. In field trials across seven Norwegian marine farm sites, SAV was detected in seawater at an earlier stage compared to traditional fish sampling methods, with a significant negative relationship between seawater viral concentration and the time until SAV was detected in fish [42]. This approach offers a non-lethal, animal welfare-friendly alternative for surveillance that can provide early warning of impending outbreaks.

The shedding dynamics of SAV are critical for understanding transmission risk. Jarungsriapisit et al. [24] demonstrated that a dose as low as 7 TCID50 L⁻¹ of seawater can establish infection in a population, confirming the highly infectious nature of SAV through horizontal transmission. Maximal shedding rates of 2.4 × 10⁴ TCID50 L⁻¹ of seawater h⁻¹ kg⁻¹ were recorded 10 days post-exposure from the highest dose group [24]. Importantly, vaccination significantly reduces shedding, as Skjold et al. [26] demonstrated that PD-vaccinated fish shed significantly less virus than unvaccinated controls, providing a mechanism for herd immunity. Thorarinsson et al. [4] further showed that DNA-vaccinated fish did not transmit SAV2 to cohabiting DNA-vaccinated fish, whereas oil-adjuvanted vaccinated fish transmitted infection to vaccinated cohabitants at levels similar to naïve fish, highlighting the superior capacity of DNA vaccines to curb viral spread.

The environmental persistence of SAV is temperature-dependent. Jarungsriapisit et al. [18] demonstrated that SAV3 shed by infected fish at a starting titer of 882 TCID50 L⁻¹ of seawater was completely inactivated after 3 weeks at 16°C and 4 weeks at 12°C. The virus was more stable in culture medium compared to seawater, and a threshold of 448 TCID50 L⁻¹ of seawater was required for successful transmission to post-smolts in vivo [18]. These data are essential for parameterizing pathogen dispersal models and implementing risk-based management strategies.

Genetic Determinants of Clinical Outcome

The clinical outcome of SAV infection is strongly influenced by host genetics. Genome-wide association studies have identified a major quantitative trait locus (QTL) on chromosome 3 that explains a significant proportion of genetic variation in resistance to PD [28]. Manousi et al. [3] refined this QTL and identified three tandemly duplicated gig1-like genes as likely causal candidates. The gig1 gene encodes a fish-specific antiviral effector, and the identification of multiple copies with differential expression patterns suggests that gene duplication has provided evolutionary flexibility in antiviral defense. Aslam et al. [17] further demonstrated moderate genomic heritability of viral load at 4 weeks post-infection (0.15–0.21) and a high positive genetic correlation with survival (0.91–0.98), indicating that selection for reduced viral load would concurrently improve survival. These genetic markers are being incorporated into selective breeding programs to enhance PD resistance, representing a sustainable, long-term approach to disease management.

The interaction between ploidy and SAV susceptibility has practical implications for aquaculture. Moore et al. [46] found that triploid Atlantic salmon post-smolts accumulated SAV3 prevalence more slowly than diploids following bath challenge, reaching 19% and 56% prevalence at 14 and 21 days post-exposure, respectively, compared to 82% and 100% in diploids. However, once infected, viral loads in heart tissue did not differ between ploidies, and triploids were not more susceptible overall [46]. Herath et al. [45] observed lower viral RNA copy numbers in triploid fry compared to diploids, particularly in the liver, along with less severe pancreatic and myocardial degeneration. These findings suggest that triploid salmon may have a degree of relative resistance, possibly related to differences in cell metabolism or immune function, though the mechanisms require further investigation.

The physiological status of the host at the time of exposure profoundly influences clinical outcome. Jarungsriapisit et al. [25] demonstrated that Atlantic salmon post-smolts challenged 2 weeks after seawater transfer were significantly more susceptible to SAV3 infection than those challenged 9 weeks after transfer, as evidenced by higher viremia, higher viral RNA in hearts, more extensive histopathological lesions, and higher virus shedding. Nuñez-Ortiz et al. [36] showed that post-smolts adapted for longer to seawater developed stronger humoral and cell-mediated immune responses, including neutralizing antibodies and upregulation of CD40, MHCII, and IL-15, indicative of robust adaptive immunity. In contrast, fish challenged shortly after transfer exhibited higher and more prolonged upregulation of the anti-inflammatory cytokine IL-10 and SOCS1, suggesting a regulatory response that may impair viral clearance [36]. These findings have direct practical implications for the timing of seawater transfer and vaccination strategies in commercial aquaculture.

Co-infections with other pathogens can dramatically alter the clinical trajectory of SAV infection. Gamil et al. [33] demonstrated that sea lice (Lepeophtheirus salmonis) infestation significantly impairs the ability of peripheral blood monocytic cells to control SAV replication ex vivo, with higher virus titers coinciding with an inability to upregulate key antiviral genes including IFIT5, IRF9, and Mx. This immunosuppressive effect of sea lice infestation likely contributes to the increased severity of PD outbreaks observed in farms with high lice burdens, and underscores the importance of integrated pest management strategies.

Diagnostic Methods for SAV Detection and Surveillance

The detection and surveillance of salmonid alphavirus (SAV) in aquaculture systems demands a multi-faceted diagnostic approach that integrates molecular, serological, histopathological, and emerging environmental monitoring techniques. Given the significant economic impact of pancreas disease (PD) and sleeping disease (SD) on European salmonid aquaculture, with the World Organisation for Animal Health (WOAH) listing SAV as a notifiable pathogen, the development and validation of robust diagnostic tools remain paramount for disease management, control, and eventual eradication strategies. The following sections provide an exhaustive analysis of the current diagnostic arsenal, emphasizing their mechanistic foundations, operational characteristics, and strategic applications in both clinical and surveillance contexts.

Molecular Detection Methods: The Cornerstone of SAV Diagnostics

Real-time reverse transcription quantitative PCR (RT-qPCR) has become the gold standard for SAV detection, owing to its exceptional sensitivity, specificity, and capacity for high-throughput analysis. The assay typically targets highly conserved regions of the viral genome, most frequently the non-structural protein 1 (nsP1) gene, which provides reliable detection across all known SAV subtypes (SAV1-6 and the proposed SAV7 genotype) [12, 25, 30]. The analytical performance of RT-qPCR has been rigorously evaluated in field settings; a comprehensive Bayesian latent class analysis conducted across three Norwegian Atlantic salmon populations demonstrated that real-time RT-PCR possesses a diagnostic sensitivity (DSe) of ≥0.977, though its diagnostic specificity (DSp) was moderate at 0.831 [51]. This moderate specificity is a critical consideration, as it implies that weak positive results may require confirmatory testing, particularly in populations where false positive results carry significant regulatory or economic consequences. The study emphasized that cell culture, with its high DSe and DSp, remains a valuable confirmatory tool for verifying viral presence in cases of ambiguous RT-qPCR results [51].

The evolution of RT-qPCR technology has yielded increasingly refined detection platforms. The EvaGreen-based real-time PCR assay represents a significant advancement, offering a dye-based alternative to probe-based systems that combines high sensitivity with reduced costs. This assay demonstrated a detection limit of 1.5 × 10¹ copies, substantially outperforming conventional RT-PCR methods (detection limit of 1.5 × 10⁶ copies) and exhibiting superior sensitivity compared to earlier SYBR Green-based assays [53]. Importantly, this platform showed high specificity for SAV1, SAV2, and SAV5, making it a practical and economical option for diagnostic laboratories with budget constraints [53]. The assay's capacity to detect extremely low viral loads is particularly relevant for surveillance programs where early detection of subclinical infections is critical for preventing widespread outbreaks.

More specialized molecular approaches include RT-droplet digital PCR (RT-ddPCR), which provides absolute quantification of viral RNA copies without the need for standard curves. This technique has proven particularly valuable in environmental surveillance applications, specifically for quantifying SAV in seawater samples. In validation studies using spiked seawater, RT-ddPCR demonstrated limits of quantification (LOQ) and detection (LOD) of 5.18 × 10³ and 2.0 × 10² SAV3 copies per liter, respectively [52]. The superior performance of RT-ddPCR in quantifying low-titer samples makes it an indispensable tool for understanding viral shedding dynamics and environmental persistence.

Advanced Genomic Characterization: Nanopore Sequencing and Beyond

The application of third-generation sequencing technologies has revolutionized our understanding of SAV genetic diversity, evolutionary dynamics, and transmission patterns. Nanopore sequencing, in particular, has emerged as a powerful tool for real-time genomic epidemiology. In a landmark study, Macqueen et al. successfully captured approximately 90% of the SAV2 genome from 68 field isolates sampled across ten Norwegian aquaculture production regions between 2018 and 2020, using nanopore sequencing [12]. This approach enabled the reconstruction of time-calibrated phylogenies that revealed the emergence of SAV2a and SAV2b clades around 2013, with distinct latitudinal distributions and evidence of long-distance transmission events [12]. The discovery of co-infections involving SAV2a and SAV3 within individual fish, including novel SAV3 lineages diverging from previously characterized strains by over 25 years, underscores the complexity of viral population dynamics that can only be resolved through genome-wide characterization [12].

