Immunological responses in Atlantic salmon (Salmo salar) to
co-infections with sea lice (Lepeophtheirus salmonis)
A Thesis
Submitted to the Graduate Faculty
in Partial Fulfillment of the Requirements
for the Degree of
MASTER OF SCIENCE
Department of Pathology and Microbiology
Faculty of Veterinary Medicine
University of Prince Edward Island
Laura A. Carvalho
Charlottetown, PE
September 2019
© 2019. L. A. CARVALHO
i Abstract
Intensive fish farming at aquaculture sites is reaching maximum exploitation of the environment, providing suitable conditions for simultaneous infections to occur in wild and farm systems. A major parasitic contributor that is negatively impacting the health of Atlantic salmon (Salmo salar) is the salmon louse, Lepeophtheirus salmonis. This thesis aimed to determine the pathogenic effects of several co-infection models with a concurrent L. salmonis infection by evaluating responses of host immune gene transcription identified by RT-qPCR. The first study investigated winter ulcer disease; a cold-water disease of salmonids caused by Moritella viscosa.
When co-infected with L. salmonis at high lice loads, mortalities were significantly higher and skin lesions were more severe than in a single infection with M. viscosa or co-infections with lower lice loads. Evidence of osmoregulatory stress from increased potassium levels and significant up-regulation in genes associated with wound healing and humoral responses in skin was also prevalent in co-infected fish with high lice loads, suggesting that co-infection may have a significant impact on the ability of lesions to resolve, leading to increased mortality. The second study investigated functional feeds, which are an alternative anti-louse management strategy. Atlantic salmon were co-infected with L. salmonis and a high virulent isolate of infectious salmon anemia virus (ISAv). Fish were administered one of four functional diets with varying EPA/DHA levels; a control feed with a low EPA/DHA mix of 0.3% (Ctrl), an
EPA/DHA enriched diet (1.0%) with pro-inflammatory supplementation (FA+I), an EPA/DHA enriched diet (1.0%) with anti-inflammatory supplementation (FA-I), and a low EPA/DHA feed
(0.3%) with an immunostimulant added (Ctrl + IS). Fish fed fatty acid enriched (FA+I, FA-I) feeds had significantly lower lice abundance during single infection and lower survival during co-infections, suggesting the improved anti-parasitic responses may not be effective in a co-
ii infection model. Reverse transcription qPCR was used to assess immune gene transcription modulation in the head kidney between single and co-infected fish fed functional diets. Pro- inflammatory and antiviral genes associated with the interferon system were modulated following co-infection with ISAv in all functional diets but the immunostimulated diet, while several genes accounted for high lice and high viral loads in diets FA+I and Ctrl + IS. The results generated from this thesis contribute to the growing body of knowledge of understanding fish immune responses during co-infections and developing alternative strategies, such as functional diets, for managing pathogenic outbreaks at farm sites.
iii Acknowledgements
I would first like to thank my thesis advisor, Dr. Mark Fast, for his continuous support of my research. Thank you for the opportunity to work alongside you, for your patience, enthusiasm, knowledge and guidance.
I would also like to thank my committee members, Drs. Dave Speare, Mike Van Den Heuvel, Sussie Dalvin and Fred Kibenge for their encouragement, insightful comments, and thought- provoking questions.
Thank you to Drs. Sara Purcell, Laura Braden, Shona Whyte and Jordan Poley for your mentorship and providing me with the invaluable skills of troubleshooting and thinking outside the box when it came to new scientific concepts. As well as the graduate students of the Hoplite lab, for being there to support me every step of the way. From the long days sampling in the aquatics facility to debates in our office, it has been a pleasure working alongside you!
I would also like to acknowledge the funding agencies that made my research possible: National Sciences and Engineering Research Council of Canada, Elanco Animal Health Canada, Novartis/Elanco Research Chair in Fish Health, Integrated Pathogen Management of Co- infection in Atlantic salmon by the Genome Canada GAPP program, Ocean Frontier Institute, University of Prince Edward Island, and Atlantic Veterinary College.
Thank you to my family for their constant support of my education and for teaching me the value of determination and persistence in everything that I do.
A very special thank you to my husband, Leo, for all of his love, support and encouragement throughout my years in school. For always believing in me and understanding my dedication to science. Lastly, to Amari – you might have only been 2 months old when I defended my thesis, but I did this all for you. Love you both!
iv Table of Contents IMMUNOLOGICAL RESPONSES IN ATLANTIC SALMON (SALMO SALAR) TO CO- INFECTIONS WITH SEA LICE (LEPEOPHTHEIRUS SALMONIS) ...... I ABSTRACT ...... II ACKNOWLEDGEMENTS ...... IV TABLE OF CONTENTS ...... V LIST OF TABLES ...... 1 LIST OF FIGURES ...... 2 LIST OF ABBREVIATIONS ...... 4 CHAPTER 1: INTRODUCTION ...... 6 1.1 OVERVIEW AND OBJECTIVES ...... 6 1.2 INTRODUCTION TO SALMON AQUACULTURE ...... 6 1.3 ATLANTIC SALMON BIOLOGY ...... 8 1.4 THE TELEOST IMMUNE SYSTEM ...... 10 1.4.1 The mucosal immune system ...... 11 1.4.2 Overview of teleost skin ...... 11 1.4.3 The innate immune response ...... 12 1.4.4 Humoral mechanisms in the mucus layer ...... 13 1.4.5 The adaptive immune response ...... 17 1.4.6 Immune signaling of the teleost head kidney ...... 18 1.5 SEA LICE: LEPEOPHTHEIRUS SALMONIS ...... 19 1.5.1 Louse development ...... 20 1.5.2 Attachment and feeding behaviour ...... 22 1.5.3 Sea lice as potential vectors ...... 24 1.6 CO-INFECTIONS ...... 25 1.6.1 Parasitic and bacterial co-infections ...... 26 1.6.2 Parasitic and viral co-infections ...... 26 1.7 MORITELLA VISCOSA ...... 27 1.8 INFECTIOUS SALMON ANEMIA VIRUS ...... 28 1.9 PREVENTION AND MANAGEMENT STRATEGIES ...... 29 1.10 FUNCTIONAL DIETS ...... 33 CHAPTER 2: WOUND HEALING AND IMMUNE RESPONSES OF ATLANTIC SALMON (SALMO SALAR) DURING CO-INFECTION WITH LEPEOPHTHEIRUS SALMONIS AND MORITELLA VISCOSA ...... 35 ABSTRACT ...... 36 2.1 INTRODUCTION...... 37 2.2 METHODS ...... 39 2.2.1 FISH MAINTENANCE AND HUSBANDRY ...... 39 2.2.2 L. SALMONIS INFECTION ...... 40 2.2.3 PREPARATION OF M. VISCOSA INOCULUM AND BACTERIAL CHALLENGE ...... 41
v 2.2.4 SAMPLING ...... 42 2.2.5 LESION GRADING ...... 43 2.2.6 OSMOLALITY ANALYSIS ...... 44 2.2.7 BACTERIOLOGY ...... 44 2.2.8 RNA ISOLATION AND CDNA SYNTHESIS...... 45 2.2.9 REVERSE TRANSCRIPTION - QUANTITATIVE PCR ...... 46 2.2.10 STATISTICAL ANALYSIS ...... 48 2.3 RESULTS ...... 48 2.3.1 ATLANTIC SALMON SURVIVAL ...... 48 2.3.2 L. SALMONIS INFECTION ...... 50 2.3.3 BACTERIOLOGY ...... 51 2.3.4 LESION GRADING ...... 52 2.3.5 OSMOLALITY: SERUM ANALYTES ...... 55 2.3.6 GENE EXPRESSION IN SKIN OVER TIME AND EXPOSURE LEVEL ...... 55 2.3.7 INFLAMMATORY AND ACUTE PHASE GENES ...... 56 2.3.8 CELLULAR EFFECTORS ...... 58 2.3.9 EXPRESSION PROFILES OF FISH POSITIVE FOR MORITELLA VISCOSA BY CULTURE ...... 61 2.4 DISCUSSION ...... 62 CHAPTER 3: FUNCTIONAL FEED IMPACTS ON ATLANTIC SALMON (SALMO SALAR) IMMUNE RESPONSES TO SINGLE INFECTION OF SEA LICE (LEPEOPHTHEIRUS SALMONIS) AND CO-INFECTION OF SEA LICE AND INFECTIOUS SALMON ANEMIA VIRUS ...... 69 ABSTRACT ...... 70 3.1 INTRODUCTION...... 72 3.2 METHODS ...... 75 3.2.1 FISH MAINTENANCE AND HUSBANDRY ...... 75 3.2.2 EXPERIMENTAL DIETS AND FEEDING TRIAL ...... 75 3.2.3 COPEPODID CHALLENGE ...... 79 3.2.4 ISAV VIRAL CULTURES AND INFECTION ...... 79 3.2.5 SAMPLING ...... 80 3.2.6 ISAV DETECTION ...... 81 3.2.7 RNA ISOLATION AND CDNA SYNTHESIS: GENE EXPRESSION ANALYSIS ...... 82 3.2.8 RT-QPCR: GENE EXPRESSION...... 83 3.2.9 STATISTICAL ANALYSIS ...... 86 3.3 RESULTS ...... 87 3.3.1 FEED CONSUMPTION AND CONDITION FACTOR ...... 87 3.3.2 ATLANTIC SALMON SURVIVAL ...... 90 3.3.3 L. SALMONIS INFECTION ...... 91 3.3.4 ISAV DETECTION ...... 92 3.3.5 LOGISTIC REGRESSION ANALYSIS AND ODDS RATIO ...... 93 3.3.6 IMPACT OF FUNCTIONAL DIET ON GENE EXPRESSION ...... 94 3.4 DISCUSSION ...... 102
vi CHAPTER 4: DISCUSSION ...... 111 REFERENCES ...... 116
vii List of Tables Table 2.1: Primer sequences and efficiencies for RT-qPCR……………………………………47
Table 2.2: Lice counts in single, medium co-infected and high co-infected fish over time…….51
Table 2.3: Lice counts per lice stage and exposure level in moribund fish and mortalities…….51
Table 2.4: Confirmation of M. viscosa by culture and qPCR in sampled fish…………………..52
Table 2.5: Confirmation of M. viscosa by culture and qPCR in mortalities…………………….52
Table 2.6: Concentration of serum analytes over time………………………………………….55
Table 3.1: Composition of functional diets…………………………………………………..….77
Table 3.2: Primer sequences and efficiencies for RT-qPCR……………..……………………..85
Table 3.3: Feed consumption and condition factor of single and co-infected fish over time…...88
Table 3.4: Feed consumption of single and co-infected fish fed functional diets between 40-47 dpi………………………………………………………………………………………………..88
Table 3.5: Condition factor of single and co-infected fish fed functional diets at 47 dpi………89
Table 3.6: Adult male and adult female L. salmonis lice counts on mortalities……………...…91
Table 3.7: Lice counts on fish exposed to L. salmonis only (single infection) over time………92
Table 3.8: Lice counts on fish exposed to L. salmonis and ISAv (co-infection) over time……..92
Table 3.9: Summary of coefficients of analyzed genes
1 List of Figures
Figure 1.1: The life cycle of Lepeophtheirus salmonis…………………...…………………….22
Figure 1.2: Gravid female lice on Atlantic salmon…………………………………………..….24
Figure 2.1: Schematic representation of the single and co-infection trials……………………...43
Figure 2.2: Percent mortality in single infection, medium co-infection and high co-infection groups…………………………………………………………………………………………….49
Figure 2.3: Kaplan-Meier survival plot of single infection, medium co-infection and high co- infection groups………………………………………………………………………………….50
Figure 2.4: Lesion grading scheme during sampling……………………………………………53
Figure 2.5: Lesion severity over time for moribund fish and mortalities in single infection, medium co-infection and high co-infection groups……………………………………………...54
Figure 2.6: Expression of il-12 and igm across L. salmonis exposure groups at 7 and 12 dpLs..56
Figure 2.7: Expression of inflammatory and acute phase genes across L. salmonis exposure groups over time…………………………………………………………………………………58
Figure 2.8: Expression of il-12, cd4, and cd83 over time……………………………………….59
Figure 2.9: Expression of cellular effector genes across L. salmonis exposure groups over time………………………………………………………………………………………………60
Figure 2.10: Principal component analysis (PCA) and heat map of samples based on gene expression………………………………………………………………………………….…….61
Figure 2.11: Expression of il-8 and hep in samples positive for M. viscosa at lesion sites…..…62
Figure 3.1: Schematic representation of single and co-infection trials………………………….78
Figure 3.2: Percent mortality in fish based on functional diet administered……………………90
2 Figure 3.3: Infectious salmon anemia virus detection in posterior kidney samples fed functional diets at 33 and 46 dpi…………………………………………………………………………….93
Figure 3.4: Expression of irf7b across each functional diet at 33 and 47 dpi………………...…95
Figure 3.5: Expression of mxb across each functional diet at 33 and 47 dpi……………………96
Figure 3.6: Expression of mhii across each functional diet at 33 and 47 dpi…………………...97
Figure 3.7: Expression of ccl19b across each functional diet at 33 and 47 dpi…………………99
Figure 3.8: Expression of isg15a across each functional diet at 33 and 47 dpi………………..100
Figure 3.9: Expression of mmp-9 across each functional diet at 33 and 47 dpi……………….101
Figure 3.10: Expression of il-1β across each functional diet at 33 and 47 dpi……………...…102
3 List of Abbreviations ACC Animal Care Committee AMP Antibacterial peptide ANOVA Analysis of variance APC Antigen presenting cells ARA Arachidonic acid AS-IFN Atlantic salmon interferon-like activity ASK Atlantic salmon kidney AVC Atlantic Veterinary College CCAC Canadian Council of Animal Care CCL Chemokine-like ligand CD Cluster of differentiation cDNA Complementary deoxyribonucleic acid Cl- Chloride ion CI Co-infection CM Cumulative mortality CpG ODN Cytosine-phosphate-guanine oligodeoxynucleotide CPP Classical complement pathway CRP C-reactive protein Ct Cycle threshold Ctrl Control diet Ctrl + IS Control diet with immunostimulant C3 Complement 3 dd Degree days DHA Docosahexaenoic acid DNA Deoxyribonucleic acid dpi Days post infection dpLi Days post lice infection dpMvc Days post Moritella viscosa challenge EF1aAb Eukaryotic translation elongation factor 1-alpha eIF3s6 Eukaryotic initiation factor 3 subunit 6 EPA Eicosapentaenoic acid FAO Food and Agricultural Organization FA+I Fatty acid enriched diet with pro-inflammatory supplementation FA-I Fatty acid enriched diet with anti-inflammatory supplementation FW Freshwater GALT Gut-associated lymphoid tissue gDNA Genomic deoxyribonucleic acid GIALT Gill-associated lymphoid tissue Hep Hepcidin HL High lice HPR Highly polymorphic region HSD Honest significant difference HV High virus IFN Interferon system Ig Immunoglobulin IHNv Haematopoietic necrosis virus IL Interleukin IRF Interferon regulatory factor IP Intraperitoneal IPC Internal positive control ISAv Infectious salmon anemia virus ISG Interferon stimulated gene K+ Potassium ion LL Low lice
4 LS Lepeophtheirus salmonis LV Low virus MALT Mucosa-associated lymphoid tissue MBL Mannose-binding lectin MH Major histocompatibility MMP Matrix metalloproteinase MS-222 Tricaine methanesulphonate MV Moritella viscosa Mv+LsM M.viscosa and medium lice exposure Mv+LsH M.viscosa and high lice exposure Mv+Ls- Single infection with M. viscosa and no lice Mv-Ls- No M. viscosa and no lice Na+ Sodium ion NB New Brunswick noRT Negative reverse transcriptase NTC No template control OR Odds ratio PAMP Pathogen-associated molecular patterns PCA Principle component analysis PCR Polymerase chain reaction PE Prince Edward Island pIC Polyriboinosinic polyribocytidylic acid PIT Passive integrated transponder PGE2 Prostaglandin E2 PRR Pattern recognition receptor RNA Ribonucleic acid RPC Research and Productivity Council RPS20 Ribosomal protein S20 RT-qPCR Reverse transcription quantitative polymerase chain reaction SAA Serum amyloid A SALT Skin-associated lymphoid tissue SHK Salmon head kidney SI Single infection SS Salmo salar ssRNA Single stranded RNA SW Seawater TCC Terminal complement complex TGF Transforming growth factor TIR Toll-interleukin receptor TLR Toll-like receptor TMS Tricaine methanesulphonate TSA Trypticase soy agar UPEI University of Prince Edward Island USD United States Dollar VHSv Viral hemorrhagic septicemia virus
5 CHAPTER 1: INTRODUCTION
1.1 Overview and objectives
The studies in this thesis were designed to investigate the immunological consequences of heterologous co-infections: parasitic and bacterial, and parasitic and viral. The studies were designed to simulate natural disease succession as close to what may be expected in aquaculture conditions as possible. The overall objectives of the thesis were to investigate the impact of co- infections in the Atlantic salmon (Salmo salar) immune system. The aim involved gaining better understanding of host immune responses while determining any additive effects on the host during simultaneous infections.
The overall theme of the thesis was to characterize immunologic responses between simultaneous infections. The first co-infection study involving Lepeophtheirus salmonis and
Moritella viscosa in Atlantic salmon focused on skin immune responses targeting immune candidates involved in wound healing, antigen presentation and lymphocyte responses at different louse exposure levels. The second co-infection study involving Lepeophtheirus salmonis and infectious salmon anemia virus (ISAv) assessed whether there was an additive effect on the host immune responses when experimental groups were fed four different functional diets. Candidate biomarkers were selected to determine responses in wound healing and systemic immune markers from head kidney samples, to ultimately determine if varying feed components can modulate host immune responses and impact louse infection and co-infection survival for future implications to the aquaculture industry.
1.2 Introduction to salmon aquaculture
As the world population continues to grow, aquaculture is gaining acceptance as a high-
6 quality source of protein for human consumption. In 2016, according to the Food and
Agricultural Organization (FAO), worldwide production from aquaculture alone was over 101 million tons ($166 billion USD), with salmonid production reaching 2.7 million tons valued at approximately $17.2 million USD (FAO 2016) and still growing. Salmonids, although only comprising approximately 2.6% of overall aquaculture production, is an attractive candidate for an alternative protein source in Western Europe and North America to livestock. Norway, which began Atlantic salmon aquaculture in 1970, is now the largest contributor in salmonid production at 1.3 million tonnes per year (11.7 billion USD) (FAO 2016). Of the world’s total salmonid tonnage in 2016, 2.3 million tons is made up of Atlantic Salmon (Salmo salar), 171,749 tonnes of coho salmon (Oncorhynchus kisutch), 12, 400 tonnes of Chinook salmon (Oncorhynchus tshawytscha), and 814,000 for rainbow trout (Oncorhynchus mykiss) (FAO 2016). Canada, Chile,
Ireland, Scotland and the Faroe Islands are also major global suppliers of salmon, specifically
Atlantic salmon, with a total production value of 105 million tons per year (FAO 2016).
The majority of Canada’s salmon farming industry on the Atlantic and Pacific coasts is carried out in marine net-pen grow-out sites. Net-pens consist of an open mesh net that is anchored to a buoy and suspended in place. A typical net-pen arrangement in an Atlantic salmon farming site includes 10 net-pens, each approximately 30 m in width and 20 m in depth (Ayer and Tyedmers 2009).
Location of farm sites, as well as stocking density, is determined based on seasonal fluctuations in water temperature, dissolved oxygen content, salinity, depth, exposure to storms, presence of harmful algae, and the movement of currents to naturally remove waste and provide a continuous supply of dissolved oxygen (Nash 2003). Typical Atlantic salmon grow-out sites maintain a maximum stocking density between 15-18 kg/m3. (RSPCA 2015). This stocking
7 density is quite high compared to other salmonids, such as coho salmon, where stocking density during harvest does not exceed 8-12kg/m3 (FAO 2016). Generally, the high stocking density in farmed Atlantic salmon is to prevent the fish from becoming territorial, which may increase aggression and result in reduced access to feed. By increasing the stocking density, fish school together and do not have the space for territorial tendencies (Turnbull et al. 2008). Alternatively, species such as coho and Chinook salmon are more aggressive at higher densities.
Proper production strategies are important to yield maximum profits. However, a number of environmental impacts are inevitable due to the marine net-pen farming system. Such impacts include: potential amplification and spread of disease between farmed and wild fish populations
(Bjørn et al. 2001; Krkošek et al. 2005; Naylor et al. 2000); ecological and genetic impacts of escaped salmon (Hindar et al. 1991; Krkošek et al. 2005; Mcginnity et al. 2003); release of chemotherapeutants into coastal waters (Ernst et al. 2001); alterations in the local ecological environment under net-pen systems (Kempf et al. 2002).
One of the major costs in the aquaculture industry, along with the cost of feed, is the concern of fish health and disease. It has become common practice throughout salmon aquaculture to vaccinate fish as a preventative management strategy, however intraperitoneal
(IP) vaccination is costly and time consuming for farmers and causes stress to the fish.
Chemicals can be introduced to the water column to aid in disease prevention. However, there are increasing concerns regarding the environmental impacts and safety of non-target species during these treatments (Saunders 1995).
1.3 Atlantic salmon biology
Atlantic salmon (Salmo salar) [Linnaeus, 1758] is a teleost, in the Salmonidae family. Wild
8 Atlantic salmon are endemic throughout the North Atlantic Ocean, in North America from Maine to the Arctic and Europe, from Portugal to Russia. Atlantic salmon are anadromous species, meaning they spawn and spend early development in freshwater before migrating to seaweater where they grow for several years before returning to spawn back in their natal streams. Salmon spawn in upstream gravel beds in freshwater rivers between October and January. It takes approximately 250-degree days (dd) for eggs to eye, followed by 250 dd to hatch into the alevin stage during the spring (Kottelat and Freyhof 2007). At this stage, alevin depend on their yolk sac for nutrition (~300 dd), until the yolk sac is fully absorbed, and they begin to feed as fry
(Kottelat and Freyhof 2007). Fry feed on zooplankton, insect larvae and small invertebrates as they become stronger and move up the water column. The next juvenile stage is parr (fingerling), identified by their prominent black markings to aid in camouflage as they increase in size.
Juvenile salmon may remain in either the fry or parr stages from 1 to 5 years, until a physiological change called smoltification occurs (Balon 1990). This process is triggered by a change in photoperiod as the smolts migrate downstream to saltwater environments between
March and June. Salmon feed on pelagic species in the marine environment such as squid, herring and krill. Upon maturation, salmon return to their original river from sea to spawn in the fall and winter after one year at sea are known as grilse (McCormick et al. 1998). Salmon often spawn only once. A fish that remains in freshwater following spawning or returns to sea is called a kelt (FAO 2016; Hansen and Quinn 1998). Atlantic salmon differ from other salmon species in that they have the ability to spawn multiple times, observed up to six times (Halttunen et al.
2009).
Within aquaculture, the production cycle has tried to parallel the natural life cycle as close as possible, through a concept called precision fish farming (Føre et al. 2018). The
9 manipulation of temperature and photoperiod allows for the life cycle to occur over a shorter period of time and to produce market sized fish year-round (Deacon and Hecht 1996). In the culture environment, broodstock are mature individuals (male and female) that are used for breeding purposes. They are kept in freshwater tanks in a facility until eggs have been produced, typically in the fall. Eggs are removed, disinfected and fertilized with milt. Unfertilized eggs are sifted out, while fertilized eggs are placed in hatchery trays for hatching. During early life stages, fish remain in freshwater tanks in flow-through or recirculation systems (Willoughby 1999).
Smoltification typically occurs in the following spring unless artificially manipulated to occur at an alternate time of year. Once fish have smoltified, they are transferred to sea cages with stocking densities up to 20 kg/m3. Fish remain at sea cage sites until market weight is reached, typically 18-22 months to reach a minimum weight of 2kg but targeted at 4-6 kg. Harvesting techniques aim to minimize stress to the fish. Fish are crowded in the sea cage and transferred to a well boat for processing or processed on site (Willoughby 1999).
1.4 The teleost immune system
The vertebrate immune system is composed of innate (non-specific) and adaptive (specific) immune responses, providing a complex network of molecules and signaling pathways to protect the host against heterologous pathogens. The following section will introduce the innate and adaptive host response, contributing organs involved in immunity, and how each of these responses are reliant on the other. It is important to note that immune responses may be influenced by nutrition and stress. In poikilothermic species like salmon, environmental factors, such as water temperature, have potential immunosuppressive/inducing effects and could affect disease susceptibility.
10
1.4.1 The mucosal immune system
The skin surface is the largest immunological organ within the fish. The skin, gills, and gut are in intimate contact with the surrounding aqueous environment, making their associated lymphoid tissue of major importance in pathogen defense. The MALT (mucosa-associated lymphoid tissue) in teleosts is subdivided into the GALT (gut-associated lymphoid tissue), SALT (skin- associated lymphoid tissue) and GIALT (gill-associated lymphoid tissue) (Salinas, Zhang, and
Sunyer 2011).
1.4.2 Overview of teleost skin
Vertebrate skin is not only important as a physical barrier responsible for ensuring systemic homeostasis with respect to acid-base and osmotic balance, but also for protecting the host from invading pathogens and by secreting mucus that is directly involved in immune function. Since the aquatic environment provides a pathogen optimal accessibility to the host, cutaneous diseases are more common in fish than in terrestrial species (Groff 2001).
According to Hawkes (1974), the structure of teleost skin is composed of the cuticle or mucosal layer, the epidermis, and the dermis. The epidermis is 5-10 cell layers thick and composed of squamous stratified epithelium and goblet cells. The majority of the cell composition is non- keratinized. The intermediate stratum of the epidermis houses mucosal goblet cells containing antimicrobial substances (Kumar, Mittal and Whitear 1979). The dermis can be characterized in two layers: the hypodermis, or stratum spongiosum, and the stratum compactum. In fish, the dermis is composed primarily of dense connective tissue and collagen fibers. The epidermis has relatively little pigmentation, while the dermis contains chromatophores, melanin, and guanine or carotenoid pigments (Kumar, Mittal and Whitear 1979). The hypodermis is encompassed by loose
11 collagen fibers and blood vessels. The stratum compactum is the innermost layer that is attached to striated muscle.
The mucus is secreted by goblet cells, unicellular glands within the epidermis (Harris and
Hunt 1975). The number of mucus-producing cells varies depending on the stress experienced by the fish in its’ environment. It has been demonstrated that the number of goblet cells is positively correlated with increased stress-levels experienced by the fish (Angeles Esteban et al. 2012).
Teleost skin is also structurally comprised of motile keratocytes. These are located along the surface of the skin and function by rapidly migrating to a wound or port of entry created by an invasive pathogen as studied in sea bream (Vieira et al. 2011). Specific cell types have been shown to internalize surface bacteria, change the architecture of the surface of the epidermis, and aid in response to injury via mucus secretion, wound healing, and regeneration by cell rearrangement (Sire and Akimenko 2004).
1.4.3 The innate immune response
The innate immune system provides a nonspecific response that acts as the first line of defense against invading pathogens. It is an evolutionarily ancient defense mechanism found in all animals that has the ability to distinguish between homologous host cells and heterologous pathogens. The innate immune response aims to prevent the pathogen from attaching, invading and multiplying within the host. To combat this, the innate immune response is typically divided into 3 components of defense: (1) providing physical barriers to pathogens through epithelial surfaces and mucus secretion (gills, skin, gut), (2) chemical barriers and humoral parameters, such as antimicrobial compounds, complement secretions, interferons, lysozymes, transferrin, lectins, etc., and (3) initiating cellular responses such as leukocytes, including monocytes (ie.
12 dendritic cells) and macrophages, neutrophils, lymphocytes, and non-specific cytotoxic cells
(Buonocore et al. 2009; Ellis 2001; Magnadottir 2010). The ability of the innate immune system to recognize a pathogen as “non-self” is accomplished by pattern-recognition receptors (PRRs)
(Ellis 2001). PRRs are proteins expressed on the surfaces of leukocytes and epithelial cells to target pathogen-associated molecular patterns (PAMPs) released/expressed by heterologous pathogens or signals released from the host’s apoptotic cells (Alvarez-Pellitero 2008; Ellis 2001;
Whyte 2007). PAMPs released from pathogens are generally polysaccharides and glycoproteins associated in viral double-stranded RNA, bacterial CpG, and peptidoglycans, to name a few.
PAMPs are recognized by transmembrane Toll-like receptors (TLRs) that span both the innate and adaptive immune responses, triggering effector molecules such as pro-inflammatory cytokines and interferons (Akira et al. 2006). The innate immune response in fishes has been shown to be more diverse in comparison to their mammalian counterparts. For example, the diversity in the response of the complement classical and alternative pathways (Magnadóttir
2006). Recent aquaculture industry efforts have focused on introducing immune enhancing compounds, such as immunostimulants, probiotics, and prebiotics, to further support the innate immune response, in order to manage, and ultimately prevent, pathogenic outbreaks
(Magnadóttir 2006).
1.4.4 Humoral mechanisms in the mucus layer
The mucosal surfaces of fish skin acts as a semipermeable barrier functioning in the exchange of gases, nutrients, water, hormones and gametes (Raj et al. 2011). As a biological barrier, the mucus layer traps particles, bacteria, and viruses and is comprised predominantly of macromolecules, such as mucins (Ellis 2001). These elements are removed from the fish surface as the fish moves through the water column and the mucus layers slough off the epithelial
13 surface (Mayer 2003). Motile agents that become entrapped in the mucus may move “upstream” through mucosal layers to the skin epithelium. Epidermal goblet cells and glycoproteins are involved in inhibiting invasion and proliferation of pathogens, ion regulation, osmoregulation, and lubrication of the skin surface (Alvarez-Pellitero 2008; Subramanian et al. 2007). If fish are stressed or injured, the mucus composition changes by increasing total protein within the mucus
(Blackstock and Pickering 1982).
Several proteins have been isolated from fish mucus that are determined to have specific immune functions. Bergsson et al. (2005) isolated histone H2B from Atlantic cod that was shown to inhibit bacterial and fungal pathogens in vitro. Other histone fragments have also been isolated and have demonstrated antimicrobial properties in the presence of alpha-defensins, lysozyme, and LL-37 (mature peptide) (Frohm et al. 1996). LL-37 was originally noted in wound healing fluids in humans, but also important in fish tissues. Lactoferrin, an iron-binding protein necessary for skin immunity and the inhibition of allergic responses has also been identified (González-Chávez et al. 2009).
To combat bacterial disease, the iron transporter transferrin provides tight regulation on the pathogen’s ability to acquire iron to establish infection. Iron availability in the skin is low with high affinity to transferrin making it difficult for the pathogen to access iron reserves
(Griffiths and Bullen 1987). However, some pathogens have evolved mechanisms of obtaining iron. Genetic polymorphism in the transferrin genotype (AA, AC, CC) may allow for greater restriction on the acquisition of iron by the pathogen. This restriction has been seen in the C- transferrin allele of coho salmon when introduced to bacterial kidney disease (Suzumoto et al.
