HUMORAL IMMUNE RESPONSE OF GREAT LAKES FISHES TO VIRAL
HEMORRHAGIC SEPTICEMIA VIRUS GENOTYPE IVB
By
Elena Virginia Millard

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Pathobiology- Doctor of Philosophy
2013

ABSTRACT
HUMORAL IMMUNE RESPONSE OF GREAT LAKES FISHES TO VIRAL
HEMORRHAGIC SEPTICEMIA VIRUS GENOTYPE IVB
By
Elena Virginia Millard

Viral hemorrhagic septicemia virus genotype IVb (VHSV-IVb) is a recently emerged
pathogen of Great Lakes fishes in North America. Following initial fish kills in Lake St.
Clair, Michigan (MI) in 2006, VHSV-IVb went undetected for nearly three years there
until another die-off ensued in the summer of 2009. This prompted questions regarding
host immunity to the virus as well as the usefulness of the virus isolation assay as a
surveillance tool. Herein, I describe evidence of humoral immunity to VHSV-IVb in
several populations of wild fishes and the development of antibodies by muskellunge
(Esox masquinongy) following experimental challenge and vaccination. Using a 50%
plaque neutralization test (PNT) and a competitive enzyme-linked immunosorbent assay
(cELISA), antibodies were detected in Lake St. Clair fish during all sampling years
between 2004-2011 (six summers). In contrast, VHS virus was only detected during the
summer of 2006 and 2009, and in both cases, fish were sampled during or immediately
following mortality events. Neutralizing (PNT) and/or binding (cELISA) antibodies were
detected in 5 of 13 fish species including muskellunge, northern pike (E. lucius),
freshwater drum (Aplodinotus grunniens), smallmouth bass (Micropterus dolomieu) and
channel catfish (Ictalurus punctatus). The greatest overall seroprevalence and antibody
titers were detected in muskellunge. Antibodies were also detected by cELISA in

muskellunge from Lower Fox River/Green Bay, Wisconsin and from Thornapple Lake,
Michigan.

Muskellunge experimentally infected with VHSV-IVb developed neutralizing antibodies
by 5 – 7 weeks (385 – 539 degree days) post-challenge. The development of
neutralizing antibodies corresponded to a decrease in virus presence and titers in sera,
indicating their role in limiting VHSV-IVb infection. A reduced neutralizing antibody
response was mounted by fish exposed with the lowest dose of the virus, suggesting a
threshold level of infection is necessary for induction. In surviving fish, neutralizing
antibodies were detectable for a longer period of time after infection then was the virus
in tissues (assessed in a parallel study). To better understand host immune response in
this species, a DNA vaccine encoding the glycoprotein (G) gene of VHSV-IVb was
developed. DNA vaccination primed the adaptive immune response to respond to
VHSV-IVb as evident by high levels of neutralizing antibodies in 100% of surviving
muskellunge by only 4 weeks (310 degree days) after challenge. Compared to plasmid
control fish, pVHSivb-G-vaccinated fish were significantly protected (relative percent
survival = 45.2%) following lethal virus challenge 7 weeks (539 degree days) postvaccination. Surviving fish from the pVHSivb-G vaccinated group also had significantly
lower infection prevalence and tissue viral loads compared to control fish. The cELISA,
developed as part of this dissertation, and the PNT are non-lethal immunological tools
that will find multiple applications in future studies for assessment of adaptive immunity
to VHSV-IVb.

Copyright by
ELENA VIRGINIA MILLARD
2013

This dissertation is dedicated to my parents, Ronald and Eileen Millard.

