EFFECTS OF CATECHOLAMINES ON ruckeri and in ATLANTIC SALMON (Salmo salar)

by

Nguyen Duc Quynh Anh

Bachelor of Science in Aquaculture

Master of Science in Aquaculture

Institute for Marine and Antarctic Studies

University of Tasmania

Submitted in fulfilment of the requirements for the Doctor of Philosophy

University of Tasmania

December 2019 DECLARATIONS BY THE AUTHOR

Statement of Originality

This thesis contains no material which has been accepted for a degree or diploma by the

University or any other institution, except by way of background information and duly acknowledged in the thesis, and to the best of my knowledge and belief no material previously published or written by another person except where due acknowledgement is made in the text of the thesis, nor does the thesis contain any material that infringes copyright.

Authority of Access

This thesis may be made available for loan and limited copying and communication in accordance with the Copyright Act of 1968.

Statement of Ethical Conduct

The research associated with this thesis abides by the international and Australia codes on animal experimentation. Animal ethics permits were obtained from the University of Tasmania Animal Ethics Committee for all aspects of the project where experiments involved live animals (Permit number A0017118)

Signed: Date: 7th December 2019

Nguyen Duc Quynh Anh

ii ACKNOWLEDGEMENTS

The completion of this thesis would not have been possible without the guidance and precious support of several individuals. I would like to thank all who have directly or indirectly contributed to shape my career life during the last four years.

First and foremost, I would like to express my deepest gratitude to my supervisors

Assoc. Prof. Andrew Bridle, Prof. Barbara Nowak, and Dr. Philip Crosbie for their invaluable expertise, continuous guidance, support and tireless patience during my PhD.

My special thank goes to Andrew, who was always available despite of his very busy schedule to give me useful feedback, suggestions in experimental designing, data analysis as well as new practical technique guidance. I will never forget how patient he was to try to explain something complicated for me. I would like to thank Barbara for her care and kind support especially at the beginning of my PhD life and her contribution in writing correction. I am also grateful to Phil for the big help in my experiment sampling (even at the midnight), careful reading and corrections for all my writing drafts as well as his touched care for my little son and my family.

I would like to express my sincere thanks to Dr Mark Adams, Heindrick and Amin for their useful help in setting up my experiment at Aquaculture centre. Special thanks to

Mei for your kindness and willingness to help whenever I asked even in research or normal life, thanks for being my good friend.

I would also like to thank the Vietnamese Government for the finance support to give me opportunity to study and complete my PhD course in Australia. My thanks also go

iii to all staffs of Faculty of Fisheries - Hue University of Agriculture and Forestry for their supporting during my study. I would also feel lucky and really appreciate to receive cares and concerns from “Thay Phuoc”, “chi Hue Linh”, “anh Tuan – chi Hang” and

“co Hong”. Thank you for sharing your experience and helping me feel better whenever

I had problems.

To friends and colleagues: Ylenia, Lucas, Zack, Jimena, Simon, Aaron, Frankie, Em,

Anna, Peter, Thu, Cuong, Trinh, Mai, Ngoc, Duc, Xuyen, Hoang, Nha, Mui and Thao, thanks for a nice friendship and social activities that made my life much more colourful during my stay in Launceston.

I would like to thank my all family members from the bottom of my heart, for their understanding, endless and unconditional support and encouragement that helped me go through all difficulties during my study. I deeply gratitude to my dear parents who have looked after my son since he was few months old. Thanks for loving him as yours another son.

Finally, to my little family, thank you my husband and my little boy for always being with me. You are always in my heart and we will be together soon.

iv TABLE OF CONTENTS

DECLARATIONS BY THE AUTHOR...... ii

ACKNOWLEDGEMENTS ...... iii

TABLE OF CONTENTS ...... v

LIST OF FIGURES...... ix

LIST OF TABLES ...... xiv

ABBREVIATIONS AND UNITS ...... xvi

EXPLANATORY NOTE REGARDING THESIS STRUCTURE ...... xx

ABSTRACT ...... xxi

Chapter 1. GENERAL INTRODUCTION ...... 1

1. 1. INFECTION ...... 1 1. 2. CHARACTERISTICS OF Yersinia ruckeri ...... 5 1.2.1. Morphological and biochemical characteristics ...... 5 1.2.2. Virulence factors...... 8 1.2.2.1. Survival outside the host...... 8 1.2.2.2. Colonisation of the host ...... 9 1.2.2.3. Extracellular products (ECPs)...... 10 1.2.2.4. Iron acquisition ...... 14 1.2.2.5 Plasmids ...... 15 1.2.2.6. Other virulence factors...... 16 1.3. HOST STRESS HORMONES AND THEIR IMPACT ON GROWTH AND VIRULENCE FACTORS OF PATHOGENS ...... 18 1.4. FISH GUT MICROBIOME...... 25 1.5. THESIS AIMS...... 33 Chapter 2. EFFECTS OF CATECHOLAMINES ON GROWTH AND VIRULENCE

OF Yersinia ruckeri ...... 34

2.1. INTRODUCTION...... 34

v 2. 2. MATERIALS AND METHODS ...... 36 2.2.1. Growth conditions for Yersinia ruckeri ...... 36 2.2.2. Calibration curve for absorbance and cell density of Y. ruckeri ...... 37 2.2.3. Hormones ...... 37 2.2.4. Growth assays ...... 37 2.2.5. Lipase, phospholipase, caseinase and haemolysin assays ...... 38 2.2.6. Siderophore assay ...... 39 2.2.7. Biofilm formation assay ...... 40 2.2.8. Antimicrobial assay...... 40 2.2.9. Statistical analysis ...... 41 2.3. RESULTS ...... 42 2.3.1. Stress hormones increase the growth of Y. ruckeri in serum – supplemented medium ...... 42 2.3.2. Lipase, phospholipase, caseinase and haemolysin assays ...... 45 2.3.2.1. Lipase ...... 45 2.3.2.2. Caseinase ...... 47 2.3.2.3. Haemolysin...... 48 2.3.3. Siderophore ...... 50 2.3.4. Biofilm formation ...... 51 2.3.5. MIC assay ...... 52 2. 4. DISCUSSION...... 55 Chapter 3. EFFECTS OF STRESS AND Yersinia ruckeri CHALLENGE ON

ATLANTIC SALMON (Salmo salar) ...... 63

3.1. INTRODUCTION...... 63 3.2. MATERIALS AND METHODS ...... 65 3.2.1. Ethics statement ...... 65 3.2.2. Fish ...... 65 3.2.3. Hormones and pathogen ...... 66 3.2.4. Stress application and Y. ruckeri challenge...... 67 3.2.5. Sample collection ...... 68 3.2.6. Plasma cortisol measurement...... 69

vi 3.2.7. DNA and RNA extraction ...... 70 3.2.8. Yersinia ruckeri load ...... 72 3.2.9. cDNA transcription ...... 73 3.2.10. Immune gene expression ...... 73 3.3. RESULTS ...... 76 3.3.1. Survival rate ...... 76 3.3.2. Plasma cortisol level ...... 78 3.3.3. Yersinia ruckeri load ...... 80 3.3.4. Immune gene expression ...... 81 3.3.4.1. The effect of stress on Atlantic salmon immune genes unchallenged with Y. ruckeri...... 81 3.3.4.2. The effect challenge with Y. ruckeri (unexposed to catecholamine) on Atlantic salmon immune genes ...... 88 3.3.4.3. The effect challenge with Y. ruckeri (exposed and unexposed to catecholamine) on unstressed Atlantic salmon immune genes ...... 95 3.4. DISCUSSION ...... 102 Chapter 4. EFFECTS OF STRESS AND Yersinia ruckeri CHALLENGE ON GUT

MICROBIOME OF ATLANTIC SALMON (Salmo salar) ...... 113

4.1. INTRODUCTION...... 113 4.2. MATERIALS AND METHODS ...... 114 4.2.1. PCR amplification and sequencing ...... 114 4.2.2. Data analysis ...... 115 4.3. RESULTS ...... 117 4.3.1. Characteristics of the sequence data ...... 117 4.3.2. Alpha diversity analyses ...... 117 4.3.3. Beta diversity analyses...... 119 4.3.4. Taxonomic composition ...... 123 4.3.5. Gut microbiome comparisons between fish challenged with Yersinia ruckeri exposed (CAT+) and not exposed to the catecholamine epinephrine (CAT- ) ...... 129 4.3.5.1. Alpha and beta diversity ...... 129 4.3.5.2. Taxonomic composition ...... 130

vii

4.4. DISCUSSION ...... 136 Chapter 5. GENERAL DISCUSSION...... 147

REFERENCE ...... 156

viii

LIST OF FIGURES

Figure 1.1. Catecholamine biosynthesis (Freestone et al., 2008b) ...... 19

Figure 2.1. Effect of epinephrine (A), norepinephrine (B) and dopamine (C) on growth of Yersinia ruckeri in serum – supplemented medium (left) and total area under the growth curve (right) ...... 44

Figure 2.2.a. Opalescent zones (red arrow) surrounding the colonies (black arrow) on lipase assay ...... 45

Figure 2.2.b. Effect of stress hormones on lipase activity of Yersinia ruckeri...... 46

Figure 2.3. Effect of stress hormones on caseinolytic activity of Yersinia ruckeri ..... 47

Figure 2.4. Effect of stress hormones on haemolytic activity of Yersinia ruckeri ...... 49

Figure 2.5. Effect of stress hormones on siderophore activity of Yersinia ruckeri...... 50

Figure 2.6. Effect of stress hormones on biofilm formation of Yersinia ruckeri after 48 hours ...... 51

Figure 2.7.a. Antibacterial activity and MIC50 for erythromycin against Yersinia ruckeri ...... 52

Figure 2.7.b. Antibacterial activity and MIC50 for florfenicol against Yersinia ruckeri

...... 53

Figure 2.7.c. Antibacterial activity and MIC50 for oxolinic acid against Yersinia ruckeri

...... 53

Figure 2.7.d. Antibacterial activity and MIC50 for oxytetracycline against Yersinia ruckeri ...... 54

Figure 2.7.e. Antibacterial activity and MIC50 for streptomycin against Yersinia ruckeri

...... 54

ix Figure 3.1. Survival rate of Atlantic salmon from groups challenged with Yersinia ruckeri exposed (CAT+, n = 171) or not exposed to catecholamine (CAT-, n =177) or unchallenged with bacterium (Ch-, n = 176)...... 77

Figure 3.3.a. Effect of stress (Ch-S+) and no stress (Ch-S-) on Th1 markers TBet and

IL 12β in Atlantic salmon...... 82

Figure 3.3.b. Effect of stress (Ch-S+) and no stress (Ch-S-) on Th1 markers TNFα 1-2 and IFNγ in Atlantic salmon...... 83

Figure 3.3.c. Effect of stress (Ch-S+) and no stress (Ch-S-) on Th2 markers in Atlantic salmon...... 84

Figure 3.3.d. Effect of stress (Ch-S+) and no stress (Ch-S-) on Th17 markers in Atlantic salmon...... 85

Figure 3.3.e. Effect of stress (Ch-S+) and no stress (Ch-S-) on T cell marker CD8α and

CD4-1 in Atlantic salmon...... 86

Figure 3.3.f. Effect of stress (Ch-S+) and no stress (Ch-S-) on Treg marker TGFb and

B cell marker IgM in Atlantic salmon...... 87

Figure 3.4.a. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on Th1 markers TBet and IL 12β in

Atlantic salmon...... 89

Figure 3.4.b. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on Th1 markers TNFα 1-2 and IFNγ in

Atlantic salmon...... 90

Figure 3.4.c. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on Th2 markers in Atlantic salmon. 91

x Figure 3.4.d. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on Th17 markers in Atlantic salmon.

...... 92

Figure 3.4.e. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on T cell marker CD8α and CD4-1 in

Atlantic salmon...... 93

Figure 3.4.f. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on Treg marker TGFb and B cell marker IgM in Atlantic salmon...... 94

Figure 3.5.a. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non- exposure (CAT-S) to epinephrine on Th1 markers TBet and IL 12β in Atlantic salmon.

...... 96

Figure 3.5.b. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non- exposure (CAT-S) to epinephrine on Th1 markers TNFα 1&2 and IFNγ in Atlantic salmon...... 97

Figure 3.5.c. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non- exposure (CAT-S) to epinephrine on Th2 markers in Atlantic salmon...... 98

Figure 3.5.d. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non- exposure (CAT-S) to epinephrine on Th17 markers in Atlantic salmon...... 99

Figure 3.5.e. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non- exposure (CAT-S) to epinephrine on T cell markers in Atlantic salmon...... 100

xi Figure 3.5.f. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non- exposure (CAT-S) to epinephrine on Treg marker TGFb and B cell marker IgM in

Atlantic salmon...... 101

Figure 4.1. Principal coordinate analysis plots based on (A) Bray-Curtis index, (B)

Jaccard index, (C) unweighted UniFrac, (D) unweighted UniFrac distance methods for different sampling time (day); (E), (F) and (G) Bray-Curtis index for challenge, stress and gut region factor...... 122

Figure 4.2. Relative abundance at phylum level in challenge group (A), different sampling time group (B), stress group (C) and gut region group (D)...... 125

Figure 4.3. Significantly greater abundance of of Sphingomonas and Burkholderia on day 40 compared to day 10 and day 0 (Univariate statistical comparisons p < 0.05).

...... 127

Figure 4.4. Abundance at genus level in different gut region (Gut), sampling time (Day), challenge group (Challenge) and stress group (Stress)...... 128

Figure 4.5. Beta diversity showed significantly different microbial community composition between fish challenged with Yersinia ruckeri prior exposed to epinephrine

(CAT+) and no exposed (CAT-) ...... 129

Figure 4.6. Relative abundance at the genus level of hindgut sampling at day 10 in fish challenged with Yersinia ruckeri which prior exposed to catecholamine epinephrine

(CAT+) (bottom) or no exposed (CAT-) (top)...... 132

Figure 4.7. Heatmap abundance at genus level of individual fish (bottom) challenged with Yesinia ruckeri exposed (CAT+) or not exposed to catecholamine epinephrine

(CAT-) ...... 133

xii Figure 4.8. Significantly greater abundance of Yersinia ruckeri in challenged fish without exposed to epinephrine (CAT-) compared to those exposed to epinephrine

(CAT+)...... 135

xiii LIST OF TABLES

Table 1.1. Host species and geographic ranges of Yersinia ruckeri ...... 3

Table 1.2. Biochemical characteristics of Yersinia ruckeri ...... 6

Table 1.3. Some characteristics and biological activities of the extracellular products from Yersinia ruckeri strains and some other aquatic pathogenic ...... 11

Table 1.4. Catecholamines and theirs effects on ...... 21

Table 1.5. Effect of different factors on Atlantic salmon gut microbiome ...... 30

Table 3.1. Experimental design to test the effects of stress (S+) or no stress (S-) and Y. ruckeri, exposed and unexposed to the catecholamine (epinephrine) at 50 µM (CAT+ or CAT- respectively), challenge in Atlantic salmon. Sampling points were at 0h, 12h,

24h (after stress); 3, 10, 20, 30 and 40 days (post challenge)...... 68

Table 3.2. Primers used for immune gene expression analysis ...... 74

Table 3.3. Plasma cortisol levels in Atlantic salmon prior to challenge and post challenge with Yersinia ruckeri exposed (CAT+) or not exposed to catecholamine (CAT-

) and subjected or not to stress (S+ or S- respectively) or unchallenged with bacterium

(Ch-). (n = 6/treatment/time point) ...... 79

Table 3.4. Detectable Yersinia ruckeri load in gut samples (cell equivalents mg-1) of

Atlantic salmon post challenge with Yersinia ruckeri exposed (CAT+) or not exposed to catecholamine (CAT-) and subjected or not to stress (S+ or S- respectively) or unchallenged with bacterium (Ch-), n = 6 per treatment and timepoint...... 80

Table 4.1. Richness and alpha diversity indexes for gut sequence libraries of Salmo salar. The groups are: stressed (S+) or unstressed (S-); challenged with Yersinia ruckeri,

xiv

exposed or not exposed to catecholamine (CAT+ or CAT- respectively); unchallenged with the bacterium (Ch-) and samples at day 0 (Control)...... 118

Table 4.2. Significantly more abundant OTUs at the genus level of hindgut sampling at day 10 in fish challenged with Yersinia ruckeri which prior exposed to catecholamine epinephrine (CAT+) or no exposed (CAT-)...... 134

xv

ABBREVIATIONS AND UNITS

µM Micromole

AFp18 antifeeding prophage 18 toxin

ANOSIM Analysis of similarities

ANOVA analysis of variance

ASVs amplicon sequence variants

Bp base pair

CAS Chrome azurol S

CAT Catecholamine

CD4-1 Cluster of differentiation 4-1

CD8α Cluster of differentiation 8 alpha cDNA complementary deoxyribonucleic acid

Cds cysteine desulfidase

CFU colony forming units

Ch Challenge

DMSO dimethyl sulfoxide

DNA deoxyribonucleic acid

Do Dopamine dpc day post challenge

ECPs extracellular products

EDTA ethylenediaminetetraacetic acid

xvi EF1α elongation factor 1α

ELISA enzyme linked immunosorbent assay

Epi Epinephrine

ERM enteric redmouth disease

FAO Food and agriculture organization of united nations

G Gram g Gravity

GATA3 GATA binding protein 3

H Hour

IFNγ Interferon gamma

IgM Immunoglobulin M

IL Interleukin

Kb Kilobyte

L Litre

LD lethal dose

LPS lipopolysaccharide

MDa Megadalton

MHA Mueller-Hinton agar

MHB Mueller-Hinton broth

MHC histocompatibility complex

MIC minimum inhibitory concentration min Minute mL Millilitre

xvii Mm Millimetre mM Millimole

MM9 minimal media 9 mRNA messenger ribonucleic acid

N normality mole

NE Norepinephrine

NE-AI noradrenaline-induced autoinducer

Nm Nanometre

NMDS Non-Metric Multidimensional Scaling

ºC degree Celsius

OD opticcal density

OTUs operational taxonomic units

PCoA Principal Coordinates Analysis

PERMANOVA Permutational multivariate analysis of variance

QIIME 2 Quantitative Insights Into Microbial Ecology qPCR quantitative real-time RT-PCR

RNA ribonucleic acid

RoRγ RAR-related orphan receptor gamma

Rpm revolutions per minute

S Stress

S Second

SDS sodium dodecyl sulphate

SE standard error

xviii

SPSS Statistical Package for the Social Sciences

Tbet T-box expressed in T cells

TCEP Tris (2-carboxyethyl) phosphine

TGFβ Transforming growth factor beta

Th T helper

TNA total nucleic acid

TNFα Tumor necrosis factor alpha

TRIS Trisaminomethane

TSA Tryptone soya agar

TSB Tryptone soya broth

TTSS type three secretion system

UK United Kingdom

USA United State of America

USD United States Dollar

Yhl Yersinia haemolysin yrIlm Y. ruckeri invasion-like molecule yrInv Y. ruckeri invasion

Yrp Yersinia metaloprotease

Ysa Yersinia secretion apparatus ysps Yersinia secreted proteins

Μm Micrometre

xix

EXPLANATORY NOTE REGARDING THESIS STRUCTURE

Chapters 2, 3 and 4 are drafts of publication manuscripts at the time of submission, hence some repetition of material between the introductory sections of the chapters has occurred. The Harvard referencing style has been used for this thesis.

xx ABSTRACT

Since its inception in the 1980s Atlantic salmon (Salmo salar) aquaculture in Tasmania,

Australia has expanded at an ever-increasing rate and is the highest value commercial fishery related industry in Australia. The growth of the Tasmanian industry is expected to continue as are concerns regarding the environmental impacts and potential diseases associated with the rapidly expanding industry. These concerns are common to all intensive culture environments where stress and its impacts on animal health cannot be avoided. The response of fish to stress is not unlike that of farmed terrestrial animals where stress induces the release of stress hormones including catecholamines that can negatively impact the animal’s immune response. Stress hormones may also affect potential bacterial pathogens as bacteria are able to sense catecholamines resulting in enhanced growth and/or virulence. Yersinia ruckeri is the aetiological agent of enteric redmouth disease or yersiniosis in Atlantic salmon. Despite the success of yersiniosis vaccination disease outbreaks are still reported implying a greater understanding of this bacterium and its relationship to the host and the aquaculture environment is needed to improve fish health management in response to this disease. This research aimed to explore the effects of catecholamines on virulence factors of Y. ruckeri in vitro and the host immune response and gut microbiome changes of Atlantic salmon exposed to catecholamine treated Y. ruckeri.

Exposure of Y. ruckeri in vitro to catecholamines was found to enhance the growth and increase activity of known virulence factors of Y. ruckeri including lipase, haemolysin,

xxi

siderophore and biofilm production while not altering the inhibitory effects of selected antibiotics on Y. ruckeri. More conclusively, epinephrine was found to increase the virulence of Y. ruckeri in vivo as demonstrated from an experimental infection challenge that found Atlantic salmon exposed to catecholamine-treated Y. ruckeri had a higher mortality compared to salmon challenged with untreated Y. ruckeri.

To investigate the effects of stress on the host immune response and Y. ruckeri infection

Atlantic salmon were subject to an acute stress by reducing the tank water levels to 50% and chasing the fish for 10 minutes. This treatment did not induce any significant changes in fish plasma cortisol when samples were taken from 12 hours onwards post- acute stressor. The survival rate of fish subjected to the acute stressor post challenge with Y. ruckeri was not significantly different to that of unstressed fish. Differential gene expression of select host immune-related genes was found to be related to both stress and/or disease challenge. Fish subjected to the acute stressor had a significant upregulation of whole blood interleukin 12β (IL12β) and downregulation of cluster of differentiation 8 alpha (CD8α) at day 2 after the stress event. When assessing the host response to disease challenge a 25-fold increase in the mRNA expression of the interferon IFNγ was found compared to unchallenged fish at day 1 post infection. The relative expression of interleukin 12 beta (IL 12β) in challenged fish was also significantly expressed at day 10 in comparison to day 1 while a significant reduction was observed in tumor necrosis factor alpha (TNFα), IFNγ and interleukin 6 (IL 6) at day 30 compared to day 1. Prior exposure Y. ruckeri to epinephrine significantly increased the expression of immunoglobin M (IgM) (day1) (2.5-fold), TNFα (3.8-fold)

xxii

and interleukin 22 (IL 22) (8.4-fold) (day 30) compared to fish challenged with control

Y. ruckeri. In contrast, the expressions of IFNγ and IL 6 were 39.5 and 17.2-fold lower in Y. ruckeri exposed to epinephrine at day 1 suggesting that a reduction in these two important immune genes is associated with increased virulence.

Microbiomes were assessed using 16S rRNA amplicon sequencing where the most abundant phylum in the gut microbiome of Atlantic salmon was (42%),

Actinobacteria (19%) and Firmicutes (19%). At genus level, the major genera were

Propionibacterium (19%), Sphingomonas (14%), Aeromonas (11%) and

Staphylococcus (8%). No significant difference in alpha diversity indices was found.

Analysis of beta diversity showed no difference in the community structure between the midgut and hindgut. Importantly, significant differences were identified the between stressed and unstressed fish, fish challenged (with and without epinephrine treated Y. ruckeri) and unchallenged as well as between different sampling times. Abundance of families Aeromonadaceae and Staphylococcaceae was more dependent on different challenge treatments while abundance of Sphingomonadaceae and Burkholderiaceae was related to sampling time.

The findings of this thesis confirm that Y. ruckeri is able to sense stress-related hormones in the form of catecholamines and provides potential mechanisms at the transcriptional and functional level of how stress to the host translates to increased susceptibility to disease. These findings improve our understanding of the interaction between stress, virulence factors of Y. ruckeri and the Atlantic salmon immune xxiii

response. Furthermore, apart from consolidating the accepted view that stress negatively impacts fish health and should be minimised it introduces the possibility of manipulating the host microbiome to potentially overcome or minimise the consequences of stressors on fish health that are common to aquaculture.

xxiv

Chapter 1. GENERAL INTRODUCTION

The general objective of this study was to evaluate the impact of catecholamine stress hormones on the virulence of Yersinia ruckeri and how the host Atlantic salmon (Salmo salar) immune genes and the gut microbial communities respond to stressors and a bacterial disease challenge. Hence, this chapter summarised current knowledge focusing on Y. ruckeri virulence and its infection (section 1.1 and 1.2) as well effects of catecholamines on growth and virulence of bacteria (section 1.3). In addition, general information about the fish gut microbiome was presented in section 1.4.

1. 1. Yersinia ruckeri INFECTION

Yersinia ruckeri is the causative agent of both yersiniosis in Atlantic salmon in the

Southern Hemisphere and enteric redmouth disease (ERM) in

(Oncorhynchus mykiss) in the Northern Hemisphere (Costa et al., 2011). Enteric redmouth disease was first described in the 1950s in the USA and is continuously reported in Europe, North America, Australia, South Africa, the Middle East and China

(Davies, 1991; Tobback et al., 2007; Kumar et al., 2015). The disease has a significant economic impact on salmonid farming worldwide as shown by an estimated 2.5 million

USD loss reported for the USA trout industry (Woo and Bruno, 2014). In the USA in

1998 ERM reportedly was responsible for 84% of the 34.3 million total trout mortalities

(Woo et al., 2002). The mortality ranged from 10-50% (over a 2-6 month period) to 30-

70% dependant on whether the disease was in a chronic or acute form (Furones et al.,

1993; Barnes, 2011). The chronic form can lead to an acute epizootic under

1

unfavourable on-farm conditions, particularly in high density stocks and stressful conditions and hence may result in heavy losses (Furones et al., 1993).

Since the first report of ERM in rainbow trout, Y. ruckeri has been isolated from many other fish all over the world (see Table 1.1) which demonstrates a wide range of hosts and a broad geographic distribution (Kumar et al., 2015). Yersiniosis can affect any fish species from all ages but is more acute in young fish up to fingerling size. In larger fish, the disease appears as a more chronic condition (Kumar et al., 2015). Affected fish may change their behaviour, including swimming slowly near the surface and have a loss of appetite. Gross clinical signs of the disease include haemorrhages on the body surface, reddening at the base of the fins and along the lateral line and on the head as well as exophthalmia and darkening of the skin (Tobback et al., 2007; Kumar et al., 2015).

Internally, petechial haemorrhage may occur on the surface of the liver, pancreas, pyloric caeca, swimbladder and in the lateral muscles and the spleen is often enlarged and almost black in colour. The lower intestine can become inflamed and filled with an opaque, yellowish fluid. Histological examination shows general septicaemia with inflammation in all tissues, particularly in the kidney, spleen, heart, liver, gills and in areas with petechial haemorrhages. (Tobback et al., 2007; Kumar et al., 2015).

2

Table 1.1. Host species and geographic ranges of Yersinia ruckeri

Host species Country Reference

Oncorhynchus mykiss Australia, Bulgaria, Croatia, Ross et al. (1966); Czechoslovakia, Denmark, Bullock et al. (1977); Finland, France, Germany, Gelev et al. (1983); Greece, Poland, Portugal, Fuhrmann et al. (1984); Spain, Switzerland, Turkey, Bragg and Henton UK, Canada, USA, Peru, (1986); Valtonen et al. India, South Africa, Iran (1992); Oraić et al. (2002); Bravo and Kojagura (2004); Fouz et al. (2006); Akhlaghi and Sharifi Yazdi (2008); Pedersen et al. (2008); Barnes (2011)

Oncorhynchus kisutch USA Arkoosh et al. (2004)

Oncorhynchus nerka USA Ross et al. (1966)

Oncorhynchus tshawytscha USA, New Zealand Arkoosh et al. (2004); Barnes et al. (2016)

Salmo clarkia USA Bullock et al. (1978)

Salmo salar Australia, Chile, Finland, Valtonen et al. (1992); Norway, UK, USA, Barnes et al. (2016)

Salmo trutta Germany, USA Fuhrmann et al. (1984); Furones et al. (1993)

Acipenser baeri France Vuillaume et al. (1987)

3

Anguilla Anguilla Germany Fuhrmann et al. (1984)

Carasius auratus Singapore McArdle and Dooley- Martyn (1985)

Coregonus artedii Canada Stevenson and Daly (1982)

Coregonus peled Finland Valtonen et al. (1992)

Coregonus muksun Finland Rintamaki et al. (1986)

Cyprinus carpio Germany, Hungary, Chile Fuhrmann et al. (1984); Enriquez and Zamora (1987); Berc et al. (1999)

Ictalurus punctatus USA Danley et al. (1999)

Lota lota Canada Dwilow et al. (1987)

Oreochromis niloticus Egypt Eissa et al. (2008)

Pimephales promelas France Michel et al. (1986)

Perca fluviatilis Finland Valtonen et al. (1992)

Rutilis rutilis Finland Valtonen et al. (1992)

4

1. 2. CHARACTERISTICS OF Yersinia ruckeri

1.2.1. Morphological and biochemical characteristics

Yersinia ruckeri belongs to the family within the subdivision. This Gram – negative bacterium has coccoid-rod cell morphology and is 1.0 µm in diameter and 2 – 3 µm in length (Barnes, 2011) however its size and morphology may change in different culture conditions. The organism can grow in a wide range of temperatures and has an optimum of 28ºC. The bacterial cells do not form endospores and have no capsule whereas some strains are able to move with peritrichous flagella (Austin et al., 1982).

Biochemically, as a member of family Enterobacteriaceae, Y. ruckeri is glucose – fermentative, catalase – positive, oxidase – negative and nitrate – reductive (Tinsley,

2010). It can be distinguished from other species based on production of lysine decarboxylase, ornithine decarboxylase and β- galactosidase, whereas hydrogen sulphide and indole are not present (Tobback et al., 2007; Tinsley, 2010) (Table 1.2).

5

Table 1.2. Biochemical characteristics of Yersinia ruckeri

Source: Tinsley (2010)

Test Result Test Result

Fermentation + Utilisation of sodium citrate +

Production of: Production of acid from:

Arginine dihydrolase - Fructose +

Catalase + Glucose +

β- galactosidase + Inositol -

Hydrogen sulphide - Lactose -

Indole - Maltose +

Lysine decarboxylase + Mannitol +

Ornithine decarboxylase + Raffinose -

Oxidase-phenylalanine - Salicin - deaminase

Phosphatase - Sorbitol -

Methyl red test v Sucrose -

Reduction of nitrate + Trehalose +

Voges-Proskauer reaction v Xylose -

Degradation of: Degradation of:

Aesculin - Tributyrin V

6

Chitin - Tween 20 V

DNA - Tween 40 V

Elastin V Tween 60 V

Gelatin V Tween 80 V

Pectin - Urea -

+: ≥80% showing positive effect, -: ≤20% showing positive effect and v: 21-79% of positive effect

This species can be divided into two biotypes based on motility and production of lipase.

Biotype 1 is positive for both motility and lipase activity whereas biotype 2 is negative for both. Isolates can also be classified by whole-cell reactions as five major serovars; serovar I – Hagerman, most virulent; serovar II – Oregon, avirulent; serovar III –

Australian, avirulent; serovar V – Colorado, avirulent and serovar VI – Ontario, avirulent (Barnes, 2011). The application of electrophoresis and immunoblots identified six O-serogroups were based on lipopolysaccharide O – antigen namely from O1 to O6

(Barnes, 2011). Another classification has been suggested by using lipopolysaccharide and membrane proteins to group the species within four serotypes (O1 – O4). Serotype

O1 was subdivided into two groups O1a (former serovar I) and O1b (former serovar III) while serotype O2 had three subgroups from the former serovar II including O2a, O2b and O2c. Serotype O3 and O4 were comprised of serovar V and VI respectively

(Barnes, 2011). Recently, a new serotype O8 was identified as the most common serotype isolated from Atlantic salmon in Scotland from 2006 to 2011 and in 2014

(Ormsby et al., 2016). Interestingly, this new serotype was also described in two

7

Scottish rainbow trout isolates, and this was explained by recombinational events in the lipopolysaccharide biosynthesis cascade (Gulla et al., 2018).

1.2.2. Virulence factors

1.2.2.1. Survival outside the host

Several studies have determined that Y. ruckeri is able to survive in both water and sediments for more than 3 months, however survival was lower in water compared to sediments and was dependant on salinity (Romalde et al., 1994a; Romalde et al.,

1994b). The bacterium can survive in fresh or brackish water (salinity 0 – 20‰) for at least 4 months but survival diminished quickly in water salinity of 35‰ (Thorsen et al.,

1992; Romalde et al., 1994b). Yersinia ruckeri is also capable of forming biofilms on different surfaces such as common fish farm materials, including polyvinylchloride, fibreglass, concrete and wood which provide a suitable reservoir for re-infection

(Coquet et al., 2002; Barnes, 2011). Interestingly, cells from immobile surfaces

(biofilms) are associated with changes in outer membrane proteins in comparison to free-swimming cells (Coquet et al., 2005).

Yersinia ruckeri can produce water-soluble antimicrobial compounds that inhibit Vibrio anguillarum, and Aeromonas salmonicida and give it a competitive advantage for substrate and host colonisation (Michel and Bernadette,

1987). This should be considered as part of a survival strategy outside the host and may be important for early stage of infection (Barnes, 2011).

8

1.2.2.2. Colonisation of the host

Adhesion and entry into the host are the first steps in the establishment of a bacterial infection (Ruwandeepika et al., 2012). Although the cell surface properties of Y. ruckeri show that the cells are neither hydrophobic nor agglutinating, the capacity of Y. ruckeri to adhere and invade fish cell lines cultured in vitro has been reported (Toranzo and

Barja, 1993). However, these characteristics are dependent on the fish cell line employed (Romalde and Toranzo, 1993; Toranzo and Barja, 1993; Tobback et al.,

2007). Kawula et al. (1996) reported lower levels of invasion rates of Y. ruckeri for rainbow trout kidney cells lines compared to those of fathead minnow (Pimephales promelas) epithelial cells. A recent study showed that two isolates of Y. ruckeri invaded different types of salmon epithelial cells (Atlantic salmon kidney, salmon head kidney and Chinook salmon embryo cells) using cell culture method. Interestingly the biotype

1 isolate ATCC 29473 was more infectious than the non-motile biotype 2 strain A7959-

11 despite the latter originated from a clinical case and was more virulent than the first one, hence this suggested that flagellum which is responsible for motility of Y. ruckeri plays important roles in attachment and entry into the host (Menanteau-Ledouble et al.,

2018). However, none of the eight strains of Y. ruckeri tested was capable of binding to carp epithelial cells (Santos et al., 1990). Later studies on Y. ruckeri infection routes indicated that this bacterium can access the interior of rainbow trout by at least three different paths; through the gills, lateral line and digestive tract (Ohtani et al., 2014).

The two putative inverse autotransporter genes, Y. ruckeri invasin (yrInv) and invasion- like molecule (yrIlm), have been found to be upregulated in a medium that mimics the environmental conditions present in the fish host (Wrobel et al., 2018), which suggests

9

that these genes may be involved in the infection process with Y. ruckeri (Guijarro et al., 2018b)

1.2.2.3. Extracellular products (ECPs)

It has been reported that extracellular products (ECPs) from Y. ruckeri contain caseinase, gelatinase, phospholipase, lipase and haemolysin (Barnes, 2011). An intraperitoneal injection of ECPs from Y. ruckeri into rainbow trout resulted in haemorrhages in the mouth which are characteristic of yersiniosis (Romalde and

Toranzo, 1993). The role of ECPs in the pathogenic mechanisms of Y. ruckeri was also shown when Romalde and Toranzo (1993), reported an LD50 for rainbow trout of 2 to

9.12 µg of Y. ruckeri ECP (protein) per g of fish. This LD50 was similar to those obtained for other fish bacterial pathogens such as Aeromonas hydrophila, A. salmonicida and Vibrio anguillarum (Table 1.3) (Romalde and Toranzo, 1993; Toranzo and Barja, 1993).