1977; Winter et al. 1980).
14 Antiproteases are another response to challenge the growth of pathogens, in the form of
α1-antiprotease and α2-antiprotease. They are found in the plasma and function in inhibiting the degradation of peptides and proteins. In a study by Ellis (1987), Aeromonas salmonicida was shown to be resistant to α1-antiprotease but inhibited by α2-antiprotease. Similarly, antiprotease activity and infection resistance was also seen in rainbow trout and brook trout exposed to A. salmonicida (Freedman 1991).
Lectins are found in the mucus, serum and ova and have the ability to bind to specific sugars which may lead to agglutination (Arason 1996). Mannose-binding lectin (MBL) plays an essential role in innate immunity by the lectin pathway, makes bacteria more susceptible to phagocytosis, and may have a role in complement activation (Kuhlman et al. 1989). MBL has been isolated in the serum of Atlantic salmon and the Japanese eel (Gercken and Renwrantz
1994). Lectin in the ova of chinook salmon has been reported in inhibiting the growth of
Aeromonas hydrophila, Yersinia ruckeri, Edwardsiella tarda and Vibrio anguillarum (Voss et al.
1978).
Lysins are another non-specific humoral factor that include antibacterial peptides, proteases, lysozymes, C-reactive protein (CRP), and complement. Antibacterial peptides (AMPs) target both Gram-negative and Gram-positive bacterial membranes, fungi, parasites and enveloped viruses. AMPs cause membrane instability, which inhibits the pathogen from infecting host cells (Edward Robinson et al. 1998). In mammalian cells, AMPs have been shown to bind to receptors on the cell surface preventing viral particle sequestration. Proteases catalyze the breakdown of peptide bonds and have been found in the skin mucus of fish. Hjelmeland et al.
(1983) reported digestive activity of proteases from rainbow trout against V. anguillarum cells.
15 Lysozymes are present in the mucus, plasma, and body fluids of fish and expressed in most tissues (Murray and Fletcher 1976). They have lytic responses against bacteria, play a major role in phagocytosis, activate the complement system and are opsonic. Lysozyme activity has a direct responsibility in conserving homeostasis, although it is dependent on stressors, disease, and environmental factors, such as season, water temperature and pH.
C-reactive proteins (CRP) are found in skin mucus and levels rise in response to inflammation, generally associated with pathogenic infections. CRP targets bacteria, parasites and fungi by reacting with their surface membranes and initiating complement (Nakanishi et al.
1991). In fishes, CRP levels are higher than in mammals due to the increased susceptibility of infection in an aqueous environment. CRP levels are also elevated in the summer and increase
20-fold during temperature increases and when inflammatory agents are administered (Kodama et al. 1989; Szalai et al. 1994; Winkelhake and Chang 1982).
Many of these non-specific humoral factors contribute to the classical (antibody- dependent), alternative (antibody-independent) and lectin pathway of the complement system.
Activation of any of these pathways results in the activation of C3 and the subsequent binding of
C3a or C3b to the surface of the pathogen (bacteria, fungi, viruses). Following binding, the terminal complement complex (TCC) is formed on the pathogen surface, ultimately leading to microbial elimination (Janeway et al. 2001). The complement system is comprised of lytic, inflammatory and opsonic activities, linking the non-specific humoral responses along with phagocytic responses for microbial removal.
Antibodies associated with specific humoral responses include anti-adhesins (prevent bacteria from entering host cells), anti-toxins (neutralize toxins released by pathogens), and anti- invasins (antibodies blocking the invasion of non-phagocytic host cells) (Gudmundsdóttir and
16 Magnadóttir 1997; Magariños et al. 1996). The activation of the classical complement pathways
(CCP) occurs when an antibody comes in contact with the microbial surface. A localized CCP response directed at the pathogen occurs and inflicts more damage than the alternative complement pathway. In bacterial infections, CCP is activated by lipopolysaccharides found in the cell wall of Gram-negative bacteria. C5a chemokine production activates the phagocytic activity of neutrophils and macrophages, which further contain receptors for C3 (Uribe et al.
2011). Cell mediated responses in conjunction with CCP act by lysing the infecting cell, promoting an influx of phagocytic cells, and increasing cytokine release by T lymphocytes (Ellis
2001).
1.4.5 The adaptive immune response
Adaptive immunity has evolved more recently in comparison to the primitive innate immune response and it is found in only 1.5% of species, mostly vertebrates (Hirano et al. 2011). The adaptive immune system is responsible for recognizing a pathogen that has been previously encountered and mounts a faster and more efficient immune response (Shanker 2010). When antigen presenting cells (APC) such as macrophages and dendritic cells are contacted, effector and memory cells are subsequently generated by the proliferation of naïve lymphocytes.
Lymphocytes differentiate into T and B cells produced in the thymus and the head kidney, respectively (Hirano et al. 2011). Cluster of differentiation (CD) are markers on the cell surface that can be used to determine different lymphocyte populations by specific antigen receptors and population-specific co-receptors (Randelli et al. 2008). Immunoglobulins in fish (IgM, IgD,
IgT/Z) are released by B-cells and have functions in neutralizing and opsonizing antigens (Xu et al. 2013). In teleosts, immunoglobulin M (IgM) is the main Ig class protein present that is
17 modified into circulating antibodies and may have roles in determining the B-cell recognition capacity (Magnadóttir 2006; Shoemaker et al. 2001). As previously mentioned, pattern recognition receptors (PRRs) activate the innate immune response when an infection has been detected. However, signals are also released to activate adaptive immunity. Although adaptive immunity is highly specific and is able to target an infection while sparing uninfected tissue, T and B cell responses need to be activated and targeted to appropriate infection sites, which occurs through innate signaling and antigen presentation. T-cell responses are initiated through specific T-cell receptors that interact with Ig proteins known as major histocompatibility (MH) receptors (Klein and Sato 2000). There are two types of MH receptors; MH class I receptors and
MH class II receptors. Class I MH receptors present peptides from intracellular pathogens and class II MH receptors present peptides from extracellular pathogens (Klein and Sato 2000).
Once the necessary immune response has been activated, granulocytes and macrophages are involved in the removal of bacteria by phagocytosis through the production of reactive oxygen species (ROS) during respiratory burst (Secombes et al. 2001). Degranulation activity of teleost neutrophils requires further examination, however, an assay to assess myeloperoxidase activity has been developed (Palić et al. 2005). This assay has shown degranulate activity when exposed to Aeromonas salmonicida (Katzenback and Belosevic 2009).
1.4.6 Immune signaling of the teleost head kidney
It is difficult to make comparisons between teleost and mammalian immunity due to the differences in morphology between the immune systems. Mammals rely on bone marrow and lymph nodes to generate blood cells and assist the body in inducing an immune response.
Instead, fish have the head kidney (pronephros) and along with the spleen and thymus are largely
18 responsible for immunity through endocrine and hemopoietic-lymphoid tissue (Gallo and
Civinini 2003; Press and Evensen 1999; Uribe et al. 2011). The head kidney is located posterio- dorsal to the gills and involved in the production of immune effector cells, and their signaling molecules including cytokines, cortisol, catecholamines and thyroid hormones (Geven and
Klaren 2017). It also contains nerve endings which may have functions in neuroimmuno- endocrine connections (Tort et al. 2003). The head kidney is responsible for functions including hemopoiesis, phagocytosis and antigen presentation in melano-macrophage centers, myelopoiesis, antigen processing as the fish matures, IgM production, immunogenic memory, and it is the primary B-cell producing organ (Alvarez-Pellitero 2008; Galindo-Villegas and
Hosokawa 2004; Press and Evensen 1999; Shoemaker et al. 2001).
1.5 Sea Lice: Lepeophtheirus salmonis
Parasitic copepods, known as sea lice, are naturally occurring pathogens in marine and freshwater environments. Within the order Copepoda, suborder Siphonostomatoida, family
Caligidae, there are approximately 36 genera, which include approximately 124 Lepeophtheirus species and 257 Caligus species, with these numbers rising as more species are being accepted
(Walter and Boxshall 2017). Lepeophtheirus salmonis and some Caligus species have become common pests in intensive salmon farming that provides suitable conditions for parasite growth and transmission compared to natural conditions. These marine parasites can cause considerable economic and ecological impacts to the aquaculture industry and wild populations of salmon, respectively.
Disease-related economic losses of L. salmonis to marine salmoniculture are estimated to exceed 1 billion USD worldwide (Abolofia et al. 2017). This is the primary species of concern in
19 the northern hemisphere. The recently sequenced salmon louse genome and biology of L. salmonis has been extensively studied to gain a better understanding of the parasite and host- parasite interactions. The most pressing issue for aquaculture is the development of resistance in the parasite against the few drug therapies available for control. The following review of the literature will focus on the biology and pathology of the salmon louse, Lepeophtheirus salmonis, host-pathogen interactions, and current control strategies within the aquaculture industry.
1.5.1 Louse development
Lepeophtheirus salmonis is a parasite that infects salmonids in the genera Salmo (salmon:
Salmo salar; sea trout: Salmo trutta), Salvelinus (Arctic charr: Salvelinus alpinus) and
Oncorhynchus (all Pacific salmon species). However, coho (Oncorhynchus kisutch) and pink
(Oncorhynchus gorbuscha) salmon are resistant to sea lice due to the initiation of strong immunological responses during initial attachment within the first week of infection (Burka et al.
2012; Wagner et al. 2008).
The life cycle of L. salmonis exhibits 8 different life stages individually separated by a single moult (Fast 2014; Hamre et al. 2013). These stages are divided into planktonic (free- swimming) and parasitic stages. The rate at which the life cycle progresses from egg to adult is dependent on the temperature and salinity (<30%) of the surrounding environment (González and Carvajal 2003; Johnson and Albright 1991; Schram 2006). The life cycle can be as short as
19 days (at 17°C) up to 93 days (at 5°C), with 43 days (at 10°C) being optimal (Burka et al.
2012; Pike and Wadsworth 1999).
The life cycle begins as eggs hatch into nauplius I, free-swimming and lecithotrophic, relying on reserved energy from yolk and lipid stores. Nauplii may also be planktotrophic by
20 feeding on other planktonic organisms (Fast 2014). There are two nauplii stages, followed by an infectious copepodid stage, where the louse attaches to the fins or scales of the host. At this stage, the copepodids rely on chemo- and mechanosensory- reception by the antennae in search of a host. These mechanisms are also involved in L. salmonis mating behavior (Hull et al. 1998).
If there is an abscission of the antennae, there is a reduced rate in which the louse finds a host and/or mate (Hull et al. 1998). The life cycle of the louse continues through 2 molts of the chalimus stages once it has attached to a host and are the first stages in which sex differentiation occurs (Hamre et al. 2013). The chalimus stages can be identified as the parasite attaches using the frontal filament structure by puncturing the epidermis of the host. Following the chalimus stages are two pre-adult stages. At this stage, the parasite is free moving on the host. The genital complex is still under development, however there is already a clear distinction in size between female (8-18mm) and male (5-6mm) lice. Once the final molt to the adult stage is complete, the lice are now mature. Females are capable of producing 10-11 pairs of egg strings with an average of 285 eggs/egg string over their lifetimes of at least 7 months (Heuch et al. 2000). Males exhibit mate guarding within the first 48 hours of becoming sexually viable (Pike and Wadsworth 1999).
In some cases, multiple males may fertilize the females’ eggs, leading to multiple paternal gametes (Todd et al. 2005).
21
Figure 1.1. The life cycle of Lepeophtheirus salmonis from Schram 1993, reproduced by Burka et al. 2012 with author’s permission.
1.5.2 Attachment and feeding behaviour
Initial attachment to the host must be made by the free-swimming copepodid stage in order to stimulate molting behavior and to progress through the life cycle. Physiologically optimal ranges of water temperature, salinity and light are necessary for successful attachment.
Preferred attachment locations of the host are likely to be in areas with the least hydrodynamic disturbances, specifically on the fins and protected areas of the scales, such as the operculum over the gills (Bron et al. 1991; Genna et al. 2005). Semiochemicals, or kairomones, are released by the host and detected by the louse to aid in locating an attachment site while also avoiding non-hosts organisms (Bailey et al. 2006). L. salmonis copepodids are attracted to isophorone and
22 6-methyl-5-hepten-2-one, two semiochemicals emitted by Atlantic salmon. Once the infectious copepod has attached to a suitable fish host, the louse moves to a well-protected area of the host body surface using their sharp maxillipeds as grip. The two antennae are brought in close contact with the host by the probing behavior of the anterior cephalothorax (Bron et al. 1991). The first antennae are the most prominent and covered in setae. At this stage of attachment, there is still the possibility that the copepod may be released from the surface of the host.
The primary attachment phase involves the hook-shaped secondary antennae repeatedly puncturing the epidermis of the host. This structure contains heavy sclerotization on the surface, allowing the antennae to go as far as the basal membrane. The louse has the capability of drawing the rostrum and dorsal cephalic shield, the structures between the antennae, in a downward motion causing the epithelium of the host to lift and aggregate over the louse (Bron et al. 1991). The secondary attachment phase involves the anchoring of the frontal filament to the host. This occurs once the louse has molted from the copepodid to the first chalimus stage (Bron et al. 1991).
Once attached to the host, L. salmonis feeds on mucus, skin, and blood. As the louse develops and matures, female lice are known to be aggressive feeders on blood, while males and earlier developmental stages primarily feed on mucus and skin tissue (Burka et al. 2012). L. salmonis is known to secrete enzymes/compounds in the host’s mucus and tissues to aid in feeding, digestion, and to avoid immune responses implemented by the host (Fast et al. 2005;
Firth et al. 2000; Wagner et al. 2008). Such compounds include trypsin to assist in feeding and digestion and prostaglandin E2 (PGE2) to regulate the feeding site (Fast et al. 2005; Wagner et al.
2008). Heavy infections of L. salmonis can lead to severe lesions on the host’s skin, underlying muscle tissue, and in some cases into the cranium (Fast 2014).
23
Figure 1.2. Gravid female salmon lice conjugated around the anus of an Atlantic salmon
1.5.3 Sea lice as potential vectors
Sea lice have been suggested to be vectors for bacterial and viral pathogens. Isolations of
L. salmonis intestines have been shown to contain infectious salmon anemia virus (ISAv) from clinically diseased fish (Nylund et al. 1994). Aeromonas salmonicida, Pseudomonas fluorescens,
Tenacibaculum maritimum, Vibrio spp., and infectious haematopoietic necrosis virus (IHNv) have also been found present in multiple species of the sea lice, including Lepeophtheirus salmonis and Caligus clemensi (Barker et al. 2019). When considering the life cycle of the salmon louse, it is suspected that the louse becomes a carrier of a viral or bacterial pathogen through the attachment and feeding of contaminated fish. Annual and seasonal fluctuations that depend on salinity and water temperature may influence the pathogen they carry, as well as the length of viability.
It is still not universally accepted as to whether sea lice are a classical vector or if the lice act merely as passive carriers of fish pathogens. However, Hammell and Dohoo (2005) reported
24 a reduced infection frequency of ISAv when regular delousing practices were put in place in
New Brunswick salmon farms. Jakob et al. (2011) demonstrated the capabilities of L. salmonis infections with IHNv through water bath exposure to further infect naïve Atlantic salmon in vitro. The short-term interaction between IHNv and L. salmonis gives evidence that the louse acts more like a mechanical vector as opposed to a biological vector. It is also important to consider that the pathogens mentioned above have the potential of targeting the host through waterborne transmission as the natural pathway of viral dissemination. A highly susceptible fish host along with a highly infectious dose of a pathogen may have the ability to spread on its own.
More research is required to determine if sea lice infections can increase viral transmission efficacy.
1.6 Co-infections
Aquatic species have an intimate relationship with their surroundings and can effectively be immersed in potential pathogens. The majority of research in fish health focuses on single infections in a controlled system to better understand the virulence effects of one pathogen and the response of the host. However, throughout the lifetime of an aquatic host, a succession of different pathogens at any given life stage is the norm. Single infection studies do not characterize a realistic host response in natural conditions. The term co-infection, or simultaneous infection, is defined by the infection of a host by genetically different pathogens
(Cox 2001). These pathogens provide their own evasion mechanisms that cause harm to the host.
In a host, the pathogenic load, parasitic abundance, disease severity, host-pathogen dynamics and host survival can be significantly different in co-infections compared to single infections. This may ultimately alter fish disease pathogenesis, prognosis and treatment needed to clear the
25 infection as synergistic or antagonistic interactions may occur between co-infecting pathogens.
Such studies have been researched in humans and reported that synergistic co-infections may cause the disease to spread faster in surrounding tissue compared to random networks (Hébert-
Dufresne and Althouse 2015). The infection of the first pathogen triggers the host response, influencing other pathogens to take advantage of the immunosuppression/immune-diversion of the host, altering typical disease outcomes (Zhu et al. 2013). Recently, there has been growing interest to investigate infections of multiple pathogens in succession and the epidemiological, ecological, immunological and evolutionary consequences resulting from co-infection (Telfer et al. 2010).
1.6.1 Parasitic and bacterial co-infections
Following the host attachment of a parasite, the risk of subsequent bacterial infections increases.
Past experimental studies have shown an increase in mortalities in fish co-infected with parasitic and bacterial pathogens (Bandilla et al. 2006; Caballero-Solares et al. 2017; Pylkko et al. 2006).
The parasite likely has the ability to immunomodulate the host and serve as a mode to transmit bacterial pathogens (Holzer and Levis 2007). Stress of the host caused by parasitic attachment may present opportunities for microbes to sequester at attachment sites. Strategies of prevention at sea cage sites in some cases aim to prevent initial parasitic attachment to reduce the chances of mortality from secondary bacterial infections.
1.6.2 Parasitic and viral co-infections
Co-infections of parasites and viruses have caused severe damage to the aquaculture industry, most notable in Chile between 2007 and 2009. Valdes-Donoso et al. ( 2013) reported the ISAv
26 outbreak may have been in succession with sea lice infestation. Previous studies have shown sea lice species, C. elongatus and L. salmonis, act as vectors for viral infections causing severe mortalities (Nylund et al. 1994). Ogut and Cavus (2014) investigated parasitic and viral loads in whiting (Merlangius merlangus euxinus) and suggest that environmental factors including spawning stress, water quality and temperature may impact mortality rates during a co-infection of ectoparasites (Trichodina spp. and Gyrodactylus alviga) and viral hemorrhagic septicemia virus (VHSv). Details of co-infections between parasites and viruses are still unclear and may continue to be a challenge due to the diversity of viral infections.
1.7 Moritella viscosa
Winter ulcer disease is a chronic and contagious infection of salmonids caused by the gram- negative bacterium, Moritella viscosa. The bacterium is psychrophilic, affecting fish marine waters below 7-8°C by producing localized inflammation, gill pallor, fin rot and the development of severe lesions (Lunder 1992). M. viscosa is responsible for significant bacterial-related loss to the aquaculture industry and is responsible for high levels of antibiotic use in North America
(MacKinnon 2017). Outbreaks of M. viscosa have also been reported along the Norwegian coast,
Iceland, Scotland and the Shetland Islands (Hoffman et al. 2012). Life stages at risk for contracting the disease are juvenile and adult salmonids in saltwater and smolts reared in freshwater hatcheries (Lunder 1992). Atlantic salmon in particular are highly susceptible to one particular isolate of M. viscosa that has high virulent host specificity to the species (Grove et al.
2010). The outer membrane of the bacterium contains lipopolysaccharides and suggested to have protective antigens against the host (Heidarsdóttir et al. 2008). The bacterium has been reported to penetrate host cells and alter epithelial structure, disrupt the cellular cytoskeleton, and cause
27 the formation of pores leading to cell lysis in vitro (Tunsjø et al. 2009). Ulcers form chronic infections ranging from scale loss and mild inflammation to severe ulcers exposing underlying muscle tissue, hemorrhaging and ascites, which may lead to septicemia and mortality of fish
(Benediktsdóttir et al. 1998). As colonization, proliferation and ulceration by M. viscosa is generally external, the mortalities are typically low, ranging between 10-40% in natural conditions (Hoffman et al. 2012). However, quality of fish at the time of harvest is low, resulting in economic losses.
1.8 Infectious salmon anemia virus
Infectious salmon anemia virus (ISAv) is the causative agent of infectious salmon anemia (ISA).
It is a highly contagious, negative-sense, single stranded RNA virus that is a member of the
Orthomyxoviridae family in the Isavirus genus (Krossøy et al. 1999; Mjaaland et al. 1997). ISAv has a genome consisting of eight segments and 10 proteins (Aspehaug et al. 2005; Falk et al.
2004). At the surface, two glycoproteins reside with segment 5 coding fusion protein and segment 6 coding hemagglutinin-esterase. ISAv targets Atlantic salmon by attacking the epithelium lining of the circulatory system (Rimstad et al. 2017). Mortality can vary between 5-
90% depending on the degree of virulence associated with the 3’ end of the ISAv haemagglutinin gene in the highly polymorphic region (HPR) located amongst segments 6, 7, and 8 (Cook-
Versloot et al. 2004; Ritchie et al. 2009). The level of virulence is determined by the deletions in the HPR in comparison to ISAV-HPR0, the non-pathogenic strain (Cunningham et al. 2002;
Gagné and LeBlanc 2018). Atlantic salmon head kidney (SHK) lines have shown up-regulation in genes involved in innate immunity upon ISAv infection (Kileng et al. 2007; Lyng Schiøtz et al. 2008). Although antiviral responses are initiated in the host, virulence factors, viral
28 replication and load quickly spread when exposed to high-virulence isolates (LeBlanc et al.
2010).
1.9 Prevention and management strategies
Multiple strategies have been proposed and recommended to maintain, but preferably prevent the outbreak of L. salmonis at salmon farm locations within Canada, Norway, Scotland, and Ireland. Control strategies within the management of the salmon farm have been implemented mainly involving anti-louse drugs and but there are other methods gaining in use due to the development of drug resistance, and these can include functional feed additives and targeting virulence factors in the louse for vaccines. (Revie et al. 2003).
It is expected that a site will have an ongoing low level of lice regardless of prevention methods as it is a commonly occurring parasite of salmon, however management plans are put in place to prevent a sea lice outbreak that would lead to health concerns at the farm and further affect wild salmon in the area. Techniques including the removal of dead fish, cleaning equipment and fallowing are basic techniques that have been shown to reduce lice levels (Revie et al. 2003). Other good farm practices include separating salmon based on year class, regular louse checks and counts, use of parasiticides when there is a noticeable increase of lice counts and monitoring pesticide resistance to keep salmon louse levels at a level below welfare issues for the host (Revie et al. 2003). On a genetic level, there is evidence that certain Atlantic salmon families are susceptible to L. salmonis and C. elongatus. Specifically, Glover et al. (2007) has linked louse susceptibility to the major histocompatibility (MH) class II.
An increasingly growing management strategy of diseases in mariculture includes the use of immunostimulant and/or functional feed additives. This can include biological factors,
29 vitamins and probiotics to kill pathogenic diseases as well as to enhance the growth and quality of the cultured fish. With respect to salmon louse infections, there are currently several widely used immunostimulants on the market to protect fish against sea lice. The immunostimulants do this by enhancing the fish’s systemic and localized inflammatory systems to aid in initiating an immune response prior to or during a salmon louse outbreak (Burka et al. 2012).
Salmon farms widely administer the use of drugs as bath treatments or in-feed treatments as methods of sea lice control. Depending on the location, specific drugs of either type may be used. This allowance is based on environmental risk assessments of proven safety to the fish and the surrounding environment (Burridge et al. 2010). There are multiple advantages and disadvantages when it comes to the efficiency of bath and in-feed treatments. Bath treatments are beneficial when it comes to administering drugs to prevent sea lice because the entire net pen is exposed to the drug. However, drug release into the environment can lead to low dose exposure of other cages and potentially farms leading to the development of resistance to the drug or negative consequences for non-target species.
Several types of bath treatments are commonly used individually or in combination as they target different mechanisms and pathways of the sea lice. Organophosphates directly kill the mobile phases of the salmon louse by inhibiting acetylcholinesterase pathways (Burka et al.
1997). Pyrethoids cause initial paralysis leading to louse death. This type of drug stimulates neuronal cell sodium channels. Drug resistance has been developed to both of these compound classes in Norway, Scotland and Chile. Mutations have been observed in the structure of the sodium channels preventing the depolarization of these channels (Fallang et al. 2005). Along with resistance to this drug, when combined with cold-water temperatures, toxic effects to the fish have been reported. Lastly, another common drug used at salmon farms is hydrogen
30 peroxide. Although environmentally friendly in solution when dissociated to water and oxygen, this chemical can have toxic effects for both the farmers, production operators and fish. This depends on the water temperature during administration and the time of exposure (Grant 2001).
Hydrogen peroxide targets mobile lice at a 98% reduction to C. rogercresseyi and a 56% reduction against the chalimus stages (Bravo et al. 2010). At the mobile stage, hydrogen peroxide promotes louse detachment, however does not prevent the louse from reattaching to another host causing further infection. Each type of bath-treatment drug provides suitable elements as a therapeutant against sea lice infections. However, drug resistance and the potential for further infection is a common theme among all the drugs described.
In-feed drug treatments involve the administration of drug-coated feed. Since the drug is lipid soluble, they can be in effect for up to 2 months with the unused drug being slowly excreted. Unlike bath treatment, in-feed treatments are less of an environmental risk and are less toxic to the fish and the farmers administering the drug. A disadvantage is if fish are already diseased or stressed, they may not be willing to feed causing an insufficient drug dose. Further, these low-feeding fish may also drive resistance in the louse due to under-dosing (Burka et al.
2012).
Several examples of in-feed drug treatments include growth regulators and avermectins.
Growth regulators prevent moulting by directly targeting chitin-synthesis pathways. This drug is most effective on the larval stages on sea lice while moulting requires the majority of energy.
Once the lice have moulted to the adult stages, growth regulators have no effect. A disadvantage of using growth regulators in a net pen farm site is the potential for the drug to target the chitin- synthesis pathways of other crustaceans that are not the target of the drug. A major advantage of growth regulators is there have yet to be any reports of resistance to these drugs (Aaen et al.
31 2015). Avermectins are the most widely used in-feed drug treatment and target glutamate-gated chloride channels in neuromuscular tissues, which lead to louse death. Ivermectin and emamectin benzoate are within the avermectin drug category, however ivermectin was generally not used due to the neurotoxic effects on fish (Merck 1988; Sanderson et al. 2007). On the other hand, emamectin benzoate has been a much safer and less toxic drug to fish, as the drug does not accumulate in the brain. Emamectin benzoate also poses much less of an environmental threat.
However, wide spread resistance to this drug has occurred in the salmon industry globally
(Igboeli et al. 2012). Due to these resistance concerns, other drugs are in formulation to find an alternative to emamectin benzoate as an in-feed drug treatment.
Prevention and management strategies for Moritella viscosa are difficult to combat and antimicrobial treatment is not always effective. Commercial vaccines that target other known bacterial infections are available, however are not fully effective. Over a third of antibiotics administered to fish in Norway were targeting winter ulcer disease (Bruno et al. 1998). Oral antimicrobial therapeutics, oxolinic acid, and a fluorinated thiamphenicol derivatives have also been studied, however are not effective against winter ulcer disease and it is suggested that antimicrobial therapy may be an inappropriate method in management (Coyne et al. 2004; Coyne et al. 2006; MacPhee 2001). Since M. viscosa is prevalent at cold water temperatures, fish tend to have difficulty fighting the infection and taking in enough feed and more susceptible to stressors.
Alternatively, when water temperature rises above 10°C, the disease is no longer prevalent.
Similar to winter ulcer disease, infectious salmon anemia has no effective treatments and only a handful of licensed vaccines, Elanco Forte V II and Pharmaq Alpha Ject®️ micro 3
(Chile) being several currently available. Efforts are being made to increase longevity of the vaccinations past the first year of production. To manage outbreaks in the European Union,
32 legislation promotes slaughter and disinfection of farms, strict biosecurity of fish and worker movement, farm sites to be at least 3 km apart, daily collection of dead fish, ultraviolet disinfections, sanitation of well boats and farm sites under surveillance (Lyngrøy 2003).
1.10 Functional diets
Functional diets were introduced to the aquaculture industry to meet the nutritional needs while also strengthening the immune system to reduce infection and mortality. Administering vaccines, antibiotics and other drugs is expensive to the industry. Although feed is also high in cost to the industry, it is a necessity. Additive stimulants aim to positively affect the immune system against common aquacultural pathogens, such as sea lice, and to promote disease prevention that is crucial to the industry. Bioactive compounds, including the addition of fatty acids, such as omega-3 and omega-6, are dosage and life stage specific to fish (Doñate et al. 2010; Tacchi et al.
2011). Omega-3 fatty acids have immunosuppressive and anti-inflammatory effects on juvenile
Atlantic salmon, while increased amounts of omega-3s have been shown to increase erythrocyte cell wall strength and lower Vibrio salmonicida infection (Erdal et al. 1991). Excessive amounts of omega-6 fatty acids increases pro-inflammatory eicosanoid production above optimal levels, increasing inflammation and inflammatory related diseases. In humans, a high omega-6/omega-3 ratio promotes the onset and pathogenesis of diseases including cardiovascular disease, inflammation, autoimmune diseases and cancer, while a lower omega-6/omega-3 ratio has a decreased risk of the aforementioned diseases (Simopoulos 2002). Other common additives in functional diets include prebiotics, probiotics, immunostimulants, and vitamins.
Immunostimulants, in particular, have been widely used in many vertebrate species and have been shown to reduce the attachment and infection of sea lice in fish (Refstie et al. 2010).
33 Immunostimulants contain pattern associated molecular patterns (PAMPs) that trick the host’s pattern recognition receptors (PRRs) into thinking the molecule is from a foreign pathogen, as
PAMPs are typically released by microbial pathogens. This activation initiates a signaling cascade to promote the release of pro-inflammatory cytokines to mimic the response of a heterologous invader, further promoting phagocytosis and the complement cascade (Cuesta et al.
2008). A typical immunostimulant used in aquaculture are β-glucans, naturally found in the cell walls of yeast, fungi, plants and bacteria (Petit and Wiegertjes 2016). A study by Guselle et al.
(2009) showed β-glucans suppress the infection of bacterial and microsporidial pathogens in salmonids. β-Glucans have also been shown to modulate innate immunity and complement in fish. However, Covello et al. (2012) reported no differences in lice counts between control fish and those with β-glucan stimulated diets.