The utility of nanopore sequencing extends to intrahost viral diversity investigations. Roh et al. employed overlapping amplicons covering 98.8% of the SAV3 genome to characterize genetic variation during experimental infections in Atlantic salmon and brown trout [5]. This approach identified a multitude of single nucleotide variants (SNVs) and deletions distributed across the genome, with distinct patterns emerging between host species. Notably, late in infection, SAV3 genomes isolated from brown trout exhibited greater variation than those from Atlantic salmon, suggesting species-specific selection pressures on viral populations [5]. The bioinformatics pipeline developed for this study, which included de novo clustering of nanopore reads at a 99% sequence identity threshold and nonmetric multidimensional scaling analysis, provides a robust framework for investigating the composition of genetic diversity during viral infections.

Target-enriched viral sequencing has further expanded our capacity to characterize SAV diversity in complex sample matrices. Gallagher et al. applied a sequence capture strategy to obtain high-coverage SAV genomes from both pooled and individual samples, revealing extensive intrahost genetic diversity that had remained undetected with conventional approaches [19]. This study uncovered mixed subtype infections in three of four studied species (farmed Atlantic salmon, rainbow trout, and wild flatfish), including evidence of co-infections within individual fish. The critical finding that pooling samples from different fish significantly underestimates viral genetic diversity has profound implications for surveillance programs, as it likely limits the power to detect transmission of novel genotypes across geographic regions or between farms [19]. The study concluded that SAV transmission and evolutionary dynamics are fundamentally more complex than previously recognized, necessitating routine genome-wide characterization to capture the true diversity associated with disease outbreaks.

Partitioned phylogenetic analyses, accounting for variation in evolutionary rates across individual genes and codon positions, have proven essential for accurate classification of novel SAV genotypes. Tighe et al. demonstrated that applying partitioned models to whole-genome sequences of a unique SAV strain isolated from ballan wrasse almost doubled the observed branch lengths compared to traditional single-model approaches, and significantly improved the statistical fit of nucleotide substitution models [14]. This rigorous analytical framework provided compelling genomic evidence for designating the wrasse isolate as a new SAV genotype (SAV7), while also supporting the proposal to amend the viral species name to the more inclusive 'piscine alphavirus' to reflect its expanding host range [14].

Non-Lethal Surveillance Methods: Water Filtration and Environmental Detection

The traditional approach to SAV surveillance in Norwegian salmon farms, which relies on monthly lethal sampling of fish for organ tissue analysis, is costly, time-consuming, and raises significant animal welfare concerns. The development of water filtration methods for SAV detection represents a paradigm shift in surveillance strategy, offering a rapid, cost-efficient, and welfare-friendly alternative [40, 42]. The underlying principle is straightforward: during infection, SAV is shed into the surrounding water, and concentrating viral particles from large volumes of seawater enables detection before the virus reaches detectable levels in fish tissues.

A systematic optimization study evaluated two filter types (electronegative MF-Millipore™ and electropositive 1 MDS Zeta Plus® Virosorb®) combined with four elution buffers for SAV3 recovery from spiked seawater. The electronegative filter paired with NucliSENS® easyMAG® Lysis Buffer achieved the highest recovery rates: 39.5 ± 1.8% by RT-ddPCR and 25.9 ± 5.7% by RT-qPCR [52]. This combination significantly outperformed the electropositive filter, which yielded only 19.0 ± 0.1% and 13.3 ± 3.8% recovery, respectively [52]. The superior performance of the electronegative filter, coupled with the requirement for only standard laboratory equipment, positions this method as the leading candidate for field validation.

The practical application of this technology was demonstrated in a landmark field trial conducted across seven Norwegian marine farm sites with no prior suspicion of SAV infection [42]. Seawater samples collected from the top layer inside net-pens were processed using the MF-Millipore™ electronegative membrane filter and NucliSENS® Lysis Buffer system, with RNA extraction followed by RT-qPCR analysis. The results were striking: SAV was detected from seawater at an earlier stage compared to traditional fish sampling at all sites where fish eventually tested positive. Furthermore, a significant negative relationship was observed between the SAV concentration in seawater and the number of days until SAV was detected in fish, meaning that lower viral concentrations in water predicted longer delays before fish became positive [42]. This temporal advantage, detecting infection days to weeks before clinical signs or tissue positivity, provides a critical window for implementing containment measures.

Experimental validation using a cohabitation challenge model further confirmed the utility of water filtration for early SAV detection [40]. SAV3 was detected in tank water as early as 7 days post-challenge (dpc), preceding detection in cohabitant fish organ tissues at 12 dpc. A significant positive correlation was established between SAV3 concentrations in water concentrates and mid-kidney tissue samples [40]. The method successfully detected SAV at concentrations as low as 7 TCID₅₀ L⁻¹ of seawater, a dose close to the minimum required to establish infection in a population, confirming the exquisite sensitivity of the approach [24, 40].

The integration of these findings with viral shedding kinetics provides a comprehensive framework for environmental surveillance. Jarungsriapisit et al. demonstrated that the outcome of SAV3 infection, including prevalence of viremic fish, viral RNA in heart tissues, and shedding rate, is positively correlated with the initial viral dose [24]. The maximal shedding rate of 2.4 × 10⁴ TCID₅₀ L⁻¹ of seawater h⁻¹ kg⁻¹ recorded 10 days post-exposure from the highest dose group provides a benchmark for estimating infection pressure in field settings [24]. Importantly, temperature significantly affects SAV survival in seawater, with higher temperatures accelerating viral inactivation. SAV3 shed by infected fish at a starting titer of 882 TCID₅₀ L⁻¹ was completely inactivated after three weeks at 16°C and four weeks at 12°C, as determined by TCID₅₀ assays [18]. These temperature-dependent survival kinetics must be incorporated into environmental surveillance models and risk assessments for virus dispersal between farms.

Serological and Immunological Diagnostic Approaches

Serological methods provide complementary information to molecular detection, offering insights into past exposure, immune status, and vaccine efficacy. The virus neutralization test (VNT) has traditionally served as the gold standard for detecting SAV-specific antibodies, particularly neutralizing antibodies that correlate with protection. However, VNT is labor-intensive, time-consuming (requiring several days), and requires specialized cell culture facilities, limiting its applicability for large-scale surveillance [50].

The development of a bead-based immunoassay represents a significant advancement in serological diagnostics for SAV. This assay utilizes detergent-treated SAV particles as antigens immobilized on fluorescent beads, enabling detection of virus-specific antibodies in plasma samples using a flow cytometric platform [50]. The assay successfully detected SAV-specific antibodies in both experimentally challenged fish and field samples from natural PD outbreaks. A critical finding was that increased antibody levels became detectable after most fish had already cleared viral RNA, indicating that seroconversion marks the convalescent phase of infection [50]. The bead-based format offers substantial advantages over VNT, including reduced turnaround time, lower sample volume requirements (enabling non-lethal testing), and suitability for high-throughput screening. The authors concluded that this serological tool is a promising complement to RT-qPCR screening, providing information on acquired immunity that molecular methods cannot offer [50].

The selective precipitation reaction (SPR) discovered by Braceland et al. represents an entirely novel diagnostic principle. When serum from SAV-infected Atlantic salmon is mixed with a sodium acetate buffer, a visible precipitation occurs that is absent in serum from healthy fish [38]. Proteomic analysis of the precipitate identified a mixture of muscle proteins (enolase, aldolase) and serum proteins (serotransferrin, complement C9), suggesting that the reaction reflects tissue damage and release of intracellular components into circulation. The assay was optimized for molarity, pH, temperature, and wavelength, enabling quantitative measurement of precipitation as change in optical density at 340 nm (Δ340). In a cohabitation trial, Δ340 values rose from undetectable levels to a maximum at 6 weeks post-infection, correlating strongly with histopathological scores of pancreas, heart, and muscle damage [38]. This simple, inexpensive test could serve as an on-farm qualitative screening tool or an in-laboratory quantitative assay for assessing tissue pathology in SAV-infected populations.

Histopathological and In Situ Detection Methods

Histopathological examination remains a cornerstone of SAV diagnosis, providing direct visualization of characteristic lesions in target organs. The classic triad of PD pathology, pancreatic necrosis, cardiomyopathy, and skeletal myopathy, is well documented [20]. However, the sensitivity and specificity of histopathology as a diagnostic test require careful consideration. Jansen et al. reported that histopathology exhibits moderate diagnostic performance compared to molecular methods, with the accuracy of lesion interpretation depending on the stage of infection and the experience of the pathologist [51].