v

ACKNOWLEDGMENTS

First, I would like to extend my sincere thanks to my major advisor, Dr. Mohamed
Faisal for his guidance, patience, and many hours of time. Thank you for the opportunity
to grow as a scientist as a part of your research team. I would also like to sincerely
thank the other members of my guidance committee: Dr. Scott LaPatra, Dr. Travis
Brenden, Dr. Scott Fitzgerald and Dr. Amber Peters. I am greatly appreciative of each
membersʼ unique contributions, advice and support throughout this process.
I would also like to thank all of my past and present colleagues, friends, mentors
and support system at the Aquatic Animal Health Laboratory, especially Dr. Thomas
Loch, Dr. Andrew Winters, Carolyn Schulz, Dr. Robert Kim, Dr. Wei Xu, Michelle Gunn,
Elizabeth Throckmorton, Isaac Standish, Dan Bjorklund, Danielle Van Vliet, Christine
Rabaut, Ashley Bourke, and Monica Lucas. A special thank you goes to Dr. Robert Kim
for his mentorship and to Ashley Bourke for being a precise, dependable, and
independent assistant. I would also like to thank undergraduate helpers Alex Prediger,
Maggie Fish and Adam Becker for all of their help.
I would also like to thank my collaborators on these projects and others including
Michigan State University Institutional Animal Care and Use Committee (IACUC), the
Fisheries Division of the Michigan Department of Natural Resources, Susan
Marcquenski and the Wisconsin DNR, and Dr. Steve Kaatarri and his lab at the Virginia
Institute of Marine Science. I am grateful to the funding agencies that supported this

vi

work including the Great Lakes Fishery Trust, U.S. Fisheries and Wildlife Services, and
U.S. Geological Survey.
With my deepest respect and gratitude, I would like to thank my parents, Ron and
Eileen Millard, and brother, Alec, for their unconditional love, encouragement, and
support throughout my life, including during this process. I would also like to sincerely
thank my boyfriend and best friend, Jonathan Lutz and our creatures: Gatsby, Tiger,
Lucy, and Goldie.
Lastly, I would like to acknowledge the rabbits and fish whose lives were taken as
part of this research, and I would like to ask for forgiveness for doing this. I acknowledge
these creatures not as research subjects but as beautiful, fascinating, necessary parts
of the earth.

vii

TABLE OF CONTENTS

LIST OF TABLES .............................................................................................................. xii
LIST OF FIGURES............................................................................................................. xiv
INTRODUCTION .............................................................................................................. 1
CHAPTER 1
REVIEW OF LITERATURE.............................................................................................. 5
VIRAL HEMORRHAGIC SEPTICEMIA VIRUS......................................................... 6
Structure and infection cycle ............................................................................. 6
Transmission and VHS disease ......................................................................... 8
Geographic distribution and genotypes ........................................................... 9
VHSV-IVb in the Great Lakes............................................................................ 11
IMMUNE RESPONSE OF FISH .............................................................................. 13
Lymphoid tissues .............................................................................................. 13
Innate immunity ................................................................................................. 14
Adaptive (acquired) immunity.......................................................................... 16
IMMUNE RESPONSE TO VIRAL HEMORRHAGIC SEPTICEMIA VIRUS ............ 19
Innate immune response of fish to VHSV ....................................................... 19
The antibody response to VHSV...................................................................... 21
VHSV immune response and temperature ..................................................... 26
DNA VACCINATION AGAINST VHSV AND THE ANTIBODY RESPONSE ......... 28
DNA vaccination technology............................................................................ 28
DNA vaccination against fish rhabdoviruses ................................................. 29
Three phases of antiviral immunity ................................................................. 32
RESEARCH QUESTIONS AND OBJECTIVES ...................................................... 35
APPENDIX .................................................................................................................... 39
CHAPTER 2
DEVELOPMENT OF NEUTRALIZING ANTIBODY RESPONSES IN MUSKELLUNGE,
ESOX MASQUINONGY (MITCHILL), EXPERIMENTALLY EXPOSED TO VIRAL
HEMORRHAGIC SEPTICEMIA VIRUS (GENOTYPE IVB)........................................... 41
ABSTRACT ............................................................................................................. 42
INTRODUCTION ..................................................................................................... 42
MATERIALS AND METHODS ................................................................................ 45
Cell culture and virus isolate ........................................................................... 45
Muskellunge....................................................................................................... 46
Experimental infection...................................................................................... 46
Assessment of VHSV in infected muskellunge serum samples .................. 48