Immunostimulatory diets may also contain another PAMP, unmethylated DNA, containing cytosine-phosphate-guanine oligodeoxynucleotide motifs (CpG ODN). CpG motifs in the unmethylated form found in bacterial and viral DNA act as a foreign signal by amplifying
TLR9 causing a trigger in antibacterial and antiviral cell signaling (Iliev et al. 2013). CpG motifs are also found in vertebrate hosts but are methylated and do not trigger toll-like receptor activation (Iliev et al. 2013). The same study by Covello et al. (2012) reported a decrease in lice infection and counts in the CpG ODN diet compared to controls. The reduction in numbers attributed to systemic inflammatory responses, particularly at the louse attachment site. Purcell et al. (2013) exposed Atlantic salmon to L. salmonis showing a reduction in lice infection when fed a diet containing CpG ODN.
34 CHAPTER 2: Wound healing and immune responses of Atlantic salmon
(Salmo salar) during co-infection with Lepeophtheirus salmonis and Moritella viscosa
Carvalho, Laura A.1, Whyte, Shona K.1, Braden, Laura M.1, Purcell, Sara L.1, Manning,
Anthony, J.2, Muckle, A.1, Fast, Mark D.1*
1Department of Pathology and Microbiology, Atlantic Veterinary College, University of Prince
Edward Island, Charlottetown, PE, Canada
2 The New Brunswick Research and Productivity Council (RPC), Food Fisheries and
Aquaculture Department, Fredericton, NB, Canada
*Corresponding author
Mark D. Fast Associate Professor Department of Pathology and Microbiology Atlantic Veterinary College – University of Prince Edward Island 550 University Avenue Charlottetown, PE, Canada C1A4P3 [email protected]
35 Abstract
This study was conducted to determine the pathogenic effects of a co-infection with M. viscosa at different exposure levels of sea lice in Atlantic salmon. Host wound healing and immune responses were studied throughout the challenge period as copepodids were introduced at medium (Mv+LsM) and high exposure (Mv+LsH) levels and salmon sampled over time (7, 12, 20,
46 days post lice infection, dpLs). M. viscosa (1.14x106 cfus/ml) was introduced to all experimental tanks at 10 days post lice infection (dpLs). Mean lice counts decreased over time in both Mv+LsM (31.5 ± 19.0 at 7 dpLs; 16.9 ± 9.3 at 46 dpLs) and Mv+LsH (62.0 ± 10.8 at 7dpLs;
37.6 ± 11.3 at 46 dpLs). There was significantly higher mortalities and more severe skin lesions in the co-infected group with high lice densities (Mv+LsH) compared to medium-infected group
(Mv+LsM) or M. viscosa only infection (Mv+Ls-). Serum analyte analysis revealed an increase in potassium in all groups over the course of the study. Quantitative gene expression analysis by
RT-qPCR was used to elucidate host responses to single infections and co-infections in skin tissue. There was significant up-regulation in Mv+LsH consistent with the wound healing genes il-8, mmp-9, hep, and saa. Skin lesions observed throughout the study were positive for M. viscosa through bacteriology, but these were rarely located in regions associated with lice (i.e. posterior to the dorsal and anal fin, posterior region of the head). Furthermore, M. viscosa was rarely identified in non-lesion skin samples by RT-qPCR. This suggests that while M. viscosa infection itself may increase potassium levels, impact osmoregulatory function and induce skin lesion development in salmon, co-infection with high numbers of lice can enhance this impact and significantly reduce the ability of these lesions to resolve, resulting in increased mortality.
36 2.1 Introduction
Fishes have an intimate relationship with their external environment that exposes them to a wide range of pathogens, including parasites and bacteria. In their natural environment, fishes are frequently suspected to be infected with a succession of different pathogens at any given time, resulting in co-infections of the host (Cox 2001). Fishes are often able to repulse pathogenic organisms before they are able to establish and initiate disease through the activation of the innate and adaptive immune systems (Workenhe et al. 2010). In the case of successive infections, one pathogen may alter the host immune response allowing for easier colonization by another pathogen; and in concurrent infections, the combination of pathogens may have additive or synergistic impacts on the health of the fishes (Cox 2001). Despite their importance and suspected commonality, our understanding of the impact of co-infections in teleosts is limited, as most studies focus on single pathogen interaction with their host.
Sea lice, such as Lepeophtheirus salmonis, mainly infect salmonids and significantly impact the aquaculture industry due to high density of hosts and limited successful treatment interventions. Disease- and treatment-related economic losses of L. salmonis to marine salmoniculture have been estimated at nearly 500 million USD in Norway alone (Abolofia et al.
2017), and therefore most likely exceeds 1 billion USD worldwide (FAO 2016). Adult female lice are aggressive feeders of blood, mucus and skin, while males and earlier developmental stages primarily feed on mucus and skin tissue (reviewed in Burka et al. 2012; Fast 2014). L. salmonis is known to secrete enzymes that aid in feeding, digestion, and modulation of the host immune response (Fast et al. 2005; Firth et al. 2000; Wagner et al. 2008). But this is likely a common strategy across ectoparasite copepods, as González et al. (2015) observed that as few as
6 adult lice, of another species of Caligid copepod (Caligus rogercresseyi), per fish may
37 negatively impact the immune function and physiology of Atlantic salmon weighing approximately 80 g. The immunomodulation by anti-parasitic defense mechanisms interferes with immune cells and signaling pathways against both L. salmonis and C. rogercressyii (Jodaa
Holm et al. 2016); similar well-established strategies are also observed in the evasion of parasites of mammalian hosts (Hewitson et al. 2009).
The physiological and economic cost of sea lice in salmon aquaculture may be compounded by the impacts of co-infection. Holm et al. (2016) reported an impact on genes involved in innate antiviral responses in skin from L. salmonis-infected Atlantic salmon suggesting reduced ability to fight off viral infections when pre-exposed to L.salmonis. Barker et al. (2018) further demonstrated this effect, whereby infection with L. salmonis significantly reduced salmon survival following infectious salmon anemia virus (ISAv) exposure. Similarly, a recent study by Figueroa et al. (2017) of Atlantic salmon co-infected with Piscirickettsia salmonis and C. rogercreesyi reported reduced vaccine efficacy, as observed by decreased survival, reduced growth, greater bacterial load and increased pathological changes in fish at sea cage sites.
Outbreaks of Moritella viscosa, the causative agent of winter ulcer disease, is responsible for elevated antibiotic use and mortalities in Atlantic Canada salmon culture (MacKinnon 2017).
M. viscosa is a gram-negative bacterium that generally affects fish in seawater below 7-8°C.
(Lunder 1992). Grove et al. (2010) documented two phenotypically and genetically different clades of M. viscosa. The variant strain has host specificity to Atlantic salmon, while the typical strain appears to have no disease effect on Atlantic salmon and rainbow trout, even though both strains appear relatively homogenous (Grove et al. 2010). The variant strain of M. viscosa may disrupt the cytoskeleton of the host cells, altering epithelial structure followed by pore formation
38 and cell lysis in vitro (Tunsjø et al. 2009). Characteristic ulcers are formed during M. viscosa infection progressing from scale loss and mild inflammation to severe ulcers exposing underlying muscle tissue, hemorrhage of internal organs and peritoneal ascites and, in some cases, septicemia (Benediktsdóttir et al. 1998; Karlsen et al. 2014). The skin has been suggested as the primary site of colonization and proliferation of M. viscosa as the ulceration resulting from infection occurs directly on the skin surface (Karlsen et al. 2012). The host skin is also the primary site of infection by the parasite L. salmonis. Furthermore, both pathogens produce secretory/virulence factors that impact inflammation and can modulate the site of interaction and the host response in their presence (Fast et al., 2007; Bjornsdottir et al., 2009).
Despite the relative importance of both L. salmonis and M. viscosa in Atlantic salmon aquaculture, there is little known about the impact of both organisms occurring simultaneously.
Mortality patterns and transcript abundance of several genes involved in wound healing, antigen presentation and lymphocyte response in the skin of salmon infected with either M. viscosa alone, or in a co-infection of M. viscosa and L. salmonis were quantified. A complete understanding of pathogenesis and host response may aid in developing countermeasures to prevent and manage outbreaks at farm sites.
2.2 Methods
2.2.1 Fish maintenance and husbandry
Atlantic salmon smolts (ca. 90 g; n = 360) were purchased from Cooke Aquaculture Inc. and transferred from Oak Bay Hatchery, New Brunswick (NB) to the Aquatic Biological
Containment Level II Facility at the Atlantic Veterinary College, Charlottetown, PE. Smolts were individually identified with passive integrated transponder (PIT) tags and were randomly distributed to 9 tanks (n = 40 /tank; 170 L water volume) and maintained at 33 ± 3 ppt in a semi-
39 closed, recirculation aquatic system. Over the course of the study, fish were maintained under a
14h:10h light:dark photoperiod and hand-fed to satiation twice daily (Skretting Spirit Supreme
Saltwater diet).
2.2.2 L. salmonis infection
Ovigerous (egg-bearing) L. salmonis were collected from harvested Atlantic salmon in the Bay of Fundy region, New Brunswick, Canada. Sea lice were transported on ice in aerated coolers to the Atlantic Veterinary College where egg strings were removed from gravid females and placed in a hatching system with light aeration until eggs hatched and developed to the infectious copepodid stage (ca. 9 days at 11oC). Prior to exposure, water was turned off to all tanks and the water level was reduced to below tank outflows. Fish were challenged with three exposure levels on the first day of the study: medium infection dose of 50 copepodids/fish (LsM) and high infection dose of 100 copepodids/fish (LsH), and 0 copepodids/fish (Ls-). This resulted in three groups, in triplicate; M.viscosa and medium lice exposure (Mv+LsM), M.viscosa and high lice exposure (Mv+LsH), and a single infection with M. viscosa only (Mv+Ls-). Fish in tanks receiving a single infection of M. viscosa were treated in the same way as fish in tanks exposed to copepodids. Supplemental oxygen was added during the sea lice infection procedure to maintain oxygen between 6.0-9.0 mg/L for a 2-hour duration. Fish exposed to sea lice exhibited behaviour associated with sea lice infections, including flashing, rubbing and jumping. After the 2-hour exposure, tank in-flows were re-established.
40 2.2.3 Preparation of M. viscosa inoculum and bacterial challenge
The Moritella viscosa isolate used in this study (Elanco Animal Health, Charlottetown, PE,
Canada) was originally isolated from Atlantic salmon from the Bay of Fundy, NB in 2012. M. viscosa cultures were produced by RPC (Fredericton, NB). M. viscosa was cultured in Tryptone
Soya Broth (TSB-2; 2% NaCl final concentration), at 8 ± 1°C in 2 L flasks with agitation. The cultures were pooled together and allowed to reach the bath challenge temperature (~10°C) before use. Atlantic salmon were challenged with M.viscosa 10 days post lice infection (dpLs) by combining all the fish from each of the three replicate tanks of the same L. salmonis infection group (Mv+Ls-, Mv+LsM, Mv+LsH; n = 3 tanks per group) into three Xactics™ holding tanks.
Fish were exposed to Moritella viscosa at 1.14 x106 colony-forming units per ml for 1 h with supplemental oxygen. The concentration of bacteria to induce a target mortality level of <30% had been informed by an in vivo challenge model developed independently at the Fisheries and
Oceans Canada, St. Andrew’s Biological Station (DFO-SABS) in St. Andrew’s, NB, Canada
(Manning et al. 2015). Following a one-hour bath treatment, fish were returned to their respective experimental tanks. The initial temperature was 10.7 ± 0.1oC, however, after 15 days post-M. viscosa challenge (dpMv), the temperature was decreased to 9.0 ± 0.04oC, and again at
22 dpMv to 7.0 ± 0.1oC to prevent the fish from resolving the bacterial infection and to maintain a steady level of disease progression throughout. Mortalities were removed upon detection.
Moribund fish, which had the inability to maintain normal dorsoventral position in the water, were removed and euthanized (tricaine methanesulfonate overdose: 250 mg/L). Fish weight, length, lesion location, lesion grade, lice number and stage were recorded. Bacteriological culture swabs were collected from the leading edge of the lesion and from the head kidney for all mortalities.
41 2.2.4 Sampling
Fish (n = 4 per tank) were sampled prior to the Moritella viscosa challenge at 7 dpLs (Figure
2.1). Subsequent sampling periods occurred at 12, 20, 46 dpLs (n = 8 fish per treatment).
Treatment groups included: no M. viscosa and no L. salmonis (Mv-Ls-), medium lice infection
(Mv-LsM) and high lice infection (Mv-LsH). Fish were fasted for 24 h prior to sampling. Four fish per tank (n = 12 per group) were opportunistically selected and euthanized by overdose with tricaine methanesulphonate (TMS, 250 mg/L). Fish weight and length were recorded for each fish. Whole blood (serum) was collected for osmolality determination and swabs from head kidney and skin lesions were collected for bacteriological analysis. Samples of head kidney, skin at louse attachment sites and corresponding gill arches were collected from each fish for gene expression. It is important to note that skin tissue for qPCR assessment was collected from a site posterior to the dorsal fin to assess the impact of sea lice infection (Braden et al. 2012). If a suspected Mv lesion was present at the sampling location, reference skin tissue was collected behind the pectoral fin so that sampled skin tissue did not include the M. viscosa lesion site as
Mv lesions were used for culture confirmation of M. viscosa. Samples were immediately placed on dry ice and subsequently stored at -80°C; replicate samples of all tissues were collected for histological analysis and stored in 10% buffered formalin. Only skin samples collected from lice attachment sites behind the dorsal fin were analysed by qPCR in this study. Finally, the number and stage of L. salmonis and presence and grade of lesion(s) were quantified.
42
Figure 2.1. Schematic outlining the design of the single and co-infection trials. Three exposure groups (40 fish/group) in triplicate. Following infection with L. salmonis from 1 to 10 dpLs experimental groups consisted of: no M. viscosa, no lice (Mv-Ls-); medium lice infection (Mv- LsM); high lice infection (Mv-LsH). Fish were further challenged with M. viscosa at 10 dpLs, after which the experimental groups consisted of: single infection M. viscosa, no lice (Mv+Ls-); medium co-infection (Mv+LsM); high co-infection (Mv+LsH). Sampling occurred at 7, 12, 20 and 46 dpLs.
2.2.5 Lesion grading
Skin lesions were classified based on severity, frequency and size using grading categories from
Manning et al. (2015). Briefly, Atlantic salmon showing no signs of infection from M. viscosa and/or L. salmonis had a grade index of 0. Grade 1 lesions were characterized by minor scale loss, epidermal and dermal erosion, and minor petechial hemorrhaging. Grade 2 lesions were characterized by minor scale loss, epidermal and dermal erosion, and moderate petechial
43 hemorrhaging. Grade 3 lesions included dermal ulceration along with visible muscle tissue and were further classified by size based on 1 cm progressions of the longest axis of the ulcer.
2.2.6 Osmolality analysis
Whole blood (n = 144) was collected in lithium heparin vacutainers (Becton Dickinson) and placed immediately on ice. Serum was separated by centrifugation within 2 hours of collection and stored at -80°C until analysis. Serum samples, corresponding with fish samples analyzed for gene expression by qPCR, were submitted to Aquatic Diagnostic Services, Atlantic Veterinary
College for determination of osmolality analysis. Ion selective electrodes were used to measure sodium, chloride and potassium levels in the serum using Cobas® p 501 (Roche Diagnostics,
North America).
2.2.7 Bacteriology
Lesions and surrounding skin were topically disinfected with 70% ethanol. An incision was made with a sterile scalpel to lift the epidermis and swab the subcutaneous tissue followed by transfer onto trypticase soy agar (TSA) plates with salt. Plates were incubated and evaluated for growth of M. viscosa by colonization.
Head kidney and skin samples were processed by RPC (Fredericton, NB) to confirm the presence of M. viscosa. DNA was extracted using the DNeasy Blood and Tissue kit (Qiagen) and quantified using Nanodrop® ND-1000 (Nanodrop Technologies). The real-time PCR assay was conducted using primers and probe which target the tonB gene of M. viscosa (GenBank
EU332345) as specified in Grove et al. (2008). Samples were analyzed in duplicate in 25 µL reactions using 2x TaqMan Universal PCR Master Mix (Life Technologies) on a BioRad CFX96
44 Real-Time PCR machine. Amplification conditions followed those determined in Løvoll et al.
(2009) using a modified TaqMan® probe-based qPCR assay targeting the tonB gene of M. viscosa (Grove 1999). DNA from the original M. viscosa isolate from the challenge culture was used as a positive control; Aliivibrio wodanis and negative reagents were used as negative controls. The assay was analyzed on a CFX96 system (Bio-Rad Laboratories Inc., US) and was allowed to run for 45 cycles.
2.2.8 RNA isolation and cDNA synthesis
Total RNA was isolated from ~100 mg of skin tissue (n = 110) using Tri-reagent and a modified phenol-chloroform method (Chomczynski and Sacchi 1987). The quantity and quality of RNA was determined using a NanoDrop 2000 Spectrophotometer (Thermo Scientific). The integrity of the RNA samples was assessed by visualization on a 3% agarose gel. To eliminate potential genomic DNA, all samples were treated with Turbo DNase DNA-free™ kit (Ambion, Austin,
TX, US). First strand synthesis was performed with 750 ng of RNA using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Cheshire, UK) in 40 μl reactions.
Negative reverse transcriptase (noRT) samples were generated by pooling up to 12 individual
DNase-treated RNA samples. In the case where a pooled noRT showed genomic DNA contamination all total RNA samples making up the pool were retreated with DNase and tested again to ensure there was no gDNA. Approximately 2 μl of each sample was added to a cDNA pool to generate standard curves. Each cDNA sample was further diluted by 1:5 for gene expression analysis. cDNA was stored at -20ºC until downstream analysis.
45 2.2.9 Reverse Transcription - Quantitative PCR
Reverse transcription quantitative PCR (RT-qPCR) was performed with the SsoAdvanced™
Universal SYBR® Green Supermix (Bio-Rad Laboratories Inc., US) in triplicate using clear 96- well Multiplate® PCR Plates™ (Bio-Rad Laboratories Inc., UK) on a CFX Connect™ Real-
Time qPCR System (Bio-Rad Laboratories Inc., US). The thermal profile was as follows: an initial denaturation step at 95°C for 30s, followed by 40 cycles of annealing at 95°C for 15 s and extension at 60°C for 15 s, with fluorescence detections following each 60°C step. Following this protocol, melt curve analyses were performed at 0.5°C intervals from 65°C to 95°C for 5 s.
Primers were diluted to a concentration of 10 mM and 1 μl of template was used, resulting in a total reaction volume of 11 μl. For every primer pair, a 6-point 10-fold serial dilution was performed to determine the efficiency, with all assays between 86.4-109.1%. No template controls (NTC) were included on every plate. An internal positive control (IPC) was used as a calibrator on each plate to compare individual plates to each other. Three reference genes were included in the study: ribosomal protein S20 (RPS20), eukaryotic initiation factor 3 subunit 6
(eIF3s6) and eukaryotic translation elongation factor 1-alpha (EF1αAb) (Table 2.1). Only
RPS20 and eIF3s6 met the stability requirements for inclusion in the analyses, with a geNorm
M-value and coefficient of variation value of 0.584 and 0.246, respectively (Vandesompele et al.
2002). Raw data was used as input into qBASE+ that generates calibrated and normalized relative quantities using sample-specific normalization factors (Hellemans et al. 2007). Genes of interest included those known to be activated during L. salmonis infection, as well as other cellular and humoral genes involved in the innate and adaptive immune response (Table 2.1).
46 Table 2.1. Primers used for RT-qPCR, indicating forward and reverse sequences, efficiency values, accession number, amplicon size and original reference.
EFFICIENCY GENE TARGET NAME ACCESSION # REFERENCE FORWARD (5-3') REVERSE (5-3') (%) AMPLICON Collagen type-1 α COL1α XM_014209886.1 Purcell et al., 2012 TCAAGGCAGAGGGAAACAAC ACACTGTCTTGCCCCATTTG 94.1 87 Cluster of 87.0 differentiation 4 CD4 EG852912 Skugor et al., 2008 TGCATTGTTCCTCTCTTCCACAGC CCGTCCCAAGGTACCATAGTACCAA 128 Cluster of 97.6 differentiation 83 CD83 microarray Sutherland et al., 2014 TCTTTCCATTGTGGGGATTG TTGGCCCCAGAGTGTAAATC 164 C-type lectin CLEC4M2 microarray Sutherland et al., 2014 CTCTGCTTACGCCTGGGTAG TTTCGGACCAGAACCAATTC 86.8 95 Hepcidin-1 Hep NM_001140849.1 Sutherland et al., 2011 TTCAGGTTCAAGCGTCAGAG AGGTCCTCAGAATTTGCAGC 93.3 125 Immunoglobulin M IgM S48652.1 Braden et al., 2015 TCATTAAGATCACCCCGCCG TTCCTCGACATCGCACACAA 97.7 77 Immunoglobulin T IgT GQ907004 Tadiso et al., 2011 CAACACTGACTGGAACAACAAGGT CGTCAGCGGTTCTGTTTTGGA 86.4 97 Interleukin 1β-1 IL-1β AY617117 Fast et al., 2004 CGTCACATTGCCAACCTCAT ACTGTGATGTACTGCTGAAC 95.1 200 Interleukin 8 IL-8 NM_001140710.2 Braden et al., 2012 GAATGTCAGCCAGCCTTGTC TCCAGACAAATCTCCTGACCG 90.7 226 Interleukin 12 IL-12 AJ548830 Skugor et al., 2008 TCTACCTACACGACATTGTCCAGCC ATCCATCACCTGGCACTTCATCC 93.1 62 Major 92.0 histocompatibility class II β-chain MH II X70166 Braden et al., 2012 AAGGCTTGAAGACACGTTGC CAGTCCAGCAGTAACGTCCA 106 Matrix 96.9 metalloproteinase 9 MMP9 CA342769 Skugor et al., 2008 ACTCTACGGTAGCAGCAATGAAGGC CGTCAAAGGTCTGGTAGGAGCGTAT 72 Serum amyloid-A SAA NM_001146565.1 Braden et al., 2015 CTCGGGGCAACTATGATGCTA AGTCCTCATGTCCTCGACCA 108.6 114 Reference Genes Eukaryotic translation 98.2 initiation factor-3 subunit-6 eIF3s6 CX040383 Skugor et al., 2008 GTCGCCGTACCAGCAGGTGATT CGTGGGCCATCTTCTTCTCGA 92 Ribosomal protein s20 RPS20 BG936672 Olsvik et al., 2005 GCAGACCTTATCCGTGGAGCTA TGGTGATGCGCAGAGTCTTG 109.1
47 2.2.10 Statistical analysis
To analyze mortality data, a Kaplan-Meier survival estimate was generated in Microsoft Excel
(2011). Ion osmolality concentration was analyzed by ANOVA (p < 0.05) and a Tukey’s honest significant difference (HSD) test. Analyses were performed in RStudio 1.0.143 (RStudio, Inc.,
Boston, MA). An analysis of variance (two-way, p < 0.05) followed by a post-hoc Tukey’s HSD was performed on resulting gene expression data to determine significance across time (12, 20,
46 dpLs) and lice exposure levels (Mv+Ls-, Mv+LsM, Mv+LsH). Samples at 7 dpLs were not included in this analysis due to an imbalance of treatment groups for accurate statistical testing.
An ANOVA (one-way, p < 0.05) was performed to determine significance of single and co- infection status within the first time point of infection with L. salmonis and/or M. viscosa (7 and
12 dpLs). A principal component analysis (PCA) was constructed to summarize variables (fish weight, fish length and genes of interest) into a linear model to determine any correlation
(visualized in Rstudio, using R 3.5, ggbiplot package).
2.3 Results
2.3.1 Atlantic salmon survival
Mortalities were first observed in the single infection of M.viscosa (Mv+Ls-) and the medium co-infection (Mv+LsM) at 6 days post M. viscosa challenge (dpMv; 16 dpLs). At 24 h post water temperature was decreased from 10.7 ± 0.1oC to 9.0 ± 0.04oC. In contrast, mortalities in the high co-infection group began at 8 dpMv/18 dpLs. Gross pathology of mortalities included multiple lesions of varying severity on the flank and dorsal surface, with scale loss, caudal fin erosion and ulceration and necrosis of the eye and nares. There were 6 total mortalities in the single infection (Mv+Ls-) fish resulting in 7.1% cumulative mortality (CM), 14 mortalities
(16.7% CM) in Mv+LsM and 28 mortalities (33.3% CM) in Mv+LsH (Figure 2.2). The final
48 mortality in the single infection fish occurred at 32 dpLs, unlike both the medium and high co- infection groups where the final mortality occurred at 44 dpLs. Using a Kaplan-Meier survival estimate, the high co-infection group had a significantly higher probability of mortality compared to the medium co-infection group (p = 0.024) and the single infection group (p =
0.028) (Figure 2.3).
Mv+Ls- Mv+LsM Mv+LsH 100 Control Medium Lice High Lice
80
60
40 10.7 ± 0.1oC 9.0 ± 0.04oC 7.0 ± 0.1oC
20
PERCENT (%) CUMULATIVE MORTALITYCUMULATIVEPERCENT(%) 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 DAYS POST-CHALLENGE WITH M. VISCOSA
Figure 2.2. Percent cumulative mortality (%) in the single infection (Mv+Ls-), medium co- infection (Mv+LsM), and high co-infection (Mv+LsH) groups. Temperature changes indicated by dashed lines. Temperature at the start of the trial was 10.7oC and was decreased to 9.0oC at 15 days post- M. viscosa challenge (15 dpMv; 25 dpLs), and again to 7.0oC (22 dpMv; 32 dpLs). Sampling occurred at 7, 12, 20 and 46 dpLs.
49 Mv+Ls- Mv+LsM Mv+LsH
Figure 2.3. Kaplan-Meier survival plot for condition estimates of the single infection (Mv+Ls-), medium co-infection (Mv+LsM), and high co-infection (Mv+LsH) groups.
2.3.2 L. salmonis infection
Louse attachment sites were characterized by scale loss and depigmentation of the epithelial tissue. Prevalence of infection was 100% in both co-infection groups with abundance in the high co-infection group approximately twice that of the medium co-infection group, suggesting similar attachment success rate per individual copepodid between the two exposure levels (Table
2.2). Lice counts and life stages were also recorded on dead and moribund fish, with higher total sea lice counts observed in the high co-infected fish compared to the medium co-infected fish
(Table 2.3).
50 Table 2.2. Average lice counts in single, medium co-infected and high co-infected groups over time (7, 12, 20, 46 dpLs). Exposure Time A Time B Time C Time D (7 dpLs) (12 dpLs) (20 dpLs) (46 dpLs)
Single infection 0 ± 0 0 ± 0 0 ± 0 0 ± 0
Medium co-infection 31.5 ± 19.0 31.7 ± 11.7 32.8 ± 16.1 16.9 ± 9.3 High co-infection 62 ± 10.8 74.7 ± 28.4 65.6 ± 13.0 37.6 ± 11.3
Table 2.3. Lice counts for mortalities and euthanized moribund fish separated by exposure level and L. salmonis life stage. Exposure Chalimus Pre-adult/Adult males Adult females Water bath Total Lice Single infection 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 Medium co-infection 2.4 ± 8.3 20.2 ± 8.7 0 ± 0 3.8 ± 7.4 343 High co-infection 1.1 ± 4.2 45.7 ± 17.7 0.0 ± 0.2 5.4 ± 11.1 1566
2.3.3 Bacteriology
Moritella viscosa was detected in 40% of skin lesions (29/72) from sampled fish at 12 dpLs and
20 dpLs (Table 2.4). While M. viscosa was detected by bacteriological culture from skin lesions representing all groups, bacterial growth following swabs of the head kidney was only observed in one fish from the medium co-infection group at 20 dpLs. No bacterial growth was observed from the head kidney or skin lesions from any infection group at 46 dpLs. In comparison, M. viscosa was detected in only 3 skin samples by qPCR: one medium co-infected fish sampled at
20 dpLs, and two high co-infected fish sampled at 20 dpLs (Table 2.4). All other head kidney and skin tissue samples were negative when tested with qPCR. Bacterial growth from swabs collected from skin lesions of mortalities revealed positive growth in 100% of fish from the single infection and high co-infection groups and positive growth in 75% of the medium co- infection group (Table 2.5).
51
Table 2.4. Confirmation of M. viscosa by culture on tryptone soy agar plus salt and by qPCR from the reference skin site (non-M. viscosa lesion site) and head kidney in sampled fish from each exposure group at 12, 20 and 46 days post-infection with L. salmonis (dpLs). Time Exposure Sampled fish qPCR Skin Head kidney Skin Head kidney 12 dpLs Single infection 6/12 0/12 0/6 0/6 Medium co-infection 6/12 0/12 0/6 0/6 High co-infection 3/12 0/12 0/6 0/6 20 dpLs Single infection 2/12 0/12 0/6 0/6 Medium co-infection 8/12 1/12 1/6 0/6 High co-infection 4/12 0/12 2/6 0/6 46 dpLs Single infection 0/12 0/12 0/6 0/6 Medium co-infection 0/12 0/12 0/6 0/6 High co-infection 0/12 0/12 0/6 0/6
Table 2.5. Confirmation of M. viscosa by culture on Tryptone Soy Agar plus salt from mortalities. Exposure Skin Head kidney
Single infection 7/7 5/7
Medium co-infection 6/8 7/12
High co-infection 14/14 21/30
2.3.4 Lesion grading
Lesion formation was recorded during sampling at 12, 20 and 46 dpLs. Lesions were predominantly present on the lateral flanks, with some lesions around the mouth, and they were not associated with lice attachment (Figure 2.4). Severity of lesions appeared to increase with time: while only grade 1 lesions were observed in fish at 12 dpLs (20% prevalence), at 20 dpLs
31.3% of fish presented with grade 1 lesions, 15.6% exhibited grade 2 lesions, and 6.3% grade 3 lesions. Of the fish sampled at 20 dpLs, 18.8% lesion positive fish belonged to the single
52 infection group, 25.0% represented the medium co-infection group, and 9.4% the high co-
infection group. By 46 dpLs, the majority of lesions had resolved such that only 2 fish were
observed with grade 1 lesions.
A a b
B a b
C a b
Figure 2.4. Lesion grading scheme during sampling. A. a-b) Grade 1 lesions: arrows indicate areas of scale loss and slight epidermal erosion. B. a-b) Grade 2 lesions: petechial hemorrhaging apparent on lateral and ventral regions of the body. C. a-b) Grade 3 lesions: severe dermal ulceration with muscle tissue visible. Ninety-six (n = 96) mortalities or euthanized moribund fish euthanized were recorded as
having lesions. When assessing all mortalities, grade 1 lesions were most prevalent (19.8% of
fish), with fewer higher-grade lesions observed, i.e. 6.3% grade 2 lesions and 1.0% grade 3
lesions. Overall, euthanized moribund fish and mortalities removed from the single infection
53 tanks exhibited 26.3% grade 2 lesions and 10.4% grade 3 lesions. Moribund and mortalities from
the medium co-infection exhibited 22.2% grade 1, 5.3% grade 2, and 25.0% grade 3 lesions
whereas, mortalities from the high co-infection exhibited 77.8% grade 1, 68.4%, grade 2 and
64.6% grade 3 lesions. Severity of lesions increased in moribund/dead fish over the course of the
trial (Figure 2.5).