Advances in in situ hybridization (ISH) technology, particularly the RNAscope® platform, have dramatically improved the sensitivity and specificity of viral detection within tissue sections. Tartor et al. employed a combination of RT-qPCR and RNAscope® to map SAV3 tropism in Atlantic salmon tissues following cohabitation challenge [30]. ISH detected SAV3 in multiple tissues during the acute phase of infection, with the heart ventricle showing the most extensive infection. Notably, the detection of SAV3 in various adipose tissues associated with internal organs suggests a specific affinity for adipocyte components [30]. The pseudobranch, a gill-derived structure of unknown function in teleosts, showed early SAV3 detection that persisted over the experimental course, suggesting a potential role in viral pathogenesis and persistence [30]. This study demonstrates that ISH can reveal cellular-level tropism and viral distribution patterns that are not apparent from bulk tissue analysis by RT-qPCR.

Host Response Biomarkers as Diagnostic Tools

The physiological response to SAV infection generates a suite of biomarkers that can be exploited for diagnostic and prognostic purposes. A comprehensive study by Hoel et al. examined plasma and fecal biomarkers during controlled and field SAV3 infections [2]. In the controlled trial, plasma creatine kinase (CK), alanine aminotransferase (ALAT), and aspartate aminotransferase (ASAT) increased significantly at 4 weeks post-challenge, peaking at 8 weeks. By 12 weeks, ALAT had returned to pre-challenge levels, while CK and ASAT remained elevated, reflecting ongoing tissue damage [2]. Weekly mortality peaked at 4.1% in week 5, concurrent with an 89% reduction in appetite. In field trials, as SAV prevalence increased, plasma CK, ALAT, ASAT, and lactate dehydrogenase (LDH) levels peaked, along with the stress markers cortisol and tetrahydrocortisone in feces, despite low mortality (<0.2% weekly) [2]. These findings demonstrate that blood chemistry panels can serve as surrogate markers of infection and tissue damage, potentially enabling earlier intervention through stress reduction, timed harvest, or clinical nutrition.

Genetic and Genomic Approaches to Surveillance

The identification of quantitative trait loci (QTL) associated with PD resistance has opened new avenues for genetic surveillance and selective breeding. Manousi et al. confirmed a major QTL on chromosome Ssa03 and identified an additional QTL on Ssa07 linked to infection-specific responses [3]. Through integration of long-read sequencing, transcriptomics, and allele-specific expression analysis, three tandemly duplicated gig1-like genes were identified as likely causal candidates within the Ssa03 QTL region. These fish-specific antiviral effectors showed significant expression differences among resistant and susceptible haplotypes [3]. As

Vaccine Development and Immunization Strategies

The control of pancreas disease (PD) and sleeping disease (SD) in European salmonid aquaculture has necessitated the development of multiple vaccine platforms, each with distinct mechanisms of action, efficacy profiles, and practical limitations. The etiological agent, salmonid alphavirus (SAV), presents unique challenges for vaccinology due to its ability to establish persistent infections, its capacity for immune evasion through nonstructural protein-mediated suppression of the interferon response, and the complex immunological landscape of the teleost host. Over the past two decades, the aquaculture industry has progressed from reliance on oil-adjuvanted inactivated vaccines to more sophisticated DNA vaccine platforms and experimental live attenuated candidates, each representing a significant leap in our understanding of fish immunology and viral pathogenesis. The World Organisation for Animal Health (WOAH) recognizes SAV as a significant pathogen of finfish, and the development of effective immunization strategies is considered a cornerstone of sustainable disease management in regions where PD is enzootic.

Innate and Adaptive Immune Targets for Vaccine Design

Understanding the immunological correlates of protection against SAV is foundational to rational vaccine design. The teleost immune response to SAV involves both the rapid, type I interferon (IFN)-mediated innate response and the slower, yet critical, adaptive humoral and cellular arms. The mitochondrial antiviral signaling (MAVS) protein serves as a central adaptor in the RIG-I-like receptor (RLR) pathway, and disruption of MAVS function in vitro leads to a 1.5 log increase in SAV-3 titers, underscoring the importance of this pathway in controlling early replication [49]. The virus, in turn, has evolved countermeasures: nonstructural protein nsP2 effectively blocks the induction of type I IFN by acting downstream of IRF3, an inhibition that depends on the nuclear localization of nsP2 and its C-terminal domain, which is both sufficient and necessary for this suppressive effect [13]. Furthermore, SAV nonstructural protein Nsp2 activates the NF-κB signaling pathway through upregulation of TLR3, TLR7, and TLR8, leading to inflammatory cytokine production, yet this activation occurs concurrently with viral evasion of the IFN system, creating a complex and partially contradictory immune environment during infection [8].

The adaptive immune response, particularly the role of B cells and neutralizing antibodies, is central to vaccine-mediated protection. Intraperitoneal infection with SAV3 induces a robust and prolonged local B cell response in the peritoneal cavity (PerC), with the frequency of IgM+ B cells and total IgM-secreting cells (ASC) peaking between 3 to 6 weeks post-infection and correlating with anti-SAV E2 and neutralizing antibody titers in serum [54, 57]. The PerC appears to serve as a secondary immune site and an ASC survival niche, with responses lasting up to nine weeks post-infection, while systemic tissues like the head kidney and spleen show more modest or transient changes [56, 57]. Importantly, the response is transcriptionally distinct across immune compartments: RNA-seq analysis of IgM+IgD+ B cells from the head kidney, spleen, and PerC identified 334, 259, and 613 differentially expressed genes, respectively, with only 34 common to all three sites, indicating that B cell subsets acquire tissue-specific functional programs during SAV infection [56]. These findings have profound implications for vaccine administration routes, suggesting that intraperitoneal injection, the standard for many salmonid vaccines, may preferentially stimulate a local PerC response that is critical for long-term humoral memory.

Cellular immunity also plays a non-redundant role. In rainbow trout vaccinated with an oil-adjuvanted inactivated SAV vaccine, challenge resulted in significantly stronger and faster specific cytotoxicity compared to vaccination alone, with reduced viral titers and pathology, indicating that the vaccine primes cytotoxic T lymphocyte (CTL) responses that are only fully revealed upon subsequent exposure [58]. The induction of cell-mediated immunity is particularly important given that SAV can infect macrophages and dendritic-like cells, as demonstrated in TO-cells, where the virus downregulates several JAK-STAT pathway genes, including type I and II receptor genes, Jak2, Tyk2, Stat3, and Stat5, as a strategy to block the transcription of antiviral genes [37]. An effective vaccine must therefore overcome these viral immune evasion mechanisms to elicit both neutralizing antibodies and a functional CTL response.

Inactivated Whole Virus Vaccines

The first generation of commercially available PD vaccines consisted of oil-adjuvanted, inactivated whole virus preparations (Oil-PD). These vaccines have been widely used in Norwegian and Scottish aquaculture and have contributed to a measurable reduction in PD outbreaks. However, despite their widespread application, field outbreaks remain frequent, indicating suboptimal protection [1, 7]. In a recent comparative efficacy study, fish vaccinated with an Oil-PD vaccine and challenged with SAV2 by cohabitation showed moderate protection, but importantly, when Oil-PD-vaccinated, pre-challenged fish were cohabited with Oil-PD-vaccinated naïve fish, the latter became infected, demonstrating that the vaccine does not prevent onward transmission [4]. This is a critical limitation, as it implies that vaccinated populations can still amplify and shed virus, maintaining infection pressure within and between farms.

The mechanism of protection conferred by Oil-PD vaccines is likely dominated by humoral immunity. However, the inactivated virus, particularly when adjuvanted with mineral oil, induces a strong inflammatory response at the injection site, which can lead to intra-abdominal adhesions and potential growth suppression. Moreover, the antibody response to inactivated SAV is comparatively weak. In a study comparing inactivated SAV1 with live SAV3, the inactivated virus induced lower ASC responses across all immune sites tested, and specific serum antibodies were only induced by the live virus, not by the inactivated preparation [54]. A boost with inactivated virus did not increase these responses, suggesting that the antigenic conformation or the lack of intracellular replication may limit the immunogenicity of killed vaccines [54]. In rainbow trout, vaccination with an oil-adjuvanted inactivated vaccine did not cause a strong cytotoxic or humoral response by itself, and only after challenge did a robust CTL response and reduced viral loads become apparent [58]. These data suggest that inactivated vaccines may prime the immune system inadequately, relying on the subsequent natural infection to drive full adaptive maturation.