viii

50% plaque neutralization test (50%PNT) ....................................................... 48
RESULTS ................................................................................................................ 49
Disease course and virus re-isolation .......................................................... 49
Neutralizing antibody response..................................................................... 50
DISCUSSION........................................................................................................... 51
ACKNOWLEDGEMENTS ....................................................................................... 54
APPENDIX .................................................................................................................... 55
CHAPTER 3
HETEROGENEITY IN LEVELS OF SERUM NEUTRALIZING ANTIBODIES AGAINST
VIRAL HEMORRHAGIC SEPTICEMIA VIRUS GENOTYPE IVB AMONG FISH
SPECIES IN LAKE ST. CLAIR, MICHIGAN, USA ........................................................ 59
ABSTRACT ............................................................................................................. 60
INTRODUCTION ..................................................................................................... 61
MATERIALS AND METHODS ................................................................................ 62
Field serum samples......................................................................................... 63
Cell line and virus isolate ................................................................................. 64
Virus isolation.................................................................................................... 64
Neutralizing antibody detection....................................................................... 65
Complement source .......................................................................................... 66
Naïve sera examination .................................................................................... 66
Specificity of VHSV-IVb neutralizing antibodies ............................................ 67
Statistical analysis ............................................................................................ 67
RESULTS ................................................................................................................ 67
Virus isolation.................................................................................................... 67
Innate neutralization of VHSV by sera of naïve muskellunge....................... 68
Neutralization specificity of fish sera.............................................................. 68
Neutralizing antibodies against VHSV-IVb...................................................... 68
DISCUSSION........................................................................................................... 70
ACKNOWLEDGEMENTS ....................................................................................... 75
APPENDIX .................................................................................................................... 76
CHAPTER 4
APPLICATION OF A COMPETITIVE ELISA FOR DETECTING CIRCULATING
ANTIBODIES AGAINST VIRAL HEMORRHAGIC SEPTICEMIA VIRUS (GENOTYPE
IVB) IN REPRESENTATIVE FISH SPECIES IN LAKE ST. CLAIR, MICHIGAN, USA. 84
ABSTRACT ............................................................................................................. 85
BODY....................................................................................................................... 85
ACKNOWLEDGEMENTS ....................................................................................... 92
APPENDIX .................................................................................................................... 94

ix

CHAPTER 5
DETECTION OF ANTIBODIES TO VIRAL HEMORRHAGIC SEPTICEMIA VIRUS-IVB
IN SERA OF MUSKELLUNGE (ESOX MASQUINONGY) USING A COMPETITIVE
ELISA .......................................................................................................................... 99
ABSTRACT ........................................................................................................... 100
INTRODUCTION ................................................................................................... 100
MATERIALS AND METHODS .............................................................................. 104
Preparation of virus stock and purified virus............................................... 104
Production of polyclonal antibody to VHSV-IVb .......................................... 105
Competitive ELISA procedure and control sera .......................................... 106
Establishment of a positive-negative threshold for muskellunge ............. 108
Use of sera from fish surviving experimental VHSV-IVb infection to test
assay................................................................................................................. 109
Application of the assay to field-collected sera of muskellunge ............... 110
Comparison of cELISA and PNT results ....................................................... 112
RESULTS .............................................................................................................. 113
Establishment of the cELISA ......................................................................... 113
Establishment of a positive-negative threshold for muskellunge ............. 115
Use of sera from fish surviving experimental VHSV-IVb infection to test
assay................................................................................................................. 115
Application of the assay to field-collected sera of muskellunge ............... 116
Comparison of cELISA with PNT results ...................................................... 118
DISCUSSION......................................................................................................... 118
ACKNOWLEDGEMENTS ..................................................................................... 125
APPENDIX .................................................................................................................. 126
CHAPTER 6
VACCINATION WITH A PLASMID CONTAINING THE VHSV-IVB GLYCOPROTEIN
GENE PROTECTS MUSKELLUNGE (ESOX MASQUINONGY) AGAINST
CHALLENGE ............................................................................................................... 134
INTRODUCTION ................................................................................................... 135
MATERIALS AND METHODS .............................................................................. 137
Ethics statement ........................................................................................... 137
Construction of eukaryotic expression plasmid pVHSivb-G .................... 137
Muskellunge vaccination and challenge .................................................... 138
Quantification of VHSV-IVb in tissues by plaque assay............................ 140
Detection of an adaptive humoral immune response ............................... 142
Antibody testing ............................................................................................ 142
Study 2: Rainbow trout vaccination and cross-protection challenge...... 143
Data analysis ................................................................................................. 144
RESULTS .............................................................................................................. 145
Efficacy of pVHSivb-G in conferring protection to muskellunge ............. 144
Quantification of VHSV-IVb in tissues of mortalities and survivors ........ 145
Analysis of the adaptive humoral immune response to pVHSivb-G........ 147