10

Table 1.3. Some characteristics and biological activities of the extracellular products from Yersinia ruckeri strains and some other aquatic pathogenic bacteria

Source: (Romalde and Toranzo, 1993; Toranzo and Barja, 1993)

Bacteria Fish LD50 (µg Enzymatic activities ECP /g fish) Cas Gel Ela Pho Lip Amy Hae

Y. ruckeri

NCMB1316 Trout 4.7 + + - + + + +

11.4 Trout 5.5 + + - + + + +

FP13 Trout 6.2 + + - + + + +

PP31 Trout 3.3 + + - + + + +

1533/87 Trout 8.7 + + - + + + +

RS54 Trout 4.3 + + - + + + +

RS2 Trout 2.0 + + - + + + +

11.29 Trout 8.3 + + - + + + +

1638/87 Trout 7.0 + + - + + + +

RS6 Trout 2.0 + + - + + + +

11.47 Trout 9.1 + + - + + + +

11.73 Trout 4.3 + + - + + + +

A. hydrophila B-32 Trout 2.1 + + + + + + +

A. salmonicida Salmon 2.5 + + V + NR NR +

Trout 9.8 + + V + NR NR +

11

V. anguillarum R-8 Trout 5.5 + + + - - + +

V. anguillarum Turbot 4.5 – 6.0 + + V - NR NR +

Salmon 5.8 – 7.3 + + V - NR NR +

Cas: caseinase, Gel: gelatinase, Ela: elastase, Pho: phospholipase, Lip: lipase, Amy: amylase, Hae: haemolysin (fish)

LD50: lethal dose needed to kill 50% of the challenged fish, V: variable results among strains, NR: not reported

Multiple molecules contribute to the virulence of Y. ruckeri ECPs including the iron- regulated -like haemolysin YhlA, the azocasein protease and the 47kDa metalloprotease yrp1 (Kumar et al., 2015). Yrp1 is a serralysin family extracellular protein, secreted through an ATP-binding cassette (ABC) transport system and able to degrade fibronectin, actin and myosin (Fernández et al., 2002; Kumar et al., 2015).

According to Tobback et al. (2007), yrp1 is produced at the end of the exponential growth phase and contributes to the virulence of the bacterium in the invasion and colonization processes of different tissues. Yrp1 protease digests a wide variety of extracellular matrix and muscle proteins, hence may cause the leaking of blood through membranes and pores in the capillary vessels which may result in typical haemorrhages especially in oral cavities and intestines (Fernandez et al., 2003; Tobback et al., 2007).

Recently, transcriptional fusion between the luxCDABE operon and the yrp1 promoter indicated that yrp1 is expressed more in fish tissues (in vivo) than in vitro (Méndez and

Guijarro, 2013; Guijarro et al., 2018b). Interestingly, yrp1 was found to be dependent on temperature as higher levels of expression were found at 18ºC (the temperature

12

around which outbreaks of the diseases usually occur) compared to 28ºC (the optimal temperature for growth) (Fernandez et al., 2003; Guijarro et al., 2015b).

The two putative peptidase enzymes yrpA and yrpB belonging to the U32 family encoded in yrpAB operon have also been identified in Y. ruckeri (Navais et al., 2014).

These genes are present in a similar genetic organisation in different pathogens including human pathogenic Yersinia species and infection study with a yrpA mutant strain showed that this enzyme contributes to virulence in a similar way to the U32 peptidases of and (Navais et al., 2014; Guijarro et al., 2018b).

The yhlA gene encodes a Serratia-type haemolysin while the gene yhlB functions in the secretion or activation of yhlA. Evidence suggests that these genes play important roles in bacterial pathogenesis as virulence attenuation of Y. ruckeri against the bluegill fry

(Lepomis macrochirus) cell line (BF-2) was obtained when mutants of yhlA and yhlB were inserted in the bacterium (Guijarro et al., 2018b). As with the yrp1 gene the YhlAB operon also displayed elevated expression at 18ºC compared to 28ºC. The presence of yhlAB operon and haemolytic activity in different Y. ruckeri strains from different origins and geographical locations suggests this operon is important for virulence of Y. ruckeri (Fernández et al., 2007b).

The involvement of two cysteine desulfidase (cds) genes, cdsA and cdsB, in uptake and further degradation of L-cysteine was also suggested to contribute to virulence of Y.

13

ruckeri (Méndez et al., 2011). The cdsAB operon was found in several anaerobic and facultative bacteria and was specifically induced by L-cysteine under low oxygen conditions. It was suggested that it could be involved in the generation of iron-sulfur- centers for Fe-S proteins or in the accumulation of glutathione which play important roles in detoxification and act as a redox buffer molecule in the bacterium (Méndez et al., 2011; Guijarro et al., 2018b).

1.2.2.4. Iron acquisition

Acquisition of iron inside the host is very important for bacterial pathogens, and their ability to grow under a low free iron concentration is an important virulence factor

(Toranzo and Barja, 1993). Yersinia ruckeri possess ruckerbactin, a catecholate siderophore which is secreted and facilitates iron uptake from the host enabling the bacterium to multiply (Fernandez et al., 2004). Siderophores are small molecules with a high Fe3+ binding affinity which allows them to bind host iron and transport it back to the bacterial cell (Fernández et al., 2007b; Tobback et al., 2007). Siderophores can recognise a specific outer membrane receptor and are translocated into the cytosol of the bacterium where the iron is discharged and utilised for different metabolic pathways.

Genes involved in the siderophore pathway of Y. ruckeri (i.e. the biosynthesis of ruckerbactin) have been shown to be up-regulated during the bacterial infection of fish however when these genes were inactivated the bacterial LD50 for trout was shown to increase a hundred-fold which highlights the significant role that this siderophore plays in virulence of Y. ruckeri (Fernandez et al., 2004).

14

1.2.2.5 Plasmids

In bacteria, virulence factors can be encoded in plasmids and in the genus Yersinia, it is shown to be associated with the plasmid pYV (40-50 MDa). This plasmid codes for several characteristics including autoagglutination and calcium dependency, hydrophobicity, cytotoxicity and resistance to the bactericidal action of host serum, adherence and invasiveness and surface antigens (Toranzo and Barja, 1993). The presence of plasmids in Y. ruckeri has been reported by many authors, however, their role in virulence is not clear (Tobback et al., 2007). A large 75 MDa plasmid is found in most isolates regardless of their geographical origin while the smaller 70 kb plasmid encodes a type three secretion system (TTSS) (Tobback et al., 2007).

A TTSS system allows direct communication between bacteria and host cells. By using this system a bacterium can inject effector proteins into the cytosol of the host and modulate the cell functions to its advantage (Tobback et al., 2007). In Y. ruckeri, a TTSS system is different from that shared by human pathogenic Yersinia species. Gunasena et al. (2003) reported that sequencing analysis of the gene encoding the most highly conserved protein of TTSS in fish pathogens (the type three ATPase and adjacent genes) revealed a significant homology with the chromosomally encoded Yersinia secretion apparatus (ysa) TTSS genes of Y. enterocolitica biovar 1B. This suggests that a ysa-like

TTSS may be present in Y. ruckeri. The ysa TTSS of Y. enterocolitica biovar 1B is implicated in the gastrointestinal stages of infection and the delivery of effector proteins, Yersinia secreted proteins (ysps), into host cells resulting in infection

15

(Tobback et al., 2007). However, more research is needed to reveal the presence and the function of the Ysa TTSS in Y. ruckeri.

Recently, the pYR1 plasmid has been described in Y. ruckeri, which carries several antibiotic resistance determinants and encodes a type IV conjugative transfer system as well (Welch et al., 2007; Méndez et al., 2009). Although, the role of this type IV system in virulence is not fully known, it may be required for intracellular survival of Y. ruckeri and regulated by temperature with elevated expression at 18ºC (as seen with yrp1 protease, yhlA haemolysin and the ruckerbactin system) (Méndez et al., 2009).

1.2.2.6. Other virulence factors

The high-affinity zinc transporter znuABC operon is widely present in bacteria such as

Brucella abortus, and and contributes to their virulence (Guijarro et al., 2018b). The znuABC is required for Y. ruckeri survival in fish and an infection study by Dahiya and Stevenson (2010b) where the bacterium had a znuA mutant infection loads were 150 to 350-fold lower in the kidney of rainbow trout compared to the parental strain. Hence, this zinc transport system has an important role in the infection process. In another study, the BarA-UvrY two component system was demonstrated to be involved in pathogenesis of Y. ruckeri by regulating invasion of epithelial cells and increasing sensitivity to oxidative stress induced by immune cells

(Dahiya and Stevenson, 2010a).

16

Recently the lpxD gene, which encodes for lipopolysaccharide (LPS) biosynthesis, has been reported to contribute to virulence of Y. ruckeri as deletion of this gene resulted in an LPS-deficient strain (Altinok et al., 2016). This lpxD mutant caused significant attenuation of virulence compared to wildtype strain in rainbow trout (Altinok et al.,

2016). Another interesting virulence factor of Y. ruckeri is the antifeeding prophage 18 toxin (Afp18). This toxin has been shown to act as a virulence factor during the infection of insects with (Hurst et al., 2007). Microinjection of

Y. ruckeri Afp18 into zebrafish (Danio rerio) embryos was reported to block cytokinesis, actin-dependent motility and cell blebbing and abrogating gastrulation

(Jank et al., 2015). However, Afp18 effects were only tested on zebrafish embryos and it could be interesting to study its role in Y. ruckeri infection of rainbow trout by generating a mutant strain (Guijarro et al., 2018b).

A recent study has indicated flagellar regulation mediated by the regulation of capsular synthesis (Rcs) pathway was required for virulence of Y. ruckeri (Jozwick et al., 2019).

This study showed expression of the flagellin locus fliC was repressed during the infection and upregulated upon host death when tested in rainbow trout. The authors suggested this expression profile may be necessary to avoid flagellin-mediated host recognition which could lead to attenuation of pathogenesis during the infection process. In addition, the Rcs pathway was demonstrated to down-regulate biosynthesis of the flagellar apparatus as deletion of RcsB resulted in overproduction of flagellin and unregulated motility. Experimental challenge with mutants of the Rcs pathway also resulted in attenuation of virulence. Hence, it could be concluded that Rcs regulatory

17

system is responsible for sensing host signals and regulating related genes to adapt to the host environment during the course of infection (Jozwick et al., 2019).

1.3. HOST STRESS HORMONES AND THEIR IMPACT ON GROWTH AND

VIRULENCE FACTORS OF PATHOGENS

A hormone is defined as a chemical substance produced in the body by an organ, or scattered cells which possess a specific regulatory effect on the activity of an organ or several organs (Burtis, 2006). Similar to mammalian hormones, fish hormones can be divided into three major classes, including proteins (or peptides), steroids (a subclass of lipidic hormones) and amino acid derivatives (or amines) (Norman and Litwack, 1997;

Burtis, 2006; Hughes and Sperandio, 2008). The structure of a hormone dictates the location of its receptor. Steroid hormones can cross plasma membranes and primarily bind to intracellular receptors, while amine and peptide hormones have to bind to cell- surface receptors (such as receptor kinases and G protein coupled receptors) since they cannot cross the cell membrane (Hughes and Sperandio, 2008). Peptide hormones include the epidermal growth factor, insulin and glucagon. Most hormones are proteins and peptides, which contain 3–200 amino acids and usually are post-translationally processed. Steroid hormones are derived from cholesterol, whereas amines are synthesized from tyrosine. Amine hormones include the catecholamines adrenaline

(epinephrine), noradrenaline (norepinephrine) and dopamine (Figure 1.1). All of these hormones are engaged in inter-kingdom signalling with microorganisms (Hughes and

Sperandio, 2008).

18

Figure 1.1. Catecholamine biosynthesis (Freestone et al., 2008b)

Microorganisms have a long co-existence with animals and plants, hence they possess detection systems for sensing host-associated chemicals such as hormones (Hughes and

Sperandio, 2008; Sharaff and Freestone, 2011). These hormone-sensors play important roles in determination of the locality of a suitable host, gene expression involved in host colonization and, in the case of pathogenic species, virulence determinants (Sharaff and

Freestone, 2011). Most of the studies in microbial endocrinology have focused on the interaction of bacteria with the hormones released during stress such as the catecholamine “fight and flight” hormones, including epinephrine, norepinephrine and dopamine (Lyte and Freestone, 2010; Sharaff and Freestone, 2011). It has been reported

19

that stress in humans and animals leads to an increased risk of infection (Freestone and

Lyte, 2010; Sharaff and Freestone, 2011). Reiche et al. (2005b) reported that an increase in stress hormone production, mainly adrenaline and noradrenaline, resulted in a significant reduction of cell based immune function in animals.

Several species of bacteria have been shown to respond to the administration of catecholamines with increased growth (in culture) and virulence (Kinney et al., 1999;

Belay et al., 2003; Sharaff and Freestone, 2011) (see Table 1.4). Lyte and Ernst (1992) indicated that both and Y. enterocolitica showed increased bacterial numbers after 24 hours culture with norepinephrine, epinephrine or dopamine compared to control cultures without the hormones. The same enhancing effect of norepinephrine has also been exhibited in Listeria monocytogenes (Coulanges et al., 1997). In addition,

Kinney et al. (1999) reported that cultures of A. hydrophila, a Gram-negative bacillus found in brackish water causing diseases in fish and frogs, when treated with norepinephrine, epinephrine, dopamine or isoproterenol showed a dramatic increase in growth with 4.5-log greater bacterial numbers than control cultures. Later, Belay et al.

(2003) indicated that supplementation of minimal medium with norepinephrine, epinephrine, dopamine or isoproterenol resulted in a significant increase in growth of

Pseudomonas aeruginosa, , E. coli and Staphylococcus aureus compared to controls. Recently, Yang et al. (2014) indicated that the addition of norepinephrine and dopamine increased the growth of Vibrio harveyi more than 2 times in marine broth containing 30% serum compared to controls. Similarly, Pande et al.

(2014) found that the presence of 100µM of norepinephrine and dopamine could

20

increase the growth of V. campbellii and V. anguillarum in medium containing 30% of

serum also relative to controls. Although the results suggested that catecholamines can

enhance growth of pathogenic bacteria, there was no uniform effect of catecholamines

on bacterial growth (Belay et al., 2003).

Table 1.4. Catecholamines and their effects on pathogenic bacteria

Hormones Bacterial species Growth Virulence References Nor- Acinetobacter lwoffii ↑ NT Freestone et al. (1999) epinephrine Aeromonas hydrophila ↑ ↑ Kinney et al. (1999); Gao et al. (2019) Bordetella bronchiseptica ↑ ↑ Anderson and Armstrong (2006) Campylobacter jejuni ↑ ↑ Roberts et al. (2002); Cogan et al. (2007) ↑ NT Freestone et al. (1999) Citrobacter rodentium NT ↑ Moreira et al. (2016) Escherichia coli ↑ ↑ Freestone et al. (2003) Hafnia alvei ↑ NT Freestone et al. (1999) Klebsiella pneumoniae ↑ NT Belay and Sonnenfeld (2002) Roberts et al. (2002) Leptotrichia buccalis ↑ NT Coulanges et al. (1998) Listeria monocytogenes ↑ NT Freestone et al. (1999) Morganella morganii ↑ NT Freestone et al. (1999) Proteus mirabilis ↑ NT Belay et al. (2003) ↑ ↑ O'Donnell et al. (2006) ↑ NT Freestone et al. (2008a) Streptococcus aureus ↑ ↑ Pande et al. (2014) Vibrio anguillarum ↑ ↑ Pande et al. (2014); Vibrio harveyi ↑ ↑ Yang et al. (2014) Lacoste et al. (2001)

21

Vibrio splendidus ↑ ↑ Nakano et al. (2007a) ↑ ↑ Freestone et al. (1999) Xanthomonas maltophilia ↑ NT Torabi Delshad et al. Yersinia ruckeri ↑ ↑ (2019)

Epinephrine Aeromonas hydrophila ↑ NT Kinney et al. (1999) Citrobacter rodentium ↑ ↑ Moreira et al. (2016) Escherichia coli ↑ ↑ Freestone et al. (2003) Leptotrichia buccalis ↑ NT Roberts et al. (2002) Pseudomonas aeruginosa ↑ NT Belay et al. (2003) Dopamine Aeromonas hydrophila ↑ NT Kinney et al. (1999); Freestone et al. (2003); Escherichia coli ↑ ↑ Freestone et al. (2008a) Pseudomonas aeruginosa ↑ ↑ Alverdy et al. (2000); Vibrio harveyi ↑ ↑ Pande et al. (2014); Yang et al. (2014) Vibrio anguillarum ↑ ↑ Pande et al. (2014) Yersinia ruckeri ↑ ↑ Torabi Delshad et al. (2019) The “↑” indicates enhanced growth or virulence of bacteria in vitro; NT: not tested

Catecholamines can enhance bacterial growth based on two mechanisms, the ability of

bacteria to utilize iron from catecholamines and the catecholamine itself as a bacterial

growth stimulator (Freestone et al., 1999; Sharaff and Freestone, 2011). Firstly, iron is

one of the essential nutrients for growth of all bacterial pathogens, and its production is

limited in blood and mucosal secretions because of the presence of transferrin and

lactoferrin (Faraldo-Gómez and Sansom, 2003; Krewulak and Vogel, 2008; Sharaff and

Freestone, 2011). It has been shown that catecholamine complex formation with

transferrin and lactoferrin results in reduction of iron, from ferric to ferrous, a valency

22

for which transferrin and lactoferrin have a much lower affinity, resulting in rapid iron loss (Sharaff and Freestone, 2011). This iron theft process by catecholamines can be considered as a growth stimulation for bacteria, resulting significant increases of bacterial cells of many species when catecholamines are added to the blood (Lyte and

Ernst, 1992; Coulanges et al., 1997; Kinney et al., 1999; Belay et al., 2003; Sharaff and

Freestone, 2011). Another mechanism by which catecholamines can induce growth of

Gram-negative bacteria, particularly enteric species, involves induction of a bacterial growth stimulator (Lyte et al., 1996; Freestone et al., 1999; Sharaff and Freestone,

2011). According to Freestone et al. (1999), this growth stimulator is termed the NE-AI

(noradrenaline-induced autoinducer) to distinguish it from homoserine lactone type autoinducers. The NE-AI induces its own synthesis and has cross-species functionality and can induce an increase in bacterial growth of a magnitude similar to that seen with the catecholamines (Freestone et al., 1999). The mechanism by which the NE-AI stimulates growth is unsure but has been shown to be independent of transferrin or lactoferrin (Freestone et al., 2003).

Many reports have demonstrated the correlation between catecholamine accumulation and the increase of production of virulence-associated factors such as toxins, adhesions, biofilm formation and quorum sensing in many Gram-negative bacteria (Lesouhaitier et al., 2009). Catecholamines have also been reported to increase the production of

Shiga toxins in E. coli O157:H7 (Lyte et al., 1996). Increase of Shiga toxin production is significant in enterohaemorrhagic E. coli, as it can cause acute renal and neurological complications via damage to microvascular endothelial cells (Lyte et al., 1996; Hughes

23

and Sperandio, 2008; Sharaff and Freestone, 2011). In addition, many in vitro studies have shown that stress hormones can drive enhancement of bacterial attachment to host tissues through increased expression of pili and fimbriae facilitators (Sharaff and

Freestone, 2011). Similar effects of catecholamines on adherence of E. coli have also been reported (Chen et al., 2003; Green et al., 2004; Vlisidou et al., 2004; Sharaff and

Freestone, 2011). Other enteric pathogens have also been shown to respond to stress hormones for example Nakano et al. (2007b) reported that in V. parahaemolyticus, norepinephrine stimulated the expression of three genes (vscQ, vscR and vscU) associated with TTSS1 which contributed to the cytotoxicity of V. parahaemolyticus.

Additionally, during an in vivo test, a high concentration of norepinephrine (100μM) was shown to stimulate the enteropathogenicity of V. parahaemolyticus in rats (Nakano et al., 2007b). Also norepinephrine was shown to increase the expression of the PA-I lectin which plays a key role in the pathogenicity of P. aeruginosa in the intestinal epithelium (Alverdy et al., 2000).

For aquaculture species, Lacoste et al. (2001) reported that stressing farmed oysters led to increased susceptibility to infection with Vibrio species, which was directly linked to an increase in production of noradrenaline in the shellfish. The injection of noradrenaline into unstressed oysters also significantly increased oyster mortality resulting from subsequent Vibrio infection (Lacoste et al., 2001). Norepinephrine and dopamine increased motility, siderophore production, and expression of genes involved in flagellar motility, biofilm formation and exopolysaccharide production of V. harveyi and V. alginolyticus (Pande et al., 2014; Yang et al., 2014). In the challenge tests on

24

brine shrimp larvae (Artemia franciscana) and Macrobrachium rosenbergii larvae, pre- treatment of V. harveyi with norepinephrine and dopamine resulted in a significant decrease of survival rate of the larvae compared to those challenged with untreated V. harveyi (Pande et al., 2014; Yang et al., 2014). Moreover, several in vivo studies have shown that stress and stress hormones can directly affect the microflora of an animal.

For example, physical stress of mice caused by surgery or a short-term period of starvation resulted in significant increases in the number of E. coli adhering to the caecal mucosa (Hendrickson et al., 1999). Prolonged restraint stressors can also result in altered microflora which can lead to overgrowth by a commensal bacterium such as E. coli and Citrobacterro dentium and maybe result in serious infection (Bailey et al.,

2006; Bailey et al., 2010; Sharaff and Freestone, 2011).

1.4. FISH GUT MICROBIOME

Microbiota can be defined as microbial flora in a particular environment whereas microbiome is the assemblage of genomes of the microorganisms within the microbiota

(Butt and Volkoff, 2019). Early microbiota studies focused on gut microbial communities in human and mammals, and then extend to all tissues or organs with an epithelial lining such as skin, respiratory tract, urogenital tract. In fish similar studies on microbiome composition have been conducted on the skin, gills, blood, internal organs and but were mainly concentrated on the gut (Table 1.5) (Larsen, 2014).

The gut microbiota is known to have vital roles in host development, physiology and health starting from the early life stage (Llewellyn et al., 2014; Tarnecki et al., 2017).

Recent studies have reported that microbial communities contribute to proper

25

development of the gastrointestinal tract and immune system of hosts and increase nutrient absorption and can also compete with opportunistic pathogens to prevent disease (Llewellyn et al., 2014; Tarnecki et al., 2017; Egerton et al., 2018). Studies on zebrafish reported significant differences in morphology of gut epithelial cells between germ-free zebrafish and normal fish (with a commensal microbiota) (Rawls et al.,

2004). The microbiota was also shown to play important roles in the differentiation of the gut epithelium and function of the digestive tract in zebrafish including maturation of brush border enzyme activity, glycoconjugate expression, secretory cell numbers and protein macromolecule uptake (Bates et al., 2006). In addition, gut microbiota assists with maturation of the gut-associated lymphoid tissues which protects hosts against pathogens and maintains gut homeostasis (Dimitroglou et al., 2011).

Many studies have reported on the roles of gut microbiota in host nutrition. For example, in grass carp (Ctenopharyngodon idella), changes in the composition of the microbiota resulted in changes in biosynthetic and metabolic pathways of carbohydrates, amino acids and lipids (Ni et al., 2014). In addition, increases in absorption of fatty acids, lipid accumulation in the intestinal epithelium and expression of related lipid metabolism genes were reported in normal zebrafish compared to those that were germ-free (Butt and Volkoff, 2019). Similarly, commensal bacteria in zebrafish increased the absorption of macromolecule proteins and fatty acids (Bates et al., 2006). Gut microbiota was found to be able to produce vitamin B12 (in freshwater fish) or short-chain fatty acids that can be used by the hosts (Larsen, 2014). Gut microbes in grass carp have been shown to be capable of metabolising cellulose and

26

complex polysaccharides making the nutrients available which otherwise would not have been due to the lack of host cellulase (Ganguly and Prasad, 2012).

It has been suggested that the microbiota can protect hosts from colonization by pathogens and from proliferation of those pathogens. Although mechanisms have not been clearly demonstrated, commensal microbes may out-compete pathogens for niches and nutrients and also produce and secrete antimicrobial peptides (Butt and Volkoff,

2019). In addition, the ability of gut microbiota to produce short-chain fatty acids (from complex sugars) which are easily absorbed by diffusion or specific receptors helps to increase resistance against pathogens (Montalban-Arques et al., 2015; Talwar et al.,

2018). In many studies, administration of beneficial microbes has been shown to inhibit pathogens, enhance digestion and fish growth, promote a stable healthy microbiota and increase immune performance (Larsen, 2014; Egerton et al., 2018).

Diversity and composition of gut microbiota are influenced by both biotic (such as genotype, physiological status, pathobiology, gender, lifestyle) and abiotic

(environmental) factors (Llewellyn et al., 2014; Butt and Volkoff, 2019). Changes in the composition of the microbiota can result in altering its function and metabolic activity, therefore affecting feeding, growth, energy storage and health of the host (Butt and Volkoff, 2019).

Environmental factors such as salinity, season, geographic location, rearing condition and the presence of toxic pollutants can affect microbial communities in fish. Variation

27

in salinity can significantly affect composition of gut microbiota for example an increase in salinity shifted the dominant bacterial taxa in gut black molly (Poecilia sphenops) (Schmidt et al., 2015) and Atlantic salmon (Dehler et al., 2017b). The gut bacterial community differs between freshwater and marine fish as well, for example the most common genera in freshwater fish are Aeromonas and Pseudomonas while

Vibrio is dominant in marine fish (Larsen, 2014; Butt and Volkoff, 2019).

Seasonal changes also influence the gut microbiota structure in fish. Differences in gut microbial communities between seasons have been reported in many fish including: wild-caught bluegill fish (Lepomis macrochirus), largemouth bass (Micropterus salmoides), spotted gar (Lepisosteus oculatus), hybrid tilapia (Oreochromis niloticus x

Oreochromis aureus) and sea-cage cultured Atlantic salmon (Hovda et al., 2012; Ray,

2016; Zhang et al., 2016; Butt and Volkoff, 2019). However, an exception is pangasius

(Pangasius hypophthalmus) where a study on effects of multiple farm locations across multiple seasons of gut microbiota indicated that bacterial communities remained constant (Le Nguyen et al., 2008).

Rearing conditions have also been reported to shift the structure of gut microbiota in

Atlantic salmon. The gut microbiome of parr reared in a recirculating laboratory aquarium was different from those maintained in cage culture in a commercial fish farm

(Dehler et al., 2017a). When considering pollutants microbial communities of the common carp (Cyprinus caprio) and zebrafish displayed disturbances, which resulted in an increased susceptibility to pathogens and inflammation and a reduced nutrient

28

absorption after exposure to waterborne copper and polystyrene microparticles (Butt and Volkoff, 2019).

The gut microbiota of fish can also vary due to several factors, including genetics and sex, age and life stages, diet and feeding habits (Butt and Volkoff, 2019) and diseases

(Larsen, 2014). Several studies have been indicated that gut microbiota is species- specific. For example, a study on gut microbiota of zebrafish collected from different culture facilities and natural populations identified large similarities of microbes between samples, and this shared microbiota remained stable despite environmental influences (Roeselers et al., 2011). Bacterial abundance and composition were different among three species of parrotfish (Chlorurus sordidus), surgeonfish (Acanthurus nigricans) and snapper (Lutjanus bohar) from the same coral reef (Smriga et al., 2010).

29

Table 1.5. Effect of different factors on Atlantic salmon gut microbiome

Study design Microbiome Dominant phyla References

Effect of alternative protein Autochthonous Firmicutes, Gajardo et al. sources and allochthonous Proteobacteria, (2017) Fusobacteria, Bacteroidetes, Actinobacteria

Effect of habitat (recirculated Allochthonous Firmicutes, Dehler et al. aquarium and open loch) Proteobacteria, (2017a) Tenericutes

Effect of geographical Allochthonous Proteobacteria, Llewellyn et distance (freshwater and Firmicutes, al. (2016) marine water) Bacteroidetes, Actinobacteria

Effects of diets Allochthonous Bacteroidetes, Zarkasi et al. Firmicutes, (2016) Proteobacteria,

Effects of alternative feeds Autochthonous Proteobacteria Schmidt et al. on fish microbiome cultured and allochthonous and others (2016) in recirculation water system

Effect of commercial diets Autochthonous Proteobacteria, Gajardo et al. and allochthonous Firmicutes, (2016) Fusobacteria, Actinobacteria

30

Sex may also influence fish gut microbiota as microbial composition was shown to be different between sexes of threespine stickleback (Gasterosteus aculeatus) and

Eurasian perch (Perca fluviatilis) (Butt and Volkoff, 2019) (Bolnick et al., 2014).

However, the gut microbiota of zebrafish was reported to be similar for both sexes (Liu et al., 2016).

The variations in gut microbiota composition and diversity has shown to be related to fish age. In Atlantic salmon, microbial communities have lower richness and diversity in the embryonic stage in comparison to hatchlings (Lokesh et al., 2018) while juvenile zebrafish have higher bacterial richness than adult fish (Stephens et al., 2016). In farmed

Malaysian mahseer (Tor tambroides), larvae, juvenile and adult stages have higher diversity of gut microbiota than fingerling and yearling stages (Mohd Nosi et al., 2018).

This might be due to the change of nutrient requirements in different life stages, however further studies are needed for clearer explanation (Butt and Volkoff, 2019).

Feeding habitats and diets have strongly impacted fish microbial composition.

Generally, gut bacterial diversity is lower in carnivores and higher in omnivores and herbivores for both marine and freshwater fish (Butt and Volkoff, 2019). This trend has been thought to be related to the abundance of the cellulose degrading bacteria such as

Anoxybacillus, Leuconostoc, Clostridium, Actinomyces and Citrobacter in omnivores and herbivores. In carnivores, lipase and proteinase – producing bacteria such as

Halomonas have been commonly identified (Talwar et al., 2018). Diet and nutrient source alterations are also found to reshape gut microbiota composition and diversity.

31

Plant-derived dietary proteins significantly decrease microbial diversity and increase relative abundance of Lactobacillales, Bacillales and while

Bacteroidales, Clostridiales, Vibrionales, Fusobacteriales and Alteromonadales are usually found in gut fish fed with animal proteins (Michl et al., 2017; Talwar et al.,

2018). However, in channel catfish (Ictalurus punctatus), different diets with different protein sources derived from plants and animals have slightly different effects on gut microbiota (Schroeter et al., 2018). Modification of diets such as fasting results in reduction of nutrient uptake, thus favour bacterial species using wide-range energy sources in limited nutrient conditions and therefore reshape gut microbiota (Butt and

Volkoff, 2019).

Administration of beneficial bacteria to fish has been demonstrated to change bacterial communities, improve nutrient uptake and innate immune responses in many species, in contrast studies on effects of disease or pathogenic challenge on fish microbiota are currently limited. Cárdenas et al. (2011) found white disease significantly shifted the composition of bacterial communities of two Caribbean reef-building coral species.

Bacterial diversity decreased significantly, and Pseudomonas and Vibrio became dominant in spleen and liver of diseased turbot (Scophthalmus maximus) (Toranzo et al., 1993). Recently, a study on enteritis in grass carp demonstrated that diversity, structure and function of gut microbiota were significantly different, with the increase of genera Dechloromonas, Methylocaldum, Planctomyces, Rhodobacter,

Caulobacter, Flavobacterium, and Pseudomonas compared to healthy fish (Tran et al.,

2018).

32

1.5. THESIS AIMS

It is well known that Y. ruckeri can survive for long time in the environment as well as reside in fish intestine and is able to establish infections when fish are stressed (Ghosh et al., 2018). Hence better understanding of pathogenicity under different environmental conditions is essential for disease management. Although there are numerous studies of the effects of environmental factors on bacterial virulence, there is a limited information about these effects on Y. ruckeri and the interaction between environment conditions especially in stressful conditions. In particular the impact of stress hormone on virulence of Y. ruckeri requires investigation. In order to minimise the impact of this disease on salmon a fuller understanding of factors which influence virulence is needed, hence the aims of this project were:

To determine the in vitro effect of culturing Y. ruckeri with different concentrations (25,

50, 100 and 200 μM) of catecholamines norepinephrine, epinephrine and dopamine on potential Y. ruckeri virulence factors (Chapter 2)

- To assess survival of stressed and unstressed Atlantic salmon and the fish immune response after challenge with Y. ruckeri some of which had been exposed to epinephrine

(50 µM) in vitro (Chapter 3)

- To determine if a disease challenge with Y. ruckeri and host stress response can alter diversity and structure of gut microbiota in Atlantic salmon (Chapter 4)

The results of this thesis will provide more knowledge on the interactions between the host, the pathogen in the certain environment, and therefore will help to develop a health management strategy against this disease and ultimately benefit to the fish farmers.

33

Chapter 2. EFFECTS OF CATECHOLAMINES ON GROWTH AND

VIRULENCE OF Yersinia ruckeri

2.1. INTRODUCTION

Yersinia ruckeri is the causative agent of enteric redmouth disease (ERM) or yersiniosis, one of the most serious bacterial haemorrhagic septicaemia in Atlantic salmon (Salmo salar) and rainbow trout (Onchorhynchus mykiss) (Tobback et al., 2009; Barnes, 2011;

Costa et al., 2011). Although development of vaccines has shown much success in protecting salmonids from yersiniosis and ERM and has reduced economic losses for salmonid aquaculture industry, outbreaks are still reported. An improved understanding of Y. ruckeri pathogenic mechanisms is needed (Tobback et al., 2009; Kumar et al.,

2015), in particular the virulence factors which determine the capacity for virulence and pathogenicity (Casadevall and Pirofski, 2001). These factors help pathogens to colonise the host, evade and inhibit the host’s immune response and utilise nutrition from the host (Casadevall and Pirofski, 2001). Previous studies have shown that Y. ruckeri can produce some extracellular products including enzymes such as protease and lipase and also haemolysin which all may have an important role in pathogenesis (Fernández et al., 2007b; Tobback et al., 2007). Another extracellular product, ruckerbactin, a catecholate siderophore of Y. ruckeri, helps the bacterium utilise iron from the host and enables the formation of biofilms which are involved in its virulence (Fernández et al.,

2007b; Tobback et al., 2010). Host and pathogens always exist in an interaction which can be described as a dynamic battle. During disease, pathogens overcome the host immune defence system by activation of several virulence factors which can be driven

34

by specific host products and/or environmental cues ultimately allowing the pathogen to survive and multiply (Mekalanos, 1992).

Microorganisms have evolved sensory systems for detecting host-associated chemicals such as hormones (Hughes and Sperandio, 2008; Sharaff and Freestone, 2011). This sensory ability of microorganisms enables them to detect potential hosts and then to instigate the expression of genes enabling host colonisation and production of virulence factors (Sharaff and Freestone, 2011). Recently, studies have focussed on the interaction of bacteria with the hormones released in fish during stress such as the catecholamines which include epinephrine, norepinephrine and dopamine (Lyte and Freestone, 2010;

Lyte et al., 2011a; Sharaff and Freestone, 2011). Increases in stress hormone production have been shown to cause a significant reduction of cell-based immune function of the hosts (Reiche et al., 2005a) and also enhancement of bacterial growth and virulence

(Kinney et al., 1999; Belay et al., 2003; Sharaff and Freestone, 2011). Lesouhaitier et al. (2009) demonstrated a positive correlation between host catecholamine levels and the production of virulence-associated factors such as extracellular products, toxins and adhesion molecules as well as biofilm formation and quorum sensing in a large number of Gram-negative bacteria (Lesouhaitier et al., 2009). The ability to respond to stress hormones exists in several species of bacteria, including Escherichia coli, (Lyte and Ernst, 1992), Listeria monocytogenes (Coulanges et al., 1997),

Aeromonas hydrophila (Kinney et al., 1999), Pseudomonas aeruginosa, Klebsiella pneumoniae, Staphylococcus aureus (Belay et al., 2003), Vibrio parahaemolyticus

35

(Nakano et al., 2007a; Pande et al., 2014), and Vibrio harveyi (Pande et al., 2014; Yang et al., 2014).