DNA Vaccines: A Paradigm Shift in PD Control

The development of DNA vaccines against SAV represents a significant advancement in the field. The most extensively characterized and commercially available construct, the DNA-PD vaccine (pSAV), encodes the structural polypeptide of SAV under the control of a eukaryotic promoter. The mechanism of action involves intramuscular injection, where the plasmid is taken up by myocytes and antigen-presenting cells, leading to endogenous expression of viral proteins within the context of MHC class I presentation, thereby stimulating both humoral and cell-mediated immunity. Transcriptome analyses of muscle at the injection site seven days post-vaccination with pSAV revealed a unique expression profile: compared to a control plasmid, pSAV caused higher upregulation of IFNγ and several IFNγ-inducible genes, as well as increased transcripts of marker genes for B cells, T cells, and antigen-presenting cells, along with the chemokine CXCL10 and the proinflammatory cytokines IL-1β and TNFα [59]. This profile is distinct from that induced by DNA vaccines against other salmonid viruses like infectious salmon anemia virus and infectious hematopoietic necrosis virus, suggesting a pSAV-specific immunostimulatory signature that favors lymphocyte attraction and activation [59].

The efficacy of the DNA-PD vaccine against SAV3 has been demonstrated in multiple controlled trials. In a comparative study, the DNA vaccine group had significantly higher SAV3 neutralizing antibody titers after the immunization period, significantly lower SAV3 viremia levels at 19 days post-challenge, significantly reduced transmission to naïve fish, higher weight gain post-challenge, and significantly reduced prevalence and severity of histopathological changes in target organs compared to the oil-adjuvanted vaccine group [15]. Crucially, in a transmission study, DNA-PD-vaccinated fish that had been pre-challenged with SAV2 did not transmit the infection to cohabiting DNA-PD-vaccinated naïve fish, whereas Oil-PD-vaccinated cohabitants residing with Oil-PD-vaccinated, pre-challenged fish showed infection levels similar to unvaccinated controls [4]. This indicates that the DNA vaccine may be capable of inducing herd immunity, a critical tool for controlling the PD epizootic at the population level [4]. The DNA vaccine also significantly reduced virus shedding into the water, as quantified by RT-droplet digital PCR, further supporting its potential to lower environmental viral load and reduce infection pressure on neighboring farms [4, 26].

However, the DNA vaccine is not without limitations. While it provides robust protection against clinical disease and growth loss, mortality levels are often low and not significantly different from control groups in experimental settings, making it difficult to demonstrate a survival benefit [15, 43]. Furthermore, the vaccine must be administered by intramuscular injection, which is labor-intensive and requires handling of individual fish. The requirement for a cold chain and the relatively high cost of plasmid production also present logistical and economic barriers. Despite these challenges, the DNA-PD vaccine has been shown to provide superior protection against the most economically damaging consequence of PD: growth impairment due to pancreatic destruction and subsequent malabsorption. By protecting the exocrine pancreas, the DNA vaccine prevents the severe feed conversion ratio increases and weight loss that characterize field infections [15, 21].

Live Attenuated Vaccines: Balancing Virulence and Immunogenicity

The pursuit of live attenuated SAV vaccines represents a high-risk, high-reward strategy. The theoretical advantage is that a replicating, attenuated virus would stimulate a broader and more durable immune response, more closely mimicking natural infection without causing severe pathology. Researchers have employed site-directed mutagenesis to create attenuated SAV3 clones with targeted mutations in the E2 glycoprotein (to disrupt viral attachment) and/or in the nuclear localization signal (NLS) of the capsid protein (to disrupt the viral suppression of cellular nucleocytoplasmic trafficking) [1, 11]. In a comprehensive evaluation, three attenuated clones were tested in Atlantic salmon. Clones with mutations in both capsid and E2 exhibited the most profound attenuation, characterized by rapid viral clearance, minimal shedding, and little transmission to cohabitants [1]. However, the most attenuated clone provided weaker protection against subsequent challenge, while a clone with only E2 mutations displayed greater residual virulence but induced stronger immunity, as evidenced by significantly reduced viral loads upon challenge [1]. This trade-off between attenuation and immunogenicity is a classic challenge in live vaccine design. In a pilot experiment, the mutated viruses replicated in fish and transmitted to naïve cohabitants without reversion to the wild-type sequence, a promising sign of genetic stability [11]. However, immunization with the single-mutant variants caused clinical signs and pathology consistent with PD prior to challenge, raising safety concerns. Only the double mutant (E2 + capsid) resulted in minimal clinical signs and higher weight gain, albeit with weaker protection [11].

A critical concern with live attenuated SAV vaccines is the potential for environmental dissemination. Given that SAV is shed in large quantities by infected fish, with maximal shedding rates of 2.4 × 10⁴ TCID₅₀ L⁻¹ h⁻¹ kg⁻¹ recorded in experimental infections [24], and can persist in seawater for weeks at lower temperatures [18], the release of genetically modified, replication-competent virus into the marine environment poses significant ecological and regulatory risks. The WOAH guidelines and European Food Safety Authority (EFSA) assessments [6] emphasize the need for stringent containment and risk evaluation for any live vaccine candidate. Despite these hurdles, the potential for a single-dose, long-lasting vaccine that induces sterilizing immunity makes this an active area of investigation. Future efforts must focus on optimizing the balance between attenuation, immunogenicity, and safety, potentially through the incorporation of additional "genetic safeguards" such as codon deoptimization or deletion of genes essential for replication in the host but not for immunogenicity.

Replicon Vectors and Heterologous Vaccine Platforms

Beyond direct SAV vaccines, the alphavirus replicon system has been exploited as a platform for heterologous antigen delivery. The salmonid alphavirus-based replicon vector (pSAV) is a self-replicating RNA or DNA-layered plasmid that retains the nonstructural protein genes necessary for RNA replication but lacks the structural genes, which are replaced by a heterologous antigen. This system harnesses the potent replication machinery of SAV to amplify the antigen-encoding RNA within the host cell, leading to high levels of antigen expression and strong immune stimulation. In an innovative application, a pSAV replicon encoding the glycoprotein of spring viraemia of carp virus (SVCV) was tested in common carp, a non-salmonid species. The DNA-layered SAV replicon resulted in 88% survival against SVCV challenge, compared to approximately 50% in groups receiving a conventional DNA vaccine or empty plasmid [55]. The replicon induced upregulation of innate immune genes at the injection site and increased IgM expression, demonstrating its potential as a cross-species vaccine platform [55].

Similarly, the pSAV replicon has been used to express the polyprotein of infectious pancreatic necrosis virus (IPNV). Single intramuscular injection of the replicon encoding the full IPNV polyprotein (pSAV/PP) conferred low to moderate protection against IPNV challenge in Atlantic salmon, while a construct encoding only pVP2 did not [60]. The modest protection observed may reflect the inherent competition between the replicon's self-amplification and the host's innate response, or suboptimal antigen processing and presentation. Nonetheless, the replicon system offers several advantages: it is non-infectious (lacking structural proteins), cannot spread to naïve fish, and can be administered as a DNA-layered plasmid, simplifying manufacturing and distribution [55, 60]. This platform holds promise for the development of multivalent vaccines that protect against multiple pathogens simultaneously, a highly desirable goal in an industry facing a complex disease landscape.

Timing of Vaccination and Host Factors Affecting Immunization Success

The developmental stage and physiological status of the host profoundly influence vaccine efficacy. Atlantic salmon post-smolts are more susceptible to SAV3 infection two weeks after seawater transfer compared to nine weeks after transfer, with the latter group developing stronger humoral and cell-mediated immune responses [25, 36]. Fish adapted for a longer period to seawater showed upregulation of CD40, MHCII, and IL-15, indicative of a robust cellular response, and were the only group in which neutralizing antibodies were detected (at 21 and 28 days post-infection) [36]. In contrast, fish infected shortly after transfer exhibited higher IL-10 expression and prolonged SOCS1 upregulation, suggesting a regulatory or immunosuppressive state that may impair viral clearance [36]. These findings have direct implications for vaccination schedules: immunization should ideally be performed after the smolt has fully acclimated to seawater, when the immune system is more competent to respond to the vaccine and mount a protective memory response.

The genetic background of the host also contributes significantly to vaccine and infection outcomes. A major quantitative trait locus (QTL) on chromosome Ssa03, encompassing three tandemly duplicated gig1-like genes, has been robustly associated with resistance to PD in multiple Atlantic salmon populations [3, 17, 28]. The gig1 gene encodes a fish-specific antiviral effector, and polymorphisms within this region are linked to viral load and survival. Markers associated with this QTL are being incorporated into selective breeding programs [28], providing a complementary strategy to vaccination. Fish with a high genetic breeding value for PD resistance may require less vaccine-induced protection to achieve the same level of clinical protection, or may respond more robustly to immunization. The integration of genetic selection with vaccination represents a powerful, synergistic approach to disease management.

Ploidy also influences susceptibility and vaccine responses. Triploid Atlantic salmon, which are sterile and thus prevent genetic introgression from farmed escapes into wild populations, accumulate SAV3 prevalence more slowly than diploids following bath challenge, although peak viral loads are similar [46, 48]. Triploid fry also showed lower median viral RNA copy numbers in heart and liver compared to diploids following SAV1 challenge, and exhibited significantly lower pancreatic and myocardial degeneration [45]. These differences may be related to alterations in cell metabolism or receptor expression in triploid cells. However, the implications for vaccination are not yet fully understood. If triploids mount a comparable or enhanced immune response, they may be equally or more effectively protected by existing vaccines, which would be a favorable outcome for the industry's shift towards sterile fish production.