3 3 3 3 3 33 3 3 3 3 3 3 33 3 3 33 3 3 3 3 3 3 2.5
Mv L
2 Mv ML 2 2 2 2 2 2 2 2 2 Mv HL 1.5
1
Lesion Severity 1 1 11 1 1 1 1 1 1 11 1 0.5
0 1 2 3 4 5 6 7 10111214141516161718192021222223232424252527272828293030303131313232 1 2 3 4 5 6 7 8 9 101112131415161720212224242526262728293031323233333334353537373838394040404141414242 dpMv; dpLs Figure 2.5. Lesion severity by grade in moribund fish and mortalities for each experimental group over the study period; single infection (Mv+Ls-), medium co-infection (Mv+LsM), and high co-infection (Mv+LsH) groups. Numbers on the bottom row of the x-axis indicate days post L. salmonis infection (dpLs) and numbers on the top row indicate days post-M. viscosa challenge (dpMv). Grade of fish recovered on each day is shown in grey on bars. Multiple mortalities have a higher cluster of bars per day. Sampling occurred on 7, 12, 20 and 46 days post-infection of L. salmonis (dpLs).
The severity of lesions between treatment groups in mortalities was also analyzed. In the
single infection group, grade 2 lesions were the most frequent (20%) compared to grade 1 and 3
lesions. The medium co-infection group had a similar frequency of lesion severity per grade at
21.1% (grade 1), 20% (grade 2), and 21.3% (grade 3). The severity of lesions was much higher
54 in the high co-infection group (68.4%) compared to the single infection (60%) and the medium co-infection) (68.1%) groups. Grade 3 lesions were further divided based on the diameter of the longest axis: lesions less than 0.5mm, 0.5 to 2.0 mm, and greater than 2.0 mm. The majority of lesions (80%) in the single infection group were greater than 2.0 mm whereas 93.8% of lesions in the high co-infection group and all of the lesions in the medium co-infection were all greater than 2.0 mm.
2.3.5 Osmolality: Serum Analytes
There were no significant differences between the treatment groups with respect to sodium and chloride, however, there was a significant increase in potassium levels at 46 dpLs in all infection groups compared to all other time points (2.8 to 16.03 mmol 1-1; Table 2.6), with levels increasing 10-fold by the final sampling period.
Table 2.6. Concentration of serum analytes over time. Data is represented as mean ± standard deviation. Significance denoted by an asterisks (*). Serum analyte Time 7 dpLs 12 dpLs 20 dpLs 46 dpLs Sodium 166.3 ± 2.8 167.1 ± 2.9 170.0 ± 4.3 166.8 ± 17.9 (mmol/L) Chloride 135.0 ± 3.2 138.1 ± 3.0 143.4 ± 6.6 143.7 ± 20.6 (mmol/L) Potassium 0.7 ± 0.1 0.7 ± 0.1 0.9 ± 0.4 7.6 ± 3.3* (mmol/L)
2.3.6 Gene expression in skin over time and exposure level
All 13-immune related genes involved in wound healing, antigen presentation and lymphocyte response showed transcriptional changes in the single infection with M. viscosa and the medium
55 and high co-infections with L. salmonis and M. viscosa. To compare the impact of a single infection with either M. viscosa or L. salmonis and co-infection with both M. viscosa and L. salmonis, control salmon were sampled 3 d prior to bacterial challenge (7 dpLs). A significant up-regulation of interleukin 12 (il-12) and immunoglobulin mu (igm) expression was observed at
12 dpLs in the high co-infection fish compared to the control fish at 7 dpLs (Figure 2.6). At 12 dpLs, there was also significant up-regulation of igm in the high co-infection group compared to the control fish.
Figure 2.6. Gene expression in skin tissue samples (n = 59) of a) interleukin 12 (il-12), and b) immunoglobulin mu (igm). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) between treatment groups: no M. viscosa, no sea lice (Mv-L-); no M. viscosa, medium sea lice exposure (Mv-ML); no M. viscosa, high sea lice exposure (Mv-HL); M. viscosa, no lice (Mv+L-); M. viscosa, medium sea lice exposure (Mv+ML); M. viscosa, high sea lice exposure (Mv+HL). Treatments analyzed over time (7 and 12 dpLs). Letters and bars denote significant differences in expression (1-way ANOVA, p < 0.05).
2.3.7 Inflammatory and Acute Phase Genes
In general, expression of collagen 1-α (col1α) did not change over time (Figure 2.7a) with the exception of 46 dpLs, where expression was significantly down regulated both in medium and
56 high co-infection groups compared to the single infection group. Interleukin 8 (il-8) expression was significantly higher in the high co-infection group at 20 and 46 dpLs compared to 12 dpLs
(Figure 2.7b). While down regulation of il-8 was observed in the medium co-infection group between 12 and 46 dpLs. Interleukin 1β (il-1β) appeared to be up-regulated in the high co- infection group over time but expression was highly variable, and there was no significant overexpression of il-1β at any time point in any treatment group (Figure 2.7c). Expression of matrix metalloproteinase-9 (mmp-9) increased in the high co-infection over time compared to the single infection and decreased over time in medium co-infection group compared to single infection (Figure 2.7d).
Significant down regulation of hepcidin (hep) was observed at 46 dpLs compared to 12 dpLs in the medium co-infection group (Figure 2.7e). Serum amyloid A (saa) was strongly induced over time, with significant down regulation of saa observed in the medium co-infection between 12 and 46 dpLs, while there was a significant increase observed in the high co-infection group between 20 and 46 dpLs (Figure 2.7f).
57
Figure 2.7. Gene expression in skin tissue samples (n = 82) of a) collagen 1-⍺ (col⍺), b) interleukin 8 (il-8), c) interleukin 1β (il-1β), d) matrix metalloproteinase-9 (mmp-9), e) hepcidin (hep), and f) serum amyloid a (saa) with functions in inflammatory and acute phase responses. Values expressed as box plots, median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) of each treatment group: M. viscosa positive, sea lice negative (L-); M. viscosa positive, medium sea lice exposure (ML); M. viscosa positive, high sea lice exposure (HL). Treatments analyzed over time (12, 20, 46 dpLs). Letters and bars denote significant differences in expression (2-way ANOVA, p < 0.05).
2.3.8 Cellular effectors
The T lymphocyte marker cluster of differentiation 4 (cd4) was down regulated at 20 dpLs in the high co-infection group, followed by a significant up-regulation at 46 dpLs (Figure 2.8b).
Expression of cluster of differentiation 83 (cd83), a marker of antigen presenting cells, such as dendritic cells, was not affected in the single and medium co-infection groups; however, there was a significant increase in expression by 46 dpLs in the high co-infection group (Figure 2.8c).
Fish with a single infection of M. viscosa exhibited an increase in immunoglobulin tau (igt) over
58 time (12 dpLs to 46 dpLs). At 46 dpLs, fish in the medium and high co-infection groups had significantly lower igt expression compared to the single infected fish (Figure 2.9b). Expression of major histocompatibility class II beta chain (mhii) was significantly down regulated in the skin over time from 12 dpLs to 46 dpLs in the high co-infection group compared to the single infection group (Figure 2.9c). No differential expression was detected in c-type lectin 4 m
(clec4m2) between treatments or over time (data not shown). Principal component analysis, as well as heat map visualization, show clustering of these cellular effector genes together (mhii, igm, igt) as well as the inflammatory genes above distinctly clustering together (il-1β, il-8, mmp-
9, hep, etc.) over the course of the study (Figure 2.10).
Figure 2.8. Gene expression in skin tissue samples (n = 82) over time (12, 20, 46 dpLs) of a) interleukin 12 (il-12), b) cluster of differentiation 4 (cd4), and c) cluster of differentiation 83 (cd83). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) over time (12, 20, 46 dpLs). Letters denote significant differences in expression (2-way ANOVA, p < 0.05).
59
Figure 2.9. Gene expression in skin tissue samples (n = 82) of cellular marker genes a) immunoglobulin m (igm), b) immunoglobulin tau (igt) and c) major histocompatibility class II beta chain (mhii). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) of each treatment group: M. viscosa positive, sea lice negative (L-); M. viscosa positive, medium sea lice exposure (ML); M. viscosa positive, high sea lice exposure (HL). Treatments analyzed over time (12, 20, 46 dpLs). Letters and bars denote significant differences in expression (2-way ANOVA, p < 0.05).
60
Figure 2.10. a) Principal component analysis (PCA) of skin samples of Atlantic salmon co- infected with sea lice and M. viscosa showing the clustering of genes. Red represents similarities between expression profiles, while blue identifies variables with differentiated values. b) Hierarchical clustering of pro-inflammatory markers and cellular effector genes represented in a heat map. Expression is mapped based on mean log2 normalized intensities of each gene, with blue representing low expressing transcripts and red representing high expressing transcripts.
2.3.9 Expression profiles of fish positive for Moritella viscosa by culture
We further separated skin samples that tested positive for bacteriology and with a lesion severity score of at least grade 1 (n=11) for additional gene analysis. Samples meeting these criteria were observed in all groups at 20 dpLs. Expression profiles of all genes were similar, with inflammatory genes hep, il-1β, il-8, and mmp-9 downregulated in co-infected fish compared to single M. viscosa-infected fish, however, only expression of il-8 and hep showed significantly different expression (Figure 2.11).
61
Figure 2.11. Gene expression in skin tissue samples (n = 11) of a) interleukin 8 (il-8), and b) hepcidin (hep). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) of each treatment group: M. viscosa positive, sea lice negative (L-); M. viscosa positive, medium sea lice exposure (ML); M. viscosa positive, high sea lice exposure (HL). Treatments analyzed at 20 dpLs. Letters denote significant differences in expression (1- way ANOVA, p < 0.05).
2.4 Discussion
Understanding interactions between co-infecting pathogens at the physiological and immunological level is important for aiding in the prevention and management of disease in commercial salmonid aquaculture. Here we investigated the host response of Atlantic salmon during co-infection with the salmon louse, Lepeophtheirus salmonis and the causative agent of ulcer disease, Moritella viscosa.
Atlantic salmon smolts challenged with M. viscosa produced characteristic clinical signs of ulcer disease, including scale loss and skin lesions (Løvoll et al. 2009; Karlsen et al. 2012).
However, mortalities were spread over a much longer period of time. This might have been due to the low dose of bath-challenged M. viscosa (1.14x106 cfus/ml). This dose is lower in
62 comparison to a study by Bjornsdottir et al (2011), where the virulence of M. viscosa was determined by IP injecting nine selected isolates with concentrations ranging between 2x104 to
2x106 cfus/ml per fish. The isolate at 2x106 cfus/ml had the lowest virulence. Additionally, the slow onset of mortalities in the present study may have resulted from water temperatures above
10°C. Previous work has shown M. viscosa-associated clinical disease occurs at temperatures below 7-8°C and fish begin to recover from the infection when temperatures are higher than 8°C
(Lunder et al. 1995). M. viscosa has shown highest adhesion at 4°C compared to 10°C and 15°C
(Tunsjø et al. 2009). Moritella viscosa was recovered from epithelial ulcer swabs, corresponding to the direct skin colonization required for ulcer formation (Karlsen et al. 2012). However, using bacteriology, we detected M. viscosa in the head kidney of dead salmon, indicating that internally, the bacterium disseminated systemically. With respect to L. salmonis, development rates were consistent with previous studies, reaching the pre-adult stage by 20 dpLs at 9.0°C
(Johnson and Albright 1992). Characteristic pathology due to L. salmonis infection was observed, including scale loss, tissue erosion and mild inflammation at louse attachment sites.
Significant mortality was seen in the high co-infection group compared to the medium co-infection and single infection groups. While there is no evidence to suggest that direct feeding by L. salmonis is involved in the dissemination of the bacterium, this requires further investigation as L. salmonis has been implicated as a vector for other bacterial pathogens (Barker et al. 2009; Novak et al. 2016) and enhance the susceptibility to, or risks for, acquiring a viral infection (Jakob et al. 2011; Valdes-Donoso et al. 2013; Barker et al. 2019).
Attachment of L. salmonis to fish resulted in observations of sloughed scales and compromised epithelia. This may have provided a portal for M. viscosa to colonize. Lesions were observed in the mouth, nares, anterior and lateral regions of the head, the dorsal fin and
63 caudal peduncle, with the high lice exposure group having the greatest lesion severity. Multi- microbial communities on fish skin or in ulcers are likely to interact with each other (Karlsen et al. 2012). With respect to winter ulcer disease, Aliivibrio wodanis is often co-isolated with M. viscosa in the field. Interestingly, A. wodanis colonization of fish surfaces impedes the growth of
M. viscosa (Hjerde et al. 2015). This may reduce the virulence of M. viscosa and result in a more chronic presentation of the disease in the field unlike the more rapid and acute mortality observed in laboratory trials when M. viscosa is used alone (Karlsen et al. 2014). This is likely the reason for the mortalities observed in this study particularly in the co-infection with L. salmonis, in which there was a higher frequency of grade 3 lesions > 2.0mm.
L. salmonis is thought to immunomodulate, and in severe cases potentially immunosuppress, the host (Fast 2014). It is possible that in the high co-infection group, there may have been an additive or synergistic effect with L. salmonis modulating the host response and increasing the severity of the M. viscosa infection and mortality rate. This is also observed in other co-infections. For example, immunosuppression by R. salmoninarum in chinook salmon
(Oncorhynchus tshawytscha) is thought to facilitate infections with pathogenic motile
Aeromonas spp. (Loch et al. 2012). Atlantic salmon previously infected with L. salmonis have also shown increased susceptibility to Loma salmonae (Mustafa et al. 2000). Furthermore, infection by Ichthyophthirius multifilis enhances mortalities in channel catfish infected concurrently with A. hydrophila (Xu et al. 2013).
Osmoregulatory stress and variations in plasma chloride, sodium and potassium is well documented in L. salmonis infections (Dawson et al. 1999; Wagner et al. 2003; Wells et al.
-1 2006). In the present study, potassium levels were reported to be as high as 16.03 mmol 1 , which is substantially higher than previous reports (Bowers et al. 2000; Dawson et al. 1999). An
64 induction in iron regulation by hepcidin expression was seen at 20 dpLs followed by an up- regulation of expression by 46 dpLs, similar to earlier observations in the skin of L. salmonis- infected pink, coho and Atlantic salmon (Braden et al. 2015; Sutherland et al. 2014). The influx of potassium at 46 dpLs may be a result of nutritional imbalance from changes in iron homeostasis as the abundance of iron in the marine environment is limited and M. viscosa competes for iron with the host by producing siderophores (Bjornsdottir et al. 2011). This may be reflected in the overexpression of hepcidin, which is involved in iron homeostasis and regulation.
We assessed the immune response of infected skin tissue focusing on genes involved in inflammation and wound healing that are acutely and/or chronically stimulated in other skin infection models. Previous research by Tadiso et al. (2011) reported increases in acute phase proteins, immunoglobulins and matrix metalloproteases (MMPs) during the early transition from copepodid to chalimus life stages of L. salmonis. Similarly, we observed significantly elevated expression of immunoglobulins (igm) that was specific to L. salmonis-infected fish at 12 dpLs, indicating the upregulation of igm was likely due to a typical response to the chalimus life stage
(Tadiso et al. 2011).
Infection with L. salmonis has been associated with dysregulation of inflammatory responses and delayed wound healing in the skin of Atlantic salmon. A decrease in expression of inflammatory cytokine il-12 was observed in co-infected salmon which follows previous work showing a downregulation of innate inflammatory genes in the host during infection with the first chalimus stage (Fast 2014; Fast et al. 2006; Skugor et al. 2008). Strong up-regulation of mmp-9 and il-8 in co-infected fish were also observed in the high co-infection group by 46 dpLs.
65 A critical component of tissue remodeling and wound repair, MMP-9 has been shown to be induced during initial infection with L. salmonis, followed by a second induction once lice reach mobile stages as wound healing responses transition from acute to chronic in heavy infections (Fast 2014; Skugor et al. 2008). In salmon exposed to medium lice infection and M. viscosa, mmp-9 showed upregulation between 12-20 dpLs followed by a downregulation at 46 dpLs. However, in the high co-infection group, there was significant upregulation throughout the study. It is postulated that the host delays tissue regeneration until the pathogen has been cleared and dead tissue removed (Ingerslev et al. 2010). The continued upregulation in the high co- infection group may be due to the high lice burden alongside the continued presence of M. viscosa.
Collagen production in L. salmonis infected skin was upregulated at 20 dpLs, likely due to the moulting of chalimus to pre-adult followed by a downregulation of col1α by 46 dpLs.
These findings are in agreement with Matsumoto and Sugimoto (2007) as type I collagen activates the migration of keratocytes to the extracellular matrix for wound remodeling. It is possible that low water temperatures used to stimulate M. viscosa colonization during the infection trial slowed epithelial repair by keratocytes, allowing M. viscosa to effectively colonize and promote lesion formation (Åsbakk 2001). This is achieved by the bacterium forming a biofilm composed of exopolysaccharides to mimic teleost cell function and avoid or inhibit the innate immune system of the host (Karlsen et al. 2012). Delays in the healing cascade caused by the immunomodulatory capabilities of these pathogens may have contributed to the late wound healing responses observed by 20 dpLs in the high co-infection fish. Severe lesion development and inhibited inflammatory signaling was seen within the M. viscosa culture positive members of
66 the population, while an overexpression of inflammatory signals was seen in fish with no lesions and that were negative for bacterial growth and PCR.
Up-regulation of acute phase protein transcription, serum amyloid A and interleukin-8, later in the infection indicates a delayed inflammatory chemotactic signal produced to attract neutrophils and other leukocytes to the site of louse attachment, previously described in lice only infections of Atlantic salmon (Johnson and Albright, 1992; Fast 2014). Serum Amyloid A is produced by macrophages and is involved in chemotactic activity for T-cells and monocytes in response to pathogen-associated gene activation, stimulating production of cytokines and MMPs
(Mullan et al. 2006; Rebl et al. 2011). Interestingly, increased expression of cd83 and cd4 was also observed at 46 dpLs, regardless of treatment group. This would suggest that there was migration and/or, activation of dendritic cells and T-helper cells in the skin in response to either
M. viscosa or L. salmonis. As expression of igm, igt and mhii was locally suppressed in the skin of co-infected fish exposed to high levels of lice at 46 dpLs, cellular recruitment to the infection site in this group likely did not involve B-cells or other leucocytes that present antigen to CD4+
T-cells, but this would need further confirmation through use of cell specific markers.
All mortalities showed a high frequency of lesion scores over time, which may indicate animals with the inability to overcome stress and modulation of pathogens, due to limited/delayed inflammatory responses may have succumbed to infection. This would also explain higher inflammatory gene expression at 46 dpLs in the high infection group, if low responders from 20 dpLs had been removed from the population.
In conclusion, we observed fish that were positive for lesions and bacterial growth showed inhibited inflammatory signals in the skin at 20 dpLs in the high co-infection group, potentially due to greater immunomodulation from L. salmonis (Fast et al., 2004; 2007). By 46
67 dpLs, lesions showed signs of resolution and mortalities had slowed, suggesting the fish were able to resolve the M. viscosa infection by this time. In fish negative for lesions and bacterial growth, strong inflammatory and APR signals were observed comparatively earlier in the study
(12 - 20 dpLs) and may have been a marker of successful resolution of infection and eventual survival of co-infection.
68 CHAPTER 3: Functional feed impacts on Atlantic salmon (Salmo salar) immune responses to single infection of sea lice (Lepeophtheirus salmonis) and co-infection of sea lice and infectious salmon anemia virus
Carvalho, Laura A.1, Whyte, Shona K. 1, Braden, Laura M. 1, Purcell, Sara1, Taylor, Richard2,
Rise, Matt3, Gagné, Nellie4, Fast, Mark D1*.
1- Department of Pathology and Microbiology, Atlantic Veterinary College, University of Prince
Edward Island, Charlottetown, PE, Canada
2- Cargill Innovation Center, Dirdal, Norway
3- Department of Biology, University of Memorial, St. John’s, NL, Canada
4- Fisheries and Oceans Canada, Moncton, NB, Canada
*Corresponding author
Mark D. Fast Associate Professor Department of Pathology and Microbiology Atlantic Veterinary College – University of Prince Edward Island 550 University Avenue Charlottetown, PE, Canada C1A4P3 [email protected]
69 Abstract
Due to the nature of open-pen farming, salmon are exposed to numerous pathogens shared by their wild counterparts. Industry must manage these outbreaks through vaccination, oral or bath treatments, and more recently through functional feed administration. Globally, the most important pathogens of farmed salmon are sea lice (Lepeophtheirus salmonis, and Caligus rogercresseyi) not only due to their direct impacts on the host, but indirectly enhancing host susceptibility to co-infection. This study aims to characterize biomarkers and molecular responses during a co-infection of L. salmonis and a highly contagious salmon orthomyxovirus
(infectious salmon anemia virus; ISAv) under administration of four functional feed diets: a control feed with a low EPA/DHA mix of 0.3% (Ctrl), an EPA/DHA enriched diet (1.0%) with pro-inflammatory supplementation (Ω6>Ω3; FA+I), an EPA/DHA enriched diet (1.0%) with anti-inflammatory supplementation (Ω3>Ω6; FA-I), and a low EPA/DHA feed (0.3%) with an immunostimulant added (Ctrl + IS). Atlantic salmon (40 fish per tank; 8 tanks per feed) were acclimated to one of the four functional diets selected at random. Sea lice copepodids were introduced to all experimental tanks and 10 fish sampled from each tank over time (-3, 11, 33, 47 days post infection). A high virulence ISAv isolate (ISAV-HPR4) was IP injected into donor fish. Donor fish were cohabited (ca. 10-15% of tank density) 10 days after injection with experimental tanks (4 tanks per feed group) to achieve peak shedding rates at time of stocking.
Diets had a significant impact on sea lice abundance during single infection in fatty acid enriched feeds, FA+I and FA-I. However, FA+I and FA-I had lower overall survivorship compared to the other functional diets. Quantitative gene expression was used to elucidate the host responses to single infections and co-infections in head kidney tissue using reverse transcriptase-qPCR. Gene expression profiles showed significant upregulation in the irf7b and mxb antiviral genes
70 associated with the interferon system in fish fed the control, FA+I and FA-I diets. High lice and high viral loads in fish fed FA+I and Ctrl + IS may have an effect on the increase in irf7b, isg15a, mmp-9, and mxb transcript levels. As feeds successful at reducing lice were the least successful in the survival of co-infected fish, there appears to be a trade-off for better anti- parasitic and anti-viral responses.
71 3.1 Introduction
As the demand for fish and other aquatic organism production increases, wild fisheries production has plateaued since the 1990s from past over-exploitation. However, the Canadian finfish aquaculture sector continues to grow with production value upwards of $1.29 billion
(151,342 tonnes) in 2017 (DFO 2017). To control disease outbreaks in the field, farmers utilize prevention and management strategies involving pre-emptive vaccination against bacteria and viruses and drug therapy for parasite control (Jackson et al. 2018). The development and production of these interventions are time consuming, labour intensive and costly. Furthermore, unforeseen environmental factors can impact efficacy. As resistance development continues to reduce efficiency of antibiotics and anti-parasiticides, efforts into researching genetic markers and the development of functional diets are being implemented to enhance innate immunity
(Aaen et al. 2015). Recent work in Atlantic salmon has shown diets enriched with a high dose of a glucosinolate provide an up-regulation of innate antiviral responses in the skin, and reduce infection of ectoparasitic sea lice (Holm et al. 2016).
Marine ectoparasitic copepods, Leopeophtheirus salmonis and Caligus rogercresseyi
(Family: Caligidae), are the most economically important pathogens affecting Atlantic salmon culture worldwide. Commonly referred to as sea lice, these parasites thrive in farm settings with high host density and limited treatment options. Sea lice have been shown to immunomodulate the host through the secretion of compounds that able to interfere with inflammatory and immunological signalling (Fast et al. 2004; Fast et al. 2007). In terrestrial and aquatic species, successful and/or resistant hosts respond to infection through rapid inflammation initiated by interleukin -1β (IL-1β) and other pro-inflammatory cytokine production and release, attracting
MH class II expressing antigen-presenting cells to the infection site. In turn, wound healing and
72 tissue remodelling mechanisms follow in response to epithelial grazing by the parasite (Braden et al. 2015; Fast 2014; Grimholt 2016). In Atlantic salmon infected with L. salmonis and infectious salmon anemia virus (ISAv), a reduction in antiviral responses has been reported, leading to significantly increased mortalities showing clinical signs of viral infection regardless of L. salmonis infection level (Barker et al. 2012).
RNA viruses, such as ISAv, have caused severe consequences and massive losses to the aquaculture industry. The first outbreak of ISAv occurred in Norway in 1984 (Thorud and
Djupvik 1988). The virus has since been a continued threat to Atlantic salmon aquaculture in
Canada (Bouchard et al. 1999; Lovely et al. 1999; Mullins et al. 1998), the United States
(Bouchard et al. 2001), Scotland (Rodger et al. 1998; Rowley et al. 1999) the Faeroe Islands
(Lyngrøy 2003), Chile (Kibenge et al. 2007; Godoy et al. 2008), and Iceland (OIE 2017). The most notable of these outbreaks was in Chile where the industry suffered a loss of 210,000 tons of Atlantic salmon between 2007 and 2010. ISAv has also impacted Atlantic Canada on a massive scale between 1996 and 2007 in the Bay of Fundy, and in 2012 Newfoundland experienced losses of up to 750,000 Atlantic salmon mortalities (Godoy et al. 2008; Asche et al.
2009; Harvell 2019).
ISAv is a negative-sense, single stranded RNA virus with a genome consisting of eight segments and 10 proteins (Aspehaug et al. 2005; Falk et al. 2004). Two glycoproteins are found at the surface, with segment 6 encoding the hemagglutinin-esterase (HE), and segment 5 encoding the fusion protein. ISAv is a member of the Orthomyxoviridae family in the Isavirus genus (Krossøy et al. 1999; Mjaaland et al. 1997). Cumulative mortality can vary between 5% to
90% depending on the strain and degree of virulence (Ritchie et al. 2009). Virulence is determined by deletions in the highly polymorphic region (HPR) of the HE protein, compared to
73 a completely intact HE protein in the non-pathogenic ISAV-HPR0 strain, with these isolates indicated by ISAV-HPRΔ (Cunningham et al. 2002; Gagné and LeBlanc 2018). Antiviral responses in the host are induced by the type I interferon (IFN) system, which releases proteins
(ISGs) that bind to cell surface receptors, initiating cell-to-cell signalling and further promoting natural killer cell proliferation to eliminate viral pathogens (Cunha et al. 1996; Recht et al. 1991).
Although antiviral responses are initiated in the host, virulence factors, viral replication and load quickly spread as the fish succumb to systemic infections in high-virulence isolates
(LeBlanc et al. 2010). Mx proteins have shown antiviral activity in mice (Mx1) and humans
(MxA) when infected with Orthomyxoviruses, making these proteins a good candidate for studying antiviral effects against ISAv (Haller et al. 1998). Up-regulation of Mx expression has been reported in viral infections such as ISA, however no subsequent reduction in viral loads was observed (Kibenge et al. 2005). A promising approach by Caballero-Solares et al. (2017) analyzed functional diet impacts on Atlantic salmon injected with a synthetic viral mimic, polyriboinosinic polyribocytidylic acid (pIC). The study found that functional diets with low arachidonic acid (ARA) proportion to EPA had highest pIC-induction of antiviral transcripts
(tlr7, mxb) and high ARA/EPA diets showed an increased expression of pro-inflammatory eicosanoids and increased chemotaxis (ccl19b). Their conclusions suggested a well-balanced diet regimen may enhance the response of the salmon immune system to viral infections.
The present study investigated co-infections of L. salmonis and ISAv to assess whether there was an additive effect on the host immune system when experimental groups were fed different functional diets. Candidate biomarkers were selected to characterize host responses using qPCR to determine if varying feed components can modulate host immune responses and impact louse infection and co-infection survival.
74 3.2 Methods
3.2.1 Fish maintenance and husbandry
Atlantic salmon smolts (70 g; n = 1200) were purchased from Cooke Aquaculture Inc. and transferred from Oak Bay Hatchery, New Brunswick to the Aquatic Biological Containment
Level II Facility at the Atlantic Veterinary College, Charlottetown, PE. Fish were maintained in
1500 L tanks on flow-through freshwater (FW) at 11oC for several weeks and PIT (passive integrated transponder)-tagged during this period. At approximately 90 g, fish were randomly distributed to 32 tanks across four modules (n=40 fish/tank; 270 L water volume), each on a biofiltration fed recirculation unit (10.5 ± 1oC, 15-17 PSU). The fish were acclimated to increasing salt, through the addition of Instant Ocean® to a reservoir, over a period of two weeks until salinity reached 33 ± 3 PSU. Fish were kept under a 14 h:10 h light-dark photoperiod and hand fed commercial pellets twice daily while on the FW system, and transitioned to the EWOS standard marine diet until the functional diets were introduced.
3.2.2 Experimental diets and feeding trial
The feeding trial consisted of four functional diets formulated by Cargill Innovation Center
(Dirdal, Norway). The content of the four diets were blinded to investigators until the completion of the trial (Table 3.1). The first diet was a basic commercial feed (Ctrl; EWOS Dynamic S).
This diet contained 0.3% eicosapentaenoic acid (EPA)/docosahexaenoic acid (DHA). The second diet was EPA/DHA enriched (1%) with pro-inflammatory supplementation (Ω6>Ω3; FA+I) having omega-6 fatty acid levels twice as high as omega-3 fatty acids; the third diet was
EPA/DHA (1%) enriched with anti-inflammatory supplementation (Ω3>Ω6; FA-I) with omega-3 fatty acid levels twice as high as omega-6 fatty acids; and the fourth diet was the control feed
75 above with an undisclosed immunostimulant (Boost ®) added. All feeds met the nutritional requirements for Atlantic salmon smolts (Jobling and National Research Council (NRC) 2012).
Fish were given the experimental diets for 28 d prior to first pathogen exposure. Each functional diet was randomly assigned to two tanks in each module, totaling 8 tanks per diet across the four modules (Figure 3.1). Each module consisted of nine tanks; 8 tanks fed experimental diets (n =
40 fish per tank) and a ninth tank in each module (n = 24 fish per tank) remained on the basic
EWOS marine diet. Throughout the trial, Atlantic salmon were hand-fed a 1% body weight ration twice daily. Feeding behaviour was monitored and recorded using a scoring system of 1-4, indicating the percent of fish demonstrating normal behaviour. Feed consumption was assessed using a feed scoring system of 1-4 and uneaten pellets were counted in the effluent and recorded following each of the two feedings per day to determine daily feed consumption.