x

Saprolegnia infection post-challenge ........................................................ .148
Study 2: Rainbow trout vaccination and cross-protection challenge...... 148
DISCUSSION......................................................................................................... 150
ACKNOWLEDGEMENTS ..................................................................................... 156
APPENDIX .................................................................................................................. 158
CHAPTER 7
CONCLUSIONS AND FUTURE RESEARCH
CONCLUSIONS .................................................................................................... 168
FUTURE RESEARCH DIRECTIONS AND APPLICATIONS................................ 174
APPENDIX .................................................................................................................. 180
REFERENCES ............................................................................................................. 182

xi

LIST OF TABLES

-1

Table 2.1. Virus isolation and titer range in plaque forming units (pfu) mL of serum of
sampled muskellunge, Esox masquinongy, experimentally infected with viral
hemorrhagic septicemia virus (VHSV, genotype IVb). “VHSV Pos.” column indicates
number of VHSV-positive serum samples/ number of samples tested. Groups were
2

-1

challenged by immersion in the following concentrations of virus: 10 pfu mL
3

-1

1), 4 × 10 pfu mL

5

-1

(Group 2), and 10 pfu mL

(Group

(Group 3) ......................................... 56

Table 3.1. Distribution of neutralizing titers against viral hemorrhagic septicemia virus
in 118 serum samples from naïve juvenile muskellunge (Esox masquinongy). Group I (~
4 mo.; 17 cm, 1 SD; 19 g, 4 SD), Group II (~ 6 mo.; 16 cm, 2 SD; 17 g, 5 SD), Group III
(~ 22 mo.; 32 cm, 1 SD; 148 g, 21 SD) .......................................................................... 77
Table 3.2. Serum neutralization specificity of Lake St. Clair muskellunge (1-6; Esox
masquinongy) and freshwater drum (7; Aplodinotus grunniens). Sera were tested in
parallel with viral hemorrhagic septicemia virus (VHSV-IVb) and infectious
hematopoietic necrosis virus (IHNV) using a 50% plaque neutralization test (50% PNT)..
....................................................................................................................................... 78
Table 3.3. Prevalence (%) of viral hemorrhagic septicemia virus (VHSV-IVb)
neutralizing antibodies among Lake St. Clair fish species ............................................. 79
Table 3.4. Maturity, sex, age, and size of Lake St. Clair muskellunge (Esox
a

masquinongy) included in this study ............................................................................ 81
Table 4.1. Viral hemorrhagic septicemia virus-genotype IVb competitive ELISA results
of 113 fish of 9 species sampled from Lake St. Clair, MI, May 2010. Bold font indicates
species for which the mean inhibition was significantly different than the overall mean.
SD = standard deviation ................................................................................................. 95
Table 4.2. Pairwise differences in least-squares means for percent inhibition values
from 13 fish species from Lake St. Clair, Michigan. Blood samples were collected in May
of 2010 and tested for VHSV-IVb antibodies by cELISA. Differences in least-squares
means (Diff.), t and p values for comparisons between species are shown. Degrees of
freedom = 104. The type-I error rate was set at 0.05. Bolded p values indicate
statistically significant comparisons. Species abbreviations are as follows: CCF =
channel catfish, ROB = rock bass, NOP = northern pike, SIR = silver redhorse, SMB =
smallmouth bass, FRD = freshwater drum, WHP = white perch, SHR = shorthead
redhorse, QUI = quillback............................................................................................... 96