Environmental and Logistical Considerations

The assessment of vaccine efficacy in the field is complicated by the high variability in challenge conditions, water temperature, co-infections, and management practices. The survival of SAV in seawater is temperature-dependent, with infectivity decaying more rapidly at higher temperatures; at 16°C, SAV shed by infected fish was completely inactivated within three weeks, whereas at 12°C, it persisted for four weeks [18]. These dynamics influence the environmental persistence of vaccine virus from live attenuated candidates and the risk of transmission post-vaccination. Additionally, co-infections are common; sea lice (Lepeophtheirus salmonis) infestation compromises the ability of peripheral blood mononuclear cells to control SAV replication ex vivo, and this is associated with an inability to upregulate IFIT5, IRF9, and Mx [33]. Vaccinated fish on farms with high lice burdens may therefore experience "vaccine breakthrough" not due to vaccine failure per se, but due to immune suppression by the parasite.

The potential for vaccine-induced herd immunity, as demonstrated for the DNA-PD vaccine [4], is a powerful tool for regional disease control. Mathematical models incorporating virus shedding rates, environmental decay, and vaccination coverage could guide the deployment of vaccines to achieve population-level protection. However, this requires high uptake and compliance across farming regions, which is challenging given the voluntary nature of vaccination in many jurisdictions. The EFSA assessment on SAV concluded that while the virus is eligible for Union intervention under the Animal Health Law, its categorization remains uncertain for categories B through E, reflecting the complexity of control at the pan-European level [6].

In conclusion, the field of SAV vaccinology has advanced from empirical oil-adjuvanted formulations to mechanistically designed DNA and replicon vaccines, with live attenuated candidates on the horizon. The DNA vaccine currently offers the best balance of efficacy, safety, and potential for herd immunity, while replicon vectors open the door to multivalent platforms. Future progress will depend on a deeper understanding of B cell biology in the PerC, the role of CD8+ T cells in viral clearance, and the genetic factors that govern host resistance. As the global demand for farmed salmon continues to rise, and as climate change alters the epidemiology of PD through shifts in temperature and host susceptibility, robust immunization strategies will remain an indispensable pillar of sustainable aquaculture.

Risk Assessment and Environmental Impact of Live Attenuated Vaccines

The deployment of live attenuated vaccines against Salmonid Alphavirus (SAV) presents a unique paradox in veterinary vaccinology: the very properties that confer robust, long-term protective immunity, namely, limited viral replication within the host, simultaneously generate the most significant environmental and biosafety concerns. Unlike inactivated or subunit vaccines, live attenuated SAV constructs retain the capacity for replication, shedding, and, critically, genetic modification that raises the specter of reversion to virulence, recombination with field strains, and unintended ecological spillover. A comprehensive risk assessment must therefore integrate molecular virology, epidemiology, environmental fate analysis, and regulatory frameworks to evaluate whether the theoretical benefits of live attenuated vaccines outweigh the potential for adverse outcomes in both farmed and wild salmonid populations.

The Paradox of Attenuation: Residual Virulence and Immunological Trade-Offs

The foundational work by Braaen et al. [1] and earlier studies by Aksnes et al. [11] established that genetically modified live attenuated SAV3 clones, engineered through site-directed mutagenesis targeting the E2 glycoprotein attachment domain and/or the capsid protein nuclear localization signal (NLS), can provide significant protection against subsequent wild-type challenge. However, a critical finding that underpins environmental risk assessment is the persistent observation of residual virulence. Even the most attenuated clones, those mutated in both capsid and E2, caused “significant reductions in weight gain” compared to uninfected controls [1]. This growth impairment, while less severe than wild-type infection, indicates that the attenuated virus still induces a measurable pathophysiological burden. The clone with only E2 mutations displayed “greater residual virulence” but also “provided stronger protection” [1], illustrating an inherent trade-off where immunogenicity is directly coupled to the degree of in vivo replication and, by extension, the potential for environmental shedding.

The implications for environmental risk are profound. If an attenuated vaccine retains the capacity to cause subclinical pathology, it is axiomatic that it retains the capacity to replicate to levels sufficient for horizontal transmission. Indeed, Braaen et al. demonstrated that all three attenuated clones transmitted to naïve cohabitant fish, albeit with reduced efficiency compared to wild-type [1]. The capsid+E2 mutant showed “little transmission,” but transmission was not entirely eliminated [1]. From a risk assessment perspective, even low-level transmission represents a non-zero probability of establishing a modified viral population in the environment, particularly under the high-density conditions characteristic of commercial aquaculture.

Reversion to Virulence and Genetic Stability: The Quasispecies Challenge

The single most critical risk associated with any live attenuated RNA virus vaccine is the potential for reversion to a virulent phenotype through the accumulation of compensatory mutations. SAV, as a positive-sense single-stranded RNA virus, exhibits the high mutation rates typical of RNA-dependent RNA polymerases, which lack proofreading activity [5, 19]. This genetic plasticity was vividly demonstrated by Roh et al. [5] using Nanopore sequencing to track SAV3 genetic variation during experimental infections in Atlantic salmon and brown trout. They identified a “multitude of single nucleotide variants (SNVs) and deletions” widespread across the SAV3 genome, with specific variants increasing in frequency over the course of infection, particularly in brown trout [5]. The observation that some minor SNVs “only showed an increase in frequency in brown trout” [5] highlights how host species context can shape viral evolution, a factor that is critically relevant when considering vaccine virus introduction into ecosystems containing multiple susceptible salmonid species.

The engineered mutations in live attenuated candidates, such as the N-glycosylation site knockout in E2 or the NLS disruption in capsid, are designed to be attenuating, but they are not necessarily stable. Gallagher et al. [19] revealed “extensive ‘hidden’ SAV diversity” in both farmed and wild fish populations, including mixed subtype infections within individual fish and evidence of recombination. The presence of co-infecting SAV subtypes in wild flatfish (dab) demonstrates that the ecological conditions for genetic exchange exist [19]. If a live attenuated vaccine virus were to co-infect a fish concurrently harboring a field strain of SAV, the potential for recombination between the engineered genome and the wild-type genome becomes a tangible risk. Macqueen et al. [12] documented co-infections of SAV2a and SAV3 in single fish in Rogaland, Norway, confirming that mixed infections are not merely theoretical but occur under natural farming conditions. The introduction of a genetically modified replicating virus into this landscape could accelerate the emergence of novel chimeric viruses with unpredictable virulence profiles.

Shedding Dynamics and Environmental Contamination

The quantification of viral shedding from vaccinated fish is a cornerstone of environmental risk assessment. Thorarinsson et al. [4] demonstrated in transmission studies that DNA-vaccinated fish (which do not shed live virus) effectively prevented infection of cohabiting DNA-vaccinated naïve fish, indicating that herd immunity can be achieved without environmental release. In contrast, the live attenuated clones described by Braaen et al. [1] were detected in cohabitant fish, confirming that shedding occurs. The magnitude of shedding is dose-dependent, as Jarungsriapisit et al. [24] established that SAV3 shedding rates can reach 2.4 × 10⁴ TCID₅₀ L⁻¹ of seawater h⁻¹ kg⁻¹ during peak infection. Even at reduced levels, the continuous shedding of a modified virus from thousands of vaccinated fish in a marine net-pen represents a substantial environmental introduction event.

The environmental persistence of shed virus is temperature-dependent, as demonstrated by Jarungsriapisit et al. [18], who showed that SAV3 stability decreases with increasing temperature. At 12°C, SAV3 shed by infected fish was completely inactivated after four weeks, while at 16°C inactivation occurred by three weeks [18]. However, Atlantic salmon production occurs across a range of temperatures, and in colder waters (4–8°C), viral persistence would be expected to extend considerably. Critically, the in vivo infectious dose was determined to be as low as 448 TCID₅₀ L⁻¹ of seawater [18], while cell culture infection required only 24 TCID₅₀ L⁻¹ [18]. This low infectious dose means that even minimal shedding from vaccinated fish could establish infection in susceptible wild fish populations if they pass through a contaminated water mass. Bernhardt et al. [40, 42] developed sensitive water filtration and RT-qPCR methods capable of detecting SAV in seawater at concentrations as low as 2.0 × 10² copies L⁻¹ [52], confirming that environmental surveillance is technically feasible but also highlighting that low-level contamination may be widespread.