76 Table 3.2. Experimental and control functional diet composition.
Experimental Diets Standard Diet EPA+DHA 1% EPA+DHA 1% EPA+DHA 0.3% EPA+DHA EWOS Ω3: Ω6 High Ω6 High Ω3 High Ω6 Dynamic S Immunostimulant - - B+S - Fish oil (% diet) * 4,320 4,250 0,090 - Soy oil (% diet) 10,100 - 12,500 - Linseed oil (% diet) - 6,450 - - Poultry fat (% diet) 0,580 4,300 2,410 - Rapeseed oil (% diet) - - - - Added oil (% diet) 15,000 15,000 15,000 - EPA+DHA (% diet) 1,00 1,00 0,34 - Saturated (% total FA) 17,0 17,0 17,0 - 18:2n-6 (% total FA) 38,1 12,3 48,1 - 18:3n-3 (% total FA) 5,9 24,6 7,2 - n-6 (% total FA) 38,4 13,0 48,3 - n-3 (% total FA) 13,4 32,3 7,4 - *Fish oil not same inclusion as EPA+DHA level to correct for analysed 0.37% of total FA in poultry fat (OilsFABudge)
77
Figure 3.7. Experimental tank layout indicating single and coinfected rooms and function diets fed per tank in duplicate. Fish in auxiliary tanks (ta nk 9) housing donor fish in each room were injected with either a high virulent ISAv isolate (ISAV- HPR4) or L-15 supplemented with 10% FBS.
78 3.2.3 Copepodid challenge
Ovigerous L. salmonis females were collected from harvested Atlantic salmon in New
Brunswick, Canada. Lice were transported on ice in aerated coolers to the Huntsman Marine
Science Center where egg strings were removed from gravid females and placed in a flow- through hatching system with light aeration until eggs hatched and developed to the copepodid stage (ca. 9 days at 10oC). Atlantic salmon in all experimental tanks (n= 32 tanks) were challenged with ~2000 copepodids (50 lice per fish) initiating the infection trial (0 dpi).
Air stones were added to each tank and dissolved oxygen was maintained between 6.0-9.0 mg/L during the 2-h static bath. Following the bath period, additional air stones were removed, and tank in-flows were re-established. Netting was put in place to prevent lice from entering the flow-through system.
3.2.4 ISAv viral cultures and infection
Viral preparations of infectious salmon anemia virus (ISAv), were obtained from the Department of Fisheries and Oceans, in Moncton, New Brunswick (Dr. N. Gagné), initially provided by the
Research and Productivity Council (RPC) in Fredericton, New Brunswick. The prepared high- virulence ISAv isolate ISAV-HPR4 (RPC/NB 04-085-1) was suspended in L-15 culture media harvested from Atlantic salmon head kidney tissue by Ritchie et al. (2009). The ISAv isolate
(TCID50 of 1x105/ml) virulence was confirmed on an Atlantic salmon kidney (ASK) cell-line monolayer in vitro using the Spearmean-Kärber method at the Atlantic Veterinary College
(Hierholzer and Killington, 1996).
Viral preparations were thawed on ice, and aliquoted into 1 ml syringes. Atlantic salmon housed in the auxiliary (donor) tank in each module (n = 48 fish) were transported to a
79 quarantined module for viral inoculation at 9 dpi. Each fish was lightly anesthetized using tricaine methanesulphonate (MS-222) and were IP-injected with 0.8 ml of the ISAv isolate
(Figure 3.1). Fish were allowed to recover from anesthesia in an auxiliary tank aerated with oxygen stones. Fish were scanned using a PIT tag reader for identification and placed into their respective tanks. Viral donor fish remained in the quarantined module until 15 dpi. Three donor fish were added to experimental diet tanks in two of the four modules (n = 16 tanks) where the cohabitation element of the trial began. Sham donor fish (injected with L-15 media), were maintained in their home module until cohabitation began, at which point, similar to the ISAv exposed modules, 3 fish per tank were added, to ensure the same design and density across all experimental groups. Fish were monitored and checked for mortalities four times daily. In the case of a mortality, fish were necropsied, externally/internally examined for clinical signs of disease and posterior kidney samples were taken and stored in RNAlater™ RNA Stabilization
Solution for 24 h at 4oC (Thermo Fisher Scientific) prior to long term storage at -80 oC. Fish care and husbandry practices were in accordance with the guidelines of the Canadian Council of
Animal Care protocol approved by the University of Prince Edward Island and Atlantic
Veterinary College Animal Care Committee.
3.2.5 Sampling
Atlantic salmon (298.4g ± 16.7g) were sampled at -3, 11, 33, and 47 dpi (n= 832). Fish were fasted 24 h prior to sampling. Two fish per tank were sampled at time 1 (-3 dpi; n = 64 fish) prior to pathogen introduction. Time 2 sampling occurred when lice were anticipated to have moulted to the chalimus stage (11 dpi; n = 320 fish, 10 per tank) and time 3 sampling occurred once all donor fish died, and cohabitants first experienced mortality (33 dpi; n=320 fish, 10 per tank).
80 Time 4 sampling occurred when mortalities began to plateau in experimental tanks (47 dpi; n =
128 fish, including all remaining survivors). Modules containing experimental tanks with single
L. salmonis infection were sampled within 24-48 h of the co-infected modules. Fish were opportunistically selected and euthanized by MS-222 overdose (250 mg/L). Whole blood serum, head kidney, spleen, skin at louse attachment site, skin away from attachment site, posterior kidney, and individual lice were collected for gene expression. Head kidney, skin at louse attachment site, spleen and liver samples were also collected for histology. Whole blood plasma was stored on ice and centrifuged for serum removal. Tissues for gene expression were immediately put on dry ice once sampled. Serum and tissues were stored at -80°C until further processing and histological samples were stored in formalin at room temperature for 24-48 h before processing and embedding. Lice were counted and staged on each fish before discarding for incineration.
3.2.6 ISAv detection
Total RNA from RNA preserved posterior kidney samples collected from donor fish mortalities was performed using Trizol™ Reagent. Approximately 100 mg of sample was removed from
RNAlater™, homogenized and adhered to the extraction protocol outlined at (Chomczynski and
Sacchi 1987). Samples were DNase treated using the DNA-free™ kit (Ambion, Austin, TX, US).
Reverse transcription was performed using iScript™ Reverse Transcription Supermix (Bio-Rad
Laboratories, Inc., US) from 2 μg of DNase treaded RNA. Detection of ISAv was determined using Taqman® Gene Expression Master Mix (Thermo Fisher Scientific) in 20 μl reactions: 10
μl of Taqman®️ Gene Expression Master Mix, 7 μl nuclease-free water, 1 μl of FAM-labelled primer probe and 2 μl of cDNA. The primer probe targeted segment 8 of the virus (ISAV-8;
81 AIS08LC; PN4331348). Negative reverse transcriptase (noRT) samples, along with a no template control (NTC) and an internal positive control (IPC) were included on each plate in triplicate. The PCR profile consisted of an initial incubation step at 50 °C for 2 min, an enzyme activation step of 95 °C for 10 min followed by 40 cycles of denaturation at 95°C for 15 s and extension at 60 °C for 1 min, with fluorescence detection following each 60 °C step. The qPCR assay was performed using a CFX Connect™ Real-Time qPCR System (Bio-Rad Laboratories
Inc., US).
Following the detection of ISAv in the posterior kidneys of donor fish mortalities, total
RNA was extracted from posterior kidney samples of all time 3 (33 dpi) and time 4 (47 dpi cohabitated fish (n = 266). Methods were consistent with those described above for ISAv detection. To account for inter-plate variation, an ISAv-infected sample was used to link and calibrate Ct values across plates. Samples with a Ct value of 37.0 or higher were considered negative for ISAv, while any samples less than 37.0 Ct were considered infected with ISAv.
3.2.7 RNA isolation and cDNA synthesis: Gene expression analysis
For gene expression analysis, fish were divided into four infection groups: low lice (LL) fish
(<10 lice per fish); high lice (HL) fish (10-20 lice per fish); low virus (LV) fish (33-35.5 Ct); high virus (HV) fish (<27 Ct). Low lice and high lice groups were determined from lice counts of single infected fish, while low virus and high virus groups were determined by the Ct value of posterior kidney samples in the co-infected fish. Total RNA was isolated from 100 mg of head kidney tissue (n = 236) using Tri-reagent and the phenol-chloroform method (Chomczynski and
Sacchi 1987). The quantity and quality of RNA was determined using a NanoDrop 2000
Spectrophotometer (Thermo Fisher Scientific). The integrity of RNA samples was assessed using
82 a 3% agarose gel (IBI Scientific, Peosta, IA, US). Samples were DNase treated using the DNA- free™ kit (Ambion, Austin, TX, US). cDNA was synthesized from 2000 ng of DNase treated total RNA using iScript™ Reverse Transcription Supermix (Bio-Rad Laboratories, Inc., US).
NoRT samples were generated by pooling up to 15 individual DNase treated RNA samples. In the case of genomic DNA contamination in a noRT pool, all RNA samples included were retreated with DNase and tested again to ensure no impact on downstream quantification. Total
RNA and cDNA were stored at -80°C and -20°C, respectively. While running the reverse transcription - quantitative PCR assay, cDNA samples were stored at 4 °C.
3.2.8 RT-qPCR: Gene expression
Reverse transcription quantitative PCR (RT-qPCR) was performed with the SsoAdvanced™
Universal SYBR® Green Supermix (Bio-Rad Laboratories Inc., US) in triplicate using clear 96- well Multiplate® PCR Plates™ (Bio-Rad Laboratories Inc., UK) on CFX Connect™ Real-Time qPCR System (Bio-Rad Laboratories Inc., US) for standard curves and CFX384 (Bio-Rad
Laboratories Inc., US) when running samples. Primers were diluted to a concentration of 10 mM and 1 μl of template was used, resulting in a total reaction volume of 11 μl. Ten-fold and five- fold serial dilution standard curves were used to determine the efficiency of the primer pairs for reference genes and genes of interest, respectively. NoRT and no template controls (NTC) were run in triplicate in conjunction with the standard curves to ensure no genomic contamination of the sample groups. An internal positive control (IPC) was used as a calibrator to link the plates of each gene. Reference genes were quantified to ensure stability of samples. Reference genes chosen were ribosomal protein S20 (rps20), 18S ribosomal RNA (18s) and eukaryotic translation elongation factor 1-alpha (ef1αb), all of which were previously validated. Only RPS20 and 18S
83 met the stability requirements, using geNORM software, to be included in the analyses
(M=0.997; CV= 0.349).
Genes of interest selected were CC chemokine-like ligand 19b (ccl19b), interferon regulatory factor 7b (irf7b), interleukin-1β (il-1β), interferon stimulated gene 15a (isg15a), major histocompatibility class II (mhii), matrix metalloproteinase-9 (mmp-9), interferon-induced GTP- binding protein or myxovirus resistance b (mxb), and toll-like receptor 7 (tlr7) (Table 3.2). Two
PCR amplification profiles were used in analysis. The first PCR profile consisted of an initial denaturation step at 95°C for 30 s, followed by 40 cycles of annealing at 95°C for 15 s and extension at 60°C for 15 s, with fluorescence detections following each 60°C step. Following this protocol, melt curve analyses were performed at 0.5°C intervals from 65°C to 95°C for 5 s.
This profile was used to analyze mchii, mmp-9, rps20, and 18s. The second PCR profile used was adapted by Caballero-Solares et al. (2017) and used to analyze the remaining genes. The
PCR profile consisted of 1 cycle of 50°C for 2 min, 1 cycle of 95°C for 10 min and 40 cycles of
95°C for 15 s and 60°C for 1 min with fluorescence detections following each 60°C step. Melt curve analysis as previously mentioned concluded the protocol.
84 Table 3.3. Primers used for RT-qPCR, indicating forward and reverse sequences, efficiency values, accession number, amplicon size and original reference.
GENE TARGET NAME REFERENCE FORWARD (5-3') REVERSE (5-3') EFFICIENCY AMPLICON (%) CC chemokine-like 19b CCL19b Caballero-Solares et al., CTGCTTGACAACGACCGATA GTTGTTCTTGGTGGCAGGAG 87.7 151 2017 Interferon regulatory factor 7b IRF7b Caballero-Solares et al., GTCAGTGGTAAAATCAACACGC CACCATCATGAAACGCTTGGT 87.3 99 2017 Interleukin 1β IL-1β Fast et al., 2004 CGTCACATTGCCAACCTCAT ACTGTGATGTACTGCTGAAC 88.5 200
Interferon stimulated gene 15a ISG15a Caballero-Solares et al., AAAGTGGCCACAACAAAGCAG ATAGGAGCGGGCTCCGTAATC 85.6 140 2017 Major histocompatibility class II β- MHC II Braden et al., 2012 AAGGCTTGAAGACACGTTGC CAGTCCAGCAGTAACGTCCA 96.7 106 chain Matrix metalloproteinase 9 MMP9 Skugor et al., 2008 ACTCTACGGTAGCAGCAATGAAGGC CGTCAAAGGTCTGGTAGGAGCGTAT 106.8 72
Interferon-induced GTP-binding Mxb Barket et al., 2019 TGCAGCTGGGAAGCAAACT CAACGTTTGGCTGATCAGATTC 97.4 protein or myxovirus resistance b Toll-like receptor 7 TLR7 Caballero-Solares et al., CACCAACACAGAGCTGGAGA GCCTTGGAAAACTTGCTGAG 93.2 184 2017 Reference Genes 18S ribosomal RNA 18S Sutherland et al., 2017 CCCCGTAATTGGAATGAGTAC ACGCTATTGGAGCTGGAATTA 93.8 98
Ribosomal protein s20 RPS20 Olsvik et al., 2005 GCAGACCTTATCCGTGGAGCTA TGGTGATGCGCAGAGTCTTG 100.3 462
85 3.2.9 Statistical analysis
Statistical analysis of gene expression data was performed using qbase+ (Biogazelle) software with geNorm implementation to fully normalize and evaluate quality control of the data set.
Statistical significance of gene expression data was determined based on relative expression of genes of interest to reference genes (rps20, ef1ab, 18a). Reference genes selected for analysis were RPS20 and 18S based on stability. An ANOVA (two-way, p < 0.05) followed by a post-hoc
Tukey’s honest significant difference (HSD) test was used to determine significance across time
(-3, 33, 47 dpi), infection level and diet. An ANOVA (one-way, p < 0.05) was performed to determine significance of control, single and coinfection status with L. salmonis and/or ISAv (-3 and 33dpi). Kaplan-Meier survival curves, logistic regression and gene expression analyses were performed in RStudio 1.1.456 (RStudio, Inc., Boston, MA). The logistic regression predictive analysis was used to describe the gene expression data and to explain the relationship within variables (functional diets, infection level, and time). To determine the relative measure of effect, an odds ratio (OR) was computed from the regression coefficient to allow comparisons within each variable and to measure the association between amplification of a gene product and no amplification. Chi-square test of independence and condition factor calculations were computed
105푊 in Microsoft Excel (2016). The condition factor (퐾 = ) was also determined for fish 퐿 sampled throughout the study, where W is the weight of the fish and L is the fork length of the fish.
86 3.3 Results
3.3.1 Feed consumption and condition factor
Fish were fasted for 24 h prior to sampling, sea lice challenge, viral inoculation and other procedures, causing a break in the continuity of the feeding regime. This caused difficulties in comparing daily feed assessments across groups over the course of the study. To account for this, average feed consumption was determined during different phases of the study, with the inclusion of the condition factor of single and co-infected fish at each sample period (Table 3.3).
Average feed consumption was also determined per functional diet in single and co-infected fish
(Table 3.4).
87 Table 3.3. Average feed consumption (%) of single and co-infected fish at different stages of the trial is shown, along with average condition factor (K) at each sampling period (-3, 11, 33, 47 dpi).
Pre-infection Time 1 Sampling Post-lice Time 2 Sampling Post-ISAv Time 3 Sampling Post-ISAv Time 4 Sampling (%) (K) infection (%) (K) inoculation (%) (K) inoculation (%) (K) -31 to -1 dpi -3 dpi 0 to 13 dpi 11 dpi 14 to 33 dpi 33 dpi 34 to 47 dpi 47 dpi
Single 98.92 1.14 98.58 1.11 93.68 1.14 95.22 1.20 Infection
Co-infection 98.50 1.20 98.27 1.08 71.47 1.19 73.41 1.22
Table 3.4. Average feed consumption (%) of single and co-infected fish fed functional diets (Crtl, FA+I, FA-I, Ctrl + IS) between 40- 47 dpi (p < 0.001).
Single infection Co-infection (%) (%)
Ctrl 97.20 70.93
FA+I 95.30 82.03
FA-I 95.48 70.60
Ctrl + IS 93.10 83.93
88 Experimental diets did not result in altered feed consumption between groups. A chi-square test of independence was performed to examine the feed consumption of each experimental diet in single and co-infected fish during the final 7 days of the trial (Table 3.5). The relation between these variables was not significant, χ2 = 0.9711, p > 0.05. However, feed consumption showed significance different between single and co-infected fish treatments (p < 0.001). Table 3.5 depicts the condition factor at 47 dpi based on diet consumed and treatment. Condition factor for salmonid fish is a qualitative measure to determine if a fish is in poor (< 1.00), fair (1.20), good
(1.40), or excellent (1.60) condition. Based on the results obtained from the study, fish sampled at 47 dpi were in fair condition. However, the average condition factor of mortalities throughout the trial was K = 0.9973, indicating poor to extremely poor condition. This can be reinforced by the increase in mortality among the co-infection group following ISAv exposure (Figure 3.2).
Table 3.5. Average condition factor (K) of single and co-infected fish fed functional diets at 47 dpi (p = 0.23) and mortalities throughout the trial.
Single Infection Co-infection Mortalities
Ctrl 1.24 1.16 -
FA+I 1.20 1.25 -
FA-I 1.15 1.27 -
Ctrl + IS 1.20 1.21 -
- - - 0.9973
89 120
100
80
Ctrl 60 FA+I FA-I
40 Ctrl+IS Percentcumulative survival (%)
20
0 0 5 10 15 20 25 30 35 40 Days post L. salmonis infection (dpi)
Figure 3.2. Percent cumulative survival of cohabitated fish over time (dpi) for experimental diets groups (Ctrl; Ω6 > Ω3, FA+I; Ω3 > Ω6, FA-I; Ctrl + IS).
3.3.2 Atlantic salmon survival
Mortalities of cohabitated fish were first observed at 30 dpi in co-infected, cohabitated fish resulting in approximately 37-50% cumulative mortality (CM) throughout the study (Figure
3.2). Gross pathology of mortalities showed external hemorrhaging, exophthalmia, ascites, darkened liver, enlarged spleen and hemorrhaging of the viscera. Mortality rates in single infected fish were very low, resulting in 4 total mortalities throughout the course of the study
(<1%). Fish infected with sea lice only showed characteristic macroscopic pathology, associated with lice infection, of scale loss, mild inflammation and slight hemorrhaging at some of the louse
90 attachment sites. All ISAv donor fish mortalities (47/48 total fish) had concluded by 34 dpi. Fish fed the standard commercial feed (Ctrl) resulted in 47.8% CM, fish fed FA+I experienced 46.7%
CM, fish fed FA-I had 37.3% CM, and fish fed Ctrl + IS resulted in 50% CM (Figure 3.2). A
Kaplan-Meier survival estimate was performed and did not show any significant differences between fish fed experimental diets.
3.3.3 L. salmonis infection
Following initial L. salmonis exposure, Atlantic salmon exhibited typical jumping and flashing behaviour associated with parasite attachment. Attachment site pathology showed depigmentation of epithelial tissue, scale loss and mild inflammation. Lice infection was confirmed in all experimental tanks throughout the study. During sampling, lice on the surface of the fish were counted and staged. Lice counts and life stages were also recorded on dead and moribund fish and categorized by functional diet (Table 3.6). A significant decrease of lice was seen in FA+I and FA-I feeds at 11 dpi compared to 47 dpi in experimental tanks exposed to sea lice only (Table 3.7). No significant differences were seen between lice counts and functional diets in co-infected tanks, or in mortalities (Table 3.7; 3.8).
Table 3.6. Average lice counts of adult male and adult female lice on mortalities based on functional diet administered.
Life Stage Ctrl FA+I FA-I Ctrl + IS
Adult female 5.3 ± 4.7 6.6 ± 0.7 7.3 ± 0.7 6.9 ± 0.7
Adult male 3.5 ± 0.7 4.8 ± 0.9 6.0 ± 0.8 4.5 ± 0.8 Total 8.7 ± 0.7 11.7 ± 0.9 12.3 ± 0.8 11.8 ± 1.0
91 Table 3.7. Average lice counts on fish exposed to sea lice only (single infection) over time.
Time A Time B Time C Time D (-3 dpi) (11 dpi) (33 dpi) (47 dpi) Ctrl 0 ± 0 24.5 ± 1.2 12.9 ± 0.8 9.2 ± 0.6
FA+I 0 ± 0 22.6 ± 1.6* 12.0 ± 1.0 6.5 ± 0.4*
FA-I 0 ± 0 27.5 ± 2.0 10.4 ± 0.8 6.6 ± 0.5 Ctrl + IS 0 ± 0 24.1 ± 1.5 11.2 ± 0.7 8.8 ± 0.6
Table 3.8. Average lice counts on fish exposed to sea lice and ISAv (co-infection) over time.
Time A Time B Time C Time D (-3 dpi) (11 dpi) (33 dpi) (47 dpi)
Ctrl 0 ± 0 25.1 ± 1.6 11.7 ± 0.9 11.7 ± 0.9
FA+I 0 ± 0 32.0 ± 1.8 13.6 ± 1.1 10.1 ± 1.0
FA-I 0 ± 0 33.4 ± 1.7 16.1 ± 0.9 11.8 ± 1.2
Ctrl + IS 0 ± 0 31.6 ± 1.7 14.1 ± 1.0 10.3 ± 1.5
3.3.4 ISAv detection
Posterior kidney samples were extracted from coinfected fish at 33 and 47 dpi to detect the presence of ISAv. Raw Ct values were plotted to illustrate the relationship between qPCR amplification, indicating ISAv detection and the functional diets administered. Data showed consistent Ct values regardless of functional diet (32.07 Ct Ctrl; 32.19 Ct FA+I; 32.35 Ct FA-I;
32.31 Ct Ctrl + IS). At 33 dpi, median values of posterior kidney samples illustrate the presence of ISAv (Figure 3.3). By 47 dpi, fish fed the control (Ctrl) diet and the diet with immunostimulant added (Ctrl + IS) were primarily infected with ISAv, and Cts were significantly lower than 33 dpi (Figure 3.3). Fish fed fatty acid enriched diets were more variable in terms of ISAv detection, although median values were still within ISAv detection range. Ct values of donor fish posterior kidneys had slightly higher Ct values compared to sampled fish
92 (34.60 Ct Ctrl; 33.58 Ct FA+I; 33.81 Ct FA-I; 33.14 Ct Ctrl + IS) likely due to natural RNA degradation of the mortalities. A significant difference in detection between the posterior kidneys of sampled fish compared to donor fish when fed the functional diets was seen (p < 0.001).
47 47 47 47
Figure 3.3. Raw Ct values of posterior kidneys sampled at 33 and 47 dpi labelled with FAM primer-probe to detect the presence of ISAv in fish fed each functional diet.
3.3.5 Logistic regression analysis and odds ratio
Logistic regression models were used to determine the amplification outcome of genes where qPCR results lead to little or no expression in fish. A common predictor of amplification outcome showed significance in fish from the high virus group; irf7b (p = 0.000004; OR =
13.30), il-1β (p = 0.002; OR = 4.23) and isg15a (p = 0.002; OR = 3.76). With respect to time, amplification was likely to occur at 47 dpi in ccl19b (p = 0.04; OR = 0.52), isg15a (p = 0.03; OR
= 2.02) and mhii (p = 0.04; OR = 0.41). The only significant predictor in terms of functional diet was FA-I showing significance in il-1β (p = 0.02; OR = 2.90) and isg15a (p = 0.01; OR = 3.16).
93 3.3.6 Impact of functional diet on gene expression
The gene expression profiles in head kidney collected at 33 and 47 dpi were analyzed using reverse transcription- qPCR (Table 3.2). Within these time points, head kidney samples generated from fish were categorized into one of four subsets: low lice (< 10 lice per fish; n=43), high lice (10-20 lice per fish; n=37), low ISAv (33-35.5 Ct; n=41), high ISAv (<27 Ct; n=65).
All eight immune related genes showed transcriptional changes. No significant differences were detected in any genes at -3 dpi before exposure to pathogens.
To determine the impacts of functional diets over the development of infection (single and co-infection), samples were compared at 33 and 47 dpi. The basic marine diet (Ctrl) showed significantly higher expression of irf7b in fish with high ISAv (HV) loads compared to low ISAv
(LV) and high lice (HL) infection at 33 dpi (Figure 3.4a). Higher expression in HV compared to low lice (LL) was also observed in the Ctrl diet at 47 dpi. There were significant differences in mxb gene expression in the Ctrl diet, where HV fish had significantly higher expression than all other categories of fish fed the Ctrl diet at both 33 dpi and 47 dpi (Figure 3.5a). expression was significantly higher in the Ctrl diet fish with high lice infection at 47 dpi compared to all other groups (Figure 3.6a).
94
Figure 3.4. Gene expression in head kidney samples (n = 190) for interferon regulatory factor 7b (irf7b) in each of the functional diets: a) control (Ctrl); b) fatty acid enriched with pro- inflammatory (FA+I); c) fatty acid enriched with anti-inflammatory (FA-I); d) control feed with added immunostimulant (Ctrl+IS), for each infection group: low lice (LL; <10 lice per fish), high lice (HL; 10-20 lice per fish), low virus (LV; 33-35.5 Ct), high virus (HV; <27 Ct). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) of each infection level: low lice (LL), high lice (HL), low viral load (LV), high viral load (HV). Treatments analyzed over time (33 and 47 dpi). Letters and bars denote significant differences in expression (2-way ANOVA, p < 0.05).
95
Figure 3.5. Gene expression in head kidney samples (n = 190) for myxovirus resistance b (mxb) in each of the functional diets: a) control (Ctrl); b) fatty acid enriched with pro-inflammatory (FA+I); c) fatty acid enriched with anti-inflammatory (FA-I); d) control feed with added immunostimulant (Ctrl+IS), for each infection group: low lice (LL; <10 lice per fish), high lice (HL; 10-20 lice per fish), low virus (LV; 33-35.5 Ct), high virus (HV; <27 Ct). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) of each infection level: low lice (LL), high lice (HL), low viral load (LV), high viral load (HV). Treatments analyzed over time (33 and 47 dpi). Letters and bars denote significant differences in expression (2-way ANOVA, p < 0.05).
96
Figure 3.6. Gene expression in head kidney samples (n = 190) for major histocompatibility class II (mhii) in each of the functional diets: a) control (Ctrl); b) fatty acid enriched with pro- inflammatory (FA+I); c) fatty acid enriched with anti-inflammatory (FA-I); d) control feed with added immunostimulant (Ctrl+IS), for each infection group: low lice (LL; <10 lice per fish), high lice (HL; 10-20 lice per fish), low virus (LV; 33-35.5 Ct), high virus (HV; <27 Ct). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) of each infection level: low lice (LL), high lice (HL), low viral load (LV), high viral load (HV). Treatments analyzed over time (33 and 47 dpi). Letters and bars denote significant differences in expression (2-way ANOVA, p < 0.05).
The base marine diet with the addition of an immunostimulant (Ctrl+IS), showed a very different profile of immune gene expression than the control diet. Expression of ccl19b was significantly higher in HL fish compared to the LV fish at 33 dpi, however, this pattern was lost by 47 dpi (Figure 3.7a). While there were no expression differences in irf7b for this diet, high viral levels impacted mxb expression compared to single infected fish (HL vs LL) in this feed grouping at both 33 and 47 dpi (Figure 3.5d). Dissimilar to the Ctrl feed, at 33 dpi, HV fish had a significantly higher level of expression of isg15a than the LV fish (Figure 3.8d).
97 In salmon fed 1.0% EPA/DHA (Ω6 > Ω3; FA+I), expression of irf7b was significantly higher in HV fish compared to HL and LL fish at 33dpi, followed by higher expression of the
HV and HL fish compared to the LL fish at 47 dpi (Figure 3.4b). Similar to the Ctrl + IS, at 33 dpi mxb expression was highest in HV fish compared to all other categories and, concurrently,
LV fish were also upregulated compared to HL and LL fish (Figure 3.7b). By 47 dpi, HV fish levels of mxb were significantly higher compared to all other groups. High expression levels of isg15a were seen in the high virus fish compared to all other categories at 33 dpi, followed by no significant expression at 47 dpi (Figure 3.8b). At 33 dpi, upregulation of mmp-9 was significantly higher in LL and LV fish compared to HV (Figure 3.9b). However, this pattern shifted by 47 dpi where transcript levels were significantly higher in HV fish compared to LL and LV fish.
In salmon fed 1.0% EPA/DHA (Ω3>Ω6; FA-I), HV fish had significantly higher irf7b compared to LL fish at 33 dpi (Figure 3.4c). However, this pattern was lost by 47 dpi. Similar to the expression of the other diets, mxb expression was significantly higher in HV fish compared to high lice fish at 33 dpi, and by 47 dpi, overexpression was seen in high viral fish compared to all other groups (Figure 3.5c). In il-1β expression, the HL, LV and HV fish were upregulated compared to the low lice fish at 33 dpi (Figure 3.10c).
98
Figure 3.7. Gene expression in head kidney samples (n = 190) for CC chemokine-like 19b (ccl19b) in each of the functional diets: a) control (Ctrl); b) fatty acid enriched with pro- inflammatory (FA+I); c) fatty acid enriched with anti-inflammatory (FA-I); d) control feed with added immunostimulant (Ctrl+IS), for each infection group: low lice (LL; <10 lice per fish), high lice (HL; 10-20 lice per fish), low virus (LV; 33-35.5 Ct), high virus (HV; <27 Ct). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) of each infection level: low lice (LL), high lice (HL), low viral load (LV), high viral load (HV). Treatments analyzed over time (33 and 47 dpi). Letters and bars denote significant differences in expression (2-way ANOVA, p < 0.05).
99
Figure 3.8. Gene expression in head kidney samples (n = 190) for interferon stimulated gene 15a (isg15a) in each of the functional diets: a) control (Ctrl); b) fatty acid enriched with pro- inflammatory (FA+I); c) fatty acid enriched with anti-inflammatory (FA-I); d) control feed with added immunostimulant (Ctrl+IS), for each infection group: low lice (LL; <10 lice per fish), high lice (HL; 10-20 lice per fish), low virus (LV; 33-35.5 Ct), high virus (HV; <27 Ct). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) of each infection level: low lice (LL), high lice (HL), low viral load (LV), high viral load (HV). Treatments analyzed over time (33 and 47 dpi). Letters and bars denote significant differences in expression (2-way ANOVA, p < 0.05).