Despite many virulence factors have been reported for Y. ruckeri (see chapter 1), an improved understanding of its virulence is still needed in particular to demonstrate how the bacterium responds to host stress hormones. Recent study showed norepinephrine and dopamine could increase growth, biofilm formation and motility of Y. ruckeri

(Torabi Delshad et al., 2019) however only one concentration of the hormones was tested. As each hormone and its concentration may have different effects on bacterial growth and virulence, the aim of this study was to determine the effects of three catecholamines at different concentrations using relevant in vitro assays.

2. 2. MATERIALS AND METHODS

2.2.1. Growth conditions for Yersinia ruckeri

A single strain of Y. ruckeri isolated from Atlantic salmon (serotype O1b, strain

UTYR001) and stored -80 ºC (General laboratory, Newnham campus) was used for all experiments. The bacterium was grown on Tryptone soya agar (TSA) (Oxoid,

Australia) or in Tryptone soya broth (TSB) (Oxoid, Australia) at 18 ºC under constant agitation of 100 rpm. Cell density was measured spectrophotometrically at 600 nm and correlated to viable cell numbers on TSA.

36

2.2.2. Calibration curve for absorbance and cell density of Y. ruckeri

A calibration curve was constructed by using absorbance data at 600 nm of ten-fold serial dilutions ranging from 10-4 to 10-9 of an overnight broth culture of Y. ruckeri and colony forming units (CFU) from each dilution on TSA plates. Briefly, 100 μL of each dilution was spread over the surface of the plates in triplicate and incubated for 24 hours at 18 ᵒC. Number of colonies on TSA were only counted from 20 to 300 for CFU calculation. For Y. ruckeri, the formula for the line of best fit from the calibration curve was: y (CFU/mL) = 3 ×1010 × (Absorbance 600 nm).

2.2.3. Hormones

In this study, three hormones were used: epinephrine (Epi), norepinephrine (NE) and dopamine (Do) (Sigma-Aldrich, Australia). Epinephrine and norepinephrine were dissolved in hydrochloric acid (HCl 0.1 N) while dopamine was dissolved in distilled water at room temperature; all were prepared to a stock concentration of 10 mM and then filtered (0.2 µm) and stored at -20 ᵒC.

2.2.4. Growth assays

A starter culture of Y. ruckeri was established from a frozen stock in TSB and incubated overnight at 18 ºC with constant agitation (100 rpm). The starter suspension was then sub-cultured in TSB supplemented with 30% (v/v) of heat inactivated Atlantic salmon serum (56 ºC for 30 minutes), with epinephrine, norepinephrine and dopamine in concentrations for each hormone of 0, 25, 50, 100 and 200 µM. To initiate the assay

37

200 µL of the cultures with each concentration of hormones were transferred to 6 wells a 96-well plate (6 wells per hormone concentration) at the same temperature and agitation as previous culture for 24 hours. To measure bacterial growth the optical density at 600 nm was measured by a microplate reader (Tecan Infinite 200 Pro,

Austria) in all wells each hour.

2.2.5. Lipase, phospholipase, caseinase and haemolysin assays

The effects of hormones on enzymatic activities of Y. ruckeri were all performed on hormone supplemented TSA plates with the addition of the appropriate substrates according to Yang et al. (2014) with some modifications. Autoclaved TSA (55ºC) was mixed thoroughly with the substrates and each of the hormones (Epi, NE and Do at 25,

50, 100 and 200 µM) prior to pouring into Petri dishes. To initiate the lipase assay an overnight culture of Y. ruckeri in TSB was diluted to an optical density (OD) of 0.5 at

600 nm and then a 5 µL drop of this diluted culture was placed in the middle of the hormone supplemented TSA plates which also contained 1% of Tween 80 (Sigma-

Aldrich, Australia). All plates were incubated for 3 days at 18 ºC, then the diameter of each opalescent zone and each colony were measured. Phospholipase, caseinase and haemolytic activities were assessed in the same way except that 2% of egg yolk emulsion (Sigma-Aldrich, Australia), 4% skim milk powder and 5% of defibrinated sheep blood (Oxoid, Australia) with and without heat inactivated Atlantic salmon serum at 10% (v/v) were added in lieu of Tween 80 respectively and clearing zones surrounding colonies were measured after the 3-day incubation. All the assays were carried out with at least 6 replicates and controls were plates with no hormone

38

supplements. The difference between hormone treatments and controls was determined from the mean of ratio between opalescent zone (or clearing zone) and colony diameter of each group.

2.2.6. Siderophore assay

Siderophore activity of Y. ruckeri was performed on serum supplemented Chrome azurol S (CAS) agar plates according to Yang et al. (2014) with some modifications.

Briefly, 60.5 mg of CAS (Sigma-Aldrich, Australia) was dissolved in 50 mL of deionised water and then mixed with 10 mL iron (III) solution (1 mM FeCl3.6H2O, 10 mM HCl). To this solution 72.9 mg of hexadecyltrimethylammonium bromide (Sigma–

Aldrich, Australia) previously dissolved in 40 mL deionised water was slowly added and stirred in by using magnetic stirring bar. The resulting dark blue solution was autoclaved at 15 psi in 15 minutes. Then 100 mL of 10x minimal media 9 (MM9) salt solution (30 g KH2PO4, 50 g NaCl, and 100 g NH4Cl in 1L of distilled water) 15 g agar,

30.24 g Pipes, and 12 g of a 50% (w/v) NaOH solution were added to 750 mL deionised water. After autoclaving (15 psi for 15 min) and cooling to 50 ºC, 10mL of sterile 20% glucose (Sigma-Aldrich, Australia) was added. The dark blue solution was added slowly with enough agitation to mix thoroughly without foaming. Finally, 10% heat inactivated salmon serum (v/v) and each of the hormones were added directly into the agar. The concentrations of hormones were 0, 25, 50, 100 and 200 µM. A 5µL drop of overnight culture of Y. ruckeri (OD600 = 1.0) was placed on CAS agar plate and plates were incubated at 28 ºC for 72 hours after which the clearing zones formed around each colony were measured.

39

2.2.7. Biofilm formation assay

Biofilm formation assay was performed according to Yang et al. (2014). Briefly, an overnight culture of Y. ruckeri was diluted in TSB to an OD of 0.1, and then hormones were added at concentrations of 0, 25, 50, 100 and 200 µM. Culture aliquots of 200 µL were transferred to 6 replicate wells of a 96-well plate and incubated at 18 ºC without agitation for 24 to 48 hours. To remove non-adherent bacteria, the liquid from each well was removed and wells were washed 3 times with 300 µL of sterile saline. The attached bacteria were fixed with 150 µL of 99% methanol per well for 20 minutes and the plate was left to air dry. Biofilms were stained with 150 µL crystal violet solution (1%) per well for 15 minutes. Excess crystal violet stain was removed by running tap water continuously through the wells until the washings were free of the stain. After drying,

150 µL of 95% ethanol were added to each well to allow the dye bound to the adherent cells to be resolubilized. Absorbance was then measured at 570 nm.

2.2.8. Antimicrobial assay

Antimicrobial assays were performed according to Bridle et al. (2011) with some modifications. Briefly, an overnight culture of Y. ruckeri in Mueller-Hinton broth

(MHB) with or without Epi 50 µM at 18 ºC was centrifuged at 1,000 x g for 5 min at 4

ºC and then the bacterial pellet was resuspended in one quarter strength (0.25X) MHB for a final working concentration of approximately 10 6 CFU/mL. The bacterial concentration was confirmed using spread plates of tenfold dilutions of the suspension onto Mueller-Hinton agar (MHA). Antibiotics tested were: erythromycin, oxytetracycline and streptomycin (diluted in sterile water) and also oxolinic acid and

40

florfenicol (diluted in NaOH 0.5N and dimethyl sulfoxide (DMSO)). All antibiotics were prepared at stock concentrations of 1 mM then all were diluted in sterile water to give final concentrations of 50, 25, 12.5, 6.3, 3.2, 1.6, 0.8 and 0.4 µM. To initiate the assays, 180 µL of the bacterial suspension were added to each well of a flat bottom 96 well plate and then 20 µM of the antibiotic dilutions were added in 5 replicates to make final antibiotic concentrations of 5, 2.5, 1.25, 0.63, 0.32, 0.16, 0.08 and 0.04 µM.

Controls were also in 5 replicates and included no-inhibition controls contained bacterial suspension (untreated with Epi or prior treated with Epi) with 20 µL of water or antibiotic solvent (NaOH/ DMSO); and no growth control contained 200 µL of 0.25

MHB only. Plates were incubated at 18 ºC for 18 hours, and then absorbances were measured at 600 nm. Bacterial growth in wells containing antibiotic was compared to the no-inhibition control wells (considered 100% growth) to determine the minimum inhibitory concentration (MIC) of each antibiotic. The MIC was defined as the lowest concentration that reduced bacterial growth by more than 50% (MIC50) compared to no-inhibition control wells.

2.2.9. Statistical analysis

Data analysis was carried out using the Statistical Package for the Social Sciences SPSS software (version IBM SPSS 23.0). Normal distribution and homogeneity of variance were verified using Shapiro-Wilk and Levene’s test before performing ANOVA. All data were compared with one-way ANOVA, followed by Tukey post hoc test or

Multiple comparisons of Dunnett T3 (for unequal variance assumption). The results were considered significant if p<0.05. For MIC assay, bacterial growth was compared

41

using independent samples t-test. In the growth assay, the difference in growth of Y. ruckeri between treatments were compared with Area under the curve by using

GraphPad Prism version 7.0. All data were graphed using GraphPad Prism version 7.0.

2.3. RESULTS

2.3.1. Stress hormones increase the growth of Y. ruckeri in serum – supplemented medium

There was a positive effect of stress hormones on Y. ruckeri growth over 24 hours for each concentration of Epi, NE and Do as seen by an increase in absorbance (Figure 2.1

A, B and C). In order to compare the growth curves between treatments and control, areas under the curves were calculated by using GraphPad Prism (version 7). These differences were shown on right graph (Figure 2.1).

42

A AUC Epi 13

***

12 e

v ***

E ffe c t o f E p i o n g ro w th o f Y e rs in ia ru c k e ri r u

c ***

0 .8 h

t 11 ***

w

o

r

g

e

h

m t

0 .6

n 10

r

0

e

0

d

6

n

t

u

a

a

e

e r

c 0 .4 9

a

n

l

a

a

b

t

r

o

o

T s b 0 .2

A 8

0 .0 7

0 3 6 9 1 2 1 5 1 8 2 1 2 4 l M M M M o     tr 5 0 0 0 n 2 5 0 0 T im e (h o u r ) o i i 1 2 C p p i i E E p p E E C o n tro l E p i 2 5  M E p i 5 0  M E p i 1 0 0  M E p i 2 0 0  M Treatment

B AUC NE 13

12 *

e *** v

E ffe c t o f N E o n g ro w th o f Y e rs in ia ru c k e ri r

u c

***

h *

0 .8 t 11

w

o

r

g

e

m

h

t n

0 .6 r

0 10

e

0

d

6

n

t

u

a

a

e

e r

c 0 .4 9

a

n

l

a

a

t

b

r

o

o

T s

b 0 .2 8 A

0 .0 7

0 3 6 9 1 2 1 5 1 8 2 1 2 4 l M M M M o     tr 5 0 0 0 T im e (h o u r) n 2 5 0 0 o E E 1 2 C N N E E N N C o n tro l N E 2 5  M N E 5 0  M N E 1 0 0  M N E 2 0 0  M Treatment

43

C AUC Do 13

** *

12

e v

E ffe c t o f D o o n g ro w th o f Y e rs in ia ru c k e ri r

u

c

0 .8 h

t 11

w

o

r

g

e

m

h

t n

0 .6 10

r

0

e

0

d

6

n

t

u

a

a

e

e r

c 0 .4 9

a

n

l

a

a

t

b

o

r

T

o s

b 0 .2 8 A

0 .0 7

0 3 6 9 1 2 1 5 1 8 2 1 2 4 l M M M M o     tr 5 0 0 0 n 2 5 0 0 T im e (h o u r) o o o 1 2 C D D o o D D C o n tro l D o 2 5  M D o 5 0  M D o 1 0 0  M D o 2 0 0  M Treatment

Figure 2.1. Effect of epinephrine (A), norepinephrine (B) and dopamine (C) on growth of Yersinia ruckeri in serum – supplemented medium (left) and total area under the growth curve (right)

Data were shown as Mean ± SE of 6 replicates. Asterisks indicate statistically significant differences from the control at *p<0.05, **p<0.01 and ***p<0.001 according to One-way ANOVA analysis.

Total area under the growth curve of Y. ruckeri incubating in serum-supplemented medium with Epi and NE at concentrations of 25, 50, 100 and 200 µM were significantly higher compared to control, and with Do only at 25 and 50 µM.

44

2.3.2. Lipase, phospholipase, caseinase and haemolysin assays

2.3.2.1. Lipase

Lipase activity resulted in formation of opalescent zones surrounding the colonies

(Figure 2.2.a). Results showed that each stress hormones at a concentration of 50 µM recorded the highest enhancement of lipase activity of Y. ruckeri. Additionally, Epi at

100 µM and NE at 200 µM significantly increased lipase activity in comparison with the control (p<0.01) (Figure 2.2.b).

Figure 2.2.a. Opalescent zones (red arrow) surrounding the colonies (black arrow) on lipase assay

45

Figure 2.2.b. Effect of stress hormones on lipase activity of Yersinia ruckeri

Data were shown as Mean ± SE of 24 replicates. Asterisks indicate statistically significant differences from the control at * p<0.05, **p<0.01 and ***p<0.001 according to One-way ANOVA analysis.

46

2.3.2.2. Caseinase

Clearing zones surrounding colonies indicated caseinolytic activity of Y. ruckeri, however, there was a significant reduction in ratio between clearing zones and colony diameters observed in most hormone treatments compared to the control except for Epi at 25 µM and NE at 200 µM.

Figure 2.3. Effect of stress hormones on caseinolytic activity of Yersinia ruckeri

Data were shown as Mean ± SE of 24 replicates. Asterisks indicate statistically significant differences from the control at *p<0.05, **p<0.01 and ***p<0.001 according to One-way ANOVA analysis.

47

2.3.2.3. Haemolysin

Haemolytic activity of Y. ruckeri after exposure to each hormone was tested using sheep blood agar with or without the addition inactivated serum. For the first trial without serum, there were only small clearing zones (< 1mm) surrounding colonies in all hormone treatments (50 and 100 µM) while no haemolytic activity was recorded in control. This indicated that stress hormones may enhance haemolytic activity of Y. ruckeri. When the haemolytic assay was repeated with the addition of serum at 10%

(v/v) in blood agar larger measurable clearing zones were seen surrounding colonies in both control and treatments in most cases (Figure 2.4).

However, only Epi at 200 µM and NE at 25 µM showed significantly larger ratios between clearing zones and colony diameters compared to the control (p<0.05). The high concentrations of hormones (Epi and NE 200 µM and Do 100 and 200 µM) changed the colour of the medium making it difficult to accurately measure the diameter of clearing zone and the colony. Therefore, no data were recorded for Do 100 µM, Do

200 µM and only 4 replicates of Epi 200 µM and NE 200 µM out of 12 replicates were measured.

48

Haemolysin Epi Haemolysin NE Haemolysin Do

2.20 2.20 2.20

****

r

r

r

e

e

e

t

t

t

e e

2.15 e 2.15 2.15

m

m

m

a a

* a

i

i

i

d

d

d

y

y

y

n

n n

o 2.10 o

o 2.10 2.10

l

l

l

o

o

o

c

c

c

d

d

d

n

n

n

a

a

a

2.05 2.05 2.05

e

e

e

n

n

n

o

o

o

z

z

z

g

g

g

n

n

n

i i

2.00 i 2.00 2.00

r

r

r

a

a

a

e

e

e

l

l

l

c

c

c

n

n

n

e e

1.95 e 1.95 1.95

e

e

e

w

w

w

t

t

t

e

e

e

b

b

b

o

o

o

i

i i

t 1.90 t

t 1.90 1.90

a

a

a

R

R R

1.85 1.85 1.85

l M M M M l M M M M l M M M M o     o     o     tr 5 0 0 0 tr 5 0 0 0 tr 5 0 0 0 n 2 5 0 0 n 2 5 0 0 n 2 5 0 0 o i i 1 2 o 1 2 o 1 2 C p p i i E E o o E E p p C N N E E C D D o o E E N N D D Treatment Treatment Treatment

Figure 2.4. Effect of stress hormones on haemolytic activity of Yersinia ruckeri

Data were shown as Mean ± SE of 12 replicates except of Epi 200 µM and NE 200 µM (4 replicates) and no data for Do 100µM and Do 200 µM due to the effect of high concentration of hormone to prevent from accurately measure. Asterisks indicate statistically significant differences from the control at *p<0.05, **p<0.01 and ***p<0.001 according to One-way ANOVA analysis.

49

2.3.3. Siderophore

Exposure of Y. ruckeri to Epi, NE and Do at all concentrations (25 to 200 µM) resulted in significant increase in siderophore activity in comparison to the controls (Figure 2.5).

Figure 2.5. Effect of stress hormones on siderophore activity of Yersinia ruckeri

Data were shown as Mean ± SE of 28 replicates. Asterisks indicate statistically significant differences from the control at *p<0.05, **p<0.01 and ***p<0.001 according to One-way ANOVA analysis.

50

2.3.4. Biofilm formation

Only the presence of Epi and Do at 200 µM significantly increased biofilm formation compared to the control (p<0.05) (Figure 2.6). Exposure Y. ruckeri to NE did not show any differences between treatments and the control.

Figure 2.6. Effect of stress hormones on biofilm formation of Yersinia ruckeri after 48 hours

Data were shown as Mean ± SE of 6 replicates. Asterisks indicate statistically significant differences from the control at *p<0.05, **p<0.01 and ***p<0.001 according to One-way ANOVA analysis.

51

2.3.5. MIC assay

There was no significant difference in MIC50 for each antibiotic between Y. ruckeri treated and untreated with Epi 50 µM (p >0.05). MIC50 value was low for oxolinic acid

(0.08 µM), streptomycin (0.16 µM) and oxytetracycline (0.32 µM) while erythromycin exhibited no inhibition effect to Y. ruckeri growth. Bacterial growth in treatments with florfenicol decreased with the increasing of concentrations of this antibiotic, and MIC50 was more than 5 µM (Figure 2.7.a. to Figure 2.7.e.).

Figure 2.7.a. Antibacterial activity and MIC50 for erythromycin against Yersinia ruckeri

52

Figure 2.7.b. Antibacterial activity and MIC50 for florfenicol against Yersinia ruckeri

Figure 2.7.c. Antibacterial activity and MIC50 for oxolinic acid against Yersinia ruckeri

53

Figure 2.7.d. Antibacterial activity and MIC50 for oxytetracycline against Yersinia ruckeri

Figure 2.7.e. Antibacterial activity and MIC50 for streptomycin against Yersinia ruckeri

54

2. 4. DISCUSSION

The secretion and circulation of catecholamine hormones in blood occurs when fish experience an acute stress. Circulating catecholamine levels could increase immediately with stress and then return to baseline values in hours (see Reid et al., 1998; Barton,

2002). However, in this thesis, the exposure time of the bacterium Y. ruckeri to catecholamines was over 24 hours. This duration was chosen to maintain consistency between the in vitro and in vivo experiments as the aims were to focus on the effects of catecholamines on virulence factors of Y. ruckeri and how virulence impacts Atlantic salmon.

In this in vitro experiment, hormone concentrations chosen ranged from 25 to 200 µM which were much higher than the resting level of catecholamines in fish plasma: from

1 to 30 nM in rainbow trout (Butler et al., 1986; Van Dijk and Wood, 1988) or from 3 to 77 nM in Atlantic salmon (Flysand et al., 1992; Deitch et al., 2006) or from 169 to

1690 pg/mL in pelagic sharks (Hight et al., 2007). However, in stressful conditions, the catecholamine level has been reported to increase 150 times in rainbow trout (Butler et al., 1986; Van Dijk and Wood, 1988) and up to 1600 times in pelagic sharks (Hight et al., 2007). In addition, concentrations of catecholamines can be higher in innervated tissues and can be up to 10mM in neuronal synapses (Lyte, 2004). Furthermore, previous reports from in vitro studies demonstrated that bacterial species such as E. coli,

Salmonella, Yersinia, Staphylococcus mostly responded at 50 µM NE or Do and 100

µM Epi, and higher concentrations (>200 µM) start to be inhibitory on bacterial growth

(Freestone and Lyte, 2008). Therefore, in this study the range of catecholamine

55

concentrations chosen was deemed to be the most appropriate to investigate effects on growth and virulence factors of Y. ruckeri.

The results presented here show that the presence of stress hormones at most of the concentrations in the range tested stimulated the growth of Y. ruckeri in serum- supplemented medium. This concurs with Freestone et al. (2008b) who reported that exposure of bacteria to stress hormones (at concentration > 5 µM) also enhanced growth of many pathogenic bacteria. Some of previous studies demonstrated hormones significantly increased growth of V. harveyi (Yang et al., 2014), E. coli (Vlisidou et al.,

2004) and V. parahaemolyticus (Nakano et al., 2007a) only in serum-supplemented media but not in un-supplemented controls, possibly bacteria response with catecholamine released inside the host and serum added may reflect host internal environment. Stress hormones enhancement of bacterial growth has been mainly explained by their ability to utilize iron, which is essential for bacterial growth in hosts, removed from the host iron-binding proteins lactoferrin and transferrin (Yang et al.,

2014). Mechanistically, catecholamines facilitate release of iron from lactoferrin and transferrin to form a direct complexes with the ferric ion and result in reduction of Fe 3+ to Fe2+ which is then available for bacteria (Lyte et al., 2011b). The process of catecholamine-facilitated removal of iron from these proteins by stress hormones can be considered as a growth stimulation for bacteria, resulting in significant increases in the numbers of bacterial cells of many species when stress hormones are added into the blood (Lyte and Ernst, 1992; Coulanges et al., 1997; Coulanges et al., 1998; Kinney et al., 1999; Belay et al., 2003; Sharaff and Freestone, 2011). Other studies also suggested

56

that catecholamines induced growth of bacteria in a serum-based medium may lead to the production of a non-LuxS dependant autoinducer of growth (termed norepinephrine - induced autoinducer – NE-AI) (Freestone and Lyte, 2008). This autoinducer is heat stable, cross-species acting and enhance the growth of many bacteria in a similar fashion to catecholamines (Freestone et al., 1999; Freestone et al., 2008b). It is induced by Gram negative bacteria after 4-6 hours exposure to norepinephrine and works independently on transferrin and lactoferrin. The finding of NE-AI may explain why catecholamines which are released and persist over only a short period of time during acute stress could still have effects on bacteria after their concentrations have returned to normal levels

(Freestone and Lyte, 2008).

Extracellular products from Y. ruckeri containing caseinase, gelatinase, phospholipase, lipase and haemolysin are considered to contribute to the virulence and pathogenic potential of Y. ruckeri (Toranzo and Barja, 1993). In this experiment, the presence of stress hormones stimulated lipase activity and haemolytic activity of Y. ruckeri. Lipase enzymes function in breaking long chain triacylglycerols into fatty acids and glycerols, therefore may have some role in host tissue damage (Defoirdt, 2013). On the other hand, lipid A is a structural component of the LPS in the outer membrane of Gram negative and has important role in pathogenesis of a bacterium (Karavolos et al., 2008), thus increase production of enzyme lipase may play some roles on lipid A synthesis and impact on bacterial virulence. However, an understanding of their involvement in pathogenesis is very limited (Defoirdt, 2013). According to Toranzo and Barja (1993), extracellular products of Y. ruckeri exhibit haemolytic activity against sheep and trout

57

erythrocytes, but haemolysin detection can be influenced by many factors. In the study of Toranzo and Barja (1993), haemolytic activity was only seen when the blood cells were exposed in a soft overlay agar and it was suggested that possibly oxygen tension or viscosity had some effect on haemolytic activity. In the experiment reported here, sheep blood was mixed with TSA to make blood agar (hard agar) (without serum), which may have made haemolytic activity difficult to detect as very little clearing was seen around colonies. However, when serum was added to the blood agar (hard agar), much bigger clearing zones surrounding colonies were seen indicating haemolytic activity of Y. ruckeri. Therefore, serum supplemented medium plays very important role in reflecting the host environment and bacterial response to catecholamines. Huang and DuPont (2005) reported similar results with the human pathogen causing , Salmonella typhi. Although S. typhi is not haemolytic on blood agar plates, exposure of S. typhi to neuroendocrine hormones resulted in the appearance of clearing zones on blood agar plates and boosted haemolysis by 40% in liquid assays (Karavolos et al., 2011; Karavolos and Khan, 2013). Karavolos and Khan (2013) reported the haemolytic response is specific to outer membrane vesicles containing the haemolysin

HlyE. Norepinephrine and epinephrine interact with the CpxAR putative adrenergic sensory system to down-regulate outer membrane protein A (OmpA) levels, and consequently increase outer membrane vesicle shedding, therefore the release of HlyE increases (Karavolos and Khan, 2013). Another study conducted on E. coli suggested that norepinephrine significantly induced the expression of slyA and rhaH genes, which are responsible for regulating expression and activating hemolysin (Dowd, 2007). In Y. ruckeri, a YhlB haemolysin activation protein and a YhlA haemolysin were found and

58

when these proteins were absent, virulence is decreased 10-fold and 100-fold in LD50 assay in rainbow trout respectively (Fernández et al., 2007b). Therefore, catecholamines may have the same impacts as in S. typhi or E. coli to activate and induce expressions of haemolysin genes in Y. ruckeri. Thus, further investigation of the effects of catecholamines on haemolytic gene expressions may find out the mechanisms. In contrast to lipase and haemolysin activity, the results of this study show that exposure to stress hormones seemed to decrease caseinase (a protease with caseinolytic activity) activity of Y. ruckeri. According to Secades and Guijarro (1999), the bacterial proteases with their proteolytic activities may provide peptide nutrients for microorganisms, and hence have important roles in damaging host tissues, thus they could contribute to the establishment of infection. In Y. ruckeri, caseinolytic activity was reported in many strains and caseinase production was varied under different culture conditions (Secades and Guijarro, 1999). As there are a lot of different proteases playing different functions, hence the reduction of caseinase enzyme may be related to the increase of the others and they may involve in different roles in bacterial growth and/or virulence.

Although iron is very important for both growth and virulence of many bacteria, most of the iron in the biological fluids is bound with protein transferrin or lactoferrin

(Williams et al., 2006). To overcome an iron-depleted environment, most bacteria produce small molecules named siderophores (catecholates, hydroxamates, and heterocyclic compounds) to sequester iron for their use. Although Y. ruckeri can only produce heterocyclic ruckerbactin, it is provided with uptake system for the two others.

Ruckerbactin has been confirmed to contribute to the pathogenicity of Y. ruckeri

59

(Fernandez et al., 2004). The results of the current study are in agreement with a previous study that catecholamines increase siderophore production of V. harveyi (Yang et al., 2014). Although many studies have reported the interaction between catecholamines and iron, the mechanism by which catecholamines enhance siderophore production of bacteria is still largely unknown. The study on Pseudomonas aeruginosa showed that this bacterium can obtain iron through two pyoverdine and pyochelin siderophores under the iron-limited conditions and the production of siderophores is negatively regulated by iron availability (Li et al., 2009). Furthermore, Li et al. (2009) also indicated that norepinephrine can repress the siderophore gene expression including pyoverdine genes pvdD and pvdE and their regulators pvdS, regA and pchR resulting in repression of siderophore pyoverdine and more iron provision for bacterial growth (Li et al., 2009).

Biofilm formation is the process of bacterial cells attaching onto surfaces and forming a matrix of extracellular polymeric substances (Donlan, 2001). Several reports have indicated that biofilm formation may play an important role in bacterial survival, virulence and stress resistance. Y. ruckeri is able to form biofilms with substrate materials commonly found in fish farms (Kumar et al., 2015), and these biofilms can be considered as a suitable reservoir for initial and repeated of outbreaks of disease (Woo and Bruno, 2011). The results of this study agree with previous reports documenting the presence of stress hormones increasing biofilm formation by Staphylococcus epidermidis (Lyte et al., 2003), Streptococcus (Sandrini et al., 2014).

According to Kumar et al. (2015) biofilm formation is usually promoted by pili and

60

flagella, and high adhesion cell efficiency is expressed by flagellar proteins. Therefore, exposure of hormones may have resulted in an upregulation of flagellar genes and stimulated biofilm formation.

There was no significant difference in MIC50 between treated and untreated Y. ruckeri with epinephrine. Pre-treated Salmonella with the same concentration of epinephrine resulted in a significant reduction of bacterial survival during exposure to antibiotic polymyxin B when compared to the control (Karavolos et al., 2008). Polymyxin B binds to lipid A of the outer membrane of bacteria and damages the cell envelope while the antibiotics used in this study mostly interfere with bacterial ribosome, therefore the different results may come from different action mechanisms of antibiotics as well as different bacterial species. Enteric redmouth disease and yersiniosis are controlled with antibiotics such as quinolones, sulphonamides, oxytetracycline and erythromycin

(Cortés et al., 2014). The MIC results in this study showed that Y. ruckeri was highly resistant to erythromycin. This is consistent with observations by De Grandis and

Stevenson (1985); Calvez et al. (2014); Kumar et al. (2015) indicating erythromycin

MIC ranging from 16 – 1024 µg/mL. Similarly, no inhibition effect of erythromycin was found in Y. ruckeri strain H01 isolated from farm-raised amur sturgeon (Acipenser schrencki) in China (Shaowu et al., 2013). Natural resistance to erythromycin was also described in other Yersinia species such as Yersinia enterocolitica, and (Stock et al., 2002). In contrast, Y. ruckeri proved sensitive to oxolinic acid, streptomycin and oxytetracycline which was in accordance with Inglis and Richards (1991); Calvez et al. (2014); Kumar et al. (2015). These

61

antibiotics interfere the functions of 30S ribosomal subunits and DNA synthesis while florfenicol is able to bind to the 50S ribosomal subunit to prevent bacterial protein synthesis. Although Y. ruckeri is susceptible to oxolinic acid, florfenicol and these antibiotics are the most commonly used to control ERM and yersiniosis, bacterial resistances have been reported (Rodgers, 2001; Miranda and Rojas, 2007; Calvez et al.,

2014).

In conclusion, generally the exposure of Y. ruckeri to catecholamines increased the growth, siderophore, lipase, haemolysin and biofilm production but had no effects on

MIC, phospholipase and decreased production of caseinase at 18 ºC. There is a lack of dose responses as the responses for each concentration tested were generally the same.

62

Chapter 3. EFFECTS OF STRESS AND Yersinia ruckeri CHALLENGE ON

ATLANTIC SALMON (Salmo salar)

3.1. INTRODUCTION

Atlantic salmon (Salmo salar) is one of the most important farmed species worldwide.

The global production of Atlantic salmon has increased continuously from about 1 million tonnes in 2001 to 2.2 million tonnes in 2016 (FAO, 2004-2019). The main producers of Atlantic salmon are Norway, Chile, Scotland, Canada, Faroe Islands,

Australia and Ireland (MOWI, 2018). In Australia, the salmon industry is mainly in

Tasmania with annual output valued at around $550 million (ISFA, 2018), however, the industry is still facing disease-related issues early in the production cycle, most notably that caused by Y. ruckeri. The bacterium was first isolated in farmed Atlantic salmon in

Tasmania in 1987 where it was responsible for mortality in fish in hatcheries and post- seawater transfer (Barnes et al., 2016). Although vaccination against Y. ruckeri was introduced in Tasmania in the late 1990s, disease outbreaks are still reported. In 2007 half a million freshwater juvenile Atlantic salmon in Australia died due to yersiniosis in a few months despite having been vaccinated against yersiniosis (Bridle et al., 2012).

Hence there remains a need to improve our knowledge of the relationship between the pathogen, the host and other factors including vaccination that impact this disease.

Disease is defined as a spectrum of responses based on the intensity of the interactions of the host, pathogen, and environment (Hedrick, 1998). In most cases, animals exposed

63

to chronic stress or immediately after acute stress are usually more susceptible to diseases due to immune suppression (Tort, 2011). For example, mortality increased by

20% and elevated lysozyme activity was positively correlated with bacterial DNA in the blood of stressed catfish (Ictalurus punctatus) challenged with Edwardsiella ictaluri after being held in a low water level for 30 minutes (Small and Bilodeau, 2005). Severe depression of humoral components of the immune system was observed in stressed gilthead seabream (Sparus aurata) (Sunyer et al., 1995; Doménech et al., 1997; Tort et al., 1998; Salati et al., 2016). Similarly, it has been reported that overcrowding stress can impact on the expression of stress and innate immune-related genes of Senegalese sole (Solea senegalensis) (Salas-Leiton et al., 2010) and rainbow trout (Onchorhynchus mykiss) (Yarahmadi et al., 2016).

Stress related hormones such as the catecholamines (norepinephrine, epinephrine and dopamine) have been reported to enhance growth and/or virulence of many pathogenic bacteria such as Escherichia coli, Yersinia enterocolitica (Lyte and Ernst, 1992),

Listeria monocytogenes (Coulanges et al., 1997), Aeromonas hydrophila (Kinney et al.,

1999), Pseudomonas aeruginosa, Klebsiella pneumoniae, Staphylococcus aureus

(Belay et al., 2003), Vibrio parahaemolyticus (Nakano et al., 2007a; Pande et al., 2014),

Actinobacillus pleuropneumoniae (Li et al., 2012), Vibrio harveyi (Pande et al., 2014;

Yang et al., 2014) and Campylobacter jejuni (Xu et al., 2015). In an in vitro experiment, exposure of Y. ruckeri to stress hormones resulted in increasing growth and virulence factors such as siderophore, lipase, haemolytic activity and biofilm formation (see chapter 2). The next step is to carry out an in vivo investigation particularly as there is

64

limited information about the effects of stress hormones on the virulence of Y. ruckeri used to challenge Atlantic salmon. In addition, the relationship between stress factors and the fish immune response after Y. ruckeri challenge (and subsequent exposure in vivo to stress hormones) is poorly understood and requires further research.

The aim of this study was to investigate the effects of challenge with Y. ruckeri (some of which had been exposed to exogenous epinephrine) on cortisol levels and survival of salmon (some of which had been subjected to an acute stressor) as well as the fish immune response at different sampling times.

3.2. MATERIALS AND METHODS

3.2.1. Ethics statement

All fish work was performed in accordance with the Australian Code of Practice for the

Care and Use of Animals for Scientific Purposes and was approved by the University of Tasmania Animal Ethics Committee (AEC permit number: A0017118).

3.2.2. Fish

Atlantic salmon (Salmo salar) weighing 5g were obtained from a commercial hatchery

(in Tasmania, Australia) and as such all fish had been previously vaccinated against Y. ruckeri (Yersinivac B) and had not been received any previous disease treatments. Fish were acclimated for 3 weeks in a 100L tank and where then moved to 24 x 80L tanks with a recirculating freshwater at 15ºC for 10 days prior to the experiment (carried out

65

at the Aquaculture Centre, University of Tasmania, Newnham campus). Fish were fed twice daily to satiation with a commercial diet (Skretting, Australia) and starved for 24 hours prior to initiation of the experiment and all sampling events. Water quality parameters including temperature, pH, total ammonia, nitrite and nitrate were monitored daily by using a commercial test kit and maintained within optimal ranges

+ -1 - -1 - -1 for the fish (pH = 7, NH4 < 2 mg.L , NO2 < 2 mg.L , NO3 < 80 mg.L ).