xii

Table 5.1. Competitive ELISA results for 200 free-ranging muskellunge shown by
location and sampling event. Fish were sampled between 2005-2012 from water bodies
in Michigan (MI) and Wisconsin (WI). Sera with percent inhibition values ≥14.6 by
cELISA are positive for antibodies to VHSV-IVb. Tissues and/or non-lethal samples
(ovarian fluid, sera) from all fish were negative for VHS virus by cell culture isolation. A
positive VHSV-IVb status indicates that the virus had been isolated from fish from that
water system prior to the first sampling event. Confidence intervals (CI) were calculated
using R (R Core Team 2012). SD = standard deviation ............................................... 132
Table 5.2. Percent inhibitions of muskellunge testing positive (≥14.6% inhibition) for
VHSV-IVb antibodies by cELISA on at least one time point sampled. Fish were
challenged at 0 and 22 weeks with VHSV-IVb by immersion and blood was sampled
from 22 survivors 67 and 79 weeks post-infection (45 and 57 weeks post-second
exposure). SD = standard deviation ............................................................................ 133
Table 6.1. A pilot study was conducted in order to determine the minimum VHSV-IVb
dose that would be lethal to ≥60% of muskellunge (Esox masquinongy) by immersion.
Five fish were infected at each dose, except for control fish (n = 3). A dose of 10
-1

plaque-forming units (pfu) mL

5

was chosen for challenge of the vaccinated fish ....... 162

Table 6.2. VHSV-IVb infection prevalence and tissue titers of mortalities and survivors.
Fish were vaccinated with an empty plasmid or the pVHSVivb-G DNA vaccine and
challenged with VHSV-IVb 7 weeks (539 degree days) post-vaccination. Data from
survivors is from fish sampled 28 days post virus-challenge. Virus titers are in plaqueforming units gram

-1

of tissue ...................................................................................... 163

Table 6.3. Antibody titers of muskellunge (Esox masquinongy) vaccinated with
pVHSivb-G or the plasmid control. Blood was drawn non-lethally from 10-12 fish from
each treatment prior to vaccination (0d) and 7 and 11 weeks later. Fish were not
challenged at any point with VHSV-IVb. Sera were tested for neutralizing G antibodies
by 50% plaque neutralization test (PNT) and for binding G antibodies by VHSV-IVb
cELISA. Parentheses indicate number of sera with the specified titer. ........................ 164
Table 6.4. Neutralizing antibody titers in vaccinated muskellunge (Esox masquinongy)
surviving challenge with VHSV-IVb. Fish were sampled 28 days post-challenge ........ 165
Table 6.5. Cumulative percent mortality (CPM) of rainbow trout (Oncorhynchus mykiss)
challenged by immersion 7 days (100 degree days) and 28 days (400 degree days)
5

-1

post-vaccination with 10 pfu mL IHN virus. Number designations 1 and 2 indicate
replicate tanks. Relative percent survival (RPS) = [1-(average CPM of vaccinated
fish/average CPM of plasmid control fish)] × 100......................................................... 166

xiii

LIST OF FIGURES

Figure 1.1. A) Schematic of the typical Novirhabdovirus virion (70 nm x 180 nm) and its
linear, negative-sense single-stranded RNA molecule of approximately 11,000
nucleotides. The genome encodes 5 structural proteins [L (RNA-dependent RNA
polymerase), G (glycoprotein), N (nucleoprotein), P (phosphoprotein) and M (matrix
protein)] and one non-virion protein (NV). B) Diagram showing the general design of a
DNA vaccine containing the G gene of viral hemorrhagic septicemia virus (VHSV)
(modified from Schirmbeck and Reimann (2001) and Kurath et al. (2007)). For
interpretation of the references to color in this and all other figures, the reader is referred
to the electronic version of this dissertation ................................................................... 40
Figure 2.1. Development of neutralizing antibodies measured by the 50% plaque
neutralization test (50%PNT) in three groups of muskellunge, Esox masquinongy,
challenged by immersion with viral hemorrhagic septicemia virus (VHSV-IVb). An
2

3

antibody titer of ≥20 is positive. Groups 1, 2 and 3 were exposed to 10 , 4 × 10 and
5