Risk to Non-Target Species and Ecological Spillover

The host range of SAV extends beyond farmed Atlantic salmon and rainbow trout. The virus has been detected in a diverse array of species, including ballan wrasse (SAV6) [14, 22], Arctic char [47], brown trout [5], and even wild flatfish such as dab [19]. Tighe et al. [14] proposed the designation of a new genotype (SAV7) isolated from ballan wrasse, noting that “SAV has been found to infect non-salmonid fish” and suggested amending the species name to the more inclusive “piscine alphavirus.” The isolation of SAV6 from wild-caught ballan wrasse, which showed “no signs of disease” [22], raises the possibility that non-salmonid species may serve as asymptomatic reservoirs capable of sustaining viral transmission cycles independently of farmed fish.

The environmental release of a live attenuated SAV vaccine therefore carries the risk of establishing a modified virus in wild reservoir species. If the attenuated virus replicates poorly in salmonids but maintains efficient replication in wrasse or flatfish, it could establish a cryptic enzootic cycle. Subsequent recombination with wild-type SAV in these reservoir species could generate novel variants that retain the attenuating mutations from the vaccine while acquiring compensatory mutations from field strains. The work of Madhun et al. [39] suggests that SAV infection in wild migrating salmon post-smolts is relatively uncommon, but the same study detected piscine orthoreovirus (PRV) in 4.6% of wild post-smolts, indicating that viral spillover from farms does occur. The introduction of a replicating genetically modified virus into this dynamic could fundamentally alter the epizootiology of SAV in wild populations.

Regulatory Frameworks and Categorization

The European Food Safety Authority (EFSA) Panel on Animal Health and Welfare conducted a comprehensive assessment of SAV infection under the Animal Health Law (Regulation (EU) 2016/429) and concluded that “infection with salmonid alphavirus does not meet the criteria in Section 1 (Category A)” with only 5–10% probability of meeting criteria for the highest risk category [6]. However, the panel expressed uncertainty (50–90% probability) regarding categorization in Sections 2–5 (Categories B through E) [6]. This regulatory ambiguity is critical because it influences the stringency of risk mitigation measures required for vaccine deployment. A Category A pathogen would mandate eradication; a Category B or C pathogen allows for control measures but does not require stamping out. The introduction of a live attenuated vaccine, which by definition involves the intentional release of a replicating, genetically modified organism, into a production system where the disease agent is not classified for eradication creates a regulatory gap. The World Organisation for Animal Health (WOAH) lists SAV as a notifiable pathogen, and its terrestrial and aquatic animal health codes emphasize the importance of safe vaccine development, yet specific guidelines for the environmental risk assessment of live attenuated aquatic RNA virus vaccines remain underdeveloped compared to those for terrestrial livestock.

Comparative Risk: Live Attenuated versus Alternative Vaccine Platforms

The risk-benefit calculus must consider that alternative vaccine platforms exist. Thorarinsson et al. [4, 15] and Skjold et al. [26] demonstrated that DNA vaccines against SAV not only protect vaccinated fish but also reduce shedding to levels that can achieve herd immunity, breaking the transmission cycle. The DNA vaccine platform does not involve live virus replication, thereby eliminating the risk of reversion, recombination, and environmental persistence. Similarly, inactivated whole-virus vaccines provide protection without replication [43, 58], although they typically require oil adjuvants that can cause injection-site reactions and may induce weaker cellular immune responses [58]. The live attenuated platform, while potentially offering superior immunity and lower production costs, introduces risks that are absent from these alternative platforms. The decision to deploy a live attenuated SAV vaccine must therefore be justified by demonstrated inability to control PD through existing means, yet Thorarinsson et al. [4] showed that the DNA vaccine “may curb the spread of SAV infection” and that “herd immunity may be achieved,” suggesting that non-replicating alternatives are viable.

Environmental Monitoring and Containment Strategies

If live attenuated SAV vaccines were to be deployed, robust environmental monitoring would be essential. The water filtration and RT-qPCR methods developed by Bernhardt et al. [40, 42] and Weli et al. [52] provide a validated framework for detecting SAV in seawater, with the ability to detect virus earlier than traditional fish sampling [42]. These methods could be adapted to specifically differentiate vaccine-derived virus from wild-type strains through the use of mutation-specific probes. However, the genetic diversity observed within SAV populations [5, 19] complicates this approach; a vaccine-specific probe might fail to detect a recombined virus that has lost the engineered marker mutation while retaining the attenuating phenotype, or, worse, retained the marker while reverting to virulence elsewhere in the genome.

Containment in marine net-pen systems is inherently challenging. Unlike terrestrial livestock facilities where waste can be treated, marine farms release effluent directly into the surrounding water column. The World Health Organization’s (WHO) guidelines for the environmental risk assessment of genetically modified organisms (GMOs) emphasize the need for physical containment, biological containment (e.g., attenuation), and surveillance. In the marine aquaculture environment, physical containment is impossible. Biological containment therefore bears the entire burden of risk mitigation. The data from Braaen et al. [1] suggest that the double mutant (capsid+E2) provides the highest level of biological containment but at the cost of reduced immunogenicity, necessitating a higher vaccine dose or repeated administration, which in turn increases production costs and handling stress.

Co-Infection Dynamics and Immune Modulation

The environmental risk of live attenuated SAV vaccines may be modulated by concurrent infections, which are ubiquitous in aquaculture. Gamil et al. [33] demonstrated that sea lice (Lepeophtheirus salmonis) infestation impairs the ability of peripheral blood mononuclear cells to control SAV replication, resulting in “significantly higher virus replication” and reduced upregulation of antiviral genes. If a vaccinated fish is infested with sea lice, a common scenario, the attenuated virus could replicate to higher titers than expected, increasing shedding and the probability of transmission. Similarly, Reid et al. [32] and Brown et al. [31] documented that SAV infection causes skin and gill microbiome dysbiosis, potentially predisposing fish to secondary bacterial infections. A live attenuated vaccine that replicates in the skin or gills could exacerbate this dysbiosis, creating a nidus for bacterial pathogens and increasing overall disease burden in the net-pen.

Long-Term Ecological Consequences and the Precautionary Principle

The precautionary principle, as articulated by the Food and Agriculture Organization (FAO) and WOAH in their guidelines for responsible vaccine use, dictates that when the potential for serious or irreversible harm exists, lack of full scientific certainty should not be used as a reason to postpone cost-effective measures to prevent environmental degradation. Given that SAV is not zoonotic, there is no evidence that any alphavirus transmitted between fish can infect mammals, the risk to human health is negligible. However, the ecological consequences for salmonid

References

[1] Braaen S, Wessel Ø, Bjørgen H, Rimstad E. Long-Term Protection in Atlantic Salmon (Salmo salar) to Pancreas Disease (PD) Can Be Achieved Through Immunization with Genetically Modified, Live Attenuated Salmonid Alphavirus 3. Vaccines. 2025. DOI: https://doi.org/10.3390/vaccines13020190

[2] Hoel E, Keitel-Gröner F, Husebø C, Knudsen D, Mullins J, Wallace C, et al.. Use of Biomarkers in Plasma and Faeces to Evaluate Physiological Status of Atlantic Salmon (Salmo salar L.) During Salmonid Alphavirus Infection.. Journal of Fish Diseases. 2025. DOI: https://doi.org/10.1111/jfd.14114

[3] Manousi D, Jaskula DM, Grammes F, Knutsen TM, Naseer S, Martin SAM, et al.. Dissecting the genetic basis of response to salmonid alphavirus in Atlantic salmon. BMC Genomics. 2025. DOI: https://doi.org/10.1186/s12864-025-11735-2

[4] Thorarinsson R, Ramstad A, Wolf JC, Sindre H, Skjerve E, Rimstad E, et al.. Effect of pancreas disease vaccines on infection levels and virus transmission in Atlantic salmon (Salmo salar) challenged with salmonid alphavirus, genotype 2. Frontiers in Immunology. 2024. DOI: https://doi.org/10.3389/fimmu.2024.1342816

[5] Roh H, Skaftnesmo K, Kannimuthu D, Madhun A, Patel S, Kvamme BO, et al.. Nanopore sequencing provides snapshots of the genetic variation within salmonid alphavirus-3 (SAV3) during an ongoing infection in Atlantic salmon (Salmo salar) and brown trout (Salmo trutta). Veterinary Research. 2024. DOI: https://doi.org/10.1186/s13567-024-01349-z

[6] Nielsen SS, Álvarez J, Calistri P, Canali E, Drewe J, Garin‐Bastuji B, et al.. Assessment of listing and categorisation of animal diseases within the framework of the Animal Health Law (Regulation (EU)2016/429): Infection with salmonid alphavirus (SAV). EFSA journal. European Food Safety Authority. 2023. DOI: https://doi.org/10.2903/j.efsa.2023.8327

[7] . Infection with salmonid alphavirus. CABI Compendium. 2022. DOI: https://doi.org/10.1079/cabicompendium.70655

[8] Gao S, Liu X, Han B, Wang N, Lv X, Guan X, et al.. Salmonid alphavirus non-structural protein 2 is a key protein that activates the NF-κB signaling pathway to mediate inflammatory responses.. Fish and Shellfish Immunology. 2022. DOI: https://doi.org/10.1016/j.fsi.2022.08.059