3.2.3. Hormones and pathogen

Based on the effects of different concentrations of stress hormones on functional virulence assays with Y. ruckeri in previous in vitro results (chapter 2), epinephrine

(Epi) (Sigma-Aldrich, Australia) 50 µM was selected for use in this experiment. Briefly, epinephrine was dissolved in hydrochloric acid (HCl 0.1 N) at room temperature to a stock concentration of 10 mM then filtered (0.2 µm) and stored at -20ºC.

Yersinia ruckeri (serovar O1b, UTYR001) was inoculated from frozen stock to tryptone soya broth (TSB) (Oxoid, Australia) and incubated overnight at 18ºC under constant agitation of 100 rpm. The starter culture then was sub-cultured in TSB with and without

50µM epinephrine at the same temperature and agitation for 24 hours. Concentration of

Y. ruckeri for the challenge experiment was 106 (CFU) mL-1.

66

3.2.4. Stress application and Y. ruckeri challenge

Stress was applied by exposing the fish to a 50% reduction of water volume (removal of 40L) and chasing them for 10 minutes with a net. Unstressed fish (control) were maintained in full water level (80L) without chasing. Twenty-four hours after exposure to the stressors, fish were challenged with Y. ruckeri either treated or untreated with epinephrine. To initiate the challenge all fish were removed from individual tanks and placed into well aerated 20L containers with freshwater at 15ºC which had been previously inoculated with the treated or untreated Y. ruckeri with the final concentration of 106 CFU mL-1. This immersion challenge was for 30 min after which all fish were rinsed in clean freshwater for a minute before being returned to respective tanks. Control fish were subjected to the same handling (removal from tanks and immersion in 20L containers) but were not exposed to any Y. ruckeri.

As there were groups of salmon which had been stressed and unstressed and challenged with Y. ruckeri exposed and unexposed to catecholamine (epinephrine) there were 6 treatments in total (see Table 3.1), 4 tanks per treatment, each tank containing from 20 to 24 fish. There were unequal numbers of fish in groups because some escaped from individual tanks and could not be replaced. After the challenge moribund and dead fish were removed from the tanks twice daily. To verify the cause of death or morbidity of challenged fish, kidney swabs from those fish were cultured on tryptone soya agar

(TSA) (Oxoid – Australia) and confirmed with a standard PCR using specific primers of Y. ruckeri (Ghosh et al., 2016). After 40 days the challenge was terminated and survival data were analysed using Kaplan-Meier survival analysis.

67

Table 3.1. Experimental design to test the effects of stress (S+) or no stress (S-) and Yersinia ruckeri, exposed and unexposed to the catecholamine (epinephrine) at 50 µM (CAT+ or CAT- respectively), challenge in Atlantic salmon. Sampling points were at 0h, 12h, 24h (after stress); 3, 10, 20, 30 and 40 days (post challenge).

Treatment Number of fish

Y. ruckeri treated with Epi Stressed fish CAT+ S+ 89

Y. ruckeri treated with Epi Unstressed fish CAT+ S- 82

Y. ruckeri only Stressed fish CAT- S+ 89

Y. ruckeri only Unstressed fish CAT- S- 88

No exposure to Y. ruckeri Stressed fish Ch-S+ 90

No exposure to Y. ruckeri Unstressed fish Ch-S- 86

3.2.5. Sample collection

To investigate the effect of stress on fish, 8 fish were taken randomly and sampled

(blood, spleen and intestine) before exposure to stress (0h) and 8 fish, both stressed and unstressed, were sampled each time at 12 hours and 24 hours post stress. At day 3, 10,

20, 30 post challenge, 2 fish were taken from each tank (8 fish/treatment) and at day 40 all the remaining fish were sampled. Before each sampling the fish were euthanised

68

with clove oil (400 mg.L-1) and at each time point blood, spleen and intestine were collected except at day 40 when only spleen and intestine were sampled.

Blood was obtained from the caudal vein using a 1 mL heparinised syringe and 25 gauge needle, and then plasma was collected after centrifugation at 10,000 x g for 3 min at

4ºC. Both plasma and retained blood pellets were stored at -80ºC for further use. Spleen and gut were kept in RNA preservation solution (4M ammonium sulphate, 25 mM sodium citrate, 10 mM EDTA, pH 5.2) overnight at 4ºC before freezing at -20ºC.

3.2.6. Plasma cortisol measurement

Cortisol levels in blood plasma were determined by using the cortisol enzyme linked immunosorbent assay kits (ELISA) (Enzo Life sciences Inc., US) following the manufacturer’s instructions. First, samples were diluted with the steroid displacement reagent 99:1 (v/v) and then were 5-fold diluted with assay buffer (from the kit). The standards provided by the kit were serially diluted in the assay buffer and mixed by vortex. One hundred µL of diluted samples, standards and blank samples were added to

96-well microtiter plate coated with anti-cortisol goat antibodies. Fifty µL of the cortisol enzyme conjugate and then 50 µL of the cortisol antibody were added into wells and the plate was incubated at room temperature for 2 hours with shaking at 500 rpm. After incubation the plate was washed 3 times by flooding all wells with 400 µL of wash buffer and then discarded by flicking the plate. Any residual wash buffer was removed by firmly tapping the plate on paper towels and 200 µL of the substrate was then added

69

and incubated for an hour without shaking. Finally, 50 µL of stop solution was added to stop the reaction and the plate was read immediately at an optical density of 405 nm using a Tecan Infinite 200 Pro microplate reader. All standards and blanks were run in duplicate and samples were run in at least 4 replicates from different individual fish per treatment.

The absorbance was used to calculate percent bound and cortisol concentrations were measured by a 4 parameter logistic curve fitting program (GraphPad Prism 7).

3.2.7. DNA and RNA extraction

Total nucleic acid from spleen and intestine was extracted using a silica spin column.

Briefly, 3 - 6 mg of spleen was homogenized in 300 µL lysis buffer (4 M Urea, 0.5%

SDS, 50 mM TRIS pH = 8, and 10 mM EDTA) and 15 µL TCEP and heated for 15 min at 55ºC, mixed by vortex and then cooled on ice for 5 min. Ammonium acetate (7.5 M,

Sigma-Aldrich, USA) was added to a final concentration of 2.5 M, mixed by vortex for

15s and spun for 5 min at 16,000 x g at 4ºC. The supernatant was removed and mixed with 100µL ammonium acetate and 350 µL 100% ethanol and transferred to spin columns and then centrifuged for 1 min at 14,000 x g. The column was washed twice with 600 µL washing buffer (80% ethanol, 10mM TRIS pH = 7) and spun dry before eluting with 100 µL of buffered water (10 mM TRIS, pH = 7, 0.05% tween- 20). Ten

µL of total nucleic acid (TNA) was taken and stored at -20ºC for performing qPCR to measure Y. ruckeri load. The remaining volume of TNA (90 µL) was added to 10µL

10X DNAse buffer and 2µL Baseline-Zero DNAse (Epicentre) and then incubated for

30 min at 37ºC. The RNA was obtained by adding lysis buffer, cooling on ice and

70

spinning for 5 min at 16,000 x g at 4ºC in ammonium acetate, then the supernatant was removed and mixed with ethanol and transferred to a new column. The column was rinsed twice, dried and RNA eluted with 50µL buffered water (10mM TRIS, pH = 7,

0.05% tween- 20) and stored at -80ºC for further use in investigation of immune gene expression.

Intestine samples were divided into 2 parts, midgut and hindgut, before processing.

Midgut was indicated from beginning at the posterior edge of the stomach and ending with an increase in tube diameter. Hindgut was from the end of midgut to the anus

(Egerton et al., 2018). Each gut sample was cut into small pieces and lysed with 1 mL of lysis buffer (4 M Urea, 0.5% SDS, 50 mM TRIS pH = 8, and 10 mM EDTA) and 5

µL proteinase K (Bioline, Australia), heated overnight at 37ºC and mixed with a vortex.

Ten mg equivalent of each gut sample (300 µL) was thoroughly mixed with 300 µL binding buffer (6 M guanidine hydrochloric, 7.5 M ammonium acetate, pH = 6) and 300

µL of 100% ethanol. The mixture was then transferred to a silica spin column and passed through column after centrifuging. The column was rinsed twice and spun to dry and then TNA eluted in 300 µL of buffered water. The TNA was treated with 50µL

RNAse buffer for 30 min at 37ºC, then DNA was obtained by precipitation after adding

150 µL of lysis buffer, cooling on ice for 5min before adding 250 µL ammonium acetate, then mixing by vortex for 15s and centrifuging at 16,000 x g at 4ºC for 5 min.

Supernatant was then removed and mixed with an equal volume of isopropanol with pink co-precipitant (Bioline, Australia) (200:1 v/v) by inversion and centrifuged at

71

16,000 x g at 4ºC for 10 min. DNA extract was rinsed with 500 µL 60% ethanol twice and resuspended in 50µL of buffered water.

RNA isolation from blood pellets was achieved using a precipitation method. Fifteen

µL of ice thawed blood pellet was taken and added into 400 µL lysis buffer and 20 µL

TCEP and heated at 55ºC for 15 min with 20s mix on a vortex every 3 min. After digestion protein, cellular debris, detergents were removed by centrifugation in 7.5 M ammonium acetate at 16,000 x g for 5 min at 4ºC. Nucleic acids were recovered by isopropanol precipitation at 16,000 x g for 10 min followed by two washes in 60% ethanol of the nucleic acid pellet. Ten µL of TNA was taken and stored at - 20ºC to perform qPCR for Y. ruckeri load. Removal of DNA was carried out by resuspending

TNA in DNAse buffer and treating with Baseline-Zero DNAse for 30 min at 37ºC followed by centrifugation in ammonium acetate. The RNA was re-precipitated in isopropanol, washed twice with ethanol and resuspended in 30 µL buffered water. Total

RNA concentration was determined using an Invitrogen Qubit fluorometer (Invitrogen,

VIC, Australia). The integrity of the RNA was estimated from gel electrophoresis on a

1% agarose gel. All RNA extracted samples were stored at -80ºC for further use.

3.2.8. Yersinia ruckeri load

The TNA extractions from spleen samples (day 40) and gut samples (day 10 and day

40) were selected to perform qPCR in CFX Connect Real-time System (Bio-Rad

Laboratories Inc., USA). The qPCR reaction (10 µL) consisted of 5 µL of 2x MyTaq

HS mix, 400 nM each of YrF and YrR primers, 100 nM of hydrolysis probe (Ghosh et al., 2016) and 2 µL of TNA extraction. The qPCR programme was 95ºC for 3 min and

72

35 cycles of 95ºC for 5s and 60ºC for 30s. Results were quantified by using raw fluorescent unit (rfu) data and the CM3 mechanistic model included in the qPCR package (v.1.4-0) for RStudio statistical computing software (Ghosh et al., 2016).

3.2.9. cDNA transcription

RNA isolated from the blood pellet samples (stored at – 80 ºC) were purified using an

Isolate II RNA mini kit (Bioline, Australia) following manufacturer’s instructions. Total purified RNA (250ng) was reverse transcribed using an iScript reverse transcription supermix for RT-qPCR (Bio-rad, US) according to manufacturer’s instructions. In order to confirm that contaminating trace amounts of genomic DNA were not present after

RNA extraction, a control lacking reverse transcriptase was also performed.

3.2.10. Immune gene expression

Primers used for immune gene expression analysis were designed using Beacon

Designer software (Premier Biosoft, CA, USA) and the respective Atlantic salmon gene sequences available in Genbank (Table 3.2). Each PCR reaction (10 μL) consisted of 5

μL 2x MyTaq HS Mix (Bioline) containing 0.5x SYBR Green (Invitrogen), 400 nM each of forward and reverse primers, and 2 μL DNA template. Cycling conditions included an initial activation of DNA polymerase at 95°C for 1 min, followed by 40 cycles of 15s at 95 °C, 20s at 55 °C and 10s at 72 °C. Melt curve analysis was performed to assess specificity of each reaction. When genes of interest were not amplified, an approximate relative expression was calculated by using the detection threshold as Cq

73

= 40 (Van Maerken et al., 2006). mRNA expression levels were standardised against mean expression levels of two reference genes (elongation factor 1α (EF1α) and β-actin) and were analysed by Mann-Whitney test between two treatments using the qBase Plus software (Biogazelle, Belgium). All the data were graphed by using GraphPad Prism 7.

Table 3.2. Primers used for immune gene expression analysis

Gene Primer sequence (5’ – 3’)

β actin Forward GTGTGACTCGTACATTAG

Reverse TAGAGGGAGCCAGAGAGG

Elongation factor 1 α Forward TGATTGTGCTGTGCTTAC

EF-1α Reverse AACGCTTCTGGCTGTAGG

T-box expressed in T cells Forward ACTGATACGCAAACTTTC

Tbet Reverse ATAATCTGGTGTCCTGAAT

Interleukin 12 beta Forward CTGACTGCCAACAATCAATC

IL-12β Reverse AAACTGTCCCCCTTATCAAC

Interferon gamma Forward CTTGGATAGTGTCAGATAC

IFNγ Reverse CTCAATAGTCCTCAATCG

Tumor necrosis factor alpha Forward CCATACATTGAAGCAGATT

TNFα Reverse CAGCGGTAAGATTAGGATT

GATA binding protein 3 Forward CAGGTGAACGAAGGTAAC

74

GATA3 Reverse CCGATACAAGTCAAGTAAGAA

Interleukin 4 and 13 Forward AATAAACCCAACCAAAGATGA

IL-4/13 Reverse GGCAGAGTTCCAGAGTTA

Interleukin 1 beta Forward AAGACACTGTTACCTACA

IL-1β Reverse ATACCTCCAGATCCAGAC

RAR-related orphan receptor gamma Forward TCACTGGTGCTCATTAAC

RoRγ Reverse CAAGTGTCTCCTCATTCTAT

Interleukin 6 Forward CAGTAACCATGACAACAC

IL-6 Reverse GCATTAGAGAATAAGACACT

Interleukin 22 Forward GGAAAGGGATAAAGAAAG

IL-22 Reverse ATGTGATGATATAGGATGT

Transforming growth factor beta Forward TCAGGATACAGAAGAGGAA

TGFβ Reverse TTGGTGCTTAATGTTGTTC

Cluster of differentiation 4-1 Forward TGTTTCTGGATTCCTTTCAAT

CD4-1 Reverse CCATACACAAGCACATCTC

Cluster of differentiation 8 alpha Forward AAGACAACGCTGGAATGG

CD8α Reverse TATCTGCTCCTCGCTGAA

Immunoglobulin M Forward TGAGGAGAACTGTGGGCTACACT

IgM Reverse TGTTAATGACCACTGAATGTGCAT

75

3.3. RESULTS

3.3.1. Survival rate

Within each group of salmon challenged with Y. ruckeri previously exposed or not exposed to the catecholamine epinephrine (CAT+ and CAT- respectively) no significant difference was seen in survival due to stress at 40 days post challenge so the data were pooled at that level. However, there were significant differences in survival evident between CAT+, CAT- and the unchallenged (Ch-) groups (see Fig 3.1) with lower survival seen in the CAT+ (86.9%) compared to the CAT- (93.7%) and 100% survival in Ch- (p<0.01).

76

Three groups aaa 100 Ch- b CAT-

CAT+ )

% 90

(

l

a v

i c

v

r

u

s

t

n

e

c

r e P 80

70 0 10 20 30 40 50

Days

Figure 3.1. Survival rate of Atlantic salmon from groups challenged with Yersinia ruckeri exposed (CAT+, n = 171) or not exposed to catecholamine (CAT-, n =177) or unchallenged with bacterium (Ch-, n = 176).

The different letters indicate statistically significant differences between two treatments according to Kaplan-Meier survival analysis (pairwise comparisons).

77

3.3.2. Plasma cortisol level

The mean cortisol concentration for 78 fish plasma samples was 4.95 ng/mL and there were not any significant differences in cortisol concentration between treatments and timepoints (Table 3.3).

78

Table 3.3. Plasma cortisol levels in Atlantic salmon prior to challenge and post challenge with Yersinia ruckeri exposed (CAT+) or not exposed to catecholamine (CAT-) and subjected or not to stress (S+ or S- respectively) or unchallenged with bacterium (Ch-). (n = 6/treatment/time point)

Cortisol concentration (ng/mL) Sampling time (post stress) Treatment Mean ±SE

0 Control 10.71 ± 4.78

S- 7.5 ± 3.14 12 h S+ 2.34 ± 0.51

S- 8.36 ± 4.34 24 h S+ 5.66 ± 2.79

Ch-S- 6.6 ± 3.44

Ch-S+ 2.05 ± 0.68

CAT-S- 2.97 ± 1.01 48 h (24 h post challenge) CAT-S+ 1.02 ± 0.18

CAT+S- 6.01 ± 2.53

CAT+S+ 4.04 ± 1.37

CAT-S- 2.18 ± 0.83 31 day (30 day post challenge) CAT-S+ 3.57 ± 2.67

79

3.3.3. Yersinia ruckeri load

Yersinia ruckeri was detected in both mid and hindgut (day 10 and day 40) but interestingly was not detected in any spleen samples at any timepoints (Table 3.4).

Table 3.4. Detectable Y. ruckeri load in gut samples (cell equivalents mg-1) of Atlantic salmon post challenge with Y. ruckeri exposed (CAT+) or not exposed to catecholamine (CAT-) and subjected or not to stress (S+ or S- respectively) or unchallenged with bacterium (Ch-), n = 6 per treatment and timepoint.

Gut DPC Ch- CAT- CAT+

region PS Load PS Load PS Load

Midgut 10 1/6 6 4/6 3048 1/6 47

40 0/6 0 1/6 679 0/6 0

Hindgut 10 2/6 736 1/6 15152 2/6 38

40 0/6 0 1/6 51 0/6 0

DPC: day post challenge, PS: positive sample, Load: Y. ruckeri cell equivalents mg-1

The bacterium was mostly detected in gut samples from day 10 and in the treatment without exposure to catecholamine CAT- than day 40 and other treatments. However,

80

the number of Y. ruckeri in detected samples were not significant different between gut regions, or sampling times or treatments (p>0.05).

3.3.4. Immune gene expression

3.3.4.1. The effect of stress on Atlantic salmon immune genes unchallenged with Y. ruckeri

The impact of stress on fish immune gene responses from blood pellet samples were calculated by comparing mRNA expression levels of immune genes between unstressed fish and stressed fish (without Y. ruckeri challenge) at a sampling time and between stressed fish treatments only at different sampling times (Figure 3.3.a to Figure 3.3.f).

The relative expression of the pro-inflammatory cytokine IL 12β was significantly upregulated (22.3 fold) in stressed fish compared to unstressed fish at day 2 post stress

(day 1 after infection) (Figure 3.3.a). In contrast, the mRNA expression of the T cell marker gene CD8α was significantly downregulated (14 fold) in stressed fish (Figure

3.3.e). The peak expression of tested genes in stressed fish treatments Ch-S+ was seen at day 10 (PI) and the mRNA expression levels were significantly higher compared to day 1 (IL 4/13A, RoRγ, CD8α, IgM) and day 30 (IL 12β, CD 4-1 and IgM).

Additionally, no differences were shown in expressions of TBet, TNFα, IFNγ, GATA3,

IL 1β, IL 6, IL 22 and TGFb.

81

y

t Tbet

i

t

n a

u 256

q

e v

i 64

t

a l

e 16 r

Control d

e 4

s i

l Ch-S+

a 1

m Ch-S- r

o 0.25

n

n

a e

M Day 0 Day 2 Day 11 Day 31

Day post stress y

t IL 12 i

t ✱✱ n

a 4096

u q

1024 e

v ✱✱ i

t 256

a l

e 64 Control

r

d 16

e Ch-S+

s i

l 4 Ch-S- a

m 1 r

o 0.25

n

n a

e Day 0 Day 2 Day 11 Day 31 M Day post stress

Figure 3.3.a. Effect of stress (Ch-S+) and no stress (Ch-S-) on Th1 markers TBet and IL 12β in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

82

y t

i TNF 1&2

t n

a 4096

u q

1024

e

v i

t 256

a l

e 64

r

d 16

e Control

s i

l 4

a Ch-S+

m 1 r Ch-S-

o 0.25

n

n a

e Day 0 Day 2 Day 11 Day 31 M Day post stress

y IFN

t

i

t

n a

u 4096

q

e 1024

v

i t

a 256

l

e r

64 d

e 16 Control

s

i l

a 4 Ch-S+ m

r 1

o Ch-S- n

0.25

n

a e

M Day 0 Day 2 Day 11 Day 31 Day post stress

Figure 3.3.b. Effect of stress (Ch-S+) and no stress (Ch-S-) on Th1 markers TNFα 1-2 and IFNγ in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

83

y t

GATA3 y i

t IL-1

t

i

t

n n

a 4096

a

u

u q

1024 256

q

e

v e

i 256

v t

i 64

a

t

l a

e 64

l

r

e 16

r d

16 Control e

d Control

s e i 4

l 4

Ch-S+ s

a i

l Ch-S+ m 1 a 1

r Ch-S- m

o Ch-S- r

n 0.25

o 0.25

n

n

a

n e

Day 0 Day 1 Day 11 Day 31 a M e Day 0 Day 2 Day 11 Day 31 Day post stress M

y Day post stress

t IL 4/13A

i t

✱

n a

u 256

q

e v

i 64

t

a l

e 16 r

Control d

e 4 s

i Ch-S+ l

a 1

Ch-S-

m r

o 0.25

n

n a

e Day 0 Day 2 Day 11 Day 31 M Day post stress

Figure 3.3.c. Effect of stress (Ch-S+) and no stress (Ch-S-) on Th2 markers in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

84

y

RoR t IL 22

i

y

t

t

i t

256 n

n a

a ✱ 4096

u

u

q q

64

e 1024

e

v

i

t

v i

a 16 ✱ l

t 256

e

a

r

l

d Control 4 e 64

e Control

r

s

i l d Ch-S+ a Ch-S+ 16

1 e

m

s

r

i l

o Ch-S- 4 Ch-S- n

0.25 a

n

m 1

a

r

e o

M 0.25 n

Day 0 Day 2 Day 11 Day 31 n

Day post stress a

e Day 0 Day 2 Day 11 Day 31

M

y

t i

t IL 6 Day post stress

n

a

u q

1024

e v

i 256

t a

l 64

e

r

d 16 Control

e s

i 4 l Ch-S+

a 1 m

r Ch-S-

o 0.25

n

n a

e Day 0 Day 2 Day 11 Day 31 M Day post stress

Figure 3.3.d. Effect of stress (Ch-S+) and no stress (Ch-S-) on Th17 markers in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group).

Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

85

y

t CD8

i

t n

✱✱ a

u 1024

q

e 256 v

i ✱✱

t a

l 64

e ✱✱ r 16

d Control

e s

i 4 l Ch-S+

a 1 m

r Ch-S-

o 0.25

n

n a

e Day 0 Day 2 Day 11 Day 31 M

Day post stress y

t CD4-1

i ✱

t n

a 4096

u q

1024

e

v i

t 256

a l

e 64

r

d 16 Control

e

s i

l 4 Ch-S+ a

m 1

r Ch-S- o

n 0.25

n a

e Day 0 Day 2 Day 11 Day 31 M Day post stress

Figure 3.3.e. Effect of stress (Ch-S+) and no stress (Ch-S-) on T cell marker CD8α and CD4-1 in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

86

y

t TGFb

i

t

n a

u 1024

q

e

v 256

i

t a

l 64

e r

16 d

e Control s

i 4 l

a Ch-S+

1 m r Ch-S-

o 0.25

n

n a

e Day 0 Day 2 Day 11 Day 31 M

Day post stress

y

t i

t IgM

n a

u 512 ✱

q ✱

e 128

v i

t 32

a l

e 8

r

d 2 Control e

s 0.5

i Ch-S+ l

a 0.125 Ch-S-

m

r

o

n

n

a Day 0 Day 2 Day 11 Day 31 e

M Day post stress

Figure 3.3.f. Effect of stress (Ch-S+) and no stress (Ch-S-) on Treg marker TGFb and B cell marker IgM in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

87

3.3.4.2. The effect challenge with Y. ruckeri (unexposed to catecholamine) on Atlantic salmon immune genes

The effect of challenge with Y. ruckeri (unexposed to catecholamine) on salmon immune responses was determined by comparing mRNA gene expression levels between challenged and unchallenged for unstressed fish only (see Figure 3.4.a to

Figure 3.4.f). The relative expression of IL 12β in challenged fish was significantly higher (32 fold) at day 10 in comparison at day 1 (Figure 3.4.a) similarly there was a

25-fold increase in mRNA expression of the cytokine interferon IFNγ gene compared to unchallenged at day 1 (Figure 3.4.b). There was a significant reduction in expression level of TNFα, IFNγ and IL 6 at day 30 compared to its level at day 1 (Figure 3.4.b and

3.4.d). There were no other differences in the expressions of TBet, GATA3, IL4/13, IL

1β, RoRγ, IL 22, CD8α, CD4, IgM and TGFb.

88

y

t Tbet

i

t

n a

u 1024

q

e

v 256

i

t a

l 64

e

r

d 16 Control

e s

i 4

l CAT-S- a

m 1

r Ch-S- o

n 0.25

n

a e

M Day 0 Day 1 Day 10 Day 30

Day post infection

y t

i IL 12

t n

a ✱ u

q 1024

e

v 256

i

t a

l 64

e r

16

d e

s Control

i 4 l

a 1 CAT-S-

m r

o 0.25 Ch-S-

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.4.a. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on Th1 markers TBet and IL 12β in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

89

y

t TNF 1&2

i t

✱ n

a 4096

u q

1024

e v

i 256

t

a l

e 64

r

d 16

e Control

s i

l 4

a CAT-S-

m 1 r

o 0.25 Ch-S-

n

n a

e Day 0 Day 1 Day 10 Day 30 M

Day post infection y

t IFN i

t ✱ n

✱ a

u 4096

q

e 1024

v

i t

a 256

l e

r 64

d

e 16 Control

s i l 4

a CAT-S- m

r 1 Ch-S- o

n 0.25

n

a e

M Day 0 Day 1 Day 10 Day 30 Day post infection

Figure 3.4.b. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on Th1 markers TNFα 1-2 and IFNγ in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

90

y

t

i t

GATA 3 y

t n

i IL-1

a

t

u

n

q a

1024 256

u

e

q

v

i

t 256

e 64

a

v

l

i e

64 t

r

a

l 16 d

16 e e

Control r

s

i d

l 4 4 a

CAT-S- e Control

s m

1 i l r Ch-S- 1

a CAT-S- o

n 0.25

m

r

n 0.25 Ch-S-

o

a

n e

Day 0 Day 1 Day 10 Day 30

M

n a

Day post infection e Day 0 Day 1 Day 10 Day 30

M y

t Day post infection i

t IL 4/13A

n a

u 256

q

e

v 64

i

t

a l

e 16

r

d Control

e 4

s i

l CAT-S-

a 1

m Ch-S- r

o 0.25

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.4.c. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on Th2 markers in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

91

RoR y

t IL 22

i

y t

t 256

i

t

n

n

a

a u u 64

q 4096

q

e

e v

i 1024 t

16 v

i

a

l

t

e r

a 256

Control l

d 4

e

e r

s 64

i

l CAT-S- d a 1

e 16

m r

Ch-S- s Control

o

i l

n 4

0.25 a

n CAT-S-

a m

e 1

r M

o 0.25 Ch-S- n

Day 0 Day 1 Day 10 Day 30 n

Day post infection a e

y Day 0 Day 1 Day 10 Day 30

t M

i IL 6 t

n ✱ Day post infection a

u 1024

q

e 256

v

i t

a 64

l

e r

16

d Control e

s 4 i

l CAT-S- a 1

m Ch-S- r

o 0.25

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.4.d. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on Th17 markers in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01)

92

y

t CD8

i

t

n

a u

q 4096

e

v 1024

i t

a 256

l

e r

64

d Control

e 16

s i

l 4 CAT-S- a

m Ch-S-

r 1 o

n 0.25

n a

e Day 0 Day 1 Day 10 Day 30 M

Day post infection y

t CD4-1

i

t n

a 4096

u q

1024

e

v i

t 256 a l Control

e 64

r

d 16 CAT-S-

e

s i

l 4 Ch-S- a

m 1 r

o 0.25

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.4.e. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on T cell marker CD8α and CD4-1 in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

93

y

t TGFb

i

t n

a 1024

u q

256

e

v i

t 64

a

l

e r

16 d

e Control

s 4

i l

a CAT-S-

1 m r Ch-S-

o 0.25

n

n a

e Day 0 Day 1 Day 10 Day 30 M

Day post infection y

t IgM

i

t n

a 512

u q

128

e v

i 32

t a

l 8

e

r

d 2 e

s Control

i 0.5 l

a 0.125 CAT-S-

m r

o Ch-S-

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.4.f. Effect of Yersinia ruckeri challenge (no exposed to catecholamine epinephrine CAT-S-) and unchallenged (Ch-S-) on Treg marker TGFb and B cell marker IgM in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

94

3.3.4.3. The effect challenge with Y. ruckeri (exposed and unexposed to catecholamine) on unstressed Atlantic salmon immune genes

When investigating immune gene expression in unstressed but challenged Atlantic salmon prior exposure of Y. ruckeri to epinephrine (CAT+S-) significantly upregulated the expression of IgM (by 2.5-fold) on day1 post infection (Figure 3.5.f). In addition

TNFα (3.8-fold) and IL 22 (8.4-fold) (day 30) were also significantly upregulated compared to those genes from salmon challenged with Y. ruckeri unexposed to epinephrine (CAT-S-) (Figures 3.5.b and 3.5.d). In contrast, the relative expression of

IFNγ (Figure 3.5.b) and IL 6 (Figure 3.5.d.) were significantly lower in the CAT+S- treatment group at day 1 (at 39.5 and 17.2-fold lower respectively) compared to those in the CAT-S- treatment. Among CAT+S- treatments at different sampling points, there were significantly higher expressions of IL 4/13 and RoRγ (day 10) and IFNγ and IL

22 (day 30) in comparison to day 1 (Figures 3.5.c and 3.5.d). There were no differences in gene expressions of TBet, IL 12β, GATA 3, IL 1β, CD8α, CD4 and TGFb.

95

y

t Tbet

i

t

n a

u 1024

q

e 256

v

i

t a

l 64

e r

16 Control

d

e s

i 4 CAT+S- l a 1

m CAT-S- r

o 0.25

n

n a

e Day 0 Day 1 Day 10 Day 30 M

Day post infection y

t IL 12

i

t

n a

u 1024

q

e 256

v

i

t a

l 64

e r 16

d Control e

s 4 i l CAT+S-

a 1 m

r CAT-S-

o 0.25

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.5.a. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non- exposure (CAT-S) to epinephrine on Th1 markers TBet and IL 12β in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

96

y t

i TNF 1&2

t n

a 4096

u q

1024

e

v i

t 256

a l

e 64 ✱

r

d 16 Control

e

s i

l 4 CAT+S- a

m 1

r CAT-S-

o 0.25

n

n a

e Day 0 Day 1 Day 10 Day 30 M

Day post infection

y t

i IFN t

✱ n

a ✱✱

u 4096

q

e 1024

v

i t

a 256

l

e r

64

Control d

e 16 s

i CAT+S-

l 4 a

m CAT-S-

r 1 o

n 0.25

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.5.b. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non- exposure (CAT-S) to epinephrine on Th1 markers TNFα 1&2 and IFNγ in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

97

y t

i GATA 3 t

y IL-1

t

n

i

t

a n

u 256 a

q 1024

u

e

q

v

i 256 64

e

t

v

a

i l

64 t e

a 16

l

r

e d

16 r

e Control 4

d

s i

4 e

l

s i a CAT+S-

l 1 Control a

m 1

r CAT-S- m

o CAT+S- r 0.25

n 0.25

o

n n

CAT-S-

a

n

e a

Day 0 Day 1 Day 10 Day 30 e

M Day 0 Day 1 Day 10 Day 30 M

Day post infection Day post infection

y

t i

t IL 4/13A

n a

u 256 ✱

q

e 64

v

i

t a

l 16

e r

Control

d 4 e

s CAT+S-

i l

a 1 CAT-S- m

r 0.25

o

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.5.c. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non-exposure (CAT-S) to epinephrine on Th2 markers in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

98

y

t

i

t

y n

RoR t IL 22

i

a t

u ✱✱

n q

256 a

u

e 4096

q

v

i 64 e

t 1024

v

a

i l

✱ t e

16 a 256

r

l

e ✱✱ d

Control r 64

e 4 Control

d

s e

i CAT+S- 16 l

s CAT+S-

i a 1 l 4

CAT-S- a m

r CAT-S-

m 1 r

o 0.25

o n

0.25

n

n

n

a

a e

Day 0 Day 1 Day 10 Day 30 e Day 0 Day 1 Day 10 Day 30

M

M y

t Day post infection

i t

n IL 6 a

u ✱✱

q 1024

e v

i 256

t a

l 64

e

r

d 16

e Control s

i 4 l CAT+S-

a 1 m

r CAT-S-

o 0.25

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.5.d. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non-exposure (CAT-S) to epinephrine on Th17 markers in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

99

y

t CD8

i

t

n

a u

q 4096

e

v 1024

i t

a 256

l

e r

64 d Control

e 16

s i

l 4 CAT+S-

a m

r 1 CAT-S- o

n 0.25

n a

e Day 0 Day 1 Day 10 Day 30 M

Day post infection y

t CD4-1

i

t n

a 4096

u q

1024

e

v i

t 256

a l

e 64

r

d 16 Control

e

s i

l 4 CAT+S- a

m 1

r CAT-S-

o 0.25

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.5.e. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non- exposure (CAT-S) to epinephrine on T cell markers in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

100

y

t TGFb

i

t n

a 1024

u q

256

e

v i

t 64

a

l

e r

16

Control d

e 4 s

i CAT+S- l a 1

m CAT-S- r

o 0.25

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

y IgM

t

i

t n

a 1024 u

q 256

✱ e

v 64

i

t a

l 16 Control

e r 4

d CAT+S- e

s 1 i

l CAT-S-

a 0.25

m

r

o

n

n a

e Day 0 Day 1 Day 10 Day 30 M Day post infection

Figure 3.5.f. Effect of Yersinia ruckeri challenge prior exposure (CAT+S-) and non- exposure (CAT-S) to epinephrine on Treg marker TGFb and B cell marker IgM in Atlantic salmon.

The mRNA gene expression is relative to the geometric mean of the two reference genes EF1-α and β-actin (scaled to control group). Asterisk denotes statistically significant different between treatment (*: p < 0.05; **: p < 0.01).

101

3.4. DISCUSSION

In this study, morbidity and/or mortality of Atlantic salmon as a result of infection with

Y. ruckeri was found to begin within 3-4 days post-challenge and had significantly subsided by 15-20 days post challenge. A pertinent finding was that prior exposure of

Y. ruckeri to epinephrine resulted a higher cumulative mortality of salmon compared to those challenged with the non-epinephrine treated bacterium. This can be interpreted as in increase in virulence and supports the results from in vitro experiments which showed that epinephrine could enhance virulence factors of Y. ruckeri (see chapter 2). The survival data are also in line with previous findings which have shown a decrease in survival rates of some crustaceans and teleosts when challenged with stress hormone- exposed bacteria (norepinephrine and dopamine) presumably due to bacterial virulence enhancement. For example Artemia franciscana and Macrobrachium rosenbergii infected with V. harveyi (Pande et al., 2014; Yang et al., 2014), rainbow trout infected with Y. ruckeri (Torabi Delshad et al., 2019) and crucian carp infected with A. hydrophila (Gao et al., 2019) all showed lower survival rates when challenged with respective bacteria previously exposed to a stress hormone compared to groups challenged with bacteria unexposed to the hormone.