-1

10 pfu mL respectively. The last sampling points were 20, 16 and 13 weeks for
Groups 1, 2 and 3, respectively...................................................................................... 57
Figure 3.1. Lake St. Clair (Lake Erie watershed, USA) showing sampling sites. Most
fish were captured using trap nets (!) positioned between 42° 39'N, 82° 46'W and 42°
37'N, 82° 46'W in Anchor Bay. Lake sturgeon (Acipenser fulvescens) were captured
using survey setlines (!: 42° 37'N, 82° 37'W). Smallmouth bass (Micropterus dolomieu)
from a mortality event in June of 2009 (!: 42° 29'N, 82° 53'W), and a single
muskellunge (Esox masquinongy) collected May 2006 (": 42° 21'N, 82° 54'W), were
caught using scoop nets................................................................................................. 82
Figure 3.2. Antibody titers measured by a 50% plaque neutralization test (50% PNT)
against viral hemorrhagic septicemia virus (IVb) from Lake St. Clair fish species that
were antibody-positive: muskellunge (# Esox masquinongy), northern pike (" E.
lucius), smallmouth bass (! Micropterus dolomieu), freshwater drum (" Aplodinotus
grunniens). Neutralization titers of >80 are considered positive for muskellunge and
northern pike. The parenthesis denote the number of sera with titers of <20/number
sampled.......................................................................................................................... 83
Figure 4.1. Distribution of percent inhibition values of sera from 113 fish of nine species
from Lake St. Clair, Michigan, May 2010. Sera were tested for antibodies against viral
hemorrhagic septicemia virus-genotype IVb using a competitive ELISA. The mean
percent inhibition value of each species is indicated by a red dash. Dashed and dotted
horizontal lines indicate the overall mean (8.8) and median (3.6) percent inhibition
values ............................................................................................................................. 98

xiv

Figure 5.1. Map of Michigan and Wisconsin, USA showing sites where sera were
collected from muskellunge (Esox masquinongy) during health surveys of wild fish, 2005
to 2012. Fish sampled from the Anchor Bay and Detroit River sites are collectively
referred to as Lake St. Clair, MI muskellunge in this study. Muskellunge sampled from
the Lower Fox River, WI are primarily residents of Green Bay waters of Lake Michigan
except during spawning................................................................................................ 127
Figure 5.2. Distribution of cELISA percent inhibition values from sera of unexposed
muskellunge (n = 60) that were used to establish a positive-negative threshold of 14.6
% inhibition (dashed line). ............................................................................................ 128
Figure 5.3. Percent inhibition values by cELISA of muskellunge (n = 27) sampled
between 1 and 15 weeks after immersion challenge with VHSV-IVb. Samples with
percent inhibition values ≥14.6 (dashed line) are positive for antibodies to VHSV-IVb.
Horizontal lines represent the mean percent inhibition ................................................ 129
Figure 5.4. Distribution of percent inhibition values by cELISA of sera from 200 wild
muskellunge by sampling location and year. Samples with percent inhibition values
≥14.6 (dashed line) are positive for antibodies to VHSV-IVb ....................................... 130
Figure 5.5. Comparison between cELISA and 50% plaque neutralization test (PNT)
using sera from 38 wild muskellunge from a Lake St. Clair, Michigan (circles), 23
muskellunge from Lower Fox River, Wisconsin (diamond) and 27 experimentally
exposed muskellunge sampled 1-15 weeks post-exposure (triangles). Titers ≥14.6
percent inhibition (cELISA) and ≥160 (PNT) are considered positive for VHSV-IVb
antibodies. In order to display all titers, PNT titers were arbitrarily transformed by adding
two and resulting values plotted on a logarithmic scale ............................................... 131
Figure 6.1. Schematic. A pilot virus challenge was first carried out with a group of 28
unvaccinated muskellunge (Esox masquinongy) in order to determine the dose to be
used for the protection challenge (Table 6.1). On day 0, two trials were started. For the
protection trial, 90 fish (3 tanks of 30 fish each) were vaccinated with the pVHSivb-G
DNA vaccine and 90 fish (3 tanks of 30 fish each) were vaccinated with the empty
5