100
Figure 3.9. Gene expression in head kidney samples (n = 190) for matrix metalloproteinase 9 (mmp-9) in each of the functional diets: a) control (Ctrl); b) fatty acid enriched with pro- inflammatory (FA+I); c) fatty acid enriched with anti-inflammatory (FA-I); d) control feed with added immunostimulant (Ctrl+IS), for each infection group: low lice (LL; <10 lice per fish), high lice (HL; 10-20 lice per fish), low virus (LV; 33-35.5 Ct), high virus (HV; <27 Ct). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) of each infection level: low lice (LL), high lice (HL), low viral load (LV), high viral load (HV). Treatments analyzed over time (33 and 47 dpi). Letters and bars denote significant differences in expression (2-way ANOVA, p < 0.05).
101
Figure 3.10. Gene expression in head kidney samples (n = 190) for interleukin 1β (il-1β) in each of the functional diets: a) control (Ctrl); b) fatty acid enriched with pro-inflammatory (FA+I); c) fatty acid enriched with anti-inflammatory (FA-I); d) control feed with added immunostimulant (Ctrl+IS), for each infection group: low lice (LL; <10 lice per fish), high lice (HL; 10-20 lice per fish), low virus (LV; 33-35.5 Ct), high virus (HV; <27 Ct). Values are median log-2 transformed data representing the calibrated normalized relative quantities (CNRQ) of each infection level: low lice (LL), high lice (HL), low viral load (LV), high viral load (HV). Treatments analyzed over time (33 and 47 dpi). Letters and bars denote significant differences in expression (2-way ANOVA, p < 0.05).
3.4 Discussion
Functional diets had no effect on the growth rates and the condition factor (K) of the co-infected fish. However, within the last 7 days of the trial (40-47 dpi), feed consumption was significantly impacted in the co-infected fish compared to fish challenged with sea lice only. Despite the lack of impacts on growth and condition, functional feed manipulation did impact multiple infection metrics. Lice prevalence and abundance remained constant over time in the co-infected groups with lower counts at 47 dpi, likely due to loss during moulting and natural mortality.
102 The effects of fatty acid levels on fish health has been a previous concern in aquaculture
(Bell et al. 1991; Thompson et al. 1996). In the present study, significant differences in lice survival were seen in the single infected fish when fed FA+I and FA-I between 11 and 47 dpi. In a study by Rosenlund et al (2001) Atlantic salmon fed diets containing different fatty acid profiles showed differences in the fatty acid composition in the muscle tissue without compromising growth rate, condition factor and mortality. Levels of DHA and oleic acid were higher in the muscle than what were supplied in the experimental diets, while EPA levels showed a reversed trend (Rosenlund et al. 2001). This may indicate differences in fatty acid metabolism and utilization when exposed to diets with different compounds. Diets with a low ratio of (n-
3)/(n-6) polyunsaturated fatty acids (PUFAs) saw greater mortalities of unvaccinated fish when exposed to Aeromonas salmonicida and Vibrio anguillarum than a higher ratio of omega fatty acids (Thompson et al. 1996). Alternately, vaccinated fish fed high PUFA diet measured greater
B-cell activation in the head kidney and spleen in response to A. salmonicida infection. In viral infections, fatty acid biosynthesis commonly contributes to the genomic replication and targets lipid signalling, synthesis and metabolism. Antiviral therapeutics target these pathways to decrease the susceptibility of host cells to viral infections through feed interventions by altering fatty acid content, however much work remains to be done in viral-lipid interactions (Heaton and
Randall 2011).
In terms of parasitic response to the functional diets, there was no impact of diet on the developmental rates of L. salmonis which were consistent with previous studies, reaching the pre-adult stage by 20 dpi at 9.0°C (Johnson and Albright 1992). This may also suggest functional differences in the feeds may have caused immunophysiological impacts but allowed the fish to maintain fair growth and condition during infection.
103 Fish survival was dependent on the level of ISAv infection, the time at which cohabitated fish encountered the virus, and any immunological effects of the functional diets. Previous work involving ISAv has shown that cohabitation exposure can infect a large portion of the population at the same time (Jørgensen et al. 2007). The cumulative mortalities of fish in each feed group were not significantly different (37.2% to 50% CM), however, higher mortality was seen in
FA+I and FA-1 tanks with significant differences in lice counts in single infected fish over time in these feeds. The Ct values of donor fish posterior kidneys had slightly higher Ct values than sampled fish which could amount to lower viral infection. The Ct values obtained from the posterior kidney of co-infected fish were higher than the mean Ct values recorded from previous studies investigating ISAv infection in salmon (Munir and Kibenge 2004). This may indicate lower viral replication overall in the present study.
In the present study, transcript levels of mxb were induced in fish with high ISAv (HV) at
33 and 47 dpi, regardless of the functional diet administered, further confirming the use of Mx as an indicator of ISAv infection, but not indicative of positive or negative clinical outcomes. The activation of genes relating to type I interferons (IFN) induces antiviral properties in vertebrate cells. IFNs are responsible for the first line of non-specific antiviral defense, however some viruses have developed the capability of avoiding proteins released by the IFN system
(Goodbourn, Didcock, and Randall 2018; Jensen and Robertsen 2002). ISAv has been shown to counteract the antiviral efforts of IFNs when stimulated with Atlantic salmon interferon-like activity (AS-IFN) or poly I:C (pIC) (Jensen and Robertsen 2002). IFN-inducible Mx proteins are antiviral immune effectors that are highly conserved in vertebrates, including mammals, birds, and fish (Staeheli et al. 1993). In teleosts, antiviral mechanisms involving IFNs have been studied in zebrafish (Altmann et al. 2003), channel catfish (Long et al. 2004), Japanese flounder
104 (Lee et al. 2000), rainbow trout (Trobridge and Leong 1995) and Atlantic salmon (Robertsen et al. 2003). The IFN system has also shown antiviral effects against infectious pancreatic necrosis virus (IPNV) in salmon, through inhibition of viral protein synthesis Jensen and Robertsen
(2002) suggest different mechanisms by which ISAv counteract the anti-viral defense, by either counteracting IFNs, insensitivity or outright inhibition to antiviral proteins. Similarly, Kibenge et al. (2005) found a 10-fold reduction in viral yield and a delay in cytopathic effects, suggesting inhibition to Atlantic salmon Mx1 protein.
Antiviral proteins induced by the IFN system bind to cell surface receptors known as
IFN-stimulated genes (ISGs). ISG15 is thought to play a role in IFN signalling. In the present study, expression of isg15a did not vary among the Ctrl and FA-I diets. Upregulation of isg15a was seen in FA+I and Ctrl + IS at 33 dpi in fish with a high viral concentration. Since these two feed groups had divergent clinical outcomes in terms of survival, this suggests that the isg15 relationship with ISAv with respect to inflammatory signaling is possibly more complicated under coinfection. However, knockout mice studies focusing on the removal of ISG15 showed signalling was uninterrupted (Kim et al. 2006). Studies have provided evidence of antiviral activity by ISG15 in mammals against HIV-1 (Künzi and Pitha 1996) and Sindbis virus
(Lenschow et al. 2005). Secreted ISG15 is believed to be involved in chemotactic activity toward neutrophils, cell-to-cell signalling, and acts as a cytokine by promoting natural killer cell proliferation through the induction of T cells by IFN-γ (Cunha et al. 1996; Recht et al. 1991).
Caballero-Solares et al. (2017) observed a strong antiviral IFN response in immune transcripts
(tlr7, mxb, stat1c, gig1, isg15a) in a diet composed of vegetable oil compared to a fish oil diet.
However, strong pIC induction was seen in diets of both oil types with differences too slight to be significant. Immune transcripts including tlr7, mxb, and stat1c had greatest modulation in the
105 vegetable oil diet compared to the fish oil diet suggesting that protein source may have a greater effect on the immune response in salmon.
Other IFN-inducible genes involved in antiviral immune responses include toll-like receptors (TLRs). TLRs, such as TLR7 and TLR8, are transmembrane glycoproteins that act as a surface receptor for viral, bacterial and synthetic ssRNA (Lee et al. 2013). In the present study, the response of tlr7 in the host when fed functional diets was very low with 21/191 individuals showing amplification by qPCR. No connections can be inferred between the fish that amplified for tlr7, the infection type, and any relation to other genes assessed. In teleost species, TLR7 has shown involvement in antiviral immunity in zebrafish (Meijer et al. 2004), common carp
(Tanekhy et al. 2010) and rainbow trout (Palti et al. 2010). In the Atlantic salmon genome, three
TLR8 genes and two TLR7 genes have been identified and expressed in immune tissues, including the head kidney (Lee et al. 2013). This is an interesting result because the signal cascade leading to the transcription of IFNs and the initiation of the innate immune system is dependent on the binding of viral nucleic acids to TLR7 brought about by plasmacytoid dendritic cells (Robertsen 2008).
Toll-like receptors also include a Toll-interleukin (IL)-1 receptor (TIR) domain with the ability to recognize and respond to pathogen-associated molecular patterns (PAMPs) to initiate a signal transduction cascade (Jault et al. 2004; Palti 2011). Similarities can be seen between TIRs and IL-1 receptors. Significant up-regulation of il-1β occurred at 47 dpi in fish with low lice levels fed the FA-I functional diet. IL-1β is a pro-inflammatory cytokine that mobilizes to the infection site to kill and digest pathogens and activates transcription factors that have been shown to bind to sites in the TLR promotor in salmon head kidney (Secombes et al. 2001; Lee et
106 al. 2013). In treatment with IL-1β, TLR7 is modulated by IL-1β and shown to be significantly up-regulated in primary head kidney leukocytes (Lee et al. 2013).
Correlations have been reported between the expression of toll-like receptors (TLR7 and
TLR3) and interferon regulatory factors (IRFs) (Caballero-Solares et al. 2017). IRFs are involved in the signal transcription and transduction factors of ISGs (Lee et al. 2013). As a major regulator of antiviral responses, IRFs contain a viral activation domain at the C-terminus (Bergan et al.
2010). As investigated by Caballero- Solares et al. (2017), IRF7a and IRF7b play a lesser role in inducing IFNs in Atlantic salmon than other analogues of the gene. However, transcript levels of
IRFs in that study did not vary among plant oil and fish oil diets with pIC-induction. A contributing factor may be IRF7b having greatest expression in the gills and intestine as opposed to the head kidney (Bergan et al. 2010). In the present study, an up-regulation of irf7b at 33 and
47 dpi was seen in in high viral load fish fed the Ctrl and FA+I diets and in fish fed Fa-I with low lice levels. In regard to the effect of diet on IRF response, fish with high viral load showed a strong induction when fed the Ctrl, FA+I and FA-I later in the infection period. It is possible that the functional diet containing the immunostimulant (Ctrl + IS) may provide the greatest enhancement to the host immune system in the case of IRF expression and an interesting marker to continue investigating for a greater understanding of functional diets with respect to co- infection outcomes.
The antiviral gene CCL19b plays a major role in T-cell chemotaxis. When activated, there is an increase in the production of T cells from the head kidney by antigen-presenting cells, such as dendritic cells and macrophages (Olson and Ley 2002). No differences were found via qPCR analysis in fish with high or low viral loads in any of the diets. However, significant up- regulation of fish with high lice concentrations compared to low lice concentrations was seen at
107 33 dpi in fish fed Ctrl + IS. Interestingly, this same diet had the highest survival to coinfection, and was the only functional feed in which, LV group had significantly higher mxb at 33 dpi compared to HV. Further examination of relationships between Mx and CCL19b should be investigated. In previous research, low expression of CCL19b was observed when salmon were pIC-induced and fed an animal by-product meal (Caballero-Solares et al. 2017). This response may be a reaction to increased T cell homing back to the head kidney. Other chemokines involved in chemotaxis of leukocytes, such as CXCL10, have shown increased expression when challenged with L. salmonis (Martin et al. 2007).
MH class II is expressed on the surface of antigen presenting cells, initiating an inflammatory response, B-cell activation, and phagocyte recruitment (Grimholt 2016). An up- regulation of mhii was systemically induced in the high lice group fed the control (Ctrl) diet at 47 dpi. During initial lice infection, subsequently followed by the transition from the copepodid to chalimus life stage, previous research has shown an induction of MMPs (Tadiso et al. 2011).
MMPs may also be induced as wound healing responses transition from acute to chronic lice infections, playing a critical role in tissue remodelling and wound repair (Fast 2014; Sanders and
Kirschnerf 1983). A reduction of mmp-9 was seen in high lice fish compared to low lice fish at
33 dpi fed the FA+I diet. A study by Rockett et al. (2010) found significant changes in lipid composition of splenocytes in mice after a single day of feeding a high fat n-3 polyunsaturated fatty acid (PUFA) diet, although viral resistance after 17 days was not detected. In bacterial infection models of mice fed a n-3 PUFA diet, disease is likely due to the progressive mechanisms of the pathogen, as these diets have been reported to affect inflammation and subsequent immune responses (Husson et al. 2016). In some cases, intracellular pathogens such as influenza virus may cause increased mortality in mice when fed n-3 PUFA diets, however this
108 is not always the case of viral infections. When mice were infected with vaccina virus and fed n-
3 fatty acids survival rates were not negatively impacted (Jones and Roper 2017). The FA+I diet, which also exhibited lower lice counts compared to other diets, may have induced the expression of mmp-9 prompting tissue regeneration and louse rejection. Typically up-regulation of MMPs accompany louse infection in Atlantic salmon (Skugor et al. 2008) and pink salmon (Sutherland et al. 2011). Furthermore, at 47 dpi an up-regulation of mmp-9 in fish with high viral loads was seen in fish fed the same diet. At 47 dpi, fish with high viral loads showed pathological signs of hemorrhaging, which may account for the increase of mmp-9 at this stage of infection.
Alternatively, at 33 dpi in fish fed Ctrl + IS, mmp-9 expression of fish with high and low viral loads was downregulated compared to fish with low lice levels. In co-infection, mmp-9 expression may be over-stimulated in the head kidney at high viral loads, due to severe hemorrhaging, specifically in the kidney caused by the virus. Feeds with greatest survival also showed a reduction in MMPs, regardless of the severity of the viral infection. This may attest to
MMPs being a noteworthy marker of wound healing mechanisms being activated within the head kidney of ISAv infected fish.
All study fish showed signs of normal feeding responses and growth, indicating proper nutrition by all functional diets. Overall, the results presented indicate consistent mortality rates between functional diets and increased expression in several antiviral genes (irf7b, mxb) when fed all functional diets, promoting detection of ISAv. An increase in transcript levels (irf7b, isg15a, mmp-9, mxb) accounted for high lice and high viral loads in diets FA+I (Ω6 > Ω3) and
Ctrl + IS. Fatty acid enriched diets showed significant impacts on lice levels, while these same diets had the lowest survival in co-infected fish. Engagement of systemic wound healing responses (i.e. il-1 and mmp-9), induced in the 1% FA feeds, may have contributed to lice
109 reductions but diverted immunity away from functional anti-ISAv responses, resulting in greater mortality. The mechanism of this is still undetermined, however, plasmacytoid dendritic cells
(pDC) are important mediators of skin re-epithelialization and viral infection resolution through type I interferon release. Future studies should further examine crosstalk of these co-infections and possible pDC involvement. Furthermore, we must continue to investigate dietary replacements and functional diets, as well as the timing and duration of administration of these feeds to further enhance the immune response of Atlantic salmon to aid in the protection against multiple pathogen challenges.
110 CHAPTER 4: DISCUSSION
In finfish aquaculture, infectious diseases continue to be a concern for the health and production of aquatic ecosystems. Disease progression, susceptibility and resistance in co-infections is understudied, however, gaining momentum as it is the most common condition in aquatic environments. Co-infections encompass a wide range of heterogeneous pathogens, including parasites, bacteria, viruses and fungi, forming host-pathogen interactions that are continually evolving and causing further economic losses to aquaculture systems (FAO 2016). Atlantic salmon production is a major contributor to worldwide aquaculture and strict management strategies are enforced to monitor disease outbreaks. Co-infections are an important piece of the puzzle in understanding salmonid disease susceptibility, disease progression and efficacy of intervention strategies in order to prevent severe outbreaks and economic losses (Marcos-López et al. 2010). More specifically the study of immune responses in fish under these co-infection conditions is essential for understanding mechanisms of disease. The current thesis investigated two experimentally induced co-infections, and analyzed host immunological gene responses by reverse transcription (RT) quantitative-(q)PCR over time and under differing feed regimes to better understand how the fish immune system is differentially modulated under single and co- infections with sea lice (Lepeophtheirus salmonis) (Cox 2001; Telfer et al. 2010).
The first study, Chapter 2.0, involved an experimental infection of Atlantic salmon with zero, medium and high loads of L. salmonis, co-infected with Moritella viscosa. Positive bacterial growth was seen in skin lesion swabs, however, only few skin and head kidney samples detected positive for M. viscosa by qPCR. The lack of bacterial detection in these samples was likely due to the location of the sample site extracted on the skin surface, away from ulcer regions but corresponding with the louse attachment site. This suggests L. salmonis infection site
111 was not an active vector, although may indirectly enhance the progression of diseases (Barker et al. 2009; Fast 2014a; Novak et al. 2016). Sampled fish showed signs of osmoregulatory stress by the end of the study, indicated by high potassium levels in the serum in all challenge groups.
This 4.7-fold change in potassium from 7 dpLs to 46 dpLs may be a result of nutritional imbalance from changes in homeostasis as M. viscosa competes with the host for accessible iron
(Bjornsdottir et al. 2011). Significant findings in immune transcript levels was seen primarily in the high lice co-infection group between 20 and 46 dpLs. Fish positive for bacteriological growth showed inhibited inflammatory signals associated with louse immunosuppression at 20 dpLs. By
46 dpLi, lesions began to resolve, and mortalities slowed, suggesting chronic disease pathology with the possibility of infection resolution. In fish negative for bacteriological growth, strong inflammatory signals were observed at 20 dpLs and lowered by the conclusion of the study. High burdens of L. salmonis and low water temperature may potentially impact antigen-presenting cells and B-cell populations of the host when co-infected with M. viscosa. In Atlantic Canada and areas along the eastern Atlantic Ocean, the likelihood of co-infection with these pathogens at aquaculture sites is becoming increasingly common, driving the importance for treatment and management strategies to control the interaction of these two pathogens in future outbreaks.
The second co-infection model included efforts to manage pathogenic infections for translation to aquaculture sites using functional diets (Chapter 3.0). This study followed a similar
L. salmonis infection process as the first study, but utilized a cohabitation model of IP injected donor fish to expose study fish to a highly virulent strain (ISAV-HPR4) of infectious salmon anemia virus (ISAv) (Ritchie et al. 2009). Highest mortality rates were seen in fish fed fatty acid diets supplemented with pro- and anti-inflammatories (FA+I; FA-I). No significant differences in lice counts were seen in co-infected fish, however single infected fish fed FA+I and FA-I saw
112 significantly lower lice numbers between 11 and 47 dpi compared to fish fed other functional diets under this infection condition. Gene expression analysis (RT-qPCR) was used to investigate the host responses to single and co-infection groups under different feeding regimes using head kidney tissue. Transcript profiles showed significant upregulation in antiviral genes associated with the interferon (IFN) system in all but the immunostimulated fish diet (Ctrl + IS). An increase in transcript levels involved in antiviral responses, inflammatory responses, B-cell activation, phagocyte recruitment, wound repair and tissue remodelling accounted for high lice and high viral loads in diets FA+I and Ctrl + IS. This indicates the inclusion of pro-inflammatory supplementation and immunostimulants in feed may be beneficial to the host when faced with viral and parasitic infections by strengthening the immune response. However, feeds successful at reducing lice (FA+I; FA-I) were the least successful in survival under the co-infection challenge. This requires further research to ensure careful selection of feed combinations dependent on target pathogens.
Several factors could have been implemented into this thesis to improve the experiments, and by outlining these limitations in study design and techniques future studies could use these suggestions to improve and strengthen co-infection models. Firstly, the study design could have been balanced by including a time-matched uninfected control group maintained throughout the study period. For logistical and animal welfare (ie. reductions in numbers) reasons, housing a large number of control fish required significant resources and beyond the scope of these studies.
Although the sample size in the L. salmonis-ISAv study was quite high, when samples were divided by time, feed and infection level, this resulted in only 6 fish per group. A larger sample size would strengthen the results per group, especially during late stage infection when greater variation in host responses were observed, likely due to stress and competency of the immune
113 systems of individual fish. To gain a greater understanding of the host’s response to infection, more tissues could have been analyzed from each fish. However, multiple tissues from each sample period were harvested and stored in the laboratory for further investigation. Samples from these studies were sent to several laboratories across Canada for testing, as well. Similar to including more tissue types, a greater number of reference genes and genes of interest for qPCR could have been chosen to supplement the generation of a big-picture scenario on the immune response of salmon. As mentioned above, limitations on resources and time were taken into consideration. Further, the use of other techniques to better inform functional differences in host responses, in supplementation to qPCR, such as in situ hybridization, electron microscopy, and immunohistochemistry, would be beneficial in observing pathogen interactions with host cells at an intimate level. This would be especially interesting, and subject of further co-infection studies, to use these techniques to observe late stage infections and/or for confirming total or partial pathogen eradication. Combinations of these techniques have been useful in the past to compare the modulation of varying responses, such as early inflammatory responses in the TH1- type followed by regulatory mechanisms by the TH2-type pathways in resistant coho salmon during a Pacific L. salmonis infection (Braden et al. 2015).
The two co-infection studies presented in this thesis have only scratched the surface of the interplay between ecto-parasitism and subsequent bacterial or viral co-infection. These studies provide the first step in developing a better understanding of the implications of co- infection and offer a preliminary management strategy using functional diets (Chapter 3.0). The main objectives of both studies were to determine the pathogenic effects of each co-infection and characterize how biomarkers and molecular responses of the host immune system respond under these conditions. In Chapter 2.0, it was found that Moritella viscosa may impact osmoregulatory
114 function in chronic infections, and when simultaneously infected with high sea lice numbers, M. viscosa may impact wound healing capabilities, resulting in increased mortality. In Chapter 3.0, co-infection with ISAv and sea lice increased the production of antiviral, inflammatory, and wound repair genes, particularly in fish with a high viral/high louse loads fed pro-inflammatory
(FA+I) and immunostimulated (Ctrl + IS) diets. The pro-inflammatory feed was successful at reducing lice levels but was least successful when protecting against viral-induced mortality. It can be determined from this study that functional diets have the potential to reduce lice numbers while also contributing to survival, however, further selection of feed composition and co- infection trials are needed.
In summation, stark differences were seen in co-infected fish compared to those with a single infection. In particular, fish with high lice and high viral loads saw the most variation in transcript levels, which is a point of concern during outbreaks at aquaculture sites. These findings can significantly impact the future management of these infections by implementing functional diets to reduce lice numbers and increase survival in a multitude of possible infection combinations.
115 REFERENCES
Aaen, Stian Mørch, Kari Olli Helgesen, Marit Jørgensen Bakke, Kiranpreet Kaur, and Tor Einar Horsberg. 2015. “Drug Resistance in Sea Lice: A Threat to Salmonid Aquaculture.” Trends in Parasitology. 31(2):72-81. Abolofia, Jay, Frank Asche, and James E. Wilen. 2017. “The Cost of Lice: Quantifying the Impacts of Parasitic Sea Lice on Farmed Salmon.” Marine Resource Economics 32(3):329– 349. Akira, Shizuo, Satoshi Uematsu, and Osamu Takeuchi. 2006. “Pathogen Recognition and Innate Immunity.” Cell 124(4):783–801. Altmann, Stephen M., Mark T. Mellon, Daniel L. Distel, and Carol H. Kim. 2003. “Molecular and Functional Analysis of an Interferon Gene from the Zebrafish, Danio Rerio.” Journal of Virology. 77(3):1992–2002. Alvarez-Pellitero, Pilar. 2008. “Fish Immunity and Parasite Infections: From Innate Immunity to Immunoprophylactic Prospects.” Veterinary Immunology and Immunopathology. 126(3– 4):171–198. Angeles Esteban, Maríá, A. Bensussan, E. Flaño, J. D. Hayball, and P. Puccetti. 2012. “An Overview of the Immunological Defenses in Fish Skin.” International Scholarly Research Network (ISRN) Immunology. (29). Anon. 2018. The State of World Fisheries and Aquaculture. Rome. Arason, Gudmundur Jóhann. 1996. “Lectins as Defence Molecules in Vertebrates and Invertebrates.” Fish & Shellfish Immunology 6(4):277–289. Åsbakk, K. 2001. “Elimination of Foreign Material by Epidermal Malpighian Cells during Wound Healing in Fish Skin.” Journal of Fish Biology. 58(4):953-966. Asche, F., H. Hansen, R. Tveteras, and SS Tveteras. 2009. “The Salmon Disease Crisis in Chile.” Marine Resource Economics 24(4):405–411. Aspehaug, Vidar, Aase B. Mikalsen, Michael Snow, Eirik Biering, and Stéphane Villoing. 2005. “Characterization of the Infectious Salmon Anemia Virus Fusion Protein Downloaded From.” Journal of Virology. 79(19):12544–12553. Ayer, Nathan W. and Peter H. Tyedmers. 2009. “Assessing Alternative Aquaculture Technologies: Life Cycle Assessment of Salmonid Culture Systems in Canada.” Journal of Cleaner Production. 17(3):362–73. Bailey, Roberta JE, Michael A. Birkett, Anna Ingvarsdóttir, A. Jennifer Mordue (Luntz), William Mordue, Bríd O’Shea, John A. Pickett, and Lester J. Wadhams. 2006. “The Role of Semiochemicals in Host Location and Non-Host Avoidance by Salmon Louse ( Lepeophtheirus Salmonis ) Copepodids.” Canadian Journal of Fisheries and Aquatic Sciences. 63(2):448–56. Balon. 1990. “Epigenesis of an Epigeneticist: The Development of Some Alternative Concepts on the Early Ontogency and Evolution of Fishes.” Guelph Ichthyology Reviews. 1–48. Bandilla, M., E. T. Valtonen, L. R. Suomalainen, P. J. Aphalo, and T. Hakalahti. 2006. “A Link between Ectoparasite Infection and Susceptibility to Bacterial Disease in Rainbow Trout.” International Journal for Parasitology. 36(9):987–91. Barker, Duane E., Laura M. Braden, Maria P. Coombs, and Brad Boyce. 2009. “Preliminary Studies on the Isolation of Bacteria from Sea Lice, Lepeophtheirus Salmonis, Infecting Farmed Salmon in British Columbia, Canada.” Parasitology Research 105(4):1173–77. Barker, S., D. Bouchard, J. Covello, MD Fast, W. Wolters, and IR Bricknell. 2012. “Are ‘lousy’
116 Fish More Susceptible to ISAV? And What Is the Role of Lice in Transmitting ISAV.” in ACFFA Annual Meeting. St. Andrew’s, New Brunswick. Barker, S. E., I. R. Bricknell, J. Covello, S. L. Purcell, W. Wolters, M. D. Fast, and D. A. Bouchard. 2019. “Sea Lice (Lepeophtheirus Salmonis, Kroyer 1837) Infected Atlantic Salmon (Salmo Salar L.) Are More Susceptible to Infectious Salmon Anemia Virus.” PloS One. 14(1):e0209178. Bell, J. Gordon, Alasdair H. McVicar, Moira T. Park, and John R. Sargent. 1991. “High Dietary Linoleic Acid Affects the Fatty Acid Compositions of Individual Phospholipids from Tissues of Atlantic Salmon (Salmo Salar): Association with Stress Susceptibility and Cardiac Lesion.” The Journal of Nutrition. 121(8):1163–72. Benediktsdóttir, E., S. Helgason, and H. Sigurjónsdóttir. 1998. “Vibrio Spp. Isolated from Salmonids with Shallow Skin Lesions and Reared at Low Temperature.” Journal of Fish Diseases. 21(1):19–28. Bergan, Veronica, Øyvind Kileng, Baojian Sun, and Børre Robertsen. 2010. “Regulation and Function of Interferon Regulatory Factors of Atlantic Salmon.” Molecular Immunology. 47(11–12):2005–14. Bergsson, Gudmundur, Birgitta Agerberth, Hans Jörnvall, and Gudmundur Hrafn Gudmundsson. 2005. “Isolation and Identification of Antimicrobial Components from the Epidermal Mucus of Atlantic Cod (Gadus Morhua).” The FEBS Journal. 272(19):4960–69. Bjørn, P. A., B. Finstad, and R. Kristoffersen. 2001. “Salmon Lice Infection of Wild Sea Trout and Arctic Char in Marine and Freshwaters: The Effects of Salmon Farms.” Aquaculture Research. 32(12):947–62. Bjornsdottir, B., T. Gudmundsdottir, and B. K. Gudmundsdottir. 2011. “Virulence Properties of Moritella Viscosa Extracellular Products.” Journal of Fish Diseases. 34(5):333-343. Bjornsdottir, Bryndis, Olafur H. Fridjonsson, Steinunn Magnusdottir, Valgerdur Andresdottir, Gudmundur O. Hreggvidsson, and Bjarnheidur K. Gudmundsdottir. 2009. “Characterisation of an Extracellular Vibriolysin of the Fish Pathogen Moritella Viscosa.” Veterinary Microbiology. 136(3-4):326–34. Blackstock, N. and A. D. Pickering. 1982. Changes in the Concentration and Histochemistry of Epidermal Mucous Cells during the Alevin and Fry Stages Ofthe Brown Trout Salmo Trutta. 197(4):463-471. Bouchard, D. A., K. A. Brockway, C. Girai, W. Keleher, and P. L. Merril. 2001. “First Report of Infectious Salmon Anemia Virus (ISA) in the United States.” Bulletin of the European Association of Fish Pathologists 21(2):86–88. Bouchard, D., W. Keleher, H. M. Opitz, S. Blake, K. C. Edwards, and B. L. Nicholson. 1999. “Isolation of Infectious Salmon Anemia Virus (ISAV) from Atlantic Salmon in New Brunswick, Canada.” Diseases of Aquatic Organisms. 35(2):131-137. Bowers, J. M., A. Mustafa, D. J. Speare, G. A. Conboy, M. Brimacombe, D. E. Sims, and J. F. Burka. 2000. “The Physiological Response of Atlantic Salmon, Salmo Salar L., to a Single Experimental Challenge with Sea Lice, Lepeophtheirus Salmonis.” Journal of Fish Diseases. 23(3):165–72. Braden, Laura M., Duane E. Barker, Ben F. Koop, and Simon R. M. Jones. 2012. “Comparative Defense-Associated Responses in Salmon Skin Elicited by the Ectoparasite Lepeophtheirus Salmonis.” Comparative Biochemistry and Physiology - Part D: Genomics and Proteomics. 7(2):100–109. Braden, Laura M., Ben F. Koop, and Simon R. M. Jones. 2015. “Signatures of Resistance to
117 Lepeophtheirus Salmonis Include a TH2-Type Response at the Louse-Salmon Interface.” Developmental and Comparative Immunology. 48(1):178-191. Bravo, S., J. Treasurer, M. Sepulveda, and C. Lagos. 2010. “Effectiveness of Hydrogen Peroxide in the Control of Caligus Rogercresseyi in Chile and Implications for Sea Louse Management.” Aquaculture. 303(1–4):22–27. Bron, J. E., C. Sommerville, M. Jones, and G. H. Rae. 1991. “The Settlement and Attachment of Early Stages of the Salmon Louse, Lepeophtheirus Salmonis (Copepoda: Caligidae) on the Salmon Host, Salmo Salar.” Journal of Zoology. 224(2):201-212. Bruno, DW, J. Griffiths, J. Petrie, and TS Hastings. 1998. “Vibrio Viscosus in Farmed Atlantic Salmon Salmo Salarin Scotland: Field and Experimental Observations.” Diseases of Aquatic Organisms. 34(3):161–66. Buonocore, F., E. Randelli, V. Scala, M. C. Belardinelli, D. Casani, A. M. Facchiano, S. Costantini, and G. Scapigliati. 2009. “Evolution of Cell‐ mediated Immune Defences: Cloning and Structural Characterisation of the T Cell Receptor Beta Chain from the Icefish Chionodraco Hamatus (Perciformes: Channichthyidae).” Italian Journal of Zoology. 76(3):258–68. Burka, J. F., K. L. Hammell, T. E. Horsberg, G. R. Johnson, D. J. Rainnie, and D. J. Speare. 1997. “Drugs in Salmonid Aquaculture - A Review.” Journal of Veterinary Therapeutics. 20(5):333–49. Burridge, Les, Judith S. Weis, Felipe Cabello, Jaime Pizarro, and Katherine Bostick. 2010. “Chemical Use in Salmon Aquaculture: A Review of Current Practices and Possible Environmental Effects.” Aquaculture. 306(1–4):7–23. Caballero-Solares, Albert, Jennifer R. Hall, Xi Xue, Khalil Eslamloo, Richard G. Taylor, Christopher C. Parrish, and Matthew L. Rise. 2017. “The Dietary Replacement of Marine Ingredients by Terrestrial Animal and Plant Alternatives Modulates the Antiviral Immune Response of Atlantic Salmon (Salmo Salar).” Fish and Shellfish Immunology 64:24–38. Canada, Department of Fisheries and Oceans. 2017. Value Added Account - Aquaculture Industry. Cervantes, Jorge L., Bennett Weinerman, Chaitali Basole, and Juan C. Salazar. 2012. “TLR8: The Forgotten Relative Revindicated.” Cellular & Molecular Immunology. 9(6):434–38. Chomczynski, P. and N. Sacchi. 1987. “Single-Step Method of RNA Isolation by Acid Guanidinium Thiocyanate-Phenol-Chloroform Extraction.” Analytical Biochemistry. 162(1):156–59. Cook-Versloot, M., S. Griffiths, R. Cusack, S. Mcgeachy, and R. Ritchie. 2004. "Identification and Characterisation of Infectious Salmon Anaemia Virus (ISAV) Haemagglutinin Gene Highly Polymorphic Region (HPR) Type 0 in North America." Bulletin-European Association of Fish Pathologists. 24(4):203-208. Covello, J., S. Barker, D. Bouchard, W. Wolters, MD Fast, and IR Bricknell. 2012. “Effects of Sea Lice Infection and ISAv on Atlantic Salmon.” in ACFFA Annual Meeting. St. Andrew’s, New Brunswick. Cox FE. 2001. “Concomitant Infections, Parasites and Immune Responses.” Parasitology 122(S1):S23-38. Coyne, Rosie, Øivind Bergh, Ole Samuelsen, Kari Andersen, Bjørn Tore Lunestad, Hanne Nilsen, Inger Dalsgaard, and Peter Smith. 2004. “Attempt to Validate Breakpoint MIC Values Estimated from Pharmacokinetic Data Obtained during Oxolinic Acid Therapy of Winter Ulcer Disease in Atlantic Salmon (Salmo Salar).” Aquaculture. 238(1–4):51–66.