The mean cortisol level of Atlantic salmon in this study was 4.95 ng/mL which was similar to that of Carey and McCormick (1998) and Sundell et al. (2003) who reported plasma cortisol ranges of 4 to 11 ng/mL and 5 to 40 ng/mL for parr and smolting

Atlantic salmon respectively. Plasma cortisol level is usually used as an indicator for stress assessment. Elevation of plasma cortisol concentration was previously reported

102

in Atlantic salmon subjected to acute and/or chronic stress (Carey and McCormick,

1998; McCormick et al., 1998; Fast et al., 2008; Pankhurst et al., 2008; Sundh et al.,

2009), seawater transfer and salinity acclimation (Sundell et al., 2003; Hvas et al.,

2018), sea lice Lepeoptheirus salmonis infection (Bowers et al., 2000; Mustafa et al.,

2000; Fast et al., 2006), and infectious pancreatic necrosis virus challenge (Sundh et al.,

2009). However, in this study cortisol levels were not significantly different in stressed fish and challenged fish in comparison with the controls. Although, most studies in

Atlantic salmon subjected to an acute stress such as handling and confinement stress

(Carey and McCormick, 1998) or hyperoxygenation and reduced water flow stress

(Sundh et al., 2009) or chasing stress (Madaro et al., 2016) found that cortisol remains elevated for hours and days, it is possible that the stress response resulted in the release of cortisol which returned to normal within a couple of hours after the initial stressor

(Einarsdóttir and Nilssen, 1996; McCormick et al., 1998; Fast et al., 2008; Pankhurst et al., 2008) and therefore any increase in cortisol was missed in this study. Interestingly,

24 hours after the application of stress the fish were challenged with the pathogenic bacteria Y. ruckeri which would also have been a potential stressful procedure, however cortisol concentrations did not show any statistical difference at 12 hours after the event.

In contrast, Sundh et al. (2009) reported that Atlantic salmon previously stressed (with hyperoxygenation and reduced water flow in freshwater) then challenged with infectious pancreatic necrosis virus after seawater transfer showed significantly higher plasma cortisol which persisted to day 14 post challenge. The dissimilar results of cortisol levels and related time response may relate to the nature and duration of the different stressors, pathogenic mechanisms, fish ages and their response (to the same

103

stressor) as well as experimental procedures. Although sampling procedures were applied as consistently as possible to treated fish, there is a possibility that cortisol levels could have been affected by undesired stressors during blood sampling including chasing and anesthetising fish before bleeding that may have masked any differences between the unstressed and stressed treatments. Therefore, a more conclusive effect of stress and Y. ruckeri challenge on plasma cortisol levels of Atlantic salmon may have been found had more sampling times post stress and the optimisation of blood sampling procedures to minimise potential stress been undertaken. Nonetheless, in this study fish were subject to an activity well known to induce stress before the bacterial challenge to see if stress-induced endogenous hormones (in particular catecholamines) had any impact on Y. ruckeri virulence, hence the study focused more on host survival data and host immune responses than the specific concentration and regulation of circulating catecholamines.

The thymus, kidney and spleen are major lymphoid organs in fish which are most commonly use in fish immunology. In this study however, due to a failure in spleen storage, whole blood cells of Atlantic salmon (red and white blood cells) were used to determine changes in expression of immune genes. The white blood cells are involved in both innate and adaptive immune responses and, although red blood cells mainly function in gas exchange, they have been recently shown to express immune relevant genes participating in stress responses, immune responses and apoptosis in rainbow trout, Atlantic salmon and tilapia (Morera et al., 2011; Dahle et al., 2015; Shen et al.,

2018).

104

A variety of immune changes have been reported after applying different kinds of stressors in fish including activating and suppressive processes (see Tort (2011)). In this study, the effect of acute chasing stress on 14 immune related genes produced by

Atlantic salmon blood cells resulted in a significant upregulation of IL-12β and downregulation of CD8α in stressed fish compared to unstressed fish. The increase in

IL-12β mRNA observed in this study is consistent with data obtained for pufferfish

(Takifugu obscurus) and common carp (Cyprinus carpio) (Jia et al., 2014; Cheng et al.,

2015). Exposure to high concentration of ammonia in pufferfish or exposure of carbon tetrachloride to common carp resulted in higher expression of innate immune genes including TNFα, IL-6 and IL-12 (Jia et al., 2014; Cheng et al., 2015). Interleukin-12 is a critical cytokine bridging innate and adaptive immunity in mammals (Langrish et al.,

2004). This cytokine is produced by inflammatory macrophages and activates dendritic cells and plays an important role in driving an efficient cellular immune response as an important regulator of Th1 cell responses (Langrish et al., 2004). Although there are limited studies about the effect of acute stress on the change of IL-12 in fish, early induction of IL-12 has been suggested to contribute to disease resistance (Robben et al.,

2004). Interestingly, the upregulation of IL-12 in Atlantic salmon infected with the parasite Kudoa thyrsites has been indicated by a positive correlation between severity of K. thyrsites and IL-12 production (Braden et al., 2018). Elevated expression of IL-

12 has been recently reported in rainbow trout infected with Y. ruckeri, viral haemorrhagic septicaemia virus and parasitic proliferative kidney disease (Wang et al.,

2014). In contrast, this study showed a reduction of mRNA production for CD8α in stressed fish. Cluster of differentiation 8 alpha has a vital role in the T cell-mediated

105

immune response and resistance to viral infections. CD8 is expressed on T cells and binds to major histocompatibility complex (MHC) class I proteins (as a co-receptor for

T-cell receptor) hence presents peptides on the cell surface. Interaction between CD8 receptor, T cell receptor and the MHC-peptide complex will induce signal transduction resulting in activation of T cells (Abbas et al., 2014). In mice, lower CD8 T-cell number was described in stressed compared to unstressed mice (Dhabhar, 2008). It has been reported that acute stress can induce redistribution of leukocytes from the blood to the skin, hence suppression of splenic and blood responses to T-cell mitogens and depletion of leukocytes in these compartments during acute stress (Dhabhar, 2008). In Atlantic salmon, depletion of CD8α was reported in different tissues infected with a highly virulent strain of infectious salmon anaemia virus compared to that found after exposure to a less virulent strain at early stages of infection, and hence it suggests that CD8α has a protective role in immune defence (at the early infection stages) (Hetland et al., 2011).

Stress has been reported to increase in expression of several immune related genes such as IL 1β, IL 8, g type lysozyme and bactericidal permeability increasing protein/lipopolysaccharide binding protein (BPI/LBP) in cod subjected to overcrowding

(Gadus morhua) (Caipang et al., 2008) expression of other genes such as lysozyme

Lyzll, TNF-1α, IL 1β, IL 8 and IFNγ in rainbow trout were downregulated after overcrowding stress (Yarahmadi et al., 2016). However in the study reported here only

IL-12 β and CD8α were impacted. Fridell et al. (2007) showed similar finding that the mRNA expression of three genes, IFN 1α, Mx and IL 1β, were not affected in Atlantic salmon subjected to hyperoxygenation and low water flow stress. It is possible that the differentiation responses of the immune genes are dependent upon species, the tissue

106

sampled and the nature of the stressor (Braden et al., 2018). The expression of immune genes under the effect of stress may be also dependant on time-course, as the mRNA transcripts at day 10 of IL 4/13A, RoRγ, CD8α, IgM and IL 12β, CD 4-1 and IgM were significantly higher compared those on day 1 and day 30. This concurs with the finding that changes in immune gene expression due to handling stress in Atlantic salmon can begin from 1 hour and be maintained for up 4 weeks after the stress event (Fast et al.,

2006).

The challenge model with Y. ruckeri (exposed or not exposed to epinephrine) showed differentially changes of the host immune genes. Overall, IFNγ and IL 6 were the most impacted in both treatments while expression of Tbet, CD4, TGFb, GATA3, IL1β and

RoRγ was not influenced by treatment or time-course. The effects of Y. ruckeri challenge on Atlantic salmon were only seen in differential expression of the IFNγ gene which was 25-fold higher in challenged compared to unchallenged fish 1 day post infection. Interferon gamma is a cytokine which is critical in regulating microbial killing activities for intracellular pathogens by activation of macrophages (Bogdan et al., 2004;

Langrish et al., 2004; Robertsen, 2006). Microbicidal activity of IFNγ induces the phagocyte oxidase components which results in increase superoxide production and other antibacterial oxygen species during phagocytosis of pathogens (Shtrichman and

Samuel, 2001). In addition, the stimulation of nitric oxide synthetase 2 (NOS2) to produce nitric oxide, several antimicrobial proteins and several enzymes also contribute to both antibacterial and antiviral activities (MacMicking et al., 1997; MacMicking,

2004; Robertsen, 2006). Protective effects of IFNγ against infection with intracellular

107

pathogens have been reported as inhibition of infectious pancreatic necrosis virus

(IPNV) in salmon cells and the activation of immune responses after Edwardsiella tarda infection in olive flounder (see Zou and Secombes (2016). The expression of IFNγ gene in this study is consistent with the finding that lipopolysaccharide injection challenge enhanced expression of IFNγ gene in Japanese flounder (Paralichthys olivaceus) spleen at early sampling time and reached a peak at 24 hours post injection before turning to normal (Zhang et al., 2017). A reduction of mRNA IFNγ transcripts at day 30 in challenge fish compared to day 1 also consistent with findings of McCarthy et al. (2016) who reported higher expression of IFNγ in the proximal intestine of infected rainbow trout by oral transmission of segmented filamentous bacteria from day 8 to day 23 but not at day 30. In addition, in the study reported here higher expression of IL 12β at day

10 and lower expression levels of TNFα, IL 6 at day 30 compared to day 1 and day 10 suggest that the challenge was no longer affecting immune responses at the late stage of the challenge. This consists with the survival analysis (Figure 3.1) and Y. ruckeri load in the gut (table 3.4). Mass mortality in challenged group was observed between day 10 to day 20, and no more dead fish were seen after day 20. Additionally, results from quantification Y. ruckeri load in the gut showed more detectable samples and higher amount of Y. ruckeri in samples from challenged fish at day 10 than day 40.

Prior exposure of Y. ruckeri to epinephrine before challenge impacted more immune related genes compared to challenge effects from the bacterium not exposed to the hormone, including elevated expression of IgM (day1), TNFα and IL22 (day 30) and depressed of IFNγ and IL 6 (day 1). This suggests that epinephrine has some effects on the virulence of Y. ruckeri and/or pretreated Y. ruckeri with epinephrine and may alter

108

the host immune response in salmon. A faint expression of IFNγ at the early stage of the infection seems to indicate that IFN was rapidly deactivated and could not stimulate

Th1 cells efficiently, which resulted in massive tissue damage and early mortality in

Mycobacterium infected mice (Lopez et al., 2003). This is consistent with survival data shown in figure 3.1 that mortality occurred earlier and at a higher rate than was seen in

CAT+ treatment. In addition, one of the most important functions of IL-6 is inducing the synthesis of hepcidin in liver cells which further the blocks the release of iron from macrophages, hepatocytes and enterocytes and thereby inhibiting bacterial propagation

(Dunn et al., 2007). Moreover, the lower expression of IFNγ and IL-6 in CAT+S- treatment may also relate to Y. ruckeri virulence enhancement due to pre-exposure to the hormone. The gene expression and survival data indicate that it is reasonable to assume the that Y. ruckeri exposed to epinephrine is more virulent than that not exposed which agrees with other studies which have shown lower expression in some immune genes in challenge models with virulent or wildtype compared those with avirulent/ less virulent or mutant strains. Although there is little published information on fish, there are some related data on mice. Challenging mice with the more virulent Beijing

Mycobacterium tuberculosis strain showed higher bacterial multiplication, early and massive pneumonia and death as well as lower expression of IFNγ in comparison with less virulent Canetti strain (Lopez et al., 2003). Similarly, IL-6 and TNFα were shown to be significantly upregulated in mice (nasal tissue) infected with less virulent

Streptococcus pneumoniae WCH16 strain compared to those infected with the more virulent D39 and WCH43 (Mahdi et al., 2008). In another study on zebrafish, the IFNγ expression level was higher in challenge treatment with heat-inactivated bacteria and

109

their extra cellular products than live bacteria A. hydrophila (Rodríguez et al., 2008).

Furthermore, the production mRNA of some immune genes seems to be related to disease severity as many genes related to IFN were identified significantly downregulated in turbot (spleen and head kidney) suffering severe myxozoan parasite

Enteromyxum scophthalmi like interferon regulatory factor IFR3, IFR4, interferon- induced very large gtpase 1 interferon regulatory factor GVINP1, interferon gamma receptor alpha chain IFNGR1 (Robledo et al., 2014). Elevated expression of IL-22 in fish in treatment CAT+S- at day 30 compared to CAT-S- is consistent with the function of this cytokine as IL-22 is known to induce the production of several classes of antimicrobial peptides and plays a critical role in regulating inflammation and promoting bacterial clearance (Kudva et al., 2011). This finding is also partly supported by the result of Y. ruckeri load (Table 3.4) showing lower detected cases of infected samples and lower numbers of Y. ruckeri cells in CAT+ compared to CAT- treatment.

Similarly, significant upregulation of IL-22 and TNFα at day 30 in CAT+ challenge treatment compared to day 1 may be due to the impairment of the immune system in trying to clear infected bacteria and protect the host from death during early stages of the infection. However, more data from further studies to confirm this. A study on immune gene expression in rainbow trout after infection with a segmented filamentous bacteria has shown similar finding to that presented here with an increase in expression of IL-22 at day 30 post infection but no change in expression of IFNγ (McCarthy et al.,

2016). Interestingly, the data from McCarthy et al. (2016) also showed no significant difference observed in expression of IL 1β at any sampling points, which is also consistent with the results presented here. Interleukin 1β is known as the pivotal early

110

response pro-inflammatory cytokine in enabling hosts to respond to infectious pathogens by activation of host pattern recognition receptors and inducing an inflammatory cascade and other defensive responses (Rojo et al., 2007; Zou and

Secombes, 2016). Therefore, upregulation of IL 1β has been usually reported in many fish after challenge such as; rainbow trout infected with Y. ruckeri (Harun et al., 2011;

Raida et al., 2011; Wangkahart et al., 2017), Atlantic salmon with Aeromonas salmonicida (Ewart et al., 2008), zebra fish with A. hydrophila (Rodríguez et al., 2008) or Listonella anguillarum (Rojo et al., 2007), seabream with Vibrio anguillarum

(Chaves-Pozo et al., 2004) and large yellow croaker (Larimichthys crocea) with V. alginolyticus (Jun et al., 2015). However, upregulation of IL 1β was varied and dependent on many factors including the balance of different immune response pathways, hence more studies related to this cytokine response should be considered in particular to vaccinated fish as fish in this study had had vaccination against Y. ruckeri before.

In conclusion, this study showed results related to the interaction between the virulence of the important pathogen Y. ruckeri and the immune responses of Atlantic salmon under the specific environment. Although stress application did not result in changing of cortisol levels, the immune response was different between stressed fish and unstressed fish evident by the significant upregulation of IL12β and CD8α in those stressed. In addition, exposure of Y. ruckeri to epinephrine prior a challenge enhanced virulence of Y. ruckeri and therefore resulted in differential regulation of immune genes.

The reduction of IL-6 and IFNγ transcripts in the early stage of infection and elevation

111

of IL-22 and TNFα at late sampling time are in line with the results from survival analysis and the load of Y. ruckeri in infected fish.

112

Chapter 4. EFFECTS OF STRESS AND Yersinia ruckeri CHALLENGE ON

GUT MICROBIOME OF ATLANTIC SALMON (Salmo salar)

4.1. INTRODUCTION

Gut microbiota is known to have vital roles in host development, physiology and health from early life stages (Llewellyn et al., 2014; Tarnecki et al., 2017). In fish, diversity and composition of gut microbiota are influenced by many factors such as genotype, physiological status, health condition, life style and environmental changes (Llewellyn et al., 2014; Butt and Volkoff, 2019). Altering the structure of gut microbiota may change function and metabolic activities, hence affecting the feeding, growth, energy storage and health of the host (Butt and Volkoff, 2019).

Yersinia ruckeri is a major pathogen impacting global salmonid farming. It has been widely isolated from different animals and the environment as well as diseased fish

(Kumar et al., 2015). This bacterium can reside in the distal fish intestine and establish a subclinical infection, resulting in asymptomatic carriers which are able to transmit infection horizontally particularly when the host is stressed (Tobback et al., 2007;

Ghosh et al., 2018). In addition, infections can be detectable in the intestine prior to becoming symptomatic (Rodgers, 1992; Ghosh et al., 2018). Therefore, more information about the composition of the gut microbiome in Atlantic salmon and how it may vary during stressful scenarios and Y. ruckeri challenge will be of benefit in yersiniosis and/or enteric redmouth disease management strategies. In addition, study

113

on gut microbiome may also provide useful information which could assist in selection of probiotic strains.

The objective of this study was to investigate the effects of stress and challenge with Y. ruckeri (including treatment with exogenous epinephrine) on the salmon microbiome

(including a group exposed to an acute stressor) in different gut regions at different time points post stress and challenge.

4.2. MATERIALS AND METHODS

Animal ethics approval, fish, experimental design and sampling and DNA extraction for intestine samples: see chapter 3.

4.2.1. PCR amplification and sequencing

As there were low molecular weight non-specific PCR products when TNA extractions from midgut and hindgut were used directly in PCR with 16S rRNA gene V1-V3 region amplification primers (27F: 5’ - AGAGTTTGATCMTGGCTCAG - 3’, 519R: 5’ –

GWATTACCGCGGCKGCTG – 3’), a semi-nested PCR was performed. The initial

PCR consisted of 2x MyTaq HS mix (Bioline), 200 nM of each 27F and 1492R 16S rRNA gene primers (Ooi et al., 2019) and 2 µL of 1:50 diluted TNA extraction in molecular grade water to a final volume of 10 µL. PCR was performed using a C1000TM

Thermal Cycler (Bio-Rad Laboratories Inc., USA). Cycling conducted of 3 min at 95ºC,

20 cycles of 95ºC for 15s, 50ºC for 30s and 72ºC for 30s, and a final extension for 3min

114

at 72ºC. PCR included negative extraction and amplification controls. PCR products were checked with 1% agarose gel electrophoresis and diluted 10 times before sending to the Australian Genome Research Facility (AGRF, VIC, Australia) for sequencing of the V1-V3 region of the 16S rRNA gene on an Illumina MiSeq with paired-end 300 bp read chemistry. The hypervariable V1-V3 region of the 16S rRNA gene was chosen due to its ability to provide discriminatory taxonomic resolution across a wide range of bacteria and the large and available 16S rRNA database (Johnson et al., 2019).

4.2.2. Data analysis

The demultiplexed FASTQ sequences from AGRF were imported into the 2019.10 version of Quantitative Insights Into Microbial Ecology (QIIME 2) software package

(Bolyen et al., 2019) for 16S rRNA marker-gene analysis. Quality plots of sequences were visualised and used to assess the maximum sequence length before the sequence quality fell below Q30. The DADA2 plugin was then used to trim sequences of the 27F and 519R primers, truncate the sequences according to the previous quality assessment before denoising and removing potential chimeras and outputting representative sequences from the pair-end reads which are the amplicon or exact sequence variants

(ASVs). The ASVs were classified using a Naive Bayes classifier pre-trained on the

Greengenes 13_8 99% OTUs database (McDonald et al., 2012). For downstream analyses such as weighted and unweighted UniFrac beta diversity analyses that require phylogenetic information a phylogenetic tree (both rooted and unrooted) was created using the phylogeny plugin and align-to-tree-mafft-fasttree pipeline.

115

For further investigations of the hindgut of challenged fish at day 10 the FASTQ sequence data from AGRF for these samples was uploaded for cloud-based analysis on the One Codex microbial genomics platform (https://www.onecodex.com/). Sequences were aligned against the One Codex Targeted Loci database that contained 247,647 records (31,633 unique species) covering 5S, 16S, 23S, gyrB, rpoB, 18S, 28S, and ITS genes from known microbes. Sequences were aligned against the Targeted Loci

Database using the Scalable Nucleotide Alignment Program (SNAP) (Zaharia et al.,

2011) and assigned a taxonomic grouping. Quality and abundance filtering were then performed to minimize the number of false positive assignments introduced by sequencing errors.

Sample metadata, ASV tables, tables and the phylogenetic tree file were uploaded and imported into MicrobiomeAnalyst (https://www.microbiomeanalyst.ca/)

(Dhariwal et al., 2017) to perform statistical analyses of alpha diversity, beta diversity, abundance, clustering and correlation. Imported data was filtered to remove low abundant ASVs (≤2 counts) and those ASVs of 10% or lower prevalence in samples.

Data were normalised by rarefying to the minimum library size. Alpha diversity refers to the measurement of diversity within a community, and was measured using different metrics including Chao1, ACE, Shannon and Simpson with each statistically assessed for significance using a non-parametric Kruskal-Wallis test. Statistical analyses of differential abundance of taxa between treatments was also assesses by classical univariate statistics using non-parametric Kruskal-Wallis tests. Beta diversity refers to the measurement of varying bacterial diversity between treatments and was analysed by

116

Bray Curtis, Jaccard, weighted and unweighted UniFrac distance based Principal

Coordinates Analysis (PCoA) and Permutational multivariate analysis of variance

(PERMANOVA). A phylogenetic tree was not uploaded to MicrobiomeAnalyst in the case of the One Codex results thus UniFrac-based analyses of beta diversity was unavailable. In this instance Bray Curtis with Non-Metric Multidimensional Scaling

(NMDS) and Analysis of Similarities (ANOSIM) statistical analysis was performed.

4.3. RESULTS

4.3.1. Characteristics of the sequence data

After quality filtering, a total of 1,597,987 reads were obtained. Average counts per samples was 20,753 and no template control, extraction control and 5 samples that had low read counts (<1,000) were excluded. Mean of Good’s coverage was 91.5% and rarefaction curves for all samples tended to plateau, indicating the sequencing of the samples approached saturation.

4.3.2. Alpha diversity analyses

Alpha diversity of bacterial community was measured by Chao1, ACE, Shannon and

Simpson indexes. There were no statistical differences between experimental groups

(stressed and unstressed fish, challenged (CAT+, CAT-) and unchallenged, different sampling points (day 0, day 10 and day 40) or different gut regions (midgut and hindgut)) (p >0.05) (Table 4.1).

117

Table 4.1. Richness and alpha diversity indexes for gut sequence libraries of S. salar. The groups are: stressed (S+) or unstressed (S-); challenged with Yersinia ruckeri, exposed or not exposed to catecholamine (CAT+ or CAT- respectively); unchallenged with the bacterium (Ch-) and samples at day 0 (Control).

Richness estimators Diversity indexes

Factors Groups Min – Max (Mean) Min – Max (Mean)

Chao ACE Shannon Simpson

S+ 1 - 25 (5) 2 – 33 (8) 0 - 2.37 (0.9) 0 - 0.88 (0.44)

Stress S- 1 – 39 (7) 2 – 31 (10) 0 - 2.62 (0.89) 0 - 0.92 (0.41)

Control 1 – 15 (7) 4 – 22 (13) 0 - 2.5 (1.27) 0 - 0.91 (0.57)

CAT+ 1 – 26 (7) 2 – 33 (11) 0 - 2.62 (1.0) 0 - 0.92 (0.46)

CAT- 1 – 39 (7) 2 – 31 (9) 0 - 2.27 (0.97) 0 - 0.88 (0.46) Challenge Ch- 1 – 16 (4) 2 – 25 (7) 0 - 2.33 (0.74) 0 - 0.88 (0.37)

Control 1 – 15 (7) 4 – 22 (13) 0 - 2.5 (1.27) 0 - 0.91 (0.57)

Day 0 1 – 15 (7) 4 – 22 (13) 0 - 2.5 (1.27) 0 - 0.91(0.57) Sampling Day 10 1 – 39 (7) 2 – 33 (11) 0 - 2.37 (0.96) 0 - 0.88 (0.44) time Day 40 1 – 26 (5) 2 – 28 (7) 0 - 2.62 (0.83) 0 - 0.92 (0.41)

Midgut 1 – 39 (6) 2 – 31 (15) 0 - 2.62 (0.79) 0 - 0.92 (0.38) Gut regions Hindgut 1 – 25 (6) 2 – 33 (9) 0 - 2.58 (1.06) 0 - 0.92 (0.5)

118

4.3.3. Beta diversity analyses

The first two axes of principal coordinate analyses (PCoA) based on Bray - Curtis,

Jaccard, unweighted and weighted Unifrac distance matrices explained 37.2%, 32.2%,

50.5% and 70.4% of the variation in abundance of operational taxonomic units (OTUs) among different sample libraries. Using PERMANOVA for statistical analyses showed significant differences in variation in Bray – Curtis and Jaccard indices among experimental groups (p <0.05), except for gut region factor. No significant differences were present for PCoA based on Unifrac for all treatment groups (p > 0.05). The most significant dissimilarity in OTUs abundance was seen in samples from different sampling times (p < 0.01) (Figure 4.1).

A.

119

B.

C

120

D

E

121

F

G

Figure 4.1. Principal coordinate analysis plots based on (A) Bray-Curtis index, (B) Jaccard index, (C) unweighted UniFrac, (D) unweighted UniFrac distance methods for

122

different sampling time (day 0, day 10, day 40); (E), (F) and (G) Bray-Curtis index for challenge (unchallenged Ch-; challenged with Y. ruckeri exposed or not exposed to catecholamine, CAT+ and CAT- respectively; and control: unstressed fish at day 0), stress (unstressed S- or stressed S+; control: unstressed fish at day 0) and gut region factor (midgut and hindgut).

4.3.4. Taxonomic composition

The four main bacterial phyla in all samples were Proteobacteria (42%), Actinobacteria

(19%), Firmicutes (19%) and OD1 (2%) while unassigned bacteria (could not be assigned a Greengenes ID) accounted for 18%. Abundance at phylum level was most significantly in challenge group and different sampling time group (Figure 4.2).

Firmicutes showed significantly lower abundance in CAT- treatment (2.4%) compared to control (38.1%), unchallenged (18.2%) and CAT+ (30.5%) (p <0.05) (Figure 4.2A).

Similarly, there were significantly fewer Firmicutes in samples at day 40 (9.8%) in comparison with day 10 (24.5%) and day 0 (38.1%) (p<0.05). Proteobacteria dominated and had significantly higher abundance at day 40 (54.8%) than day 10 (30.4%) (p<0.05)

(Figure 4.2B). Stress and gut region had no significant effect on the abundance of bacteria (p>0.05).

123

A

B

124

C

D

Figure 4.2. Relative abundance at phylum level in challenge group (A), different sampling time group (B), stress group (C) and gut region group (D). The groups are: stressed (S+) or unstressed (S-); challenged with Yersinia ruckeri, exposed or not exposed to catecholamine (CAT+ or CAT- respectively); unchallenged with the bacterium (Ch-) and samples at day 0 (Control).

The top five bacterial classes detected were (21%), Bacilli (19%),

Actinobacteria (19%), Gammaproteobacteria (17%) and (4%) while others accounted for 21%. Alphaproteobacteria were significantly more abundant and

125

lower abundance of Bacilli in gut samples collected at day 40 (30.6% and 9.8%) than at day 10 (14.4% and 24.5%) and day 0 (0% and 38.1%) (p<0.05). Betaproteobacteria showed significantly higher abundance in samples at day 40 (7.7%) than day 10 (0.7%).

In challenge group, Bacilli and Gammaproteobacteria had significantly lower abundance in CAT- (2.4%) and CAT+ (0.9%), respectively, compared to other treatments (p<0.05).

At the family level, there was a relative high abundance of Propionibacteriaceae (19%),

Aeromonadaceae (16%), Sphingomonadaceae (14%), followed by Staphylococcaceae

(8%), Caulobacteraceae (6%) and Burkholderiaceae (4%). Other families and not assigned families accounted for 33% of the total. Abundance of families

Aeromonadaceae and Staphylococcaceae was more dependent on different challenge treatments while abundance of Sphingomonadaceae and Burkholderiaceae was related to sampling time. Aeromonadaceae was more common in CAT- (28.6%) and the control

(37.1%) than in CAT+ treatment (0.9%) (p<0.05). In contrast, Staphylococcaceae were significantly less abundant in CAT- (1.2%) compared to CAT+ (16.2%).

Sphingomonadaceae and Burkholderiaceae were significantly more abundant at day 40

(27.6% and 7.7%) than day 10 (3.8% and 0.7%) (p<0.05). Similarly, at genus level, significantly lower proportion of Staphylococcus in CAT- and a large proportion of

Sphingomonas and Burkholderia were observed on day 40 in comparison with CAT+ and day 10 (Figure 4.3, Figure 4.4).

126

Figure 4.3. Significantly greater abundance of of Sphingomonas and Burkholderia on day 40 compared to day 10 and day 0 (Univariate statistical comparisons p < 0.05).

The major genera were Propionibacterium (19%), Sphingomonas (14%), Aeromonas

(11%) and Staphylococcus (8%). Their relative abundances across experimental factors and individual samples are presented in Figure 4.4.

127

Figure 4.4. Abundance at genus level in different gut region (Gut), sampling time (Day), challenge group (Challenge) and stress group (Stress). The groups are: stressed (S+) or unstressed (S-); challenged with Yersinia ruckeri, exposed or not exposed to catecholamine (CAT+ or CAT- respectively); unchallenged with the bacterium (Ch-) and samples at day 0 (Control).

128

4.3.5. Gut microbiome comparisons between fish challenged with Yersinia ruckeri exposed (CAT+) and not exposed to the catecholamine epinephrine (CAT-)

4.3.5.1. Alpha and beta diversity

There were no differences between CAT+ and CAT- treatments in alpha diversity analysis (p>0.05), but a significant difference was shown in beta diversity (Figure 4.5).

Two treatments presented significantly distinct microbial community composition and structure (ANOSIM R=0.55625 and p < 0.014).

Figure 4.5. Beta diversity showed significantly different microbial community composition between fish challenged with Y. ruckeri previously exposed to epinephrine (CAT+) and no exposed (CAT-)

129

4.3.5.2. Taxonomic composition

In CAT+ compared to CAT- Fusobacteria (14.4 versus 0%) and Firmicutes (19.5 versus

0.02%) were more abundant and Actinobacteria was less abundant (1.36 versus 21.5%) at the phylum level, (p <0.05).

The top five of abundant classes in the challenged fish were Alphaproteobacteria (55%),

Gammaproteobacteria (14%), Actinobacteria (12%), Fusobacteria (7%) and Bacilli

(6%). Among these only Clostridia and Fusobacteria showed significantly more abundant in CAT+ compared to CAT- treatment.

At the family level, there was a higher abundance of , Alcanivoracaceae,

Shewanellaceae, Microbacteriaceae, Enterobacteriaceae, Hyphomonadaceae

(p<0.001), Aeromonadaceae, Micrococcaceae (p<0.01) and Chromatiaceae (p<0.05) in CAT- in comparison with CAT+ treatment. In contrast, CAT+ treatment showed significantly more abundant of Rhizobiaceae, Rhodobacteraceae (p<0.001),

Fusobacteriaceae, Ruminococcaceae, Peptostreptococcaceae, Moraxellaceae,

Actinomycetaceae (p<0.01), Pseudomonadaceae, Streptococcaceae,

Staphylococcaceae, Flavobacteriaceae and Oxalobacteraceae (p<0.05) compared to

CAT-.

At the genus level, the top five dominant genera in the gut microbial community in

CAT- treatment from hindgut samples at day 10 were: Caulobacter (25.1%),

130

Bradyrhizobium (23.8%), Algimonas (14.9%), Microbacterium (13.7%) and Yersinia

(8.6%) while Fusobacterium (14.4%), Streptococcus (12.9%), Bradyrhizobium

(12.9%), Rhizobium (12.5%) and Paracoccus (10.2%) had higher proportion in CAT+ treatment (Figure 4.6 and Figure 4.7). Additionally, there were 23 significant abundant differences at the genus level between these treatments (Table 4.2).

131

Figure 4.6. Relative abundance at the genus level of hindgut sampling at day 10 in fish challenged with Y. ruckeri exposed (CAT+) (bottom) or not exposed to the catecholamine epinephrine (CAT-) (top)

132

Figure 4.7. Heatmap abundance at genus level of individual fish (bottom) challenged with Y. ruckeri exposed (CAT+) and not exposed to the catecholamine epinephrine (CAT-)

133

Table 4.2. Significantly more abundant ASVs at the genus level of hindgut sampling at day 10 in fish challenged with Y. ruckeri exposed (CAT+) and not exposed to the catecholamine epinephrine (CAT-).

CAT- p value CAT+ p value

Marinicella 0.0003 Rhizobium 0.0059

Yersinia 0.0003 Fusobacterium 0.0075

Microbacterium 0.0004 Paracoccus 0.0075

Algimonas 0.0005 Faecalibacterium 0.0076

Enterobacter 0.0005 Flavobacterium 0.0086

Kocuria 0.0005 Acinetobacter 0.0103

Myroides 0.0007 Sinorhizobium 0.0112

Shewanella 0.0008 Romboutsia 0.0114

Serratia 0.0012 Trueperella 0.0114

Aeromonas 0.0065 Pseudomonas 0.0163

Pararheinheimera 0.0161 Staphylococcus 0.0276

Streptococcus 0.0336

134

There was a significantly higher abundance of Marinicella, Yersinia, Microbacterium,

Algimonas, Enterobacter, Kocuria, Myroides, Shewanella, Serratia, Aeromonas and

Pararheinheimera in CAT- treatment compared to CAT+. Whilst there was a significantly higher abundance of Rhizobium, Fusobacterium, Paracoccus,

Faecalibacterium, Flavobacterium, Acinetobacter, Sinorhizobium, Romboutsia,

Trueperella, Pseudomonas, Staphylococcus and Streptococcus in the CAT+ treatment compared to CAT-. Interestingly, at species level Y. ruckeri had a significantly higher abundance in the CAT- treatment compared to CAT+ (Figure 4.8).

Figure 4.8. Significantly greater abundance of Yersinia ruckeri in challenged fish without exposure to the catecholamine epinephrine (CAT-) compared to those exposed to epinephrine (CAT+).

135

4.4. DISCUSSION

The intestinal microbiota is important to the host as it can influence health, growth and disease status (Li et al., 2014; Llewellyn et al., 2014; Givens et al., 2015; Tran et al.,

2018). In this study the microbiome from midgut and hindgut of Atlantic salmon was examined in stressful and disease challenge conditions at different timepoints. The microbial composition was not affected by gut regions but was influenced by stress, challenge and sampling time.

The lack of difference in composition of gut microbiota in midgut and hindgut is consistent with reports of similar bacterial composition of microbiota in the digestive tract of grouper Epinephelus coioides (Yang et al., 2012), rainbow trout (Merrifield et al., 2009) and juvenile Atlantic salmon (Navarrete et al., 2009). The midgut and the hindgut have similar physiological conditions hence they harbour similar microbiota

(Yang et al., 2012). However, dissimilar bacterial compositions have been found in the different intestinal regions of many marine fish such as rabbit fish Siganus fuscescens

(Nielsen et al., 2017), Atlantic salmon (seawater) (Gajardo et al., 2016), juvenile E. coioides (Sun et al., 2011) and yellow grouper (Epinephelus awoara) (Zhou et al., 2009) which the effect of gut region on microbiome was investigated here. These differences may be due to species-specificity or experimental designs or processing methods

(Talwar et al., 2018).

136

Stress has been shown to affect the microbiota of numerous hosts. A model of chronic depression in mice resulted in changes in the gut microbiota or early life stress in rats was associated with intestinal permeability and bacterial translocation to the liver and spleen (Park et al., 2013; Moussaoui et al., 2014; Kelly et al., 2015). In fish, acute handling stress reduced the number of cultivable adherent bacteria in the proximal intestine of Arctic charr (Salvelinus alpinus) and in distal intestine of Atlantic salmon and rainbow trout but not in Atlantic cod (Gadus morhua) (Ringø et al., 2014).