-1

plasmid vaccine. Fish were then challenged with 10 plaque-forming units mL VHSVIVb by immersion 7 weeks post-vaccination (PV). After 28 days, blood and tissues were
collected from surviving fish for virus titration (Table 6.2) and serology (Table 6.4). For
the antibody response trial, another group of fish was used. Blood was drawn nonlethally from 10-12 fish from each vaccine treatment group prior to vaccination, and on
week 7 and 11 PV for serological testing (Table 6.3). Fish were not challenged at any
point.............................................................................................................................. 159
Figure 6.2. Development of cumulative percent mortality (CPM) in triplicate tanks of
muskellunge (Esox masquinongy) vaccinated with pVHSivb-G DNA vaccine (squares)
or the empty plasmid (triangles) and challenged with VHSV-IVb 7 weeks (539 degree

xv

days) post-vaccination. The day of virus challenge is considered day 0. The pVHSivb-G
significantly reduced mortality compared to the control (p = 0.0089). The relative percent
a

survival (RPS) of pVHSivb-G vaccinated muskellunge was 45.2% ........................... 160
Figure 6.3. Virus titers in tissues of vaccinated muskellunge (Esox masquinongy) 28
days after virus challenge with VHSV-IVb. A-C designations represent triplicate tanks
within each treatment. Titers were transformed by adding one in order to depict titers of
-1

0 plaque forming units gram ...................................................................................... 161
Figure 7.1. Percent of sampled fish with virus in sera, tissues (kidney, spleen and/or
heart), and neutralizing antibodies over the course of VHSV-IVb infection and recovery.
3

-

Muskellunge were infected by immersion in a medium dose of the virus (4 × 10 pfu mL
1

). Four to five fish are represented at each time-point, except early in the disease
course (0 week = 3 fish, 0.5 week = 20 fish, 1 week = 8 fish). Mortality rate of fish dying
with clinical signs consistent with VHSV-IVb infection peaked between day 6 and 16
post-infection. This graph was added to emphasize that the combination of both
conventional virus isolation assays along with antibody testing allowed more consistent
detection of exposed fish throughout the disease course. ........................................... 181

xvi

INTRODUCTION

Viral hemorrhagic septicemia (VHS) is an emerging disease of wild freshwater fish in
the Great Lakes region of the United States and Canada. The hemorrhagic disease is
caused by VHS virus genotype IVb (VHSV-IVb), an RNA virus within the family
Rhabdoviridae and genus Novirhabdovirus (Dietzgen et al. 2012). Within the past
decade, the virus caused large-scale mortality events as it spread throughout naïve
populations of susceptible hosts (Kim & Faisal 2011a, Faisal et al. 2012). The virus was
isolated from many ecologically and economically important fish species such as
muskellunge (Esox masquinongy), freshwater drum (Aplodinotus grunniens) and
smallmouth bass (Micropterus dolomieu) (Faisal et al. 2012). Following initial fish kills in
Lake St. Clair in 2006, VHSV-IVb was not detected from fish in this lake during sampling
efforts for nearly three years until another die-off ensued (Faisal et al. 2012). A similar
trend was apparent in other water bodies. This raised questions both about the host
immune response to this virus and the diagnostic power of the traditional virus isolation
test.

Initial studies on VHSV-IVb were crucial for elucidating aspects of the disease course,
pathogenesis, and host range of this emerging virus (Al-Hussinee et al. 2010, Kim &
Faisal 2010a, b, Al-Hussinee et al. 2011, Goodwin & Merry 2011, Groocock et al. 2012).
Experimental studies confirmed the wide host range of the virus and revealed a great
diversity of susceptibilities among species and between individuals of the same species.

1

Such differences could reflect differences in host immune responses to the virus. Fish
that do survive VHS are protected against subsequent exposures (Bernard et al. 1983,
Kocan et al. 2001, Hershberger et al. 2010, Kim & Faisal 2012). If fish develop
protective immunity to VHSV-IVb, it might be hypothesized as the reason that mortality
events have generally diminished over time in wild populations where the virus is
established.