118 Coyne, Rosie, Peter Smith, Inger Dalsgaard, Hanne Nilsen, Heidi Kongshaug, Øivind Bergh, and Ole Samuelsen. 2006. “Winter Ulcer Disease of Post-Smolt Atlantic Salmon: An Unsuitable Case for Treatment?” Aquaculture 253(1-4):171-178. Cuesta, A., M. Esteban, and J. Meseguer. 2008. “The Expression Profile of TLR9 MRNA and CpG ODNs Immunostimulatory Actions in the Teleost Gilthead Seabream Points to a Major Role of Lymphocytes.” Cellular and Molecular Life Sciences. 65(13):2091–2104. Cunha, Jonathan D., Ernest Knight, Arthur L. Haast, Robert L. Truitt, and Ernest C. Borden. 1996. "Immunoregulatory Properties of ISG15, an Interferon-Induced Cytokine." Proceedings of the National Academy of Sciences. 93(1):211-215. Cunningham, CO, A. Gregory, J. Black, I. Simpson, and RS Raynard. 2002. “A Novel Variant of Infectious Salmon Anemia Virus (ISAV) Haemagglutinin Gene Suggests Mechanisms for Virus Diversity.” Bulletin of the European Association of Fish Pathologists. 22(6):366–374. Dawson, Leigh H., Alan W. Pike, and Dominic F. Houlihan. 1999. “Changes in Physiological Parameters and Feeding Behaviour of Atlantic Salmon Salmo Salar Infected with Sea Lice Lepeophtheirus Salmonis.” Diseases of Aquatic Organisms. 35(2):89–99. Deacon, N. and T. Hecht. 1996. “The Effect of Temperature and Photoperiod on the Growth of Juvenile Spotted Grunter Pomadasys Commersonnii (Pisces: Haemulidae).” South African Journal of Marine Science. 17(1):55–60. Doñate, Carmen, Joan Carles Balasch, Agnes Callol, Julien Bobe, Lluis Tort, and Simon MacKenzie. 2010. “The Effects of Immunostimulation Through Dietary Manipulation in the Rainbow Trout; Evaluation of Mucosal Immunity.” Marine Biotechnology. 12(1):88–99. Edward Robinson, W., Brenda McDougall, Dat Tran, and Michael E. Selsted. 1998. "Anti-HIV- 1 Activity of Indolicidin, an Antimicrobial Peptide from Neutrophils." Journal of leukocyte biology. 63(1):94-100. Ellis, A.E. 2001. “Innate Host Defense Mechanisms of Fish against Viruses and Bacteria.” Developmental & Comparative Immunology. 25(8–9):827–39. Ellis, Anthony E. 1987. “Inhibition of the Aeromonas Salmonicida Extracellular Protease by Α2- Macroglobulin in the Serum of Rainbow Trout.” Microbial Pathogenesis. 3(3):167–77. Erdal, Jon Inge, Øystein Evensen, Ole Kristian Kaurstad, Atle Lillehaug, Roar Solbakken, and Kristin Thorud. 1991. “Relationship between Diet and Immune Response in Atlantic Salmon (Salmo Salar L.) after Feeding Various Levels of Ascorbic Acid and Omega-3 Fatty Acids.” Aquaculture. 98(4):363–79. Ernst, W., P. Jackmanà, K. Doeà, F. Page, G. Julien, K. Mackay, and T. Sutherland. 2001. “Dispersion and Toxicity to Non-Target Aquatic Organisms of Pesticides Used to Treat Sea Lice on Salmon in Net Pen Enclosures.” Marine Pollution Bulletin. 42(6):432–43. Burka, J., M. D Fast, and Crawford Revie. 2012. “22 Lepeophtheirus Salmonis and Caligus Rogercresseyi.” Fish Parasites. p.350. Falk, Knut, Vidar Aspehaug, Reinhard Vlasak, and Curt Endresen. 2004. “Identification and Characterization of Viral Structural Proteins of Infectious Salmon Anemia Virus.” Journal of virology. 78(6):3063–3071. Fallang, A., I. Denholm, T. E. Horsberg, and M. S. Williamson. 2005. “Novel Point Mutation in the Sodium Channel Gene of Pyrethroid-Resistant Sea Lice Lepeophtheirus Salmonis (Crustacea: Copepoda).” Diseases of Aquatic Organisms. 65:129–136. FAO. 2016. “The State of World Fisheries and Aquaculture 2016.” Fast, M. D., S. C. Johnson, T. D. Eddy, D. Pinto, and N. W. Ross. 2007. “Lepeophtheirus Salmonis Secretory/Excretory Products and Their Effects on Atlantic Salmon Immune Gene
119 Regulation.” Parasite Immunology. 29(4):179–189. Fast, M. D., D. M. Muise, R. E. Easy, N. W. Ross, and S. C. Johnson. 2006. “The Effects of Lepeophtheirus Salmonis Infections on the Stress Response and Immunological Status of Atlantic Salmon (Salmo Salar).” Fish & Shellfish Immunology. 21(3):228–241. Fast, M. D., N. W. Ross, C. A. Craft, S. J. Locke, S. L. Mackinnon, and S. C. Johnson. 2004. “Lepeophtheirus Salmonis: Characterization of Prostaglandin E 2 in Secretory Products of the Salmon Louse by RP-HPLC and Mass Spectrometry.” Experimental Parasitology 107(1-2):5–13. Fast, M. D., N. W. Ross, and S. C. Johnson. 2005. “Prostaglandin E 2 Modulation of Gene Expression in an Atlantic Salmon (Salmo Salar) Macrophage-like Cell Line (SHK-1).” Developmental and Comparative Immunology. 29(11):951–963. Fast, Mark D. 2014a. “Fish Immune Responses to Parasitic Copepod (Namely Sea Lice) Infection.” Developmental and Comparative Immunology. 43(2):300-312. Figueroa, Carolina, Paulina Bustos, Débora Torrealba, Brian Dixon, Carlos Soto, Pablo Conejeros, and José A. Gallardo. 2017. “Coinfection Takes Its Toll: Sea Lice Override the Protective Effects of Vaccination against a Bacterial Pathogen in Atlantic Salmon.” Scientific Reports. 7(1):1–8. Firth, Kara J., Stewart C. Johnson, and Neil W. Ross. 2000. “Characterization of Proteases in the Skin Mucus of Atlantic Salmon (Salmo Salar) Infected with the Salmon Louse (Lepeophtheirus Salmonis) and in Whole-Body Louse Homogenate.” Journal of Parasitology. 86(6):1199–1205. Føre, Martin, Kevin Frank, Eirik Svendsen, Jo Arve Alfredsen, Tim Dempster, Harkaitz Eguiraun, Win Watson, Annette Stahl, Leif Magne Sunde, Christian Schellewald, Kristoffer R. Skøien, and Morten O. Alver. 2018. “Precision Fish Farming: A New Framework to Improve Production in Aquaculture.” Biosystems Engineering. 173:176–93. Freedman, S. J. 1991. “The Role of Alpha 2-Macroglobulin in Furunculosis: A Comparison of Rainbow Trout and Brook Trout.” Comparative Biochemistry and Physiology Part B: Comparative Biochemistry. 98(4):549–553. Frohm, Margareta, Hans Gunne, Ann-Charlotte Bergman, Birgitta Agerberth, Tomas Bergman, Anita Boman, Sture Liden, Hans Jornvall, and Hans G. Boman. 1996. "Biochemical and Antibacterial Analysis of Human Wound and Blister Fluid." European Journal of Biochemistry. 237(1):86-92. Gagné, N. and F. LeBlanc. 2018. “Overview of Infectious Salmon Anaemia Virus (ISAV) in Atlantic Canada and First Report of an ISAV North American-HPR0 Subtype.” Journal of Fish Diseases. 41(3):421–30. Galindo-Villegas, J. and H. Hosokawa. 2004. “Immunostimulants: Towards Temporary Prevention of Diseases in Marine Fish.” Advances En Nutricion, Acuicola VII Memorias Del VII Simposium Internationale de Nutricion Acuicola. p.16–19. Gallo, V. P. and A. Civinini. 2003. “Survey of the Adrenal Homolog in Teleosts.” International Review of Cytology. p.89–187. Genna, R. L., W. Mordue, A. W. Pike, and A. J. Mordue. 2005. “Light Intensity, Salinity, and Host Velocity Influence Presettlement Intensity and Distribution on Hosts by Copepodids of Sea Lice, Lepeophtheirus Salmonis.” Canadian Journal of Fisheries and Aquatic Sciences. 62(12):2675–2682. Gercken, J. and L. Renwrantz. 1994. “A New Mannan-Binding Lectin from the Serum of Eel (Anguilla Anguilla L.): Isolation, Characterization and Comparison with the Fucose-
120 Specific Serum Lectin.” Comparative Biochemistry and Physiology Part B: Comparative Biochemistry. 108(4):449–461. Geven, Edwin J. W. and Peter H. M. Klaren. 2017. “The Teleost Head Kidney: Integrating Thyroid and Immune Signalling.” Developmental & Comparative Immunology. 66:73–83. Glover, K. A., U. Grimholt, H. G. Bakke, F. Nilsen, A. Storset, and O. Skaala. 2007. “Major Histocompatibility Complex (MHC) Variation and Susceptibility to the Sea Louse Lepeophtheirus Salmonis in Atlantic Salmon Salmo Salar.” Diseases of Aquatic Organisms. 76(1):57–65. Godoy, Marcos G., Alejandra Aedo, Molly J. Kibenge, David B. Groman, Carmencita V Yason, Horst Grothusen, Angelica Lisperguer, Marlene Calbucura, Fernando Avendano, Marcelo Imilan, Miguel Jarpa, and Frederick S. Kibenge. 2008. “First Detection, Isolation and Molecular Characterization of Infectious Salmon Anaemia Virus Associated with Clinical Disease in Farmed Atlantic Salmon (Salmo Salar) in Chile.” BMC Veterinary Research. 4(1):28. González-Chávez, Susana A., Sigifredo Arévalo-Gallegos, and Quintín Rascón-Cruz. 2009. “Lactoferrin: Structure, Function and Applications.” International Journal of Antimicrobial Agents. 33(4):301-e1. González, Laura and Juan Carvajal. 2003. “Life Cycle of Caligus Rogercresseyi, (Copepoda: Caligidae) Parasite of Chilean Reared Salmonids.” Aquaculture. 220(1–4):101–117. González, Margarita P., Sandra L. Marín, and Luis Vargas-Chacoff. 2015. “Effects of Caligus Rogercresseyi (Boxshall and Bravo, 2000) Infestation on Physiological Response of Host Salmo Salar (Linnaeus 1758): Establishing Physiological Thresholds.” Aquaculture. 438:47–54. Goodbourn, S., L. Didcock, and R. E. Randall. 2018. "Interferons : Cell Signalling, Immune Modulation, Antiviral Responses and Virus Countermeasures." Journal of general virology. 81(10):2341-2364. Grant, Andrew N. 2001. “Medicines for Sea Lice.” Pest Management Science: formally Pesticide Science. 58(6):521-527. Griffiths, E. and J. Bullen. 1987. “Iron-Binding Proteins and Host Defence.” Iron and Infection. p.171–210. Grimholt, Unni. 2016. “MHC and Evolution in Teleosts.” Biology. 5(1):6. Groff, Joseph M. 2001. “Cutaneous Biology and Diseases of Fish.” Veterinary Clinics of North America: Exotic Animal Practice. 4(2):321–411. Grove, Deborah S. 1999. “Quantitative Real-Time Polymerase Chain Reaction for the Core Facility Using TaqMan and the Perkin-Elmer/Applied Biosystems Division 7700 Sequence Detector.” Journal of biomolecular techniques. 10(1):11. Grove, S., L. J. Reitan, T. Lunder, and D. Colquhoun. 2008. “Real-Time PCR Detection of Moritella Viscosa, the Likely Causal Agent of Winter-Ulcer in Atlantic Salmon Salmo Salar and Rainbow Trout Oncorhynchus Mykiss.” Diseases of Aquatic Organisms. 82(2):105- 109. Grove, S., C. R. Wiik-Nielsen, T. Lunder, H. S. Tunsjø, N. M. Tandstad, L. J. Reitan, A. Marthinussen, M. Sørgaard, A. B. Olsen, and D. J. Colquhoun. 2010. “Previously Unrecognised Division within Moritella Viscosa Isolated from Fish Farmed in the North Atlantic.” Diseases of Aquatic Organisms. 93(1):51-61. Gudmundsdottir, Bjarnheidur, K. and Magnadottir, B. 1997. “Protection of Atlantic Salmon (Salmo SalarL.) against an Experimental Infection Of Aeromonas Salmonicida ssp.
121 Achromogenes.” Fish & Shellfish Immunology. 7(1):55–69. Guselle, Nicole J., David J. Speare, R. J. Fred Markham, and Shane Patelakis. 2009. “Efficacy of Intraperitoneally and Orally Administered ProVale, a Yeast b-(1,3)/(1,6)-D-Glucan Product, in Inhibiting Xenoma Formation by the Microsporidian Loma Salmonae on Rainbow Trout Gills.” North American Journal of Aquaculture. 72(1):65-72. Hindar, Kjeti I., Nils Ryman, and Fred Utter. 1991. "Genetic Effects of Cu Tured Fish on Natura Fish Populations.” Canadian Journal of Fisheries and Aquatic Sciences. 48(5):945-957. Haller, O., M. Frese, and G. Kochs. 1998. “Mx Proteins: Mediators of Innate Resistance to RNA Viruses.” Revue Scientifique et Technique (International Office of Epizootics). 17(1):220– 230. Halttunen, Elina, Audun H. Rikardsen, Jan G. Davidsen, Eva B. Thorstad, and J. Brian Dempson. 2009. “Survival, Migration Speed and Swimming Depth of Atlantic Salmon Kelts During Sea Entry and Fjord Migration.” Tagging and tracking of marine animals with electronic devices. Springer, Dordrecht. p.35-49. Hammell, K. L. and I. R. Dohoo. 2005. "Risk Factors Associated with Mortalities Attributed to Infectious Salmon Anaemia Virus in New Brunswick, Canada." Journal of Fish Diseases. 28(11):651-661. Hamre, Lars A., Christiane Eichner, Christopher Marlowe A. Caipang, Sussie T. Dalvin, James E. Bron, Frank Nilsen, Geoff Boxshall, and Rasmus Skern-Mauritzen. 2013. “The Salmon Louse Lepeophtheirus Salmonis (Copepoda: Caligidae) Life Cycle Has Only Two Chalimus Stages.” PloS one. 8(9):e73539. Hansen, L. P. and T. P. Quinn. 1998. "The Marine Phase of the Atlantic Salmon (Salmo Salar) Life Cycle, with Comparisons to Pacific Salmon." Canadian Journal of Fisheries and Aquatic Sciences. 55(S1):104-118. Harris, J. E. and S. Hunt. 1975. “The Fine Structure of the Epidermis of Two Species of Salmonid Fish, the Atlantic Salmon (Salmo Salar L.) and the Brown Trout (Salmo Trutta L.).” Cell and Tissue Research. 163(4):535–543. Harvell, C. Drew. 2019. "Ocean outbreak: confronting the rising tide of marine disease." University of California Press. Hawkes, Joyce W. 1974. “The Structure of Fish Skin.” Cell and Tissue Research. 149(2):147– 158. Heaton, Nicholas S. and Glenn Randall. 2011. “Multifaceted Roles for Lipids in Viral Infection.” Trends in Microbiology. 19(7):368–375. Hébert-Dufresne, Laurent and Benjamin M. Althouse. 2015. “Complex Dynamics of Synergistic Coinfections on Realistically Clustered Networks.” Proceedings of the National Academy of Sciences of the United States of America. 112(33):10551–10556. Heidarsdóttir, K. J., K. Gravningen, and E. Benediktsdóttir. 2008. “Antigen Profiles of the Fish Pathogen Moritella Viscosa and Protection in Fish.” Journal of Applied Microbiology. 104(4):944–951. Hellemans, Jan, Geert Mortier, Anne De Paepe, Frank Speleman, and Jo Vandesompele. 2007. “QBase Relative Quantification Framework and Software for Management and Automated Analysis of Real-Time Quantitative PCR Data.” Genome Biology. 8(2):R19. Heuch, P. A., J. R. Nordhagen, and T. A. Schram. 2000. "Egg Production in the Salmon Louse [Lepeophtheirus Salmonis (Krùyer)] in Relation to Origin and Water Temperature." Aquaculture Research. 31(11):805-814. Hewitson, James P., John R. Grainger, and Rick M. Maizels. 2009. “Helminth
122 Immunoregulation: The Role of Parasite Secreted Proteins in Modulating Host Immunity.” Molecular and Biochemical Parasitology. 167(1):1–11. Hirano, Masayuki, Sabyasachi Das, Peng Guo, and Max D. Cooper. 2011. “The Evolution of Adaptive Immunity in Vertebrates.” Advances in Immunology. 109:125–157. Hjelmeland, K., M. Christie, and J. Raa. 1983. “Skin Mucus Protease from Rainbow Trout, Salmo Gairdneri Richardson, and Its Biological Significance.” Journal of Fish Biology. 23(1):13–22. Hjerde, Erik, Christian Karlsen, Henning Sørum, Julian Parkhill, Nils Peder Willassen, and Nicholas R. Thomson. 2015. “Co-Cultivation and Transcriptome Sequencing of Two Co- Existing Fish Pathogens Moritella Viscosa and Aliivibrio Wodanis.” BMC Genomics. 16(1):447. Hoffman, James, Jarl Bøgwald, Rolf Andersson, and Lennart Kenne. 2012. “Structural Studies of the Lipopolysaccharide of Moritella Viscosa Strain M2-226.” Carbohydrate Research. 347:164–67. Holzer, Aton M. and William R. Levis. 2007. “Antiviral Immunotherapy: Emerging Approaches with Relevance to Cutaneous Disease.” Expert Review of Dermatology. 2(5):619–27. Hull, M. Q., A. W. Pike, A. J. Mordue, and G. H. Rae. 1998. "Patterns of Pair Formation and Mating in an Ectoparasitic Caligid Copepod Lepeophtheirus Salmonis (Krùyer 1837): Implications for Its Sensory and Mating Biology." Philisophical Transactions of the Royal Society of London. Series B: Biological Sciences. 353(1369):753-764. Husson, Marie-odile, Delphine Ley, Madeleine Gottrand, Thomas Hueso, and Jean-luc Desseyn. 2016. “Modulation of Host Defence against Bacterial and Viral Infections by Omega-3 Polyunsaturated Fatty Acids.” Journal of Infection. 73(6):523–35. Igboeli, Okechukwu O., Mark D. Fast, Jan Heumann, and John F. Burka. 2012. “Role of P- Glycoprotein in Emamectin Benzoate (SLICE®) Resistance in Sea Lice, Lepeophtheirus Salmonis.” Aquaculture. 344:40–47. Iliev, Dimitar B., Ingrid Skjæveland, and Jorunn B. Jørgensen. 2013. “CpG Oligonucleotides Bind TLR9 and RRM-Containing Proteins in Atlantic Salmon (Salmo Salar).” BMC Immunology. 14(1):12. Ingerslev, H. C., T. Lunder, and M. E. Nielsen. 2010. “Inflammatory and Regenerative Responses in Salmonids Following Mechanical Tissue Damage and Natural Infection.” Fish and Shellfish Immunology. 29(3):440-450. Jackson, D., O. Moberg, E. M. Stenevik Djupevag, F. Kane, and F. Kane H. Hareide. 2018. “The Drivers of Sea Lice Management Policies and How Best to Integrate Them into a Risk Management Strategy : An Ecosystem Approach to Sea Lice Management.” Journal of Fish Diseases. 41(6):927–933. Janeway, Jr, Paul Travers, Mark Walport, and Mark J. Shlomchik. 2001. “The Complement System and Innate Immunity.” Immunobiology: The Immune System in Health and Disease. 5th edition. Garland Science. Jakob, E., D. E. Barker, and K. A. Garver. 2011. “Vector Potential of the Salmon Louse Lepeophtheirus Salmonis in the Transmission of Infectious Haematopoietic Necrosis Virus (IHNV).” Diseases of Aquatic Organisms. 97(2):155-165. Jault, Cyril, Laurent Pichon, and Johanna Chluba. 2004. “Toll-like Receptor Gene Family and TIR-Domain Adapters in Danio Rerio.” Molecular Immunology. 40(11):759–771. Jensen, Ingvill and Børre Robertsen. 2002. “Effect of Double-Stranded RNA and Interferon on the Antiviral Activity of Atlantic Salmon Cells against Infectious Salmon Anemia Virus and
123 Infectious Pancreatic Necrosis Virus.” Fish & Shellfish Immunology. 13(3):221–41. Jobling, M. and National Research Council (NRC). 2012. “Nutrient Requirements of Fish and Shrimp.” Aquaculture International. p.601–602. Jodaa Holm, Helle. 2017. "Immunological Response Profiles to Salmon Lice Infections in Atlantic Salmon: Modulation by nutrition and selective breeding." Jodaa Holm, Helle, Simon Wadsworth, Anne Kari Bjelland, Aleksei Krasnov, Øystein Evensen, and Stanko Skugor. 2016. “Dietary Phytochemicals Modulate Skin Gene Expression Profiles and Result in Reduced Lice Counts after Experimental Infection in Atlantic Salmon.” Parasites and Vectors. 9(1):1–15. Johnson, S. C. and L. J. Albright. 1991. “Development, Growth, and Survival of Lepeophtheirus Salmonis (Copepoda: Caligidae) Under Laboratory Conditions.” Journal of the Marine Biological Association of the United Kingdom. 71(02):425. Johnson, S. C. and L. J. Albright. 1992. “Comparative Susceptibility and Histopathology of the Response of Naive Atlantic, Chinook and Coho Salmon to Experimental Infection with Lepeophtheirus Salmonis (Copepoda: Caligidae).” Diseases of Aquatic Organisms. 14:179– 193. Jones, Gwendolyn J. B. and Rachel L. Roper. 2017. “The Effects of Diets Enriched in Omega-3 Polyunsaturated Fatty Acids on Systemic Vaccinia Virus Infection.” Scientific Reports. 7(15999):1–6. Jones, M. W., C. Sommerville, and J. Bron. 1990. “The Histopathology Associated with the Juvenile Stages of Lepeophtheirus Salmonis on the Atlantic Salmon, Salmo Salar L.” Journal of Fish Diseases. 13(4):303–310. Jonsdottir, H., J. E. Bron, R. Wootten, and J. F. Turnbull. 1992. “The Histopathology Associated with the Pre-Adult and Adult Stages of Lepeophtheirus Salmonis on the Atlantic Salmon, Salmo Salar L.” Journal of Fish Diseases, 15(6):521–527. Jørgensen, Sven Martin, Dyveke Lem Hetland, Charles McL Press, Unni Grimholt, and Tor Gjøen. 2007. “Effect of Early Infectious Salmon Anaemia Virus (ISAV) Infection on Expression of MHC Pathway Genes and Type I and II Interferon in Atlantic Salmon (Salmo Salar L.) Tissues.” Fish & Shellfish Immunology. 23(3):576–88. Karlsen, Christian, Henning Sørum, Nils Peder Willassen, and Kjetil Åsbakk. 2012a. “Moritella Viscosa Bypasses Atlantic Salmon Epidermal Keratocyte Clearing Activity and Might Use Skin Surfaces as a Port of Infection.” Veterinary Microbiology. 154(3-4):353-362. Karlsen, Christian, Christin Vanberg, Helene Mikkelsen, and Henning Sørum. 2014a. “Co- Infection of Atlantic Salmon (Salmo Salar), by Moritella Viscosa and Aliivibrio Wodanis, Development of Disease and Host Colonization.” Veterinary Microbiology. 171(1–2):112– 121. Katzenback B.A., Belosevic, M. 2009. "Isolation and functional characterization of neutrophil- like cells from goldfish (Carassius auratus L.) kidney." Developmental and Comparative Immunology. 33:601-611. Kempf, M., M. Merceron, G. Cadour, H. Jeanneret, Y. Méar, and P. Miramand. 2002. “Environmental Impact of a Salmonid Farm on a Well-Fushed Marine Site: II. Biosedimentology.” Journal of Applied Ichthyology. 18(1):51–60. Kibenge, F. S. B., M. J. T. Kibenge, Y. Wang, B. Qian, S. Hariharan, and S. McGeachy. 2007. “Mapping of Putative Virulence Motifs on Infectious Salmon Anemia Virus Surface Glycoprotein Genes.” Journal of General Virology. 88(11):3100–3111. Kibenge, Molly JT, Khalid Munir, and Frederick SB Kibenge. 2005. “Constitutive Expression of
124 Atlantic Salmon Mx1 Protein in CHSE-214 Cells Confers Resistance to Infectious Salmon Anaemia Virus.” Virology Journal. 2(1):75. Kileng, Øyvind, Marthe Iren Brundtland, and Børre Robertsen. 2007. “Infectious Salmon Anemia Virus Is a Powerful Inducer of Key Genes of the Type I Interferon System of Atlantic Salmon, but Is Not Inhibited by Interferon.” Fish & Shellfish Immunology. 23(2):378–89. Kim, Keun Il, Ming Yan, Oxana Malakhova, Jiann-Kae Luo, Mei-Feng Shen, Weiguo Zou, Juan Carlos De La Torre, and Dong-Er Zhang. 2006. “Ube1L and Protein ISGylation Are Not Essential for Alpha/Beta Interferon Signaling.” Molecular and Cellular Biology. 26(2):472– 479. Klein, Jan and Akie Sato. 2000. “The HLA System” edited by I. R. Mackay and F. S. Rosen. New England Journal of Medicine. 343(10):702–709. Kodama, Hiroshi, Fumitaka Yamada, Takashi Murai, Yuki Nakanishi, Takeshi Mikami, and Hisao Izawa. 1989. “Activation of Trout Macrophages and Production of CRP after Immunization with Vibrio Angillarum.” Developmental & Comparative Immunology. 13(2):123–132. Kottelat, Maurice. and Jörg. Freyhof. 2007. Handbook of European Freshwater Fishes. Publications Kottelat. Krkošek, Martin, Mark A. Lewis, John P. Volpe, Martin Krko ’ek1 ’, Mark A. Lewis1, and John P. Volpe. 2005. “Transmission Dynamics of Parasitic Sea Lice from Farm to Wild Salmon.” Proceedings of the Royal Society B: Biological Sciences. 272(1564):689–696. Krossøy, Bjørn, Ivar Hordvik, and Frank Nilsen. 1999. "The Putative Polymerase Sequence of Infectious Salmon Anemia Virus Suggests a New Genus within the Orthomyxoviridae." Journal of virology. 73(3):2136:2142. Kuhlman, Marcella, Keith Joiner, And R. Alan, and B. Ezekowitz. 1989. “The Human Mannose- Binding Protein Functions as an Opsonin.” Journal of Experimental Medicine. 169(5):1733–1745. Kumar Mittal, Ajay and Mary Whitear. 1979. “Keratinization of Fish Skin with Special Reference to the Catfish Bagarius Bagarius.” Cell and Tissue Research. 202(2):213–230. Künzi, Myrium S. and Paula M. Pitha. 1996. “Role of Interferon-Stimulated Gene ISG-15 in the Interferon-ω-Mediated Inhibition of Human Immunodeficiency Virus Replication.” Journal of Interferon & Cytokine Research. 16(11):919–927. Larsen, Rannveig, Torunn P. Røkenes, and Børre Robertsen. 2004. “Inhibition of Infectious Pancreatic Necrosis Virus Replication by Atlantic Salmon Mx1 Protein.” Journal of virology. 78(15):7938–7944. LeBlanc, F., M. Laflamme, and N. Gagné. 2010. “Genetic Markers of the Immune Response of Atlantic Salmon (Salmo Salar) to Infectious Salmon Anemia Virus (ISAV).” Fish & Shellfish Immunology. 29(2):217–232. Lee, Jong-Young, Ikuo Hirono, and Takashi Aoki. 2000. “Cloning and Analysis of Expression of Mx CDNA in Japanese Flounder, Paralichthys Olivaceus.” Developmental & Comparative Immunology. 24(4):407–415. Lee, P. T., J. Zou, J. W. Holland, S. A. M. Martin, T. Kanellos, and C. J. Secombes. 2013. “Identification and Characterization of TLR7, TLR8a2, TLR8b1 and TLR8b2 Genes in Atlantic Salmon (Salmo Salar).” Developmental and Comparative Immunology. 41:295– 305. Lenschow, Deborah J., Nadia V Giannakopoulos, Lacey J. Gunn, Christine Johnston, Andy K.