Transportation stress increased 10-fold the number of total culturable skin-associated bacteria in trout (Tacchi et al., 2015). In another study stress associated with presence of a predator modified the gut microbiome of perch (Perca fluviatilis) (Zha et al., 2018).

Changes in mucus production and increased gut permeability could be the mechanisms of stress-induced alteration of host microbiota (Tacchi et al., 2015), (Kelly et al., 2015).

The primary response to stress is the release of stress hormones and the start of an endocrine response that has a close interaction with the host immune system which has previously been suggested to affect microbiota composition (Moloney et al., 2014). In addition, the gut itself has been shown to induce secretion of various enzymes depending on what stimulus it encounters, therefore it may change enzymatic profiles and influence the stability of microbiota communities (Bailey et al., 2010; Sullam et al.,

2015).

The gut microbiota plays important roles in protection against pathogen invasion, however the interactions between various components of the gut microbiota and pathogenic infections or diseases are still unclear. In this study, Y. ruckeri infection has

137

been shown to modulate the bacterial composition of the gut in Atlantic salmon. This finding is consistent with Nie et al. (2017) who reported that Vibrio anguillarum infection resulted in alteration of the gut microbiome of infected ayu (Plecoglossus altivelis). Modifications of gut bacterial composition have been observed in largemouth bronze gudgeon (Coreius guichenoti) suffering from furunculosis (caused by

Aeromonas salmonicida) (Li et al., 2016) as well as in enteritis diseased grass carp compared to healthy fish (Tran et al., 2018). However, in contrast to those studies, in the study reported here fish challenged with Y. ruckeri did not show any differences in alpha microbial diversity in comparison to unchallenged fish. This may be because only surviving fish were sampled and as all the fish had previously vaccinated against yersiniosis, they may not have had the disease. Higher alpha diversity is frequently detected in healthy fish compared to diseased fish due to the invading pathogen outcompeting the gut commensals, hence resulting in reduction of microbial diversity which is closely associated with diseased fish (Nie et al., 2017; Xiong et al., 2019). In addition, after attachment to the host intestine, pathogens can produce toxins to damage the intestinal lining and induce inflammatory response of the host which may increase production of reactive oxygen species (Xiong et al., 2019) which could also influence diversity. Changes in internal environment could favour specific bacteria overgrowth resulting in a shift in community composition (Winter and Bäumler, 2014; Xiong et al.,

2019). Interestingly, the gut microbial diversity of juvenile brown trout (Salmo trutta) was shown to increase with parasite (Tetracapsuloides bryosalmonae) load (Vasemägi et al., 2017).

138

Investigation of the gut microbial composition in Atlantic salmon in this study revealed three dominant phyla: Proteobacteria, Actinobacteria and Firmicutes. These three phyla have been recognised as the characteristic components of the gut microbiome of not only Atlantic salmon (Gajardo et al., 2016; Rudi et al., 2018; Gupta et al., 2019) but also for grouper Epinephelus fuscoguttatus, Epinephelus sexfasciatus and yellowtail scad Atule mate (Ingerslev et al., 2014; Hennersdorf et al., 2016), rainbow trout (Lyons et al., 2017) and gilthead seabream (Sparus aurata) (Estruch et al., 2015). Despite variations in relative abundance of these phyla in different studies, their presence in multiple fish species indicates that they are involved in important host functions such as food digestion, nutrient absorption and immune responses (Ghanbari et al., 2015). In this study, the relative abundance of Firmicutes significantly decreased with time while

Proteobacteria showed the opposite trend. This is consistent with the finding that abundance of Firmicutes was reduced during V. anguillarum infection especially at late stages (Nie et al., 2017). Reduction of Firmicutes and enrichment of Proteobacteria has also been reported and was attributed to the critical physiological changes in Atlantic salmon on transfer to a seawater environment (Lokesh et al., 2018) and also in salmon infected with Aeromonas salmonicida (Wang et al., 2018). Proteobacteria enrichment in the human and animal gut can be an indicator of gut microbiota imbalance and is associated with disease (Shin et al., 2015). In addition, taxa belonging to the phylum

Proteobacteria are involved in metabolic pathways such as carbon and nitrogen fixation and in the stress response regulatory system (Gupta et al., 2019), hence the longer duration of the experiment may lead to accumulation of stress to the fish and result in greater abundance of this phylum. Interestingly, the second most abundant phylum,

139

Actinobacteria, was found after the stress and Y. ruckeri challenge events (at days 10 and 40) but absent in the control group (day 0). In contrast, the phylum Actinobacteria was reported to be more abundant in healthy Atlantic salmon than unhealthy ones

(Wang et al., 2018).

The Proteobacteria found in the gut samples comprised mainly Alphaproteobacteria,

Gammaproteobacteria and Betaproteobacteria which are in broad agreement with a previous study on microflora of rainbow trout intestine. The higher significant abundance of Alphaproteobacteria and Betaproteobacteria and lower abundance of

Bacilli at the later sampling times in the study reported here may have resulted from the

Y. ruckeri infection and the stress event leading to physiological changes and host immune responses. These changes result in an imbalance in microbial community structure which can lead to the disruption in fermentation of complex carbohydrates and amino acids necessary to support anaerobic bacteria and therefore provide a growth advantage for facultative anaerobic bacteria (Proteobacteria) (Winter and Bäumler,

2014).

The genus Propionibacterium (family Propionibacteriaceae) was the most abundant genus found in this study (19%). This genus is commonly found in numerous freshwater and seawater fish species including Atlantic salmon (Godoy et al., 2015; Dehler et al.,

2017a, b), rainbow trout (Lyons et al., 2017), sea bass (Dicentrachus labrax) (Carda-

Diéguez et al., 2014; Nikouli et al., 2018), sea bream (Kormas et al., 2014; Estruch et al., 2015; Nikouli et al., 2018), zebra fish (Rurangwa et al., 2015) and coral Porites

140

lutea (Kuang et al., 2015). This bacterium possesses a fermentative metabolism and is responsible for the production of vitamin B12, hence the presence of this bacterium in the gut microbiota of fish may have a probiotic role (Lyons et al., 2017). Interestingly, this bacterium is the most prevalent in human skin, and hence there is a possibility given that the presence of this genus maybe represent contamination (Salter et al., 2014;

Mollerup et al., 2016; Neumann, 2017) though care was taken to avoid contamination.

However, the genus Propionibacterium has been previously been reportedly isolated from different studies in Atlantic salmon, and this bacterium is thought to be one of the core microorganisms in salmon gut (Godoy et al., 2015; Dehler et al., 2017a, b).

Aeromonas was the second most abundant genus in this study. This finding is in agreement with other studies which reported that Aeromonas was one of the most commonly found in the intestinal tract of trout (Salmo trutta trutta) (Skrodenytė-

Arbačiauskienė et al., 2008), rainbow trout (Heikkinen et al., 2006), Atlantic salmon

(Dehler et al., 2017a), turbot (Scophthalmus maximus) (Ringø et al., 1996) and zebra fish (Cantas et al., 2012). In particular, this genus was more abundant in diseased largemouth bronze gudgeon (Li et al., 2016), red-operculum crucian carp (Carassius auratus) (Li et al., 2017) and Saprolegnia infected Atlantic salmon eggs (Liu et al.,

2014) in comparison with the heathy ones. It was revealed that Aeromonas possesses high adhesive capability to colonise the surface of the intestinal mucosa, and possibly the intestine is a primary site for stress-induced infections (Namba et al., 2007).

However, in this study Aeromonadaceae was more common in control (37.1%) and challenged group CAT- (28.6%) than in CAT+ treatment (0.9%), suggesting

141

Aeromonadaceae are normal components of gut Atlantic salmon microbial communities. In addition, prior exposure of Y. ruckeri to the stress hormone (the CAT+ treatment) resulted in a dramatic decrease in the number of Aeromonadaceae in the fish gut. This may be because the challenge event in particular to CAT+ treatment has affected the gut bacterial balance and shifted to more abundant opportunistic bacteria and reduced the relative abundance of Aeromonadaceae.

The genus Staphylococcus is widespread and has been detected in fish eggs and fish digestive tracts (Ringø and Olsen, 1999). Staphylococcus epidermidis is a fish pathogen in both freshwater and seawater environments and has been found in: red seabream

(Chrysophrys major), yellow tail (Seriola quinqueradiata), grass carp, tilapia

(Oreochromis), sea bream and sea bass (Kubilay and Ulukoy, 2004) and in intestine of

Atlantic salmon (Ringø and Olsen, 1999). The highest relative abundance of this pathogenic bacterium was in CAT+ may be due to changes in the gut of infected fish challenged with more virulent Y. ruckeri (prior exposure to stress hormone).

Both genera Sphingomonas and Burkholderia were the most abundant when sampled at day 40. Sphingomonas has been shown to be present in intestinal samples of many aquatic animals such as rainbow trout (Heikkinen et al., 2006; Lyons et al., 2017), Asian seabass (Lates calcarifer) (Ali et al., 2016), gilthead seabream (Floris et al., 2013), abalone (Lee et al., 2016). Although Sphingomonas can exhibit remarkable biodegradative and biosynthetic capabilities by consuming a lot of organic compounds

(Balkwill et al., 2006), its functions in fish intestine are still unclear, possibly this genus aids in digestion of complex compounds (Ali et al., 2016). Interestingly, a very high

142

abundance of Sphingomonas (91.6%) was detected in skin mucus of brook charr

(Salvelinus fontinalis) based on its high competitiveness for nutrients (Boutin et al.,

2013). The domination of Sphingomonas seems to be the basis for the homeostasis of the skin microbiome and plays a role in protection of the host from pathogens (Boutin et al., 2013). Sphingomonas was also detected in skin and gut of Atlantic salmon

(Gajardo et al., 2016; Lokesh and Kiron, 2016; Zarkasi et al., 2016; Gajardo et al., 2017;

Zarkasi et al., 2017; Wang et al., 2018) and it was more abundant in healthy compared to unhealthy fish, therefore suggesting that this genus is the main component of a healthy Atlantic salmon gut microbiome (Wang et al., 2018). Thus, to explain why

Sphingomonas was significantly dominant in samples at day 40 further studies on the function of bacteria in the gut should be considered.

The genus Burkholderia has been detected as a component of the skin mucus of rainbow trout (Lowrey et al., 2015), skin and/or intestine of Atlantic salmon (Gajardo et al.,

2016; Lokesh et al., 2018; Wang et al., 2018). Members of this genus are usually known as mostly plant, human and terrestrial animal pathogens (Minniti et al., 2017). Although

Burkholderia genus has not been linked to pathogenesis in fish, the higher relative abundance of this genus after handling stress compared to control fish has been reported

(Minniti et al., 2017). In this study, the highest abundance of the genus Burkholderia was detected at day 40, therefore, the presence of this genus may indicate that the fish were stressed at the later stage of experiment.

143

Surprisingly, although fish were challenged with Y. ruckeri, Yersinia was not detected as a dominant genus. Ohtani et al. (2014) reported that Y. ruckeri were present in rainbow trout intestine at 30 minutes post infection and persisted for at least 6 hours, before mortality started at day 3 and peaked day 7, however in this study the gut samples were collected at later stages of the infection (at day 10 and day 40). In addition, the fish used in the challenge test were previously vaccinated to protect against yersiniosis, potentially reducing Y. ruckeri loads so they were not high enough to be dominant at these sampling times.

Clearer views of the effects of pre-treatment Y. ruckeri with catecholamine epinephrine in a challenge event were seen by reanalysing data by using different method for taxonomic assignment (One Codex). Generally, bacterial composition was shown more diversity by using One Codex for taxonomic identification compared to QIIME 2 with

Greengenes database. Additionally, One Codex also presented a lower proportion of not assigned bacteria than QIIME 2 (see Figure 4.7 and Figure 4.4). This is likely due to difference in size between the reference databases (Minot et al., 2015; Siegwald et al.,

2017) and the different working pipelines in taxonomic assignment. At the time of analysis the One Codex Targeted Loci database contained 247,647 records (31,633 unique species) while the Greengenes database used in this study was released in 2013 and a therefore has far fewer records. Moreover, ASVs taxonomically classified to the

Greengenes database clustered at 99% (QIIME 2) or best assignment using a highly sensitive algorithm (One Codex) can lead to different results. Clustering approaches start with an OTU clustering step by gathering reads into OTUs based on their sequence

144

similarities. Then a representative sequence for each cluster is extracted and aligned to each of the 16S rDNA sequences of a reference database before finally assigning to a taxonomic group (Siegwald et al., 2017). In contrast, the latter method does not group the reads, but first compares all of them to a reference database to assign the lowest taxonomy or a lower common ancestor for a group of sequences of the same taxonomy.

Different taxonomic units are then grouped based on the reads annotations (Siegwald et al., 2017). These approaches may result in mislabelling of OTUs due to the reference sequence selected for this taxon in clustering-first, however this method allows the discrimination of unclassified reads (Siegwald et al., 2017). In addition, One Codex has been reported as a more accurate and robust data platform for microbial identification compared to the popular software programs QIIME 1 and MOTHUR (Kopylova et al.,

2016). However, in this study no matter which method was employed, no significant difference in alpha diversity but significant changes in beta diversity were shown when the gut microbiome of challenged fish was compared between two treatments (CAT+ and CAT-).

The beta diversity analysis showed there was a distinct difference in bacterial communities of challenged fish from CAT+ and CAT- treatment groups. Generally, many pathogenic bacteria were identified in the gut microbiome of challenged fish including Fusobacterium, Flavobacterium, Streptococcus, Pseudomonas, Yersinia,

Serratia, Enterobacter. The presence of these bacteria suggests that Y. ruckeri challenge disturbed the gut microbial balance and favoured opportunistic pathogens (Li et al.,

2016). The results are consistent with the data from diseased grass carp study which had

145

a significantly higher abundance of Pseudomonas, Flavobacterium, Streptococcus (and

Caulobacter) compared to healthy fish (Tran et al., 2018) or higher number of Y. ruckeri after bath challenge in rainbow trout (Ingerslev et al., 2014). Interestingly, Y. ruckeri was classified as the most significant abundant in challenged fish without exposure to catecholamine (CAT-) by using One codex but not QIIME 2 for taxonomic identification. This finding is in line with the results from Li et al. (2016) who reported

Aeromonas was more abundant in largemouth bronze gudgeon suffering from furunculosis (caused by A. salmonicida). Detection of Y. ruckeri in the gut microbiome in CAT- treatment (Figure 4.7 and Figure 4.8) is completely consistent with the data from Y. ruckeri load in the hindgut samples at day 10 by using qPCR (chapter 3, table

3.4), this can confirm the sensitive and accuracy of One codex in taxonomic identification.

In conclusion, the stress and the challenge with Y. ruckeri significantly affected the gut microbiome of Atlantic salmon. Significant differences in microbiota profiles were also seen between different sampling times while gut regions showed similar compositions.

However, the mechanisms behind the gut microbiome changes are not clear, hence further studies on the functions of microbiome and their links to host responses should address these concerns.

146

Chapter 5. GENERAL DISCUSSION

Understanding host – pathogen - environment interactions is vital for management of infectious disease including treatment and prevention. This thesis focused on the effects of catecholamines on the virulence of Y. ruckeri and Atlantic salmon immunity and gut microbiome.

The presence of catecholamines including norepinephrine, epinephrine and dopamine enhanced the growth of Y. ruckeri in serum supplemented medium which might be explained by the contribution of catecholamine in facilitating iron supply directly for Y. ruckeri growth and/or indirectly supply by stimulation of siderophore production. In this study, siderophore production increased significantly in the presence of catecholamines which was consistent with Y. ruckeri growth enhancement. Other virulence factors such as caseinase, lipase and haemolysin activity and biofilm formation were also induced by catecholamines. Some of these are in agreement with previous studies investigating the effects of norepinephrine (100 µM) on virulence of

A. hydrophila (Gao et al., 2019). Although motility and haemolytic activity were not significantly different between norepinephrine treatment and control in agar functional assays, the motility and haemolysin gene expressions were significantly induced by NE

(Gao et al., 2019). However, another study on the effects of norepinephrine and dopamine on virulence of Y. ruckeri indicated that only biofilm and motility were increased (Torabi Delshad et al., 2019). This may be because only one concentration of catecholamines was used (100 µM) and experiments were conducted at a different temperature (28ºC) while in this study (see chapter 2) effects on all virulence functional

147

assays were investigated at 18ºC with 4 concentrations of catecholamines ranging from

25, 50, 100 and 200 µM.

It is well known that the optimum temperature for Y. ruckeri growth is 28 ºC but the disease outbreaks in fish occur at a lower temperature of around 18 ºC (Guijarro et al.,

2015a; Mendez et al., 2018). There is some evidence of enhancement of virulence at the lower temperature as different genes of Y. ruckeri involved in pathogenesis, for example those controlling siderophore production (rucC-rupDGC), haemolysin (YhlA), metalloprotease (yrp1) and type IV secretion system (traH-N), all had higher expression at 18ºC compared to 28ºC (Fernandez et al., 2004; Fernández et al., 2007a; Fernández et al., 2007b; Méndez et al., 2009; Méndez and Guijarro, 2013). Recently, a Tn5 – based transposon cassette containing the promoterless luxCDABE and lacZY operons, randomly inserted into the genome of Y. ruckeri caused interruption of gene and produced 168 different clones of Y. ruckeri which exhibited higher β-galactosidase activity at 18 ºC than at 28 ºC (Mendez et al., 2018). The interrupted genes included those involved in the synthesis of legionaminic (a component of the LPS structure), the yrp1 metalloprotease, environmental changes regulated genes (the diguanylate cyclase, a glycosyltransferase and a S-adenosyl methione-dependent methyltransferase) and genes induced under osmotic shock which have been previously implicated in virulence in other pathogens (Mendez et al., 2018; Wrobel et al., 2019).

Temperature has been shown to induce virulence changes of many bacteria and and is related to most of outbreaks of many other bacterial fish diseases are associated with

148

temperature changes (Mendez et al., 2018). Several genes linked to virulence have been shown to be more higher expression at lower temperatures; for example genes predicted to encode a two-component system sensor histidine kinase, an ATP-dependent RNA helicase, a multidrug ABC transporter permease/ATPase, an outer membrane protein/protective antigen OMA87, an M43 cytophagalysin zinc-dependent metalloprotease and a hypothetical protein in Flavobacterium psychrophilum were more upregulated at 8 ºC compared to 20 ºC (Hesami et al., 2011). Similarly, a delta subunit of RNA polymerase rpoE, a cold-responsive potABCD and three genes encoding autolytic enzymes were more upregulated in Lactococcus garvieae strain

Lg8331 isolated from rainbow trout suffering from lactococcosis at 18 ºC rather than

37 ºC, (Aguado-Urda et al., 2013). In A. hydrophila the expressions of polar and lateral flagellins genes were temperature dependent and were higher at 25 ºC compared to 37

ºC (Yu et al., 2007). Genes encoding for haemolysin, type 6 secretion system and exopolysaccharides synthesis and siderophore piscibactin have been shown to preferentially expressed at 15 ºC rather than 25 ºC in V. anguillarum cultured in iron limited medium (Lages et al., 2019). Expression of siderophore piscibactin genes in this bacterium was also reported higher at 18 compared to 25 ºC (Balado et al., 2018). As temperature can induce a lot of virulence genes or virulence factors in bacteria, studies on bacterial virulence should consider the effect of temperature on bacteria (in vitro) and on the host as well (in vivo) because fish are poikilothermic ectotherms.

Although catecholamines can influence the outcome of an infection as confirmed by in vivo tests in different hosts, there is limited information about the response of the host

149

to bacteria treated with catecholamine except for the survival rate. In this study, prior treatment of Y. ruckeri with catecholamine epinephrine caused significantly higher mortality compared to untreated Y. ruckeri challenged fish. Further, this was consistent with the change in the immune related gene expression in the blood of Atlantic salmon.

It is possible that epinephrine enhances virulence of Y. ruckeri, as shown by subsequent down regulation of IL-6 and IFNγ at the early stage of infection which may have led to a higher rate of mortality. However, in order to confirm this, more studies are required that investigate gene transcriptions of the pathogen and the host. Future research using the next generation sequencing (RNA-seq), comparisons of virulence gene expression in Y. ruckeri exposed and unexposed to epinephrine (or other catecholamines) and incubated at a range of temperatures (including 18 ºC and 28 ºC) will yield more information related to virulence or pathogenesis of this bacterium. For the host, transcriptomic and proteomic studies in infected Atlantic salmon would also offer a clearer picture of the host response to Y. ruckeri.

In aquaculture environments, stress of the farmed organisms cannot be avoided due to physical and physiological disturbances arising from transportation, crowding, handling and variations in the water quality (Eissa and Wang, 2016). Fish exhibit several stress responses through three regulatory systems: neural, endocrine and immune system to mitigate the stressors and maintain haemostasis (Tort, 2011; Eissa and Wang, 2016).

This study shows that although acute chasing fish and lowering water level did not influence the post challenge survival rate or plasma cortisol level, it did however induce the expression of two immune genes, IL 12 and CD8α, in blood at the early stage of

150

experiment (chapter 3) and altered gut microbiome (chapter 4). This indicates that stress had some effects on the fish, however it was not evident at the plasma cortisol level. As plasma cortisol was first measured at 12 hours post stress, either the stressor in this case was not sufficiently acute to trigger a cortisol response or the response was short lived and not detectable at this time point. Cortisol is a traditional stress biomarker, however in some circumstances it is difficult to interpret results because the stress response is controlled by both intrinsic and extrinsic factors (Pottinger, 2008; Eissa and Wang,

2016). To date most of studies have used stress related genes in combination with hormonal levels to more fully understand the stress response (Eissa and Wang, 2016).

Although there are several studies on the Atlantic salmon immune response, this is the first one observing the effects of Y. ruckeri (treated in vitro with epinephrine) on the fish immune response at the gene expression level. Exposure of Atlantic salmon fingerlings to Y. ruckeri at the concentration of 106 CFUmL-1 showed a significantly lower survival rates particularly in epinephrine treated Y. ruckeri. In addition, pre- treatment of the bacterium with this catecholamine resulted in expression of more immune genes in Atlantic salmon compared to salmon challenged with untreated Y ruckeri. Based on the significant downregulation of IFNγ and IL 6 in the fish challenged with the epinephrine treated bacterium (chapter 3), and the results in in vitro (chapter

2) it can be concluded that epinephrine enhanced the virulence of Y. ruckeri and the subsequent fish immune system responses to the more virulent challenge was altered in that some genes were downregulated. However, there was large individual variability

151

in the activation of some immune genes and some genes could not be amplified which indicate some limitations of the study.

Although it is interesting to investigate the blood immune gene response to Y. ruckeri infection as the ERM results in lots of haemorrhage in fish, the use of whole blood for quantification of gene expression is limited. Firstly, blood samples may contain PCR inhibitors, such as heme compounds, haemoglobin and lactoferrin, immunoglobin G, hematin and etc (Akane et al., 1994; Al-Soud et al., 2000; Al-Soud and Rådström, 2001;

Sidstedt et al., 2018), hence decreasing the quantification efficiency or quenching of florescence (Sidstedt et al., 2018). Although in this study the inhibition was minimised by RNA repurification and use of higher dilution template, it still was a challenge due to weak amplification of some genes. Secondly, red blood cells and white blood cells were not separated leading to an unequal proportion of these cells between samples and their differential functions should impact gene expression responses. Thus, observation of gene expression in red blood cells or white blood cells only combined similar analyses using spleen or head kidney samples should be included in future studies.

The individual variation may be due to the different levels of infection resulting from the challenge method and/or differences in genetic makeup of individuals (Ingerslev et al., 2009). In this study, although a bath infection model was used to mimic natural infection routes, it may have resulted in variation of infection pressure for individual fish. As Y. ruckeri has been reported to invade the host by different portals through the gills, intestines and skin surface on the lateral line and successful entry seems to be dependent on a breach of the epithelium (Ohtani et al., 2014; Guijarro et al., 2018a),

152

hence infection can be different for individuals. It is also important to note that all of fish in this study had been previously vaccinated against yersiniosis, and therefore each fish may have responded to Y. ruckeri differently due to different protective effects of vaccine. Immune gene responses following Y. ruckeri infection in vaccinated rainbow trout and V. anguillarum infection in vaccinated in zebrafish were weaker than in naïve fish (Harun et al., 2011; Zhang et al., 2013). As most of investigated immune genes (IL

11, TNFα, IL 2, IL 10, IL 22) in the spleen and gills of rainbow trout challenged with

Y. ruckeri were more highly expressed in naïve than vaccinated fish (Harun et al., 2011), only a smaller change in gene transcription levels of challenged fish compared to the control could be expected in this study due to the vaccination (chapter 3).

This is the first study showing that the composition of the gut microbiome can be shaped by a stress effect, Y. ruckeri challenges (with or without exposure to epinephrine) and sampling time (chapter 4). The core gut microbiome of Atlantic salmon fingerling in this study consisted of the phyla Proteobacteria, Actinobacteria and Firmicutes and the dominant genera were Propionibacterium, Sphingomonas, Aeromonas and

Staphylococcus. Reanalysing data from the hindgut of challenged fish (at day 10 post infection) with a different taxonomic identification method using One Codex provided a clearer view of what bacterial communities changes between the two treatments

(CAT+ and CAT-). The presence of many opportunistic pathogens in challenged fish

(chapter 4) may indicate the disruption of the balance of the gut microbiome caused by challenge event, and therefore showed the link between microbial imbalances and their host health status. Additionally, as the fish gut microbiota also contributes to the host

153

immune responses (see Xiong et al. (2019)), differential expression of immune genes between challenged fish with Y. ruckeri previously exposed or not exposed to epinephrine may not only occur due to the enhancement of virulence of this bacterium by epinephrine (chapter 3) but also due to the alteration in the gut bacterial composition.

However, further studies on gut microbiome and whole host genome responses under the same challenge model are needed to confirm this. In terms of presence and quantity of Y. ruckeri, the detection of the bacterium in the hindgut results were more consistent

(chapter 4) with quantifying bacterium load (using specific primers in qPCR - chapter

3) when the microbiome was analysed using One Codex compared to QIIME 2 (chapter

4).

Although this research described compositional changes in gut microbial communities of salmon after challenge with epinephrine treated and untreated Y. ruckeri in order to get an improved understanding about the representative bacteria for each treatment, detailed studies on bacterial functions should be conducted. Studies of the relationship between the gut microbiome and the host immune response will assist in developing a health programme that incorporates knowledge of the impact of stressors with the potentially beneficial aspects of pre and probiotics, nutrigenomics and vaccination to improve fish health and prevent disease.

Overall, this thesis has increased our knowledge of the interaction between virulence factors of Y. ruckeri and the Atlantic salmon immune response. Exposure of Y. ruckeri to catecholamines enhanced growth and virulence of this bacterium as shown in in vitro

154

functional assays. Enhancement of virulence was further confirmed after an in vivo infection experiment showed higher mortality and impacted immune gene expression in Atlantic salmon exposed to catecholamine-treated Y. ruckeri. Although an acute stress event did not induce a change in cortisol levels, it did result in differential expression of immune genes. Finally, the gut microbiome of Atlantic salmon did not vary between midgut and hindgut, but the composition was influenced by stress, Y. ruckeri challenge and time post challenge. Although this research has limitations, it undoubtedly provides a greater understanding about the interaction effects between the host and pathogen. Furthermore, apart from consolidating the accepted view that stress negatively impacts fish health it highlights the possibility of manipulating the host microbiome to potentially overcome or minimise the consequences of stressors on fish health that are common to aquaculture.

155

REFERENCE

Abbas, A.K., Lichtman, A.H., Pillai, S., 2014. Cellular and molecular immunology E- book. Elsevier Health Sciences.

Aguado-Urda, M., Gibello, A., del Mar Blanco, M., Fernández-Garayzábal, J.F., López- Alonso, V., López-Campos, G.H., 2013. Global transcriptome analysis of Lactococcus garvieae strains in response to temperature. PloS one 8, e79692.

Akane, A., Matsubara, K., Nakamura, H., Takahashi, S., Kimura, K., 1994. Identification of the heme compound copurified with deoxyribonucleic acid (DNA) from bloodstains, a major inhibitor of polymerase chain reaction (PCR) amplification. Journal of Forensic Science 39, 362-372.

Akhlaghi, M., Sharifi Yazdi, H., 2008. Detection and identification of virulent Yersinia ruckeri: the causative agent of enteric redmouth disease in rainbow trout (Oncorhynchus mykiss) cultured in Fars province, Iran. Iranian Journal of Veterinary Research 9, 347-352.

Al-Soud, W.A., Jönsson, L.J., Rådström, P., 2000. Identification and characterization of immunoglobulin G in blood as a major inhibitor of diagnostic PCR. Journal of clinical microbiology 38, 345-350.

Al-Soud, W.A., Rådström, P., 2001. Purification and characterization of PCR- inhibitory components in blood cells. Journal of clinical microbiology 39, 485- 493.

Ali, S.S.R., Ambasankar, K., Nandakumar, S., Praveena, P.E., Syamadayal, J., 2016. Effect of dietary prebiotic inulin on growth, body composition and gut microbiota of Asian seabass (Lates calcarifer). Animal Feed Science and Technology 217, 87-94.

Altinok, I., Ozturk, R.C., Kahraman, U.C., Capkin, E., 2016. Protection of rainbow trout against yersiniosis by lpxD mutant Yersinia ruckeri. Fish & shellfish immunology 55, 21-27.

156

Alverdy, J., Holbrook, C., Rocha, F., Seiden, L., Licheng, R., 2000. Gut-derived sepsis occurs when the right pathogen with the right virulence genes meets the right host: evidence for in vivo virulence expression in Pseudomonas aeruginosa. Annals of surgery 232, 480.

Anderson, M.T., Armstrong, S.K., 2006. The Bordetella bfe system: growth and transcriptional response to siderophores, catechols, and neuroendocrine catecholamines. Journal of bacteriology 188, 5731-5740.

Arkoosh, M.R., Clemons, E., Kagley, A.N., Stafford, C., Glass, A., Jacobson, K., Reno, P., Myers, M.S., Casillas, E., Loge, F., 2004. Survey of pathogens in juvenile salmon Oncorhynchus spp. migrating through Pacific Northwest estuaries. Journal of Aquatic Animal Health 16, 186-196.

Austin, B., Green, M., Rodgers, C., 1982. Morphological diversity among strains of Yersinia ruckeri. Aquaculture 27, 73-78.

Bailey, M.T., Dowd, S.E., Parry, N.M., Galley, J.D., Schauer, D.B., Lyte, M., 2010. Stressor exposure disrupts commensal microbial populations in the intestines and leads to increased colonization by Citrobacter rodentium. Infection and immunity 78, 1509-1519.

Bailey, M.T., Engler, H., Sheridan, J.F., 2006. Stress induces the translocation of cutaneous and gastrointestinal microflora to secondary lymphoid organs of C57BL/6 mice. Journal of neuroimmunology 171, 29-37.

Balado, M., Lages, M.A., Fuentes-Monteverde, J.C., Martínez-Matamoros, D., Rodríguez, J., Jiménez, C., Lemos, M.L., 2018. The Siderophore Piscibactin Is a Relevant Virulence Factor for Vibrio anguillarum Favored at Low Temperatures. Frontiers in Microbiology 9.

Balkwill, D.L., Fredrickson, J.K., Romine, M.F., 2006. Sphingomonas and related genera. The Prokaryotes: Volume 7: Proteobacteria: Delta, Epsilon Subclass, 605-629.

Barnes, A.C., 2011. 13 Enteric red mouth disease (ERM)(Yersinia ruckeri). Fish Diseases and Disorders Vol. 3: Viral, Bacterial and Fungal Infections, 484-511.

157

Barnes, A.C., Delamare-Deboutteville, J., Gudkovs, N., Brosnahan, C., Morrison, R., Carson, J., 2016. Whole genome analysis of Yersinia ruckeri isolated over 27 years in Australia and New Zealand reveals geographical endemism over multiple lineages and recent evolution under host selection. Microbial genomics 2, e000095-e000095.

Bates, J.M., Mittge, E., Kuhlman, J., Baden, K.N., Cheesman, S.E., Guillemin, K., 2006. Distinct signals from the microbiota promote different aspects of zebrafish gut differentiation. Developmental biology 297, 374-386.

Belay, T., Aviles, H., Vance, M., Fountain, K., Sonnenfeld, G., 2003. Catecholamines and in vitro growth of pathogenic bacteria: enhancement of growth varies greatly among bacterial species. Life Sciences 73, 1527-1535.

Belay, T., Sonnenfeld, G., 2002. Differential effects of catecholamines on in vitro growth of pathogenic bacteria. Life sciences 71, 447-456.

Berc, A., Kozariæ, Z., Petrinec, Z., 1999. Yersinia ruckeri septicaemia in experimentally infected carp (Cyprinus carpio L.) fingerlings. Acta Veterinaria Hungarica 47, 161-172.

Bogdan, C., Mattner, J., Schleicher, U., 2004. The role of type I interferons in non‐viral infections. Immunological reviews 202, 33-48.

Bolnick, D.I., Snowberg, L.K., Hirsch, P.E., Lauber, C.L., Org, E., Parks, B., Lusis, A.J., Knight, R., Caporaso, J.G., Svanbäck, R., 2014. Individual diet has sex- dependent effects on vertebrate gut microbiota. Nature communications 5, 4500.

Bolyen, E., Rideout, J.R., Dillon, M.R., Bokulich, N.A., Abnet, C.C., Al-Ghalith, G.A., Alexander, H., Alm, E.J., Arumugam, M., Asnicar, F., 2019. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nature biotechnology 37, 852-857.

Boutin, S., Audet, C., Derome, N., 2013. Probiotic treatment by indigenous bacteria decreases mortality without disturbing the natural microbiota of Salvelinus fontinalis. Canadian journal of microbiology 59, 662-670.

158

Bowers, J., Mustafa, A., Speare, D.J., Conboy, G.A., Brimacombe, M., Sims, D.E., Burka, J.F., 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, 165-172.

Braden, L.M., Rasmussen, K.J., Purcell, S.L., Ellis, L., Mahony, A., Cho, S., Whyte, S.K., Jones, S.R., Fast, M.D., 2018. Acquired protective immunity in Atlantic salmon Salmo salar against the myxozoan Kudoa thyrsites involves induction of MHIIβ+ CD83+ antigen-presenting cells. Infection and immunity 86, e00556- 00517.

Bragg, R., Henton, M., 1986. Isolation of Yersinia ruckeri from rainbow trout in South Africa. Bulletin of the European Association of Fish Pathologists 6, 5-6.

Bravo, S., Kojagura, V., 2004. First isolation of Yersinia ruckeri from rainbow trout (Oncorhynchus mykiss) in Peru. BULLETIN-EUROPEAN ASSOCIATION OF FISH PATHOLOGISTS 24, 104-108.

Bridle, A., Nosworthy, E., Polinski, M., Nowak, B., 2011. Evidence of an antimicrobial- immunomodulatory role of Atlantic salmon cathelicidins during infection with Yersinia ruckeri. PloS one 6, e23417.

Bridle, A.R., Koop, B.F., Nowak, B.F., 2012. Identification of surrogates of protection against yersiniosis in immersion vaccinated Atlantic salmon. PLoS One 7, e40841.

Bullock, G., Stuckey, H., Shotts, E., 1978. Enteric redmouth bacterium: comparison of isolates from different geographic areas. Journal of Fish Diseases 1, 351-356.

Bullock, G., Stuckey, H., Shotts Jr, E., 1977. Early records of North American and Australian outbreaks of enteric redmouth disease. Fish Health News 6, 96-97.

Burtis, C.A., 2006. Hormones. Tietz Textbook of Clinial Chemistry and Molecular Diagnostics, 5/e. Elsevier India, 837.

Butler, P., Metcalfe, J., Ginley, S., 1986. Plasma catecholamines in the lesser spotted dogfish and rainbow trout at rest and during different levels of exercise. Journal of Experimental Biology 123, 409-421.