[9] Biacchesi S, Fauvet C, Mérour E, Bernard J, Lamoureux A, Lallias D, et al.. In silico reconstruction of a salmonid alphavirus virion reveals distinctive molecular features implicated in virulence. iScience. 2026. DOI: https://doi.org/10.1016/j.isci.2026.115070

[10] Hjortaas M, Fringuelli E, Monjane A, Mikalsen A, Jonassen C, Savage P, et al.. Emergence of Salmonid Alphavirus Genotype 2 in Norway, Molecular Characterization of Viral Strains Circulating in Norway and Scotland. Viruses. 2021. DOI: https://doi.org/10.3390/v13081556

[11] Aksnes I, Braaen S, Markussen T, Åkesson CP, Villoing S, Rimstad E. Genetically modified attenuated salmonid alphavirus: A potential strategy for immunization of Atlantic salmon.. Journal of Fish Diseases. 2021. DOI: https://doi.org/10.1111/jfd.13352

[12] Macqueen D, Eve O, Gundappa MK, Daniels RR, Gallagher M, Alexandersen S, et al.. Genomic Epidemiology of Salmonid Alphavirus in Norwegian Aquaculture Reveals Recent Subtype-2 Transmission Dynamics and Novel Subtype-3 Lineages. Viruses. 2021. DOI: https://doi.org/10.3390/v13122549

[13] Jami R, Mérour E, Bernard J, Lamoureux A, Millet J, Biacchesi S. The C-Terminal Domain of Salmonid Alphavirus Nonstructural Protein 2 (nsP2) Is Essential and Sufficient To Block RIG-I Pathway Induction and Interferon-Mediated Antiviral Response. Journal of Virology. 2021. DOI: https://doi.org/10.1128/JVI.01155-21

[14] Tighe A, Gallagher M, Carlsson J, Matějusová I, Swords F, Macqueen D, et al.. Nanopore whole genome sequencing and partitioned phylogenetic analysis supports a new salmonid alphavirus genotype (SAV7).. Diseases of Aquatic Organisms. 2020. DOI: https://doi.org/10.3354/dao03546

[15] Thorarinsson R, Wolf J, Inami M, Phillips L, Jones G, Macdonald A, et al.. Effect of a novel DNA vaccine against pancreas disease caused by salmonid alphavirus subtype 3 in Atlantic salmon (Salmo salar).. Fish and Shellfish Immunology. 2020. DOI: https://doi.org/10.1016/j.fsi.2020.12.002

[16] Aksnes I, Markussen T, Braaen S, Rimstad E. Mutation of N-glycosylation Sites in Salmonid Alphavirus (SAV) Envelope Proteins Attenuate the Virus in Cell Culture. Viruses. 2020. DOI: https://doi.org/10.3390/v12101071

[17] Aslam ML, Robledo D, Krasnov A, Moghadam H, Hillestad B, Houston RD, et al.. Quantitative trait loci and genes associated with salmonid alphavirus load in Atlantic salmon: implications for pancreas disease resistance and tolerance. Scientific Reports. 2020. DOI: https://doi.org/10.1038/s41598-020-67405-8

[18] Jarungsriapisit J, Nuñez-Ortiz N, Nordbø J, Moore L, Mæhle S, Patel S. The effect of temperature on the survival of salmonid alphavirus analysed using in vitro and in vivo methods. Aquaculture. 2020. DOI: https://doi.org/10.1016/j.aquaculture.2019.734647

[19] Gallagher M, Matějusová I, Ruane N, Macqueen D. Genome-wide target enriched viral sequencing reveals extensive ‘hidden’ salmonid alphavirus diversity in farmed and wild fish populations. Aquaculture. 2020. DOI: https://doi.org/10.1016/j.aquaculture.2020.735117

[20] Deperasińska I, Schulz P, Siwicki A. Salmonid Alphavirus (SAV). Journal of Veterinary Research. 2018. DOI: https://doi.org/10.2478/jvetres-2018-0001

[21] Røsæg M, Garseth Å, Brynildsrud O, Jansen MD. Pancreas disease caused by Salmonid alphavirus subtype 2 reduces growth and feed conversion in farmed Atlantic salmon.. Preventive Veterinary Medicine. 2019. DOI: https://doi.org/10.1016/J.PREVETMED.2019.104699

[22] Ruane N, Swords D, Morrissey T, Geary M, Hickey C, Collins E, et al.. Isolation of salmonid alphavirus subtype 6 from wild-caught ballan wrasse, Labrus bergylta (Ascanius).. Journal of Fish Diseases. 2018. DOI: https://doi.org/10.1111/jfd.12870

[23] Soares SMC, Elwenn SA, Campbell M, White P, Still N, Munro E. Salmonid alphavirus subtype I isolated from clinically-diseased Atlantic salmon, Salmo salar, in freshwater culture. Aquaculture. 2019. DOI: https://doi.org/10.1016/J.AQUACULTURE.2019.06.006

[24] Jarungsriapisit J, Moore L, Mæhle S, Skår C, Einen AC, Fiksdal I, et al.. Relationship between viral dose and outcome of infection in Atlantic salmon, Salmo salar L., post-smolts bath-challenged with salmonid alphavirus subtype 3. Veterinary Research. 2016. DOI: https://doi.org/10.1186/s13567-016-0385-2

[25] Jarungsriapisit J, Moore L, Taranger G, Nilsen T, Morton HC, Fiksdal I, et al.. Atlantic salmon (Salmo salar L.) post-smolts challenged two or nine weeks after seawater-transfer show differences in their susceptibility to salmonid alphavirus subtype 3 (SAV3). Virology Journal. 2016. DOI: https://doi.org/10.1186/s12985-016-0520-8

[26] Skjold P, Sommerset I, Frost P, Villoing S. Vaccination against pancreas disease in Atlantic salmon, Salmo salar L., reduces shedding of salmonid alphavirus. Veterinary Research. 2016. DOI: https://doi.org/10.1186/s13567-016-0362-9

[27] Biacchesi S, Jouvion G, Mérour E, Boukadiri A, Desdouits M, Ozden S, et al.. Rainbow trout (Oncorhynchus mykiss) muscle satellite cells are targets of salmonid alphavirus infection. Veterinary Research. 2016. DOI: https://doi.org/10.1186/s13567-015-0301-1

[28] Gonen S, Baranski M, Thorland I, Norris A, Grove H, Arnesen P, et al.. Mapping and validation of a major QTL affecting resistance to pancreas disease (salmonid alphavirus) in Atlantic salmon (Salmo salar). Heredity. 2015. DOI: https://doi.org/10.1038/hdy.2015.37

[29] Herath T, Ferguson H, Weidmann M, Bron J, Thompson K, Adams A, et al.. Pathogenesis of experimental salmonid alphavirus infection in vivo: an ultrastructural insight. Veterinary Research. 2016. DOI: https://doi.org/10.1186/s13567-015-0300-2

[30] Tartor H, Bernhardt L, Mohammad S, Kuiper RV, Weli S. In Situ Detection of Salmonid Alphavirus 3 (SAV3) in Tissues of Atlantic Salmon in a Cohabitation Challenge Model with a Special Focus on the Immune Response to the Virus in the Pseudobranch. Viruses. 2023. DOI: https://doi.org/10.3390/v15122450

[31] Brown RM, Moore L, Mani A, Patel S, Salinas I. Effects of ploidy and salmonid alphavirus infection on the skin and gill microbiome of Atlantic salmon (Salmo salar). bioRxiv. 2020. DOI: https://doi.org/10.1371/journal.pone.0243684

[32] Reid KM, Patel S, Robinson AJ, Bu L, Jarungsriapisit J, Moore L, et al.. Salmonid alphavirus infection causes skin dysbiosis in Atlantic salmon (Salmo salar L.) post-smolts. PLoS ONE. 2017. DOI: https://doi.org/10.1371/journal.pone.0172856

[33] Gamil A, Gadan K, Gislefoss E, Evensen Ø. Sea Lice (Lepeophtheirus salmonis) Infestation Reduces the Ability of Peripheral Blood Monocytic Cells (PBMCs) to Respond to and Control Replication of Salmonid Alphavirus in Atlantic Salmon (Salmo salar L.). Viruses. 2020. DOI: https://doi.org/10.3390/v12121450

[34] Andreassen R, Woldemariam NT, Egeland IØ, Agafonov O, Sindre H, Høyheim B. Identification of differentially expressed Atlantic salmon miRNAs responding to salmonid alphavirus (SAV) infection. BMC Genomics. 2017. DOI: https://doi.org/10.1186/s12864-017-3741-3

[35] Schachner O, Dinhopl N, Friedl A, Soliman H, El-Matbouli M. Advanced vacuolation indicates propagation of various salmonid alphavirus type 2 isolates in Acholeplasma-infected BF-2 cells.. Diseases of Aquatic Organisms. 2020. DOI: https://doi.org/10.3354/dao03481