125 O’guin, Robert E. Schmidt, Beth Levine, and Herbert W. Virgin Iv. 2005. “Identification of Interferon-Stimulated Gene 15 as an Antiviral Molecule during Sindbis Virus Infection In Vivo.” Journal of virology. 79(22):13974–13983. Loch, T. P., K. Scribner, R. Tempelman, G. Whelan, and M. Faisal. 2012. “Bacterial Infections of Chinook Salmon, Oncorhynchus Tshawytscha (Walbaum), Returning to Gamete Collecting Weirs in Michigan.” Journal of Fish Diseases. 35(1):39-50. Long, Scott, Melanie Wilson, Eva Bengten, Locke Bryan, L. W. Clem, N. W. Miller, and V. G. Chinchar. 2004. “Identification of a CDNA Encoding Channel Catfish Interferon.” Developmental & Comparative Immunology. 28(2):97–111. Lovely, J. E., B. H. Dannevig, K. Falk, L. Hutchin, A. M. Mackinnon, K. J. Melville, and S. G. Griffiths. 1999. “First Identification of Infectious Salmon Anemia Virus in North America with Haemorrhagic Kidney Syndrome.” Diseases of Aquatic Organisms. 35(2):145–148. Løvoll, M., C.R. Wiik-Nielsen, H. S. Tunsjø, D. Colquhoun, T. Lunder, H. Sørum, and S. Grove. 2009. “Atlantic Salmon Bath Challenged with Moritella Viscosa – Pathogen Invasion and Host Response.” Fish & Shellfish Immunology. 26(6):877–884. Ltd, Division of Merck &. Co. 1988. Poison Control Monograph: Ivermectin. West Point, Pennsylvannia. p.18. Lunder, T. 1992. “‘Winter Ulcer’ in Atlantic Salmon: A Study of Pathological Changes, Transmissibility, and Bacterial Isolates.” Norwegian College of Veterinary Medicine, National Veterinary Institute. Lunder, T., Ø. Evensen, G. Holstad, T. Hastein. 1995. "Winter ulver in the Atlantic salmon Salmo salar. Pathological and bacteriological investigations and transmission experiments." Diseases of Aquatic Organisms. 23(1):39-49. Lunder, T., H. Sørum, G. Holstad, AG Steigerwalt, P. Mowinckel, and DJ Brenner. 2000. “Phenotypic and Genotypic Characterization of Vibrio Viscosus Sp. Nov. and Vibrio Wodanis Sp. Nov. Isolated from Atlantic Salmon (Salmo Salar) with ‘Winter Ulcer.’” International Journal of Systemic and Evolutionary Microbiology. 50(2):427–450. Lyng Schiøtz, Berit, Sven Martin Jørgensen, Caird Rexroad, Tor Gjøen, and Aleksei Krasnov. 2008. “Transcriptomic Analysis of Responses to Infectious Salmon Anemia Virus Infection in Macrophage-like Cells.” Virus Research. 136(1-2):65–74. Lyngrøy, C. 2003. “Infectious Salmon Anemia in Norway and the Faeroe Islands: An Industrial Approach." International Response to Infectious Salmon Anemia: Prevention, Control and Eradication: Proceedings of a symposium. Tech Bull 1902. Washington, DC: US Department of Agriculture, Animal and Plant Health Inspection Service; US Department of the Interior, US Geological Survey; US Department of Commerce, National Marine Fisheries Service. p.97–109. MacKinnon, E. Brett. 2017. “The Epidemiology and Laboratory Investigation of Ulcer Disease in Atlantic Salmon (Salmo Salar) in Atlantic Canada.” University of Prince Edward Island. MacPhee. 2001. “Ulcerative Skin Disease Affecting Atlantic Salmon during Summer and Fall.” in 9th Annual New England Farmed Fish Health Management Workshop. Machias, Maine. Magariños, B., R. Bonet, J. L. Romalde, M. J. Martınez, F. Congregado, and A. E. Toranzo. 1996. “Influence of the Capsular Layer on the Virulence OfPasteurella Piscicidafor Fish.” Microbial Pathogenesis. 21(4):289–297. Magnadottir, Bergljot. 2010. “Immunological Control of Fish Diseases.” Marine Biotechnology. 12(4):361–379. Magnadóttir, Bergljót. 2006. “Innate Immunity of Fish (Overview).” Fish & Shellfish
126 Immunology. 20(2):137–151. Manning, AJ, S. Leadbeater, A. MacKinnon, and L. Hawkins. 2015. Winter Ulcer Project - Comparison of Field Isolates of Moritella Viscosa. In: ACFFA Annual Technical Forum and Research Review. St. Andrew’s, New Brunswick. Marcos-López, M., P. Gale, B. C. Oidtmann, and E. J. Peeler. 2010. “Assessing the Impact of Climate Change on Disease Emergence in Freshwater Fish in the United Kingdom.” Transboundary and emerging diseases. 57(5):293-304. Martin, Samuel A. M., Bimal P. Mohanty, Phillip Cash, Dominic F. Houlihan, and Christopher J. Secombes. 2007. “Proteome Analysis of the Atlantic Salmon (Salmo Salar) Cell Line SHK- 1 Following Recombinant IFN-γ Stimulation.” Proteomics. 7(13):2275–2286. Matsumoto, Reiko and Masazumi Sugimoto. 2007. “Dermal Matrix Proteins Initiate Re- Epithelialization but Are Not Sufficient for Coordinated Epidermal Outgrowth in a New Fish Skin Culture Model.” Cell and Tissue Research. 347(2):249-265. Mayer, L. 2003. “Mucosal Immunity.” Pediatrics. 111(6):1595-1600. McCormick, Stephen D., Lars P. Hansen, Thomas P. Quinn, and Richard L. Saunders. 1998. “Movement, Migration, and Smolting of Atlantic Salmon (Salmo Salar).” Canadian Journal of Fisheries and Aquatic Sciences. 55(S1):77–92. Mcginnity, Philip, Paulo Prodö Hl, Andy Ferguson, Rosaleen Hynes, Niall Maoiléidigh, Natalie Baker, Deirdre Cotter, Brendan O ’hea, Declan Cooke, Ger Rogan, John Taggart, and Tom Cross. 2003. “Fitness Reduction and Potential Extinction of Wild Populations of Atlantic Salmon, Salmo Salar, as a Result of Interactions with Escaped Farm Salmon.” Proceedings of the Royal Society of London. Series B: Biological Sciences. 270(1532):2442-2450. Meijer, Annemarie H., S. Gabby Krens, Indira A. Medina Rodriguez, Shuning He, Wilbert Bitter, B. Ewa Snaar-Jagalska, and Herman P. Spaink. 2004. “Expression Analysis of the Toll-like Receptor and TIR Domain Adaptor Families of Zebrafish.” Molecular Immunology. 40(11):773–83. Mjaaland, S., E. Rimstad, K. Falk, and B. H. Dannevig. 1997. “Genomic Characterization of the Virus Causing Infectious Salmon Anemia in Atlantic Salmon (Salmo Salar L.): An Orthomyxo-like Virus in a Teleost.” Journal of Virology. 71(10):7681–86. Mullan, Ronan H., Barry Bresnihan, Lucy Golden-Mason, Trevor Markham, Rosemary O’Hara, Oliver FitzGerald, Douglas J. Veale, and Ursula Fearon. 2006. “Acute-Phase Serum Amyloid A Stimulation of Angiogenesis, Leukocyte Recruitment, and Matrix Degradation in Rheumatoid Arthritis through an NF-ĸB-Dependent Signal Transduction Pathway.” Arthritis and Rheumatism: Official Journal of the American College of Rheumatology. 54(1):105-114. Mullins, J. E., David B. Groman, and D. Wadowska. 1998. "Infectious Salmon Anaemia in Salt Water Atlantic Salmon (Salmo Salar L.) in New Brunswick, Canada." Bulletin of the European Association of Fish Pathologists. 18(4):110-114. Munir, Khalid and Frederick S. Kibenge. 2004. “Detection of Infectious Salmon Anaemia Virus by Real-Time RT–PCR.” Journal of Virological Methods. 117(1):37–47. Murray, C. K. and T. C. Fletcher. 1976. “The Immunohistochemical Localization of Lysozyme in Plaice (Pleuronectes Platessa L.) Tissues.” Journal of Fish Biology. 9(4):329–34. Mustafa, Ahmed, Chris Macwilliams, Nicole Fernandez, Kelly Matchett, Gary A. Conboy, and John F. Burka. 2000. “Effects of Sea Lice (Lepeophtheirus Salmonis Kröyer, 1837) Infestation on Macrophage Functions in Atlantic Salmon (Salmo Salar L.).” Fish & Shellfish Immunology. 10(1):47–59.
127 Nakanishi, Y., H. Kodama, T. Murai, and H. Izawa. 1991. “Activation of Rainbow Trout Complement by C-Reactive Protein.” American Journal of Veterinary Research. 52(3):397– 401. Nash, Colin E. 2003. “Interactions of Atlantic Salmon in the Pacific Northwest VI. A Synopsis of the Risk and Uncertainty.” Fisheries Research. 62(3):339–47. Naylor, Rosamond L., Rebecca J. Goldburg, Jurgenne H. Primavera, Nils Kautsky, Malcolm C. M. Beveridge, Jason Clay, Carl Folke, Jane Lubchenco, Harold Mooney, and Max Troell. 2000. “Effect of Aquaculture on World Fish Supplies.” Nature. 405(29):1017-1024. Nolan, D. T., P. Reilly, and S. E. Wendelaar Bonga. 1999. “Infection with Low Numbers of the Sea Louse Lepeophtheirus Salmonis Induces Stress-Related Effects in Postsmolt Atlantic Salmon (Salmo Salar).” Canadian Journal of Fisheries and Aquatic Sciences. 56(6):947– 959. Novak, C. W., D. L. Lewis, B. Collicutt, K. Verkaik, and D. E. Barker. 2016. “Investigations on the Role of the Salmon Louse, Lepeophtheirus Salmonis (Caligidae), as a Vector in the Transmission of Aeromonas Salmonicida Subsp. Salmonicida.” Journal of Fish Diseases. 39(10):1165–1178. Nylund, A., T. Hovland, K. Hodneland, F. Nilsen, and P. Levik. 1994. “Mechanisms for Transmission of Infectious Salmon Anaemia (ISA).” Diseases of Aquatic Organisms. 19:95. Ogut, H. and N. Cavus. 2014. “A Comparison of Ectoparasite Prevalence and Occurance of Viral Haemorrhagic Septicemia Virus (VHSV) in Whiting Merlangius Merlangus Euxinus.” Revista de Biologia Marina y Oceanografia. 49(1). OIE. 2017. “World Animal Health Information Database (WAHIS).” Olson, Timothy S. and Klaus Ley. 2002. “Chemokines and Chemokine Receptors in Leukocyte Trafficking.” American Journal of Physiology. 283(1):R7–R28. Palic, D., Andreasen C.B., Menzel B.W., Roth J.A. 2005. "A rapid, direct assay to measure degranulation of primary granules in neutrophils from kidney of fathead minnow (Pimephales promelas Rafinesque, 1820). Fish and Shellfish Immunology. 19:217-227. Palti, Yniv. 2011. “Toll-like Receptors in Bony Fish: From Genomics to Function.” Developmental & Comparative Immunology. 35(12):1263–1272. Palti, Yniv, Scott A. Gahr, Maureen K. Purcell, Sima Hadidi, Caird E. Rexroad, and Gregory D. Wiens. 2010. “Identification, Characterization and Genetic Mapping of TLR7, TLR8a1 and TLR8a2 Genes in Rainbow Trout (Oncorhynchus Mykiss).” Developmental & Comparative Immunology. 34(2):219–233. Petit, Jules and Geert F. Wiegertjes. 2016. “Long-Lived Effects of Administering β-Glucans: Indications for Trained Immunity in Fish.” Developmental & Comparative Immunology. 64:93–102. Pike, A. W. and S. L. Wadsworth. 1999. “Sealice on Salmonids: Their Biology and Control.” Advances in Parasitology. 44:233–337. Press, C. McL. and Ø. Evensen. 1999. “The Morphology of the Immune System in Teleost Fishes.” Fish & Shellfish Immunology. 9(4):309–318. Purcell, S. L., S. E. Friend, J. M. Covello, A. Donkin, D. B. Groman, J. Poley, and M. D. Fast. 2013. “CpG Inclusion in Feed Reduces Sea Lice, Lepeophtheirus Salmonis, Numbers Following Re-Infection.” Journal of Fish Diseases. 36(3):229–240. Pylkko, P., L. R. Suomalainen, M. Tiirola, and E. Tellervo Valtonen. 2006. “Evidence of Enhanced Bacterial Invasion during Diplostomum Spathaceum Infection in European Grayling, Thymallus Thymallus (L.).” Journal of Fish Diseases. 29(2):79–86.
128 Raj, Victor, Guillaume Fournier, Krzysztof Rakus, Maygane Ronsmans, Ping Ouyang, Benjamin Michel, Cédric Delforges, Bérénice Costes, Frédéric Farnir, Baptiste Leroy, Ruddy Wattiez, Charles Melard, Jan Mast, François Lieffrig, and Alain Vanderplasschen. 2011. “Skin Mucus of Cyprinus Carpio Inhibits Cyprinid Herpesvirus 3 Binding to Epidermal Cells.” Veterinary Research. 42(1):92. Randelli, Elisa, Francesco Buonocore, and Giuseppe Scapigliati. 2008. “Cell Markers and Determinants in Fish Immunology.” Fish and Shellfish Immunology. 25(4):326-340. Rebl, Alexander, Henrike Rebl, Shuzhen Liu, Tom Goldammer, and Hans Martin Seyfert. 2011. “Salmonid Tollip and MyD88 Factors Can Functionally Replace Their Mammalian Orthologues in TLR-Mediated Trout SAA Promoter Activation.” Developmental and Comparative Immunology. 35(1):81-87. Recht, M., E. C. Borden, and E. Knight. 1991. “A Human 15-kDa IFN-Induced Protein Induces the Secretion of IFN-Gamma.” Journal of Immunology. 147(8):2617–23. Refstie, Ståle, Grete Baeverfjord, Rudi Ripman Seim, and Odd Elvebø. 2010. “Effects of Dietary Yeast Cell Wall β-Glucans and MOS on Performance, Gut Health, and Salmon Lice Resistance in Atlantic Salmon (Salmo Salar) Fed Sunflower and Soybean Meal.” Aquaculture. 305(1–4):109–116. Revie, C. W., G. Gettinby, J. W. Treasurer, and C. Wallace. 2003. “Identifying Epidemiological Factors Affecting Sea Lice Lepeophtheirus Salmonis Abundance on Scottish Salmon Farms Using General Linear Models.” Diseases of Aquatic Organisms. 57(1-2):85–95. Rimstad, E., O. B. Dale, B. H. Dannevig, and K. Falk. 2017. “Infectious Salmon Anaemia.” Fish diseases and disorders. 3:143-165. Ritchie, R. J., J. T. McDonald, B. Glebe, W. Young-Lai, E. Johnsen, and N. Gagné. 2009. “Comparative Virulence of Infectious Salmon Anaemia Virus Isolates in Atlantic Salmon, (Salmo Salar L.).” Journal of Fish Diseases. 32(2):157–171. Robertsen, Børre. 2008. “Expression of Interferon and Interferon-Induced Genes in Salmonids in Response to Virus Infection, Interferon-Inducing Compounds and Vaccination.” Fish & Shellfish Immunology. 25(4):351–357. Robertsen, Børre, Veronica Bergan, Torunn Røkenes, Rannveig Larsen, and Artur Albuquerque. 2003. "Atlantic Salmon Interferon Genes: Cloning, Sequence Analysis, Expression, and Biological Activity." Journal of Interferon and Cytokine Research. 23(10):601-612. Rockett, Benjamin Drew, Muhammad Salameh, Kristen Carraway, Kaitlin Morrison, and Saame Raza Shaikh. 2010. “N-3 PUFA Improves Fatty Acid Composition , Prevents Palmitate- Induced Apoptosis , and Differentially Modifi Es B Cell Cytokine Secretion in Vitro and Ex Vivo.” Journal of Lipid Research. 51(6):1284–1297. Rodger, HD, T. Turnbull, F. Muir, S. Millar, and RH Richards. 1998. “Infectious Salmon Anemia (ISA) in the United Kingdom.” Bulletin of the European Association of Fish Pathologists. 18(4):115–16. Rosenlund, G., A. Obach, M. G. Sandberg, H. Standal, and K. Tveit. 2001. “Effect of Alternative Lipid Sources on Long-Term Growth Performance and Quality of Atlantic Salmon (Salmo Salar L.).” Aquaculture Research. 32:323–28. Rowley, HM, SJ Campbell, WL Curran, T. Turnbull, and DG Bryson. 1999. “Isolation of Infectious Salmon Anemia Virus (ISAV) from Scottish Farmed Atlantic Salmon, Salmo Salar L.” Journal of Fish Diseases. 22(6):483–87. RSPCA. 2015. “RSPCA Welfare Standards for Farmed Atlantic Salmon.” RSPCA Welfare Standards for Farmed Atlantic Salmon. p.1–80.
129 Salinas, Irene, Yong-An Zhang, and J. Oriol Sunyer. 2011. “Mucosal Immunoglobulins and B Cells of Teleost Fish.” Developmental & Comparative Immunology. 35(12):1346–1365. Sanders, Martin J. and Leonard B. Kirschnerf. 1983. “Potassium Metabolism in Seawater Teleosts: II. Evidence for Active for Active Potassium Extrusion across the Gill.” Journal of Experimental Biology. 104(1):29-40. Sanderson, Hans, Brian Laird, Louise Pope, Richard Brain, Christian Wilson, David Johnson, Gareth Bryning, Andrew S. Peregrine, Alistair Boxall, and Keith Solomon. 2007. “Assessment of the Environmental Fate and Effects of Ivermectin in Aquatic Mesocosms.” Aquatic Toxicology. 85(4):229–240. Saunders, R. L. 1995. “Salmon Aquaculture: Present Status and Prospects for the Future.” Cold- Water Aquaculture in Canada. p.35–81. Schram, T. A. 2006. “Supplementary Descriptions of the Developmental Stages of Lepeophtheirus salmonis (Krøyer, 1837) (Copepoda: Caligidae).” Pathogens of wild and farmed fish: sea lice. 1:30-47. Secombes, C. J., T. Wang, S. Hong, M. Peddie, K. J. Crampe, C. Laing, Cunningham, and J. Zou. 2001. “Cytokines and Innate Immunity of Fish.” Developmental & Comparative Immunology. 25(8–9):713–723. Shanker, Anil. 2010. “Adaptive Control of Innate Immunity.” Immunology Letters. 131(2):107– 12. Shoemaker, C. A., P. H. Klesius, and Chhorn Lim. 2001. "Immunity and Disease Resistance in Fish." Nutrition and fish health. p.149-162. Simopoulos, A. 2002. “The Importance of the Ratio of Omega-6/Omega-3 Essential Fatty Acids.” Biomedicine & Pharmacotherapy. 56(8):365–379. Sire, Jean-Yves and Marie-Andrée Akimenko. 2004. “Scale Development in Fish: A Review, with Description of Sonic Hedgehog (Shh) Expression in the Zebrafish (Danio Rerio).” The International Journal of Developmental Biology. 48(2–3):233–247. Skugor, Stanko, Kevin Alan Glover, Frank Nilsen, and Aleksei Krasnov. 2008. “Local and Systemic Gene Expression Responses of Atlantic Salmon (Salmo Salar L.) to Infection with the Salmon Louse (Lepeophtheirus Salmonis).” BMC Genomics. 9(1):498. Staeheli, P., F. Pitossi, and J. Pavlovic. 1993. “Mx Proteins: GTPases with Antiviral Activity.” Trends in Cell Biology. 3(8):268–72. Statistics, FAO Fishery and Aquaculture. 2015. “Global Aquaculture Production 1950-2013.” FishstatJ. Subramanian, Sangeetha, Shawna L. MacKinnon, and Neil W. Ross. 2007. “A Comparative Study on Innate Immune Parameters in the Epidermal Mucus of Various Fish Species.” Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology. 148(3):256–63. Sutherland, Ben J. G., Stuart G. Jantzen, Dan S. Sanderson, Ben F. Koop, and Simon R. M. Jones. 2011. “Differentiating Size-Dependent Responses of Juvenile Pink Salmon (Oncorhynchus Gorbuscha) to Sea Lice (Lepeophtheirus Salmonis) Infections.” Comparative Biochemistry and Physiology Part D: Genomics and Proteomics. 6(2):213– 223. Sutherland, Ben J. G., Kim W. Koczka, Motoshige Yasuike, Stuart G. Jantzen, Ryosuke Yazawa, Ben F. Koop, and Simon R. M. Jones. 2014. “Comparative Transcriptomics of Atlantic Salmo Salar, Chum Oncorhynchus Keta and Pink Salmon O. Gorbuscha during Infections with Salmon Lice Lepeophtheirus Salmonis.” BMC Genomics. 15(1):200.
130 Suzumoto, B. K., C. Schreck, and J. D. McIntyre. 1977. “Relative Resistances of Three Transferrin Genotypes of Coho Salmon (Oncorhynchus Kisutch) and Their Hematological Responses to Bacterial Kidney Disease.” Journal of Fisheries Board of Canada. 34(1):1–8. Szalai, Alexander J., J. E. Bly, and L. W. Clem. 1994. “Changes in Serum Concentrations of Channel Catfish (Ictalurus Punctatus Rafinesque) Phosphorylcholine-Reactive Protein (PRP) in Response to Inflammatory Agents, Low Temperature-Shock and Infection by the Fungus Saprolegnia Sp.” Fish & Shellfish Immunology. 4(5):323–336. Tacchi, Luca, Ralph Bickerdike, Alex Douglas, Christopher J. Secombes, and Samuel A. M. Martin. 2011. “Transcriptomic Responses to Functional Feeds in Atlantic Salmon (Salmo Salar).” Fish & Shellfish Immunology. 31(5):704–715. Tadiso, Tariku Markos, Aleksei Krasnov, Stanko Skugor, Sergey Afanasyev, Ivar Hordvik, and Frank Nilsen. 2011. “Gene Expression Analyses of Immune Responses in Atlantic Salmon during Early Stages of Infection by Salmon Louse (Lepeophtheirus Salmonis) Revealed Bi- Phasic Responses Coinciding with the Copepod-Chalimus Transition.” BMC Genomics. 12(1):141. Tanekhy, Mahmoud, Tomoya Kono, and Masahiro Sakai. 2010. “Cloning, Characterization, and Expression Analysis of Toll-like Receptor-7 CDNA from Common Carp, Cyprinus Carpio L.” Comparative Biochemistry and Physiology Part D: Genomics and Proteomics. 5(4):245–55. Telfer, Sandra, Xavier Lambin, Richard Birtles, Pablo Beldomenico, Sarah Burthe, Steve Paterson, and Mike Begon. 2010. “Species Interactions in a Parasite Community Drive Infection Risk in a Wildlife Population.” Science. 330(6001):243–246. Thompson, K. D., M. F. Tatner, and R. J. Henderson. 1996. “Effects of Dietary (n-3) and (n-6) Polyunsaturated Fatty Acid Ratio on the Immune Response of Atlantic Salmon, Salmo Salar L.” Aquaculture Nutrition. 2(1):21–31. Thorud, K. E. and H. O. Djupvik. 1988. "Infectious Salmon Anemia in Atlantic Salmon (Salmo Salar L.)." Bulletin of the European Association of Fish Pathology. 8(5):109-111. Todd, C. D., R. J. Stevenson, H. Reinardy, and M. G. Ritchie. 2005. “Polyandry in the Ectoparasitic Copepod Lepeophtheirus Salmonis despite Complex Precopulatory and Postcopulatory Mate-Guarding.” Marine Ecology Progress Series. 303:225–234. Tort, L., J. C. Balasch, and S. Mackenzie. 2003. “Fish Immune System. A Crossroads between Innate and Adaptive Responses.” Immunologia. 22(3):277–286. Trobridge, Grant D. and Jo-Ann C. Leong. 1995. “Characterization of a Rainbow Trout Mx Gene.” Journal of Interferon & Cytokine Research. 15(8):691–702. Tunsjø, Hege Smith, Steinar Martin Paulsen, Kristin Berg, Henning Sørum, and Trine Marie L’Abée-Lund. 2009. “The Winter Ulcer Bacterium Moritella Viscosa Demonstrates Adhesion and Cytotoxicity in a Fish Cell Model.” Microbial Pathogenesis. 47(3):134-142. Turnbull, J., B. P. North, C. E. Adams, J. Bron, C. M. MacIntyre, and F. Huntingford. 2008. “Stocking Density and Welfare of Farmed Salmonids.” Fish Welfare. 10:9780470697610. Uribe, C., H. Folch, R. Enriquez, and G. Moran. 2011. "Innate and Adaptive Immunity in Teleost Fish: A Review." Veterinarni Medicina. 56(10):486-503. Valdes-Donoso, P., F. O. Mardones, M. Jarpa, M. Ulloa, T. E. Carpenter, and A. M. Perez. 2013. “Co-Infection Patterns of Infectious Salmon and Sea Lice in Farmed Atlantic Salmon, Salmo Salar L, in Southern Chile (2007-2009).” Journal of Fish Diseases. 36(3):353-360. Vandesompele, J., K. De Preter, F. Pattyn, B. Poppe, N. Van Roy, A. De Paepe, and F. Speleman. 2002. “Accurate Normalization of Real-Time Quantitative RT-PCR Data by
131 Geometric Averaging of Multiple Internal Control Genes.” Genome Biology. 3(7):research0034-1. Vieira, Florbela A., Silvia F. Gregório, Serena Ferraresso, Michael AS Thorne, Rita Costa, Massimo Milan, Luca Bargelloni, Melody S. Clark, Adelino VM Canario, and Deborah M. Power. 2011. “Skin Healing and Scale Regeneration in Fed and Unfed Sea Bream, Sparus Auratus.” BMC Genomics. 12(1):490. Voss, Edward W., J. L. Fryer, and G. M. Banowetz. 1978. “Isolation, Purification, and Partial Characterization of a Lectin from Chinook Salmon Ova.” Archives of Biochemistry and Biophysics. 186(1):25–34. Wagner, G. N., R. S. McKinley, P. A. Bjørn, and B. Finstad. 2003. “Physiological Impact of Sea Lice on Swimming Performance of Atlantic Salmon.” Journal of Fish Biology. 62(5):1000– 1009. Wagner, Glenn N., Mark D. Fast, and Stewart C. Johnson. 2008. “Physiology and Immunology of Lepeophtheirus Salmonis Infections of Salmonids.” Trends in Parasitology. 24(4):176- 183. Walter, T. C. and G. Boxshall. 2017. “World of Copepods Database.” Ergasilus von Nordmann, 1832. Wells, Alan, Christal E. Grierson, Monique Mackenzie, Iain J. Russon, Helena Reinardy, Claire Middlemiss, Pål A. Bjørn, Bengt Finstad, Sjoerd E. Wendelaar Bonga, Christopher D. Todd, and Neil Hazon. 2006. “Physiological Effects of Simultaneous, Abrupt Seawater Entry and Sea Lice (Lepeophtheirus Salmonis) Infestation of Wild, Sea-Run Brown Trout (Salmo Trutta) Smolts.” Canadian Journal of Fish and Aquatic Sciences. 63(12):2809– 2821. Whyte, Shona K. 2007. “The Innate Immune Response of Finfish – A Review of Current Knowledge.” Fish & Shellfish Immunology. 23(6):1127–1151. Willoughby, S. 1999. Manual of Salmonid Farming. Oxford, England: Blackwell Science. Winkelhake, Jeffrey L. and Roxanne J. Chang. 1982. “Acute Phase (C-Reactive) Protein-Like Macromolecules From Rainbow Trout (Salmo Gairdneri).” Developmental & Comparative Immunology. 6(3):481–489. Winter, G., C. Schreck, and J. D. McIntyre. 1980. “Resistance of Different Stocks and Transferrin Genotypes of Coho Salmon, Oncorhynchus Kisutch, and Steelhead Trout, Salmo Gairdneri, to Bacterial Kidney Disease and Vibriosis.” Fish Bulletin. 77(4):795–802. Workenhe, Samuel T., Matthew L. Rise, Molly J. T. Kibenge, and Frederick S. B. Kibenge. 2010. “The Fight between the Teleost Fish Immune Response and Aquatic Viruses.” Molecular Immunology. 47(16):2525–36. Xu, Zhen, David Parra, Daniela Gómez, Irene Salinas, Yong-An Zhang, Louise Von Gersdorff Jørgensen, Rasmus Demuth Heinecke, Kurt Buchmann, Scott Lapatra, and J. Oriol Sunyer. 2013. “Teleost Skin, an Ancient Mucosal Surface That Elicits Gut-like Immune Responses.” Proceedings of the National Academy of Sciences. 110(32):13097-13102. Zhu, Lv-yun, Li Nie, Guan Zhu, Li-xin Xiang, and Jian-zhong Shao. 2013. “Advances in Research of Fish Immune-Relevant Genes: A Comparative Overview of Innate and Adaptive Immunity in Teleosts.” Developmental & Comparative Immunology. 39(1–2):39– 62.
132