159

Butt, R.L., Volkoff, H., 2019. Gut Microbiota and Energy Homeostasis in Fish. Frontiers in Endocrinology 10.

Caipang, C.M.A., Brinchmann, M.F., Berg, I., Iversen, M., Eliassen, R., Kiron, V., 2008. Changes in selected stress and immune‐related genes in Atlantic cod, Gadus morhua, following overcrowding. Aquaculture research 39, 1533-1540.

Calvez, S., Gantelet, H., Blanc, G., Douet, D.-G., Daniel, P., 2014. Yersinia ruckeri biotypes 1 and 2 in France: presence and antibiotic susceptibility. Diseases of aquatic organisms 109, 117-126.

Cantas, L., Sørby, J.R.T., Aleström, P., Sørum, H., 2012. Culturable gut microbiota diversity in zebrafish. Zebrafish 9, 26-37.

Carda-Diéguez, M., Mira, A., Fouz, B., 2014. Pyrosequencing survey of intestinal microbiota diversity in cultured sea bass (Dicentrarchus labrax) fed functional diets. FEMS microbiology ecology 87, 451-459.

Cárdenas, A., Rodriguez-R, L.M., Pizarro, V., Cadavid, L.F., Arévalo-Ferro, C., 2011. Shifts in bacterial communities of two caribbean reef-building coral species affected by white plague disease. The Isme Journal 6, 502.

Carey, J.B., McCormick, S.D., 1998. Atlantic salmon smolts are more responsive to an acute handling and confinement stress than parr. Aquaculture 168, 237-253.

Casadevall, A., Pirofski, L.a., 2001. Host‐Pathogen Interactions: The Attributes of Virulence. Journal of Infectious Diseases 184, 337-344.

Chaves-Pozo, E., Pelegrín, P., García-Castillo, J., García-Ayala, A., Mulero, V., Meseguer, J., 2004. Acidophilic granulocytes of the marine fish gilthead seabream (Sparus aurata L.) produce interleukin-1β following infection with Vibrio anguillarum. Cell and tissue research 316, 189-195.

Chen, C., Brown, D.R., Xie, Y., Green, B.T., Lyte, M., 2003. Catecholamines Modulate Escherichia coli O157:H7 Adherence to Murine Cecal Mucosa. Shock 20, 183- 188.

160

Cheng, C.-H., Yang, F.-F., Ling, R.-Z., Liao, S.-A., Miao, Y.-T., Ye, C.-X., Wang, A.- L., 2015. Effects of ammonia exposure on apoptosis, oxidative stress and immune response in pufferfish (Takifugu obscurus). Aquatic Toxicology 164, 61-71.

Cogan, T.A., Thomas, A.O., Rees, L.E., Taylor, A.H., Jepson, M.A., Williams, P.H., Ketley, J., Humphrey, T.J., 2007. Norepinephrine increases the pathogenic potential of Campylobacter jejuni. Gut 56, 1060-1065.

Coquet, L., Cosette, P., Dé, E., Galas, L., Vaudry, H., Rihouey, C., Lerouge, P., Junter, G.-A., Jouenne, T., 2005. Immobilization Induces Alterations in the Outer Membrane Protein Pattern of Yersinia r uckeri. Journal of proteome research 4, 1988-1998.

Coquet, L., Cosette, P., Junter, G.-A., Beucher, E., Saiter, J.-M., Jouenne, T., 2002. Adhesion of Yersinia ruckeri to fish farm materials: influence of cell and material surface properties. Colloids and surfaces B: Biointerfaces 26, 373-378.

Cortés, Y., Hormazábal, E., Leal, H., Urzúa, A., Mutis, A., Parra, L., Quiroz, A., 2014. Novel antimicrobial activity of a dichloromethane extract obtained from red seaweed Ceramium rubrum (Hudson)(Rhodophyta: Florideophyceae) against Yersinia ruckeri and Saprolegnia parasitica, agents that cause diseases in salmonids. Electronic Journal of Biotechnology 17, 126-131.

Costa, A.A., Leef, M.J., Bridle, A.R., Carson, J., Nowak, B.F., 2011. Effect of vaccination against yersiniosis on the relative percent survival, bactericidal and lysozyme response of Atlantic salmon, Salmo salar. Aquaculture 315, 201-206.

Coulanges, V., Andre, P., Vidon, D.J.-M., 1998. Effect of Siderophores, Catecholamines, and Catechol Compounds onListeriaspp. Growth in Iron- Complexed Medium. Biochemical and biophysical research communications 249, 526-530.

Coulanges, V., Andre, P., Ziegler, O., Buchheit, L., Vidon, D., 1997. Utilization of iron- catecholamine complexes involving ferric reductase activity in Listeria monocytogenes. Infection and immunity 65, 2778-2785.

161

Dahiya, I., Stevenson, R., 2010a. The UvrY response regulator of the BarA–UvrY two- component system contributes to Yersinia ruckeri infection of rainbow trout (Oncorhynchus mykiss). Archives of microbiology 192, 541-547.

Dahiya, I., Stevenson, R., 2010b. The ZnuABC operon is important for Yersinia ruckeri infections of rainbow trout, Oncorhynchus mykiss (Walbaum). Journal of fish diseases 33, 331-340.

Dahle, M.K., Wessel, Ø., Timmerhaus, G., Nyman, I.B., Jørgensen, S.M., Rimstad, E., Krasnov, A., 2015. Transcriptome analyses of Atlantic salmon (Salmo salar L.) erythrocytes infected with piscine orthoreovirus (PRV). Fish & shellfish immunology 45, 780-790.

Danley, M.L., Goodwin, A.E., Killian, H.S., 1999. Epizootics in farm‐raised channel catfish, Ictalurus punctatus (Rafinesque), caused by the enteric redmouth bacterium Yersinia ruckeri. Journal of Fish Diseases 22, 451-456.

Davies, R.L., 1991. Outer membrane protein profiles of Yersinia ruckeri. Veterinary Microbiology 26, 125-140.

De Grandis, S.A., Stevenson, R., 1985. Antimicrobial susceptibility patterns and R plasmid-mediated resistance of the fish pathogen Yersinia ruckeri. Antimicrobial agents and chemotherapy 27, 938-942.

Defoirdt, T., 2013. Virulence mechanisms of bacterial aquaculture pathogens and antivirulence therapy for aquaculture. Reviews in Aquaculture, n/a-n/a.

Dehler, C.E., Secombes, C.J., Martin, S.A., 2017a. Environmental and physiological factors shape the gut microbiota of Atlantic salmon parr (Salmo salar L.). Aquaculture 467, 149-157.

Dehler, C.E., Secombes, C.J., Martin, S.A., 2017b. Seawater transfer alters the intestinal microbiota profiles of Atlantic salmon (Salmo salar L.). Scientific reports 7, 13877.

Deitch, E., Fletcher, G., Petersen, L., Costa, I., Shears, M., Driedzic, W., Gamperl, A., 2006. Cardiorespiratory modifications, and limitations, in post-smolt growth

162

hormone transgenic Atlantic salmon Salmo salar. Journal of Experimental Biology 209, 1310-1325.

Dhabhar, F.S., 2008. Enhancing versus suppressive effects of stress on immune function: implications for immunoprotection versus immunopathology. Allergy, Asthma & Clinical Immunology 4, 2.

Dhariwal, A., Chong, J., Habib, S., King, I.L., Agellon, L.B., Xia, J., 2017. MicrobiomeAnalyst: a web-based tool for comprehensive statistical, visual and meta-analysis of microbiome data. Nucleic acids research 45, W180-W188.

Dimitroglou, A., Merrifield, D.L., Carnevali, O., Picchietti, S., Avella, M., Daniels, C., Güroy, D., Davies, S.J., 2011. Microbial manipulations to improve fish health and production–a Mediterranean perspective. Fish & shellfish immunology 30, 1-16.

Doménech, A., Fernández-Garayzábal, J., Lawson, P., García, J., Cutuli, M., Blanco, M., Gibello, A., Moreno, M., Collins, M., Domínguez, L., 1997. Winter disease outbreak in sea-bream (Sparus aurata) associated with Pseudomonas anguilliseptica infection. Aquaculture 156, 317-326.

Donlan, R.M., 2001. Biofilm formation: a clinically relevant microbiological process. Clinical Infectious Diseases 33, 1387-1392.

Dowd, S.E., 2007. Escherichia coli O157:H7 gene expression in the presence of catecholamine norepinephrine. FEMS Microbiol Lett 273, 214-223.

Dunn, L.L., Rahmanto, Y.S., Richardson, D.R., 2007. Iron uptake and metabolism in the new millennium. Trends in cell biology 17, 93-100.

Dwilow, A., Souter, B., Knight, K., 1987. Isolation of Yersinia ruckeri from burbot. Lota lota (L.), from the Mackenzie River, Canada. Journal of Fish Diseases 10, 315-317.

Egerton, S., Culloty, S., Whooley, J., Stanton, C., Ross, R.P., 2018. The Gut Microbiota of Marine Fish. Frontiers in Microbiology 9.

163

Einarsdóttir, I.E., Nilssen, K.J., 1996. Stress responses of Atlantic salmon (Salmo salar L.) elicited by water level reduction in rearing tanks. Fish Physiology and Biochemistry 15, 395-400.

Eissa, A., Moustafa, M., Abdelaziz, M., Ezzeldeen, N., 2008. Yersinia ruckeri infection in cultured Nile tilapia, Oreochromis niloticus, at a semi-intensive fish farm in lower Egypt. African Journal of Aquatic Science 33, 283-286.

Eissa, N., Wang, H.P., 2016. Transcriptional stress responses to environmental and husbandry stressors in aquaculture species. Reviews in Aquaculture 8, 61-88.

Enriquez, R., Zamora, J., 1987. Aislamiento de Yersinia ruckeri de carpas (Cyprinus carpio) en Valdivia. Arch. Med. Vet 19, 33-36.

Estruch, G., Collado, M., Peñaranda, D., Vidal, A.T., Cerdá, M.J., Martínez, G.P., Martinez-Llorens, S., 2015. Impact of fishmeal replacement in diets for gilthead sea bream (Sparus aurata) on the gastrointestinal microbiota determined by pyrosequencing the 16S rRNA gene. PLoS One 10, e0136389.

Ewart, K.V., Williams, J., Richards, R.C., Gallant, J.W., Melville, K., Douglas, S.E., 2008. The early response of Atlantic salmon (Salmo salar) macrophages exposed in vitro to Aeromonas salmonicida cultured in broth and in fish. Developmental & Comparative Immunology 32, 380-390.

FAO, 2004-2019. Cultured Aquatic Species Information Programme. Salmo salar. Cultured Aquatic Species Information Programme. FAO Fisheries and Aquaculture Department [online], Rome.

Faraldo-Gómez, J.D., Sansom, M.S., 2003. Acquisition of siderophores in gram- negative bacteria. Nature Reviews Molecular Cell Biology 4, 105-116.

Fast, M., Muise, D., Easy, R., Ross, N., Johnson, S., 2006. The effects of Lepeophtheirus salmonis infections on the stress response and immunological status of Atlantic salmon (Salmo salar). Fish & Shellfish Immunology 21, 228- 241.

164

Fast, M.D., Hosoya, S., Johnson, S.C., Afonso, L.O., 2008. Cortisol response and immune-related effects of Atlantic salmon (Salmo salar Linnaeus) subjected to short-and long-term stress. Fish & shellfish immunology 24, 194-204.

Fernandez, L., Lopez, J., Secades, P., Menendez, A., Marquez, I., Guijarro, J., 2003. In vitro and in vivo studies of the Yrp1 protease from Yersinia ruckeri and its role in protective immunity against enteric red mouth disease of salmonids. Applied and environmental microbiology 69, 7328-7335.

Fernandez, L., Marquez, I., Guijarro, J., 2004. Identification of specific in vivo-induced (ivi) genes in Yersinia ruckeri and analysis of ruckerbactin, a catecholate siderophore iron acquisition system. Applied and environmental microbiology 70, 5199-5207.

Fernández, L., Méndez, J., Guijarro, J.A., 2007a. Molecular virulence mechanisms of the fish pathogen Yersinia ruckeri. Veterinary Microbiology 125, 1-10.

Fernández, L., Prieto, M., Guijarro, J.A., 2007b. The iron- and temperature-regulated haemolysin YhlA is a virulence factor of Yersinia ruckeri. Microbiology 153, 483-489.

Fernández, L., Secades, P., Lopez, J., Márquez, I., Guijarro, J., 2002. Isolation and analysis of a protease gene with an ABC transport system in the fish pathogen Yersinia ruckeri: insertional mutagenesis and involvement in virulencea. Microbiology 148, 2233-2243.

Floris, R., Manca, S., Fois, N., 2013. Microbial ecology of intestinal tract of gilthead sea bream (Sparus aurata Linnaeus, 1758) from two coastal lagoons of Sardinia (Italy). Transitional Waters Bulletin 7, 4-12.

Flysand, R., Ask, J., Serck‐Hanssen, G., Helle, K., 1992. Plasma catecholamines and accumulation of adrenaline in the atrial cardiac tissue of aquacultured Atlantic salmon (Salmo salar) during stress. Journal of Fish Biology 41, 103-111.

Fouz, B., Zarza, C., Amaro, C., 2006. First description of non‐motile Yersinia ruckeri serovar I strains causing disease in rainbow trout, Oncorhynchus mykiss (Walbaum), cultured in Spain. Journal of fish diseases 29, 339-346.

165

Freestone, P., Lyte, M., 2010. Stress and microbial endocrinology: prospects for ruminant nutrition. Animal: an International Journal of Animal Bioscience 4, 1248.

Freestone, P.P., Haigh, R.D., Lyte, M., 2008a. Catecholamine inotrope resuscitation of antibiotic-damaged staphylococci and its blockade by specific receptor antagonists. The Journal of infectious diseases 197, 1044-1052.

Freestone, P.P., Haigh, R.D., Williams, P.H., Lyte, M., 1999. Stimulation of bacterial growth by heat‐stable, norepinephrine‐induced autoinducers. FEMS microbiology letters 172, 53-60.

Freestone, P.P., Haigh, R.D., Williams, P.H., Lyte, M., 2003. Involvement of enterobactin in norepinephrine‐mediated iron supply from transferrin to enterohaemorrhagic Escherichia coli. FEMS microbiology letters 222, 39-43.

Freestone, P.P., Lyte, M., 2008. Microbial endocrinology: experimental design issues in the study of interkingdom signalling in infectious disease. Advances in applied microbiology 64, 75-105.

Freestone, P.P., Sandrini, S.M., Haigh, R.D., Lyte, M., 2008b. Microbial endocrinology: how stress influences susceptibility to infection. Trends Microbiol 16, 55-64.

Fridell, F., Gadan, K., Sundh, H., Taranger, G., Glette, J., Olsen, R., Sundell, K., Evensen, Ø., 2007. Effect of hyperoxygenation and low water flow on the primary stress response and susceptibility of Atlantic salmon Salmo salar L. to experimental challenge with IPN virus. Aquaculture 270, 23-35.

Fuhrmann, H., Böhm, K., Schlotfeldt, H., 1984. On the importance of enteric bacteria in the bacteriology of freshwater fish. Bulletin of the European Association of Fish Pathologists 4, 42-46.

Furones, M.D., Rodgers, C.J., Munn, C.B., 1993. Yersinia ruckeri, the causal agent of enteric redmouth disease (ERM) in fish. Annual Review of Fish Diseases 3, 105- 125.

166

Gajardo, K., Jaramillo-Torres, A., Kortner, T.M., Merrifield, D.L., Tinsley, J., Bakke, A.M., Krogdahl, Å., 2017. Alternative protein sources in the diet modulate microbiota and functionality in the distal intestine of Atlantic salmon (Salmo salar). Appl. Environ. Microbiol. 83, e02615-02616.

Gajardo, K., Rodiles, A., Kortner, T.M., Krogdahl, Å., Bakke, A.M., Merrifield, D.L., Sørum, H., 2016. A high-resolution map of the gut microbiota in Atlantic salmon (Salmo salar): A basis for comparative gut microbial research. Scientific Reports 6, 30893.

Ganguly, S., Prasad, A., 2012. Microflora in fish digestive tract plays significant role in digestion and metabolism. Reviews in fish biology and fisheries 22, 11-16.

Gao, J., Xi, B., Chen, K., Song, R., Qin, T., Xie, J., Pan, L., 2019. The stress hormone norepinephrine increases the growth and virulence of Aeromonas hydrophila. MicrobiologyOpen 8, e00664.

Gelev, I., Khvoĭnev, A., Dimitrov, K., Strashimirova, N., 1983. [Hemorrhagic septicemia in rainbow trout due to Yersinia ruckeri sp. nov.]. Veterinarno- meditsinski nauki 21, 84-90.

Ghanbari, M., Kneifel, W., Domig, K.J., 2015. A new view of the fish gut microbiome: advances from next-generation sequencing. Aquaculture 448, 464-475.

Ghosh, B., Crosbie, P.B., Nowak, B.F., Bridle, A.R., 2018. A highly sensitive, non‐ invasive qPCR‐based strategy for direct quantification of Yersinia ruckeri in fish faeces. Journal of fish diseases 41, 1421-1428.

Ghosh, B., Nguyen, T.D., Crosbie, P.B.B., Nowak, B.F., Bridle, A.R., 2016. Oral vaccination of first-feeding Atlantic salmon, Salmo salar L., confers greater protection against yersiniosis than immersion vaccination. Vaccine 34, 599-608.

Givens, C.E., Ransom, B., Bano, N., Hollibaugh, J.T., 2015. Comparison of the gut microbiomes of 12 bony fish and 3 shark species. Marine Ecology Progress Series 518, 209-223.

Godoy, F.A., Miranda, C.D., Wittwer, G.D., Aranda, C.P., Calderón, R., 2015. High variability of levels of Aliivibrio and lactic acid bacteria in the intestinal

167

microbiota of farmed Atlantic salmon Salmo salar L. Annals of microbiology 65, 2343-2353.

Green, B.T., Lyte, M., Chen, C., Xie, Y., Casey, M.A., Kulkarni-Narla, A., Vulchanova, L., Brown, D.R., 2004. Adrenergic modulation of Escherichia coli O157: H7 adherence to the colonic mucosa. American Journal of Physiology- Gastrointestinal and Liver Physiology 287, G1238-G1246.

Guijarro, J.A., Cascales, D., García-Torrico, A.I., García-Domínguez, M., Méndez, J., 2015a. Temperature-dependent expression of virulence genes in fish-pathogenic bacteria. Frontiers in Microbiology 6.

Guijarro, J.A., Cascales, D., García-Torrico, A.I., García-Domínguez, M., Méndez, J., 2015b. Temperature-dependent expression of virulence genes in fish-pathogenic bacteria. Frontiers in Microbiology 6, 700.

Guijarro, J.A., García-Torrico, A.I., Cascales, D., Méndez, J., 2018a. The infection process of Yersinia ruckeri: reviewing the pieces of the jigsaw puzzle. Frontiers in cellular and infection microbiology 8, 218.

Guijarro, J.A., García-Torrico, A.I., Cascales, D., Méndez, J., 2018b. The Infection Process of Yersinia ruckeri: Reviewing the Pieces of the Jigsaw Puzzle. Frontiers in Cellular and Infection Microbiology 8.

Gulla, S., Barnes, A.C., Welch, T.J., Romalde, J.L., Ryder, D., Ormsby, M.J., Carson, J., Lagesen, K., Verner-Jeffreys, D.W., Davies, R.L., 2018. Multilocus variable - number tandem-repeat analysis of Yersinia ruckeri confirms the existence of host specificity, geographic endemism, and anthropogenic dissemination of virulent clones. Appl. Environ. Microbiol. 84, e00730-00718.

Gunasena, D.K., Komrower, J.R., Macintyre, S., 2003. The fish pathogen Yersinia ruckeri possesses a TTS system. The Genus Yersinia. Springer, 105-107.

Gupta, S., Fečkaninová, A., Lokesh, J., Koščová, J., Sørensen, M., Fernandes, J., Kiron, V., 2019. Lactobacillus Dominate in the Intestine of Atlantic Salmon Fed Dietary Probiotics. Frontiers in Microbiology 9.

168

Harun, N.O., Wang, T., Secombes, C.J., 2011. Gene expression profiling in naïve and vaccinated rainbow trout after Yersinia ruckeri infection: insights into the mechanisms of protection seen in vaccinated fish. Vaccine 29, 4388-4399.

Hedrick, R., 1998. Relationships of the host, pathogen, and environment: implications for diseases of cultured and wild fish populations. Journal of Aquatic Animal Health 10, 107-111.

Heikkinen, J., Vielma, J., Kemiläinen, O., Tiirola, M., Eskelinen, P., Kiuru, T., Navia- Paldanius, D., von Wright, A., 2006. Effects of soybean meal based diet on growth performance, gut histopathology and intestinal microbiota of juvenile rainbow trout (Oncorhynchus mykiss). Aquaculture 261, 259-268.

Hendrickson, B.A., Guo, J., Laughlin, R., Chen, Y., Alverdy, J.C., 1999. Increased type 1 fimbrial expression among commensal Escherichia coli isolates in the murine cecum following catabolic stress. Infection and immunity 67, 745-753.

Hennersdorf, P., Kleinertz, S., Theisen, S., Abdul-Aziz, M.A., Mrotzek, G., Palm, H.W., Saluz, H.P., 2016. Microbial diversity and parasitic load in tropical fish of different environmental conditions. PLoS One 11, e0151594.

Hesami, S., Metcalf, D.S., Lumsden, J.S., MacInnes, J.I., 2011. Identification of cold- temperature-regulated genes in Flavobacterium psychrophilum. Appl. Environ. Microbiol. 77, 1593-1600.

Hetland, D.L., Dale, O.B., Skjødt, K., Press, C.M., Falk, K., 2011. Depletion of CD8 alpha cells from tissues of Atlantic salmon during the early stages of infection with high or low virulent strains of infectious salmon anaemia virus (ISAV). Developmental & Comparative Immunology 35, 817-826.

Hight, B.V., Holts, D., Graham, J.B., Kennedy, B.P., Taylor, V., Sepulveda, C.A., Bernal, D., Ramon, D., Rasmussen, R., Lai, N.C., 2007. Plasma catecholamine levels as indicators of the post-release survivorship of juvenile pelagic sharks caught on experimental drift longlines in the Southern California Bight. Marine and Freshwater Research 58, 145-151.

169

Hovda, M.B., Fontanillas, R., McGurk, C., Obach, A., Rosnes, J.T., 2012. Seasonal variations in the intestinal microbiota of farmed Atlantic salmon (Salmo salar L.). Aquaculture Research 43, 154-159.

Huang, D.B., DuPont, H.L., 2005. Problem pathogens: extra-intestinal complications of serotype Typhi infection. The Lancet infectious diseases 5, 341-348.

Hughes, D.T., Sperandio, V., 2008. Inter-kingdom signalling: communication between bacteria and their hosts. Nature reviews. Microbiology 6, 111-120.

Hurst, M.R., Beard, S.S., Jackson, T.A., Jones, S.M., 2007. Isolation and characterization of the Serratia entomophila antifeeding prophage. FEMS microbiology letters 270, 42-48.

Hvas, M., Nilsen, T.O., Oppedal, F., 2018. Oxygen uptake and osmotic balance of Atlantic salmon in relation to exercise and salinity acclimation.

Ingerslev, H.C., Rønneseth, A., Pettersen, E., Wergeland, H., 2009. Differential expression of immune genes in Atlantic salmon (Salmo salar L.) challenged intraperitoneally or by cohabitation with IPNV. Scandinavian Journal of Immunology 69, 90-98.

Ingerslev, H.C., von Gersdorff Jørgensen, L., Lenz Strube, M., Larsen, N., Dalsgaard, I., Boye, M., Madsen, L., 2014. The development of the gut microbiota in rainbow trout (Oncorhynchus mykiss) is affected by first feeding and diet type. Aquaculture 424-425, 24-34.

Inglis, V., Richards, R., 1991. The in vitro susceptibility of Aeromonas salmonicida and other fish‐pathogenic bacteria to 29 antimicrobial agents. Journal of Fish Diseases 14, 641-650.

ISFA, 2018. SALMON FARMING - SUSTAINING COMMUNITIES AND FEEDING THE WORLD.

Jank, T., Eckerle, S., Steinemann, M., Trillhaase, C., Schimpl, M., Wiese, S., Van Aalten, D.M., Driever, W., Aktories, K., 2015. Tyrosine glycosylation of Rho by

170

Yersinia toxin impairs blastomere cell behaviour in zebrafish embryos. Nature communications 6, 7807.

Jia, R., Cao, L.-P., Du, J.-L., Wang, J.-H., Liu, Y.-J., Jeney, G., Xu, P., Yin, G.-J., 2014. Effects of carbon tetrachloride on oxidative stress, inflammatory response and hepatocyte apoptosis in common carp (Cyprinus carpio). Aquatic toxicology 152, 11-19.

Johnson, J.S., Spakowicz, D.J., Hong, B.-Y., Petersen, L.M., Demkowicz, P., Chen, L., Leopold, S.R., Hanson, B.M., Agresta, H.O., Gerstein, M., Sodergren, E., Weinstock, G.M., 2019. Evaluation of 16S rRNA gene sequencing for species and strain-level microbiome analysis. Nature Communications 10, 5029.

Jozwick, A.K., LaPatra, S.E., Graf, J., Welch, T.J., 2019. Flagellar regulation mediated by the Rcs pathway is required for virulence in the fish pathogen Yersinia ruckeri. Fish & shellfish immunology 91, 306-314.

Jun, W., Yu-Hong, S., Xue-Heng ZHANG, C.-H.L., Ming-Yun, L., Jiong, C., 2015. Molecular characterization of an IL-1β gene from the large yellow croaker (Larimichthys crocea) and its effect on fish defense against infection. Zoological Research 36, 133.

Karavolos, M., Spencer, H., Bulmer, D., Thompson, A., Winzer, K., Williams, P., Hinton, J., Khan, C.A., 2008. Adrenaline modulates the global transcriptional profile of Salmonella revealing a role in the antimicrobial peptide and oxidative stress resistance responses. BMC genomics 9, 458.

Karavolos, M.H., Bulmer, D.M., Spencer, H., Rampioni, G., Schmalen, I., Baker, S., Pickard, D., Gray, J., Fookes, M., Winzer, K., Ivens, A., Dougan, G., Williams, P., Khan, C.M., 2011. Salmonella Typhi sense host neuroendocrine stress hormones and release the toxin haemolysin E. EMBO reports 12, 252-258.

Karavolos, M.H., Khan, C.M.A., 2013. Host Neuroendocrine Stress Hormones Driving Bacterial Behaviour and Virulence. 7, 387-398.

171

Kawula, T.H., Lelivelt, M.J., Orndorff, P.E., 1996. Using a new inbred fish model and cultured fish tissue cells to study Aeromonas hydrophila and pathogenesis. Microbial pathogenesis 20, 119-125.

Kelly, J., Kennedy, P., Cryan, J., Dinan, T., Clarke, G., Hyland, N., 2015. Breaking Down the Barriers: The Gut Microbiome, Intestinal Permeability and Stress- related Psychiatric Disorders. Frontiers in Cellular Neuroscience 9.

Kinney, K.S., Austin, C.E., Morton, D.S., Sonnenfeld, G., 1999. Catecholamine enhancement of growth. Microbial Pathogenesis 26, 85-91.

Kopylova, E., Navas-Molina, J., Mercier, C., Xu, Z., Mahé, F., He, Y., Knight, R., 2016. Open-source sequence clustering methods improve the state of the art. mSystems 13 (1): e00003–15.

Kormas, K.A., Meziti, A., Mente, E., Frentzos, A., 2014. Dietary differences are reflected on the gut prokaryotic community structure of wild and commercially reared sea bream (Sparus aurata). Microbiologyopen 3, 718-728.

Krewulak, K.D., Vogel, H.J., 2008. Structural biology of bacterial iron uptake. Biochimica et Biophysica Acta (BBA) - Biomembranes 1778, 1781-1804.

Kuang, W., Li, J., Zhang, S., Long, L., 2015. Diversity and distribution of Actinobacteria associated with reef coral Porites lutea. Frontiers in Microbiology 6.

Kubilay, A., Ulukoy, G., 2004. First isolation of Staphylococcus epidermidis from cultured gilthead sea bream (Sparus aurata) in Turkey. BULLETIN- EUROPEAN ASSOCIATION OF FISH PATHOLOGISTS. 24, 137-143.

Kudva, A., Scheller, E.V., Robinson, K.M., Crowe, C.R., Choi, S.M., Slight, S.R., Khader, S.A., Dubin, P.J., Enelow, R.I., Kolls, J.K., 2011. Influenza A inhibits Th17-mediated host defense against bacterial pneumonia in mice. The Journal of Immunology 186, 1666-1674.

Kumar, G., Menanteau-Ledouble, S., Saleh, M., El-Matbouli, M., 2015. Yersinia ruckeri, the causative agent of enteric redmouth disease in fish. Veterinary research 46, 1-10.

172

Lacoste, A., Jalabert, F., Malham, S.K., Cueff, A., Poulet, S.A., 2001. Stress and stress- induced neuroendocrine changes increase the susceptibility of juvenile oysters (Crassostrea gigas) to Vibrio splendidus. Applied and environmental microbiology 67, 2304-2309.

Lages, M.A., Balado, M., Lemos, M.L., 2019. The Expression of Virulence Factors in Vibrio anguillarum Is Dually Regulated by Iron Levels and Temperature. Frontiers in Microbiology 10.

Langrish, C.L., McKenzie, B.S., Wilson, N.J., de Waal Malefyt, R., Kastelein, R.A., Cua, D.J., 2004. IL‐12 and IL‐23: master regulators of innate and adaptive immunity. Immunological reviews 202, 96-105.

Larsen, A., 2014. Studies on the Microbiota of Fishes and the Factors Influencing Their Composition.

Le Nguyen, D.D., Ngoc, H.H., Dijoux, D., Loiseau, G., Montet, D., 2008. Determination of fish origin by using 16S rDNA fingerprinting of bacterial communities by PCR-DGGE: An application on Pangasius fish from Viet Nam. Food control 19, 454-460.

Lee, M.-J., Lee, J.-J., Chung, H.Y., Choi, S.H., Kim, B.-S., 2016. Analysis of microbiota on abalone (Haliotis discus hannai) in South Korea for improved product management. International journal of food microbiology 234, 45-52.

Lesouhaitier, O., Veron, W., Chapalain, A., Madi, A., Blier, A.S., Dagorn, A., Connil, N., Chevalier, S., Orange, N., Feuilloley, M., 2009. Gram-negative bacterial sensors for eukaryotic signal molecules. Sensors 9, 6967-6990.

Li, J., Ni, J., Li, J., Wang, C., Li, X., Wu, S., Zhang, T., Yu, Y., Yan, Q., 2014. Comparative study on gastrointestinal microbiota of eight fish species with different feeding habits. Journal of applied microbiology 117, 1750-1760.

Li, L., Xu, Z., Zhou, Y., Sun, L., Liu, Z., Chen, H., Zhou, R., 2012. Global effects of catecholamines on pleuropneumoniae gene expression. PloS one 7, e31121.

173

Li, T., Li, H., Gatesoupe, F.-J., She, R., Lin, Q., Yan, X., Li, J., Li, X., 2017. Bacterial signatures of “Red-Operculum” disease in the gut of crucian carp (Carassius auratus). Microbial ecology 74, 510-521.

Li, T., Long, M., Ji, C., Shen, Z., Gatesoupe, F.-J., Zhang, X., Zhang, Q., Zhang, L., Zhao, Y., Liu, X., 2016. Alterations of the gut microbiome of largemouth bronze gudgeon (Coreius guichenoti) suffering from furunculosis. Scientific reports 6, 30606.

Li, W., Lyte, M., Freestone, P.P., Ajmal, A., Colmer‐Hamood, J.A., Hamood, A.N., 2009. Norepinephrine represses the expression of toxA and the siderophore genes in Pseudomonas aeruginosa. FEMS microbiology letters 299, 100-109.

Liu, Y., De Bruijn, I., Jack, A.L., Drynan, K., Van Den Berg, A.H., Thoen, E., Sandoval-Sierra, V., Skaar, I., Van West, P., Diéguez-Uribeondo, J., 2014. Deciphering microbial landscapes of fish eggs to mitigate emerging diseases. The ISME journal 8, 2002.

Liu, Y., Yao, Y., Li, H., Qiao, F., Wu, J., Du, Z.-y., Zhang, M., 2016. Influence of endogenous and exogenous estrogenic endocrine on intestinal microbiota in zebrafish. PloS one 11, e0163895.

Llewellyn, M.S., Boutin, S., Hoseinifar, S.H., Derome, N., 2014. Teleost microbiomes: the state of the art in their characterization, manipulation and importance in aquaculture and fisheries. Frontiers in microbiology 5, 207.

Llewellyn, M.S., McGinnity, P., Dionne, M., Letourneau, J., Thonier, F., Carvalho, G.R., Creer, S., Derome, N., 2016. The biogeography of the atlantic salmon (Salmo salar) gut microbiome. The ISME journal 10, 1280.

Lokesh, J., Kiron, V., 2016. Transition from freshwater to seawater reshapes the skin- associated microbiota of Atlantic salmon. Scientific reports 6, 19707.

Lokesh, J., Kiron, V., Sipkema, D., Fernandes, J.M., Moum, T., 2018. Succession of embryonic and the intestinal bacterial communities of Atlantic salmon (Salmo salar) reveals stage‐specific microbial signatures. MicrobiologyOpen, e00672.

174

Lopez, B., Aguilar, D., Orozco, H., Burger, M., Espitia, C., Ritacco, V., Barrera, L., Kremer, K., HERNANDEZ‐PANDO, R., Huygen, K., 2003. A marked difference in pathogenesis and immune response induced by different Mycobacterium tuberculosis genotypes. Clinical & Experimental Immunology 133, 30-37.

Lowrey, L., Woodhams, D.C., Tacchi, L., Salinas, I., 2015. Topographical mapping of the rainbow trout (Oncorhynchus mykiss) microbiome reveals a diverse bacterial community with antifungal properties in the skin. Appl. Environ. Microbiol. 81, 6915-6925.

Lyons, P.P., Turnbull, J.F., Dawson, K.A., Crumlish, M., 2017. Exploring the microbial diversity of the distal intestinal lumen and mucosa of farmed rainbow trout Oncorhynchus mykiss (W albaum) using next generation sequencing (NGS). Aquaculture research 48, 77-91.

Lyte, M., 2004. Microbial endocrinology and infectious disease in the 21st century. Trends in Microbiology 12, 14-20.

Lyte, M., Ernst, S., 1992. CATECHOLAMINE INDUCED GROWTH OF GRAM- NEGATIVE BACTERIA. Life Sciences 50, 203-212.

Lyte, M., Frank, C.D., Green, B.T., 1996. Production of an autoinducer of growth by norepinephrine cultured Escherichia coli O157: H7. FEMS microbiology letters 139, 155-159.

Lyte, M., Freestone, P.P., 2010. Microbial Endocrinology: Interkingdom Signaling in Infectious Disease and Health. Springer.

Lyte, M., Freestone, P.P., Neal, C.P., Olson, B.A., Haigh, R.D., Bayston, R., Williams, P.H., 2003. Stimulation of Staphylococcus epidermidis growth and biofilm formation by catecholamine inotropes. The Lancet 361, 130-135.

Lyte, M., Vulchanova, L., Brown, D.R., 2011a. Stress at the intestinal surface: catecholamines and mucosa-bacteria interactions. Cell and tissue research 343, 23-32.

175

Lyte, M., Vulchanova, L., Brown, D.R., 2011b. Stress at the intestinal surface: catecholamines and mucosa–bacteria interactions. Cell and tissue research 343, 23-32.