[36] Nuñez-Ortiz N, Moore L, Jarungsriapisit J, Nilsen T, Stefansson S, Morton HC, et al.. Atlantic salmon post‐smolts adapted for a longer time to seawater develop an effective humoral and cellular immune response against Salmonid alphavirus. Fish and Shellfish Immunology. 2018. DOI: https://doi.org/10.1016/j.fsi.2018.08.059

[37] Xu C, Evensen Ø, Munang'andu H. A de novo transcriptome analysis shows that modulation of the JAK-STAT signaling pathway by salmonid alphavirus subtype 3 favors virus replication in macrophage/dendritic-like TO-cells. BMC Genomics. 2016. DOI: https://doi.org/10.1186/s12864-016-2739-6

[38] Braceland M, Tinsley J, Cockerill D, Bickerdike R, McLoughlin M, Eckersall P. Selective precipitation reaction: a novel diagnostic test for tissue pathology in Atlantic salmon, Salmo salar, infected with salmonid alphavirus (SAV3). Journal of Fish Diseases. 2016. DOI: https://doi.org/10.1111/jfd.12587

[39] Madhun A, Nilsen R, Barlaup B, Karlsen Ø, Karlsbakk E. Occurrence of salmonid alphavirus and piscine orthoreovirus-1 infections in migrating salmon (Salmo salar L.) post-smolt in western Norway.. Journal of Fish Diseases. 2023. DOI: https://doi.org/10.1111/jfd.13874

[40] Bernhardt L, Myrmel M, Lillehaug A, Qviller L, Weli SC. Filtration, concentration and detection of salmonid alphavirus in seawater during a post-smolt salmon (Salmo salar) cohabitant challenge.. Diseases of Aquatic Organisms. 2021. DOI: https://doi.org/10.3354/dao03572

[41] . salmonid alphavirus. CABI Compendium. 2022. DOI: https://doi.org/10.1079/cabicompendium.96430

[42] Bernhardt L, Lillehaug A, Qviller L, Weli S, Grønneberg E, Nilsen H, et al.. Early detection of salmonid alphavirus in seawater from marine farm sites of Atlantic salmon Salmo salar.. Diseases of Aquatic Organisms. 2021. DOI: https://doi.org/10.3354/dao03618

[43] Thorarinsson R, Wolf J, Inami M, Sindre H, Skjerve E, Evensen Ø, et al.. Effects of a DNA and multivalent oil-adjuvanted vaccines against pancreas disease in Atlantic salmon (Salmo salar) challenged with salmonid alphavirus subtype 3. Fish and Shellfish Immunology Reports. 2022. DOI: https://doi.org/10.1016/j.fsirep.2022.100063

[44] Madhun A, Isachsen CH, Omdal LM, Einen AC, Maehle S, Wennevik V, et al.. Prevalence of piscine orthoreovirus and salmonid alphavirus in sea-caught returning adult Atlantic salmon (Salmo salar L.) in northern Norway.. Journal of Fish Diseases. 2018. DOI: https://doi.org/10.1111/jfd.12785

[45] Herath T, Ashby A, Jayasuriya N, Bron J, Taylor J, Adams A, et al.. Impact of Salmonid alphavirus infection in diploid and triploid Atlantic salmon (Salmo salar L.) fry. PLoS ONE. 2017. DOI: https://doi.org/10.1371/journal.pone.0179192

[46] Moore L, Nilsen T, Jarungsriapisit J, Fjelldal P, Stefansson S, Taranger G, et al.. Triploid atlantic salmon (Salmo salar L.) post-smolts accumulate prevalence more slowly than diploid salmon following bath challenge with salmonid alphavirus subtype 3. PLoS ONE. 2017. DOI: https://doi.org/10.1371/journal.pone.0175468

[47] Lewisch E, Frank T, Soliman H, Schachner O, Friedl A, El-Matbouli M. First confirmation of salmonid alphavirus infection in Arctic char Salvelinus alpinus and in Austria.. Diseases of Aquatic Organisms. 2018. DOI: https://doi.org/10.3354/dao03265

[48] Moore L, Nilsen T, Jarungsriapisit J, Fjelldal P, Stefansson S, Taranger G, et al.. Correction: Triploid atlantic salmon (Salmo salar L.) post-smolts accumulate prevalence more slowly than diploid salmon following bath challenge with salmonid alphavirus subtype 3. PLoS ONE. 2017. DOI: https://doi.org/10.1371/journal.pone.0177250

[49] Xu C, Gamil A, Wang X, Munang'andu H, Evensen Ø. MAVS disruption impairs downstream signaling and results in higher virus replication levels of salmonid alphavirus subtype 3 but not infectious pancreatic necrosis virus in vitro. Frontiers in Immunology. 2024. DOI: https://doi.org/10.3389/fimmu.2024.1401086

[50] Teige LH, Aksnes I, Røsæg M, Jensen I, Jørgensen J, Sindre H, et al.. Detection of specific Atlantic salmon antibodies against salmonid alphavirus using a bead-based immunoassay.. Fish and Shellfish Immunology. 2020. DOI: https://doi.org/10.1016/j.fsi.2020.07.055

[51] Jansen MD, Guarracino M, Carson M, Modahl I, Taksdal T, Sindre H, et al.. Field Evaluation of Diagnostic Test Sensitivity and Specificity for Salmonid Alphavirus (SAV) Infection and Pancreas Disease (PD) in Farmed Atlantic salmon (Salmo salar L.) in Norway Using Bayesian Latent Class Analysis. Frontiers in Veterinary Science. 2019. DOI: https://doi.org/10.3389/fvets.2019.00419

[52] Weli S, Bernhardt L, Qviller L, Myrmel M, Lillehaug A, Kjellin ML. Development and evaluation of a method for concentration and detection of salmonid alphavirus from seawater.. Journal of Virological Methods. 2020. DOI: https://doi.org/10.1016/j.jviromet.2020.113990

[53] Shi W, Wang Y, Ren X, Gao S, Hua X, Guo M, et al.. EvaGreen-based real-time PCR assay for sensitive detection of salmonid alphavirus.. Molecular and Cellular Probes. 2018. DOI: https://doi.org/10.1016/j.mcp.2018.03.002

[54] Jenberie S, Nordli H, Strandskog G, Greiner-Tollersrud L, Peñaranda MMD, Jørgensen J, et al.. Virus-specific antibody secreting cells reside in the peritoneal cavity and systemic immune sites of Atlantic salmon (Salmo salar L.) challenged intraperitoneally with salmonid alphavirus.. Developmental and Comparative Immunology. 2024. DOI: https://doi.org/10.2139/ssrn.4757648

[55] Abonyi F, Eszterbauer E, Baska F, Hardy T, Doszpoly A. First Experimental Application of DNA-Layered Salmonid Alphavirus-Based Replicon Vaccine in Non-Salmonid Fish: Induced Early Semi-Specific Protection against Spring Viraemia of Carp Virus (SVCV) in Common Carp (Cyprinus carpio). Animals. 2024. DOI: https://doi.org/10.3390/ani14182698

[56] Jenberie S, Sandve S, To T, Kent M, Rimstad E, Jørgensen J, et al.. Transcriptionally distinct B cell profiles in systemic immune tissues and peritoneal cavity of Atlantic salmon (Salmo salar) infected with salmonid alphavirus subtype 3. Frontiers in Immunology. 2024. DOI: https://doi.org/10.3389/fimmu.2024.1504836

[57] Jenberie S, Peñaranda M, Thim H, Styrvold MB, Strandskog G, Jørgensen J, et al.. Salmonid Alphavirus Subtype 3 Induces Prolonged Local B Cell Responses in Atlantic Salmon (Salmo salar) After Intraperitoneal Infection. Frontiers in Immunology. 2020. DOI: https://doi.org/10.3389/fimmu.2020.01682

[58] Veenstra KA, Hodneland K, Fischer S, Takehana K, Belmonte R, Fischer U. Cellular Immune Responses in Rainbow Trout (Onchorhynchus mykiss) Following Vaccination and Challenge Against Salmonid Alphavirus (SAV). Vaccines. 2020. DOI: https://doi.org/10.3390/vaccines8040725

[59] Sobhkhez M, Krasnov A, Robertsen B. Transcriptome analyses of Atlantic salmon muscle genes induced by a DNA vaccine against salmonid alphavirus, the causative agent of salmon pancreas disease (PD). PLoS ONE. 2018. DOI: https://doi.org/10.1371/journal.pone.0204924

[60] Abdullah A, Olsen CM, Hodneland K, Rimstad E. A Polyprotein-Expressing Salmonid Alphavirus Replicon Induces Modest Protection in Atlantic Salmon (Salmo Salar) Against Infectious Pancreatic Necrosis. Viruses. 2015. DOI: https://doi.org/10.3390/v7010252