MacMicking, J., Xie, Q.-w., Nathan, C., 1997. Nitric oxide and macrophage function. Annual review of immunology 15, 323-350.

MacMicking, J.D., 2004. IFN-inducible GTPases and immunity to intracellular pathogens. Trends in immunology 25, 601-609.

Madaro, A., Olsen, R.E., Kristiansen, T.S., Ebbesson, L.O.E., Flik, G., Gorissen, M., 2016. A comparative study of the response to repeated chasing stress in Atlantic salmon (Salmo salar L.) parr and post-smolts. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology 192, 7-16.

Mahdi, L.K., Ogunniyi, A.D., LeMessurier, K.S., Paton, J.C., 2008. Pneumococcal virulence gene expression and host cytokine profiles during pathogenesis of invasive disease. Infection and immunity 76, 646-657.

McArdle, J., Dooley-Martyn, C., 1985. Isolation of Yersinia ruckeri type I (Hagerman strain) from goldfish Carassius auratus (L.). Bulletin of the European Association of Fish Pathologists 5, 10-11.

McCarthy, U., Pettinello, R., Feehan, L., Ho, Y., White, P., 2016. Experimental transmission of segmented filamentous bacteria (SFB) in rainbow trout Oncorhynchus mykiss. Diseases of aquatic organisms 119, 45-57.

McCormick, S., Shrimpton, J., Carey, J., O'dea, M., Sloan, K., Moriyama, S., Björnsson, B.T., 1998. Repeated acute stress reduces growth rate of Atlantic salmon parr and alters plasma levels of growth hormone, insulin-like growth factor I and cortisol. Aquaculture 168, 221-235.

McDonald, D., Price, M.N., Goodrich, J., Nawrocki, E.P., DeSantis, T.Z., Probst, A., Andersen, G.L., Knight, R., Hugenholtz, P., 2012. An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. The ISME journal 6, 610.

176

Mekalanos, J.J., 1992. Environmental signals controlling expression of virulence determinants in bacteria. Journal of bacteriology 174, 1.

Menanteau-Ledouble, S., Lawrence, M.L., El-Matbouli, M., 2018. Invasion and replication of Yersinia ruckeri in fish cell cultures. BMC veterinary research 14, 81.

Mendez, J., Cascales, D., García Torrico, A.I., Guijarro, J.A., 2018. Temperature - dependent gene expression in Yersinia ruckeri: tracking specific genes by bioluminescence during in vivo colonization. Frontiers in Microbiology 9, 1098.

Méndez, J., Fernández, L., Menéndez, A., Reimundo, P., Pérez-Pascual, D., Navais, R., Guijarro, J., 2009. A chromosomally located traHIJKCLMN operon encoding a putative type IV secretion system is involved in the virulence of Yersinia ruckeri. Applied and environmental microbiology 75, 937-945.

Méndez, J., Guijarro, J., 2013. In vivo monitoring of Yersinia ruckeri in fish tissues: progression and virulence gene expression. Environmental microbiology reports 5, 179-185.

Méndez, J., Reimundo, P., Pérez-Pascual, D., Navais, R., Gómez, E., Guijarro, J.A., 2011. A novel cdsAB operon is involved in the uptake of L-cysteine and participates in the pathogenesis of Yersinia ruckeri. Journal of bacteriology 193, 944-951.

Merrifield, D.L., Burnard, D., Bradley, G., Davies, S.J., Baker, R.T., 2009. Microbial community diversity associated with the intestinal mucosa of farmed rainbow trout (Oncoryhnchus mykiss Walbaum). Aquaculture research 40, 1064-1072.

Michel, C., Bernadette, F., 1987. In vitro and in vivo study of an antimicrobial activity displayed by the redmouth disease agent, Yersinia ruckeri. In, Annales de recherches vétérinaires, 43-46.

Michel, C., Faivre, B., DE KJNKELIN, P., 1986. A CLINICAL CASE OF ENTERIC REDMOUTH IN. Bull. Eur. Ass. Fish Pathol 6, 97.

Michl, S.C., Ratten, J.-M., Beyer, M., Hasler, M., LaRoche, J., Schulz, C., 2017. The malleable gut microbiome of juvenile rainbow trout (Oncorhynchus mykiss):

177

Diet-dependent shifts of bacterial community structures. PLOS ONE 12, e0177735.

Minniti, G., Hagen, L.H., Porcellato, D., Jørgensen, S.M., Pope, P.B., Vaaje-Kolstad, G., 2017. The Skin-Mucus Microbial Community of Farmed Atlantic Salmon (Salmo salar). Frontiers in Microbiology 8.

Minot, S.S., Krumm, N., Greenfield, N.B., 2015. One codex: a sensitive and accurate data platform for genomic microbial identification. BioRxiv, 027607.

Miranda, C.D., Rojas, R., 2007. Occurrence of florfenicol resistance in bacteria associated with two Chilean salmon farms with different history of antibacterial usage. Aquaculture 266, 39-46.

Mohd Nosi, M.Z., Syed Jamil Fadaak, S.N.E., Muhammad, M.D.D., Iehata, S., 2018. Assessment of gut microbiota in different developmental stages of Malaysian Mahseer (Tor tambroides). Aquaculture Research 49, 2977-2987.

Mollerup, S., Friis-Nielsen, J., Vinner, L., Hansen, T.A., Richter, S.R., Fridholm, H., Herrera, J.A.R., Lund, O., Brunak, S., Izarzugaza, J.M., 2016. Propionibacterium acnes: disease-causing agent or common contaminant? Detection in diverse patient samples by next-generation sequencing. Journal of clinical microbiology 54, 980-987.

Moloney, R.D., Desbonnet, L., Clarke, G., Dinan, T.G., Cryan, J.F., 2014. The microbiome: stress, health and disease. Mammalian Genome 25, 49-74.

Montalban-Arques, A., De Schryver, P., Bossier, P., Gorkiewicz, G., Mulero, V., Gatlin III, D.M., Galindo-Villegas, J., 2015. Selective manipulation of the gut microbiota improves immune status in vertebrates. Frontiers in Immunology 6, 512.

Moreira, C.G., Russell, R., Mishra, A.A., Narayanan, S., Ritchie, J.M., Waldor, M.K., Curtis, M.M., Winter, S.E., Weinshenker, D., Sperandio, V., 2016. Bacterial Adrenergic Sensors Regulate Virulence of Enteric Pathogens in the Gut. mBio 7, e00826-00816.

178

Morera, D., Roher, N., Ribas, L., Balasch, J.C., Donate, C., Callol, A., Boltana, S., Roberts, S., Goetz, G., Goetz, F.W., 2011. RNA-Seq reveals an integrated immune response in nucleated erythrocytes. PloS one 6, e26998.

Moussaoui, N., Braniste, V., Ait-Belgnaoui, A., Gabanou, M., Sekkal, S., Olier, M., Théodorou, V., Martin, P.G., Houdeau, E., 2014. Changes in intestinal glucocorticoid sensitivity in early life shape the risk of epithelial barrier defect in maternal-deprived rats. PloS one 9, e88382.

MOWI, 2018. Intergrated Annual Report.

Mustafa, A., MacWilliams, C., Fernandez, N., Matchett, K., Conboy, G.A., Burka, J.F., 2000. Effects of sea lice (Lepeophtheirus salmonis Kröyer, 1837) infestation on macrophage functions in Atlantic salmon (Salmo salar L.). Fish & shellfish immunology 10, 47-59.

Nakano, M., Takahashi, A., Sakai, Y., Kawano, M., Harada, N., Mawatari, K., Nakaya, Y., 2007a. Catecholamine-induced stimulation of growth in Vibrio species. Letters in applied microbiology 44, 649-653.

Nakano, M., Takahashi, A., Sakai, Y., Nakaya, Y., 2007b. Modulation of pathogenicity with norepinephrine related to the type III secretion system of Vibrio parahaemolyticus. The Journal of infectious diseases 195, 1353-1360.

Namba, A., Mano, N., Hirose, H., 2007. Phylogenetic analysis of intestinal bacteria and their adhesive capability in relation to the intestinal mucus of carp. Journal of applied microbiology 102, 1307-1317.

Navais, R., Méndez, J., Pérez-Pascual, D., Cascales, D., Guijarro, J.A., 2014. The yrpAB operon of Yersinia ruckeri encoding two putative U32 peptidases is involved in virulence and induced under microaerobic conditions. Virulence 5, 619-624.

Navarrete, P., Espejo, R., Romero, J., 2009. Molecular analysis of microbiota along the digestive tract of juvenile Atlantic salmon (Salmo salar L.). Microbial ecology 57, 550.

179

Neumann, L., 2017. Molecular detection of pathogens in southern bluefin tuna. University of Tasmania.

Ni, J., Yan, Q., Yu, Y., Zhang, T., 2014. Factors influencing the grass carp gut microbiome and its effect on metabolism. FEMS Microbiology Ecology 87, 704- 714.

Nie, L., Zhou, Q.-J., Qiao, Y., Chen, J., 2017. Interplay between the gut microbiota and immune responses of ayu (Plecoglossus altivelis) during Vibrio anguillarum infection. Fish & shellfish immunology 68, 479-487.

Nielsen, S., Walburn, J.W., Verges, A., Thomas, T., Egan, S., 2017. Microbiome patterns across the gastrointestinal tract of the rabbitfish Siganus fuscescens. PeerJ 5, e3317.

Nikouli, E., Meziti, A., Antonopoulou, E., Mente, E., Kormas, K., 2018. Gut bacterial communities in geographically distant populations of farmed sea bream (Sparus aurata) and sea bass (Dicentrarchus labrax). Microorganisms 6, 92.

Norman, A.W., Litwack, G., 1997. Hormones. Academic Press.

O'Donnell, P.M., Aviles, H., Lyte, M., Sonnenfeld, G., 2006. Enhancement of in vitro growth of pathogenic bacteria by norepinephrine: importance of inoculum density and role of transferrin. Appl Environ Microbiol 72, 5097-5099.

Ohtani, M., Villumsen, K.R., Strøm, H.K., Raida, M.K., 2014. 3D visualization of the initial Yersinia ruckeri infection route in rainbow trout (Oncorhynchus mykiss) by optical projection tomography. PLoS One 9, e89672.

Ooi, M.C., Goulden, E.F., Smith, G.G., Bridle, A.R., 2019. Haemolymph microbiome of the cultured spiny lobster Panulirus ornatus at different temperatures. Scientific reports 9, 1677.

Oraić, D., Zrnčić, S., Šoštarić, B., et al., 2002. Occurrence of enteric redmouth disease in rainbow trout (Oncorhynchus mykiss) on farms in Croatia. Acta Veterinaria Hungarica 50, 283-291.

180

Ormsby, M.J., Caws, T., Burchmore, R., Wallis, T., Verner-Jeffreys, D.W., Davies, R.L., 2016. Yersinia ruckeri isolates recovered from diseased Atlantic Salmon (Salmo salar) in Scotland are more diverse than those from rainbow trout (Oncorhynchus mykiss) and represent distinct subpopulations. Appl. Environ. Microbiol. 82, 5785-5794.

Pande, G.S.J., Suong, N.T., Bossier, P., Defoirdt, T., 2014. The catecholamine stress hormones norepinephrine and dopamine increase the virulence of pathogenic Vibrio anguillarum and Vibrio campbellii. FEMS microbiology ecology 90, 761- 769.

Pankhurst, N.W., Ludke, S.L., King, H.R., Peter, R.E., 2008. The relationship between acute stress, food intake, endocrine status and life history stage in juvenile farmed Atlantic salmon, Salmo salar. Aquaculture 275, 311-318.

Park, A., Collins, J., Blennerhassett, P., Ghia, J., Verdu, E., Bercik, P., Collins, S., 2013. Altered colonic function and microbiota profile in a mouse model of chronic depression. Neurogastroenterology & Motility 25, 733-e575.

Pedersen, K., Skall, H.F., Lassen‐Nielsen, A.M., Nielsen, T., Henriksen, N., Olesen, N.J., 2008. Surveillance of health status on eight marine rainbow trout, Oncorhynchus mykiss (Walbaum), farms in Denmark in 2006. Journal of fish diseases 31, 659-667.

Pottinger, T.G., 2008. The stress response in fish-mechanisms, effects and measurement. Fish Welfare. Blackwell Publishing Ltd, UK, 32-48.

Raida, M.K., Holten-Andersen, L., Buchmann, K., 2011. Association between Yersinia ruckeri infection, cytokine expression and survival in rainbow trout (Oncorhynchus mykiss). Fish & Shellfish Immunology 30, 1257-1264.

Rawls, J.F., Samuel, B.S., Gordon, J.I., 2004. Gnotobiotic zebrafish reveal evolutionarily conserved responses to the gut microbiota. Proceedings of the National Academy of Sciences 101, 4596-4601.

Ray, C., 2016. Characterization of the gut and skin microbiomes of wild-caught fishes from Lake Guntersville, Alabama.

181

Reiche, E.M., Morimoto, H.K., Nunes, S.M., 2005a. Stress and depression-induced immune dysfunction: implications for the development and progression of cancer. Int Rev Psychiatry 17.

Reiche, E.M.V., Morimoto, H.K., Nunes, S.M.V., 2005b. Stress and depression- induced immune dysfunction: implications for the development and progression of cancer. International Review of Psychiatry 17, 515-527.

Ringø, E., Birkbeck, T., Munro, P., Vadstein, O., Hjelmeland, K., 1996. The effect of early exposure to Vibrio pelagius on the aerobic bacterial flora of turbot, Scophthalmus maximus (L.) larvae. Journal of applied bacteriology 81, 207-211.

Ringø, E., Olsen, R., 1999. The effect of diet on aerobic bacterial flora associated with intestine of Arctic charr (Salvelinus alpinus L.). Journal of Applied Microbiology 86, 22-28.

Ringø, E., Zhou, Z., He, S., Olsen, R.E., 2014. Effect of stress on intestinal microbiota of Arctic charr, Atlantic salmon, rainbow trout and Atlantic cod: a review. African journal of microbiology research 8, 609-618.

Rintamaki, P., Valtonen, E.T., Frerichs, G., 1986. Occurrence of Yersinia ruckeri infection in farmed whitefish, Coregonus peled Gmelin and Coregonus muksun Pallas, and Atlantic salmon, Salmo salar L., in northern Finland. Journal of Fish Diseases 9, 137-140.

Robben, P.M., Mordue, D.G., Truscott, S.M., Takeda, K., Akira, S., Sibley, L.D., 2004. Production of IL-12 by macrophages infected with Toxoplasma gondii depends on the parasite genotype. The Journal of Immunology 172, 3686-3694.

Roberts, A., Matthews, J., Socransky, S., Freestone, P., Williams, P., Chapple, I., 2002. Stress and the periodontal diseases: effects of catecholamines on the growth of periodontal bacteria in vitro. Oral microbiology and immunology 17, 296-303.

Robertsen, B., 2006. The interferon system of teleost fish. Fish & Shellfish Immunology 20, 172-191.

Robledo, D., Ronza, P., Harrison, P.W., Losada, A.P., Bermúdez, R., Pardo, B.G., Redondo, M.J., Sitjà-Bobadilla, A., Quiroga, M.I., Martínez, P., 2014. RNA-seq

182

analysis reveals significant transcriptome changes in turbot (Scophthalmus maximus) suffering severe enteromyxosis. Bmc Genomics 15, 1149.

Rodgers, C., 1992. Development of a selective‐differential medium for the isolation of Yersinia ruckeri and its application in epidemiological studies. Journal of Fish Diseases 15, 243-254.

Rodgers, C., 2001. Resistance of Yersinia ruckeri to antimicrobial agents in vitro. Aquaculture 196, 325-345.

Rodríguez, I., Novoa, B., Figueras, A., 2008. Immune response of zebrafish (Danio rerio) against a newly isolated bacterial pathogen Aeromonas hydrophila. Fish & Shellfish Immunology 25, 239-249.

Roeselers, G., Mittge, E.K., Stephens, W.Z., Parichy, D.M., Cavanaugh, C.M., Guillemin, K., Rawls, J.F., 2011. Evidence for a core gut microbiota in the zebrafish. The ISME journal 5, 1595.

Rojo, I., de Ilárduya, Ó.M., Estonba, A., Pardo, M.Á., 2007. Innate immune gene expression in individual zebrafish after Listonella anguillarum inoculation. Fish & Shellfish Immunology 23, 1285-1293.

Romalde, J., Magarinos, B., Pazos, F., Silva, A., Toranzo, A., 1994a. Incidence of Yersinia ruckeri in two farms in Galicia (NW Spain) during a one‐year period. Journal of Fish Diseases 17, 533-539.

Romalde, J.L., Barja, J.L., Magariños, B., Toranzo, A.E., 1994b. Starvation-survi val processes of the bacterial fish pathogen Yersinia ruckeri. Systematic and applied microbiology 17, 161-168.

Romalde, J.L., Toranzo, A.E., 1993. Pathological activities of Yersinia ruckeri, the Enteric Redmouth (ERM) bacterium. FEMS Microbiology Letters 112, 291-299.

Ross, A., Rucker, R., Ewing, W., 1966. Description of a bacterium associated with redmouth disease of rainbow trout (Salmo gairdneri). Canadian journal of microbiology 12, 763-770.

183

Rudi, K., Angell, I.L., Pope, P.B., Vik, J.O., Sandve, S.R., Snipen, L.-G., 2018. Stable Core Gut Microbiota across the Freshwater-to-Saltwater Transition for Farmed Atlantic Salmon. Applied and Environmental Microbiology 84, e01974-01917.

Rurangwa, E., Sipkema, D., Kals, J., ter Veld, M., Forlenza, M., Bacanu, G.M., Smidt, H., Palstra, A.P., 2015. Impact of a novel protein meal on the gastrointestinal microbiota and the host transcriptome of larval zebrafish Danio rerio. Frontiers in physiology 6, 133.

Ruwandeepika, H.A.D., Jayaweera, T.S.P., Bhowmick, P.P., Karunasagar, I., Bossier, P., Defroidt, T., 2012. Pathogenesis, virulence factors and virulence regulation of vibriosis belonging to the Harveyi clade. Aquaculture 4, 59-74.

Salas-Leiton, E., Anguis, V., Martín-Antonio, B., Crespo, D., Planas, J.V., Infante, C., Cañavate, J.P., Manchado, M., 2010. Effects of stocking density and feed ration on growth and gene expression in the Senegalese sole (Solea senegalensis): potential effects on the immune response. Fish & shellfish immunology 28, 296- 302.

Salati, F., Roncarati, A., Angelucci, G., Fenza, A., Meluzzi, A., 2016. Stress and humoral innate immune response of gilthead seabream Sparus aurata cultured in sea cages. Journal of aquatic animal health 28, 166-172.

Salter, S.J., Cox, M.J., Turek, E.M., Calus, S.T., Cookson, W.O., Moffatt, M.F., Turner, P., Parkhill, J., Loman, N.J., Walker, A.W., 2014. Reagent and laboratory contamination can critically impact sequence-based microbiome analyses. BMC biology 12, 87.

Sandrini, S., Alghofaili, F., Freestone, P., Yesilkaya, H., 2014. Host stress hormone norepinephrine stimulates pneumococcal growth, biofilm formation and virulence gene expression. BMC microbiology 14, 180.

Santos, Y., Bandin, I., Nieto, T., Bruno, D., Ellis, A., Toranzo, A.E., 1990. Comparison of the cell surface hydrophobicity of bacterial fish pathogens by different procedures. Pathology in Marine Science, 101-115.

184

Schmidt, V., Amaral-Zettler, L., Davidson, J., Summerfelt, S., Good, C., 2016. Influence of fishmeal-free diets on microbial communities in Atlantic salmon (Salmo salar) recirculation aquaculture systems. Appl. Environ. Microbiol. 82, 4470-4481.

Schmidt, V.T., Smith, K.F., Melvin, D.W., Amaral‐Zettler, L.A., 2015. Community assembly of a euryhaline fish microbiome during salinity acclimation. Molecular ecology 24, 2537-2550.

Schroeter, J.C., Peterson, B.C., Bledsoe, J., Li, M., Small, B.C., 2018. Targeted gene expression panels and microbiota analysis provide insight into the effects of alternative production diet formulations on channel catfish nutritional physiology. Aquaculture 489, 46-55.

Secades, P., Guijarro, J., 1999. Purification and characterization of an extracellular protease from the fish pathogen Yersinia ruckeri and effect of culture conditions on production. Applied and environmental microbiology 65, 3969-3975.

Shaowu, L., Di, W., Hongbai, L., Tongyan, L., 2013. Isolation of Yersinia ruckeri strain H01 from farm-raised amur sturgeon Acipenser schrencki in China. Journal of aquatic animal health 25, 9-14.

Sharaff, F., Freestone, P., 2011. Microbial Endocrinology. Central European Journal of Biology 6, 685-694.

Shen, Y., Wang, D., Zhao, J., Chen, X., 2018. Fish red blood cells express immune genes and responses. Aquaculture and Fisheries 3, 14-21.

Shin, N.-R., Whon, T.W., Bae, J.-W., 2015. Proteobacteria: microbial signature of dysbiosis in gut microbiota. Trends in biotechnology 33, 496-503.

Shtrichman, R., Samuel, C.E., 2001. The role of gamma interferon in antimicrobial immunity. Current opinion in microbiology 4, 251-259.

Sidstedt, M., Hedman, J., Romsos, E.L., Waitara, L., Wadsö, L., Steffen, C.R., Vallone, P.M., Rådström, P., 2018. Inhibition mechanisms of hemoglobin, immunoglobulin G, and whole blood in digital and real-time PCR. Analytical and bioanalytical chemistry 410, 2569-2583.

185

Siegwald, L., Touzet, H., Lemoine, Y., Hot, D., Audebert, C., Caboche, S., 2017. Assessment of common and emerging bioinformatics pipelines for targeted metagenomics. PLoS One 12, e0169563.

Skrodenytė-Arbačiauskienė, V., Sruoga, A., Butkauskas, D., Skrupskelis, K., 2008. Phylogenetic analysis of intestinal bacteria of freshwater salmon Salmo salar and sea trout Salmo trutta trutta and diet. Fisheries Science 74, 1307-1314.

Small, B.C., Bilodeau, A.L., 2005. Effects of cortisol and stress on channel catfish (Ictalurus punctatus) pathogen susceptibility and lysozyme activity following exposure to Edwardsiella ictaluri. General and comparative endocrinology 142, 256-262.

Smriga, S., Sandin, S.A., Azam, F., 2010. Abundance, diversity, and activity of microbial assemblages associated with coral reef fish guts and feces. FEMS Microbiology Ecology 73, 31-42.

Stephens, W.Z., Burns, A.R., Stagaman, K., Wong, S., Rawls, J.F., Guillemin, K., Bohannan, B.J., 2016. The composition of the zebrafish intestinal microbial community varies across development. The ISME journal 10, 644.

Stevenson, R., Daly, J., 1982. Biochemical and serological characteristics of Ontario isolates of Yersinia ruckeri. Canadian Journal of Fisheries and Aquatic Sciences 39, 870-876.

Stock, I., Henrichfreise, B., Wiedemann, B., 2002. Natural antibiotic susceptibility and biochemical profiles of Yersinia enterocolitica-like strains: Y. bercovieri, Y. mollaretii, Y. aldovae and ‘Y. ruckeri'. Journal of medical microbiology 51, 56- 69.

Sullam, K.E., Dalton, C.M., Russell, J.A., Kilham, S.S., El-Sabaawi, R., German, D.P., Flecker, A.S., 2015. Changes in digestive traits and body nutritional composition accommodate a trophic niche shift in Trinidadian guppies. Oecologia 177, 245- 257.

Sun, Y.Z., Yang, H.L., Ma, R.L., Song, K., Lin, W.Y., 2011. Molecular analysis of autochthonous microbiota along the digestive tract of juvenile grouper

186

Epinephelus coioides following probiotic Bacillus pumilus administration. Journal of applied microbiology 110, 1093-1103.

Sundell, K., Jutfelt, F., Agústsson, T., Olsen, R.-E., Sandblom, E., Hansen, T., Björnsson, B.T., 2003. Intestinal transport mechanisms and plasma cortisol levels during normal and out-of-season parr–smolt transformation of Atlantic salmon, Salmo salar. Aquaculture 222, 265-285.

Sundh, H., Olsen, R.E., Fridell, F., Gadan, K., Evensen, Ø., Glette, J., Taranger, G.L., Myklebust, R., Sundell, K., 2009. The effect of hyperoxygenation and reduced flow in fresh water and subsequent infectious pancreatic necrosis virus challenge in sea water, on the intestinal barrier integrity in Atlantic salmon, Salmo salar L. Journal of fish diseases 32, 687-698.

Sunyer, J., Gomez, E., Tort, L., Navarro, V., Quesada, J., 1995. Physiological responses and depression of humoral components of the immune system in gilthead sea bream (Sparus aurata) following daily acute stress. Canadian Journal of Fisheries and Aquatic Sciences 52, 2339-2346.

Tacchi, L., Lowrey, L., Musharrafieh, R., Crossey, K., Larragoite, E.T., Salinas, I., 2015. Effects of transportation stress and addition of salt to transport water on the skin mucosal homeostasis of rainbow trout (Oncorhynchus mykiss). Aquaculture 435, 120-127.

Talwar, C., Nagar, S., Lal, R., Negi, R.K., 2018. Fish Gut Microbiome: Current Approaches and Future Perspectives. Indian journal of microbiology, 1-18.

Tarnecki, A.M., Burgos, F.A., Ray, C.L., Arias, C.R., 2017. Fish intestinal microbiome: diversity and symbiosis unravelled by metagenomics. Journal of applied microbiology 123, 2-17.

Thorsen, B., Enger, Ø., Norland, S., Hoff, K.A., 1992. Long-term starvation survival of Yersinia ruckeri at different salinities studied by microscopical and flow cytometric methods. Appl. Environ. Microbiol. 58, 1624-1628.

Tinsley, J., 2010. Studies on the pathogenicity of Yersinia ruckeri biotype 2 to rainbow trout (Oncorhynchus mykiss, Walbaum). Heriot-Watt University.

187

Tobback, E., Decostere, A., Hermans, K., Haesebrouck, F., Chiers, K., 2007. Yersinia ruckeri infections in salmonid fish. Journal of Fish Diseases 30, 257-268.

Tobback, E., Decostere, A., Hermans, K., Ryckaert, J., Duchateau, L., Haesebrouck, F., Chiers, K., 2009. Route of entry and tissue distribution of Yersinia ruckeri in experimentally infected rainbow trout Oncorhynchus mykiss. Diseases of aquatic organisms 84, 219-228.

Tobback, E., Decostere, A., Hermans, K., Van Den Broeck, W., Haesebrouck, F., Chiers, K., 2010. In vitro markers for virulence in Yersinia ruckeri. Journal of fish diseases 33, 197-209.

Torabi Delshad, S., Soltanian, S., Sharifiyazdi, H., Bossier, P., 2019. Effect of catecholamine stress hormones (dopamine and norepinephrine) on growth, swimming motility, biofilm formation and virulence factors of Yersinia ruckeri in vitro and an in vivo evaluation in rainbow trout. Journal of fish diseases 42, 477-487.

Toranzo, A., Novoa, B., Romalde, J., Nunez, S., Devesa, S., Marino, E., Silva, R., Martinez, E., Figueras, A., Barja, J., 1993. Microflora associated with healthy and diseased turbot (Scophthalmus maximus) from three farms in northwest Spain. Aquaculture 114, 189-202.

Toranzo, A.E., Barja, J.L., 1993. Virulence factors of bacteria pathogenic for coldwater fish. Annual Review of Fish Diseases 3, 5-36.

Tort, L., 2011. Stress and immune modulation in fish. Developmental & Comparative Immunology 35, 1366-1375.

Tort, L., Rotllant, J., Rovira, L., 1998. Immunological suppression in gilthead sea bream Sparus aurata of the North-West Mediterranean at low temperatures. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology 120, 175-179.

Tran, N.T., Zhang, J., Xiong, F., Wang, G.-T., Li, W.-X., Wu, S.-G., 2018. Altered gut microbiota associated with intestinal disease in grass carp (Ctenopharyngodon idellus). World Journal of Microbiology and Biotechnology 34, 71.

188

Valtonen, E.T., Rintamäki, P., Koskivaara, M., 1992. Occurrence and pathogenicity of Yersinia ruckeri at fish farms in northern and central Finland. Journal of Fish Diseases 15, 163-171.

Van Dijk, P., Wood, C., 1988. The effect of β‐adrenergic blockade on the recovery process after strenuous exercise in the rainbow trout, Salmo gairdneri Richardson. Journal of fish biology 32, 557-570.

Van Maerken, T., Speleman, F., Vermeulen, J., Lambertz, I., De Clercq, S., De Smet, E., Yigit, N., Coppens, V., Philippé, J., De Paepe, A., 2006. Small-molecule MDM2 antagonists as a new therapy concept for neuroblastoma. Cancer research 66, 9646-9655.

Vasemägi, A., Visse, M., Kisand, V., 2017. Effect of Environmental Factors and an Emerging Parasitic Disease on Gut Microbiome of Wild Salmonid Fish. mSphere 2, e00418-00417.

Vlisidou, I., Lyte, M., Van Diemen, P.M., Hawes, P., Monaghan, P., Wallis, T.S., Stevens, M.P., 2004. The neuroendocrine stress hormone norepinephrine augments Escherichia coli O157: H7-induced enteritis and adherence in a bovine ligated ileal loop model of infection. Infection and immunity 72, 5446-5451.

Vuillaume, A., Brun, R., Chene, P., Sochon, E., Lesel, R., 1987. First isolation of Yersinia ruckeri from sturgeon, Acipenser baeri Brandt, in south west of France. Bulletin of the European Association of Fish Pathologists 7, 18-19.

Wang, C., Sun, G., Li, S., Li, X., Liu, Y., 2018. Intestinal microbiota of healthy and unhealthy Atlantic salmon Salmo salar L. in a recirculating aquaculture system. Journal of Oceanology and Limnology 36, 414-426.

Wang, T., Husain, M., Hong, S., Holland, J.W., 2014. Differential expression, modulation and bioactivity of distinct fish IL‐12 isoforms: Implication towards the evolution of Th1‐like immune responses. European journal of immunology 44, 1541-1551.

189

Wangkahart, E., Secombes, C.J., Wang, T., 2017. Dissecting the immune pathways stimulated following injection vaccination of rainbow trout (Oncorhynchus mykiss) against enteric redmouth disease (ERM). Fish & shellfish immunology.

Welch, T.J., Fricke, W.F., McDermott, P.F., White, D.G., Rosso, M.-L., Rasko, D.A., Mammel, M.K., Eppinger, M., Rosovitz, M., Wagner, D., 2007. Multiple antimicrobial resistance in plague: an emerging public health risk. PloS one 2, e309.

Williams, P., Rabsch, W., Methner, U., Voigt, W., Tschäpe, H., Reissbrodt, R., 2006. Catecholate receptor proteins in Salmonellaenterica: role in virulence and implications for vaccine development. Vaccine 24, 3840-3844.

Winter, S.E., Bäumler, A.J., 2014. Why related bacterial species bloom simultaneously in the gut: principles underlying the ‘Like will to like’concept. Cellular microbiology 16, 179-184.

Woo, P.T.K., Bruno, D.W., 2011. Fish Diseases and Disorders, Volume 3 : Viral, Bacterial and Fungal Infections. CABI, Wallingford.

Woo, P.T.K., Bruno, D.W., Lim, L.S., 2002. Diseases and disorders of finfish in cage culture. CABI Wallingford.

Woo, T.K., Bruno, D.W., 2014. Diseases and Disorders of Finfish in Cage Culture. CABI, Wallingford.

Wrobel, A., Leo, J.C., Linke, D., 2019. Overcoming Fish Defences: The Virulence Factors of Yersinia ruckeri. Genes 10, 700.

Wrobel, A., Ottoni, C., Leo, J.C., Gulla, S., Linke, D., 2018. The repeat structure of two paralogous genes, Yersinia ruckeri invasin (yrInv) and a “Y. ruckeri invasin-like molecule”,(yrIlm) sheds light on the evolution of adhesive capacities of a fish pathogen. Journal of structural biology 201, 171-183.

Xiong, J.-B., Nie, L., Chen, J., 2019. Current understanding on the roles of gut microbiota in fish disease and immunity. Zoological research 40, 70.

190

Xu, F., Wu, C., Guo, F., Cui, G., Zeng, X., Yang, B., Lin, J., 2015. Transcriptomic analysis of Campylobacter jejuni NCTC 11168 in response to epinephrine and norepinephrine. Frontiers in microbiology 6, 452.

Yang, H.L., Sun, Y.Z., Ma, R.L., Ye, J.D., 2012. PCR‐DGGE analysis of the autochthonous gut microbiota of grouper Epinephelus coioides following probiotic Bacillus clausii administration. Aquaculture Research 43, 489-497.

Yang, Q., Anh, N.D., Bossier, P., Defoirdt, T., 2014. Norepinephrine and dopamine increase motility, biofilm formation, and virulence of Vibrio harveyi. Frontiers in microbiology 5.

Yarahmadi, P., Miandare, H.K., Fayaz, S., Caipang, C.M.A., 2016. Increased stocking density causes changes in expression of selected stress-and immune-related genes, humoral innate immune parameters and stress responses of rainbow trout (Oncorhynchus mykiss). Fish & shellfish immunology 48, 43-53.

Yu, H.B., Kaur, R., Lim, S., Wang, X.H., Leung, K.Y., 2007. Characterization of extracellular proteins produced by Aeromonas hydrophila AH‐1. Proteomics 7, 436-449.

Zaharia, M., Bolosky, W.J., Curtis, K., Fox, A., Patterson, D., Shenker, S., Stoica, I., Karp, R.M., Sittler, T., 2011. Faster and more accurate sequence alignment with SNAP. arXiv preprint arXiv:1111.5572.

Zarkasi, K.Z., Taylor, R.S., Abell, G.C., Tamplin, M.L., Glencross, B.D., Bowman, J.P., 2016. Atlantic salmon (Salmo salar L.) gastrointestinal microbial community dynamics in relation to digesta properties and diet. Microbial ecology 71, 589-603.

Zarkasi, K.Z., Taylor, R.S., Glencross, B.D., Abell, G.C., Tamplin, M.L., Bowman, J.P., 2017. In vitro characteristics of an Atlantic salmon (Salmo salar L.) hind gut microbial community in relation to different dietary treatments. Research in microbiology 168, 751-759.

Zha, Y., Eiler, A., Johansson, F., Svanbäck, R., 2018. Effects of predation stress and food ration on perch gut microbiota. Microbiome 6, 28.

191

Zhang, F., Qiu, X., Liu, Y., Wang, J., Li, X., Wang, X., 2017. Expression analysis of three immune genes Interferon-gamma, Mx and Interferon regulatory factor-1 of Japanese flounder (Paralichthys olivaceus). Brazilian Archives of Biology and Technology 60.

Zhang, M., Sun, Y., Liu, Y., Qiao, F., Chen, L., Liu, W.-T., Du, Z., Li, E., 2016. Response of gut microbiota to salinity change in two euryhaline aquatic animals with reverse salinity preference. Aquaculture 454, 72-80.

Zhang, Z., Wu, H., Xiao, J., Wang, Q., Liu, Q., Zhang, Y., 2013. Immune responses evoked by infection with Vibrio anguillarum in zebrafish bath-vaccinated with a live attenuated strain. Veterinary immunology and immunopathology 154, 138-144.

Zhou, Z., Liu, Y., Shi, P., He, S., Yao, B., Ringø, E., 2009. Molecular characterization of the autochthonous microbiota in the gastrointestinal tract of adult yellow grouper (Epinephelus awoara) cultured in cages. Aquaculture 286, 184-189.

Zou, J., Secombes, C.J., 2016. The function of fish cytokines. Biology 5, 23.

192