The study of the immune response is also critical from a pathogen surveillance
standpoint. Virus isolation is the currently accepted international standard for testing for
VHS. Tissues (and/or ovarian fluids) are infected on susceptible cell lines and isolated
viruses are confirmed, usually by molecular methods [World Organization for Animal
Health (OIE), 2012]. The process takes up to 28 days and furthermore, the sensitivity of
this method for detecting surviving or carrier fish is unknown (OIE 2012). It is known that
VHS virus is isolatable for a shorter time after infection compared to the duration of the
antibody response (Enzmann & Konrad 1993), and this is especially the case at warmer
water temperatures (Jørgensen 1992, Goodwin & Merry 2011). As such, screening fish
for virus-specific antibodies can increase the diagnostic ability for detecting rhabdoviral
exposure in fish that survive the disease (LaPatra 1996, Schyth et al. 2012). Following
outbreaks of VHS among cultured rainbow trout (Oncorhynchus mykiss) in Denmark,
Schyth et al. (2012) found that while virus isolation remained superior during the
outbreak, survivors were more likely to be detected as such based on the presence of

2

antibodies in the 4-5 month period that followed. Serological testing can also be done
non-lethally which is ideal for valuable and long-lived species.

The aim of this project was to develop serological reagents and indirect screening
assays for VHSV-IVb and use these tools to study the host immune response. In
particular, studies focused on the adaptive, humoral immune response of muskellunge
(Esox masquinongy). Muskellunge, a native Great Lakes species, are highly prized as a
sport fish and serve an important ecological role as a top predator in ecosystems where
they exist (Crossman 1986, Bozek et al. 1999). The species is considered vulnerable to
overexploitation and habitat alterations (Scott & Crossman 1973, Crossman 1986) and
the sustainability of some populations is reliant on continued human intervention
through stocking (Battige 2011). Muskellunge were one of the species that died in initial
die-offs due to VHSV-IVb in Lake St. Clair in 2006 (Elsayed et al. 2006) and Lake
Ontario (Lumsden et al. 2007) and there were concerns regarding the long-term effects
of VHSV-IVb on this species. Kim & Faisal (2010b,c) later confirmed experimentally that
VHSV-IVb is indeed highly virulent to this species.

The overall project goal was to better understand the host immune response to VHSV
genotype IVb. Chapter 1 presents a review of the literature on host immune response to
VHS virus and vaccination. In Chapter 2, the kinetics of the neutralizing antibody
response of experimentally exposed muskellunge was studied over the course of
disease and recovery. In Chapter 3, neutralizing antibody responses among 13 fish

3

species residing in a VHSV-IVb enzootic lake, Lake St. Clair, Michigan were assessed
over a multi-year period (2004-2010). Fish were sampled during, and in the months and
years following, VHSV-IVb positive mortality events and in some cases, comparisons
were made between serological and virus isolation results. Nine species from Lake St.
Clair were also screened for evidence of an immune response using a competitive
enzyme-linked immunosorbent assay (cELISA) (Chapter 4). The production of
polyclonal antiserum to VHSV-IVb and the development of the cELISA assay are
described in Chapter 5. Using muskellunge as an indicator for VHSV exposure based
on the presence of antibodies by cELISA, fish were tested from five water systems in
Michigan and Wisconsin (including VHSV negative and positive areas). As an additional
tool to study host immune responses, a VHSV-IVb DNA vaccine was developed
(Chapter 6). The vaccine was tested for its ability to induce innate and adaptive
protective immunity in rainbow trout (Oncorhynchus mykiss) and muskellunge,
respectively. Overall conclusions, areas of future research, and management
applications are addressed in Chapter 7.

4

CHAPTER 1
REVIEW OF LITERATURE

Viral hemorrhagic septicemia virus (VHSV) causes VHS, an internationally reportable
disease of fishes of coastal North America, Western Europe, Japan and Korea. VHSV
has an exceptionally broad host range compared to other fish rhabdoviruses, and
includes at least 80 species of cold, cool and warm-water fishes from fresh, brackish
and marine ecosystems (OIE 2012). Some fish species can become infected and
remain carriers of the virus without showing any clinical signs. The virus is highly
virulent in other species causing severe hemorrhagic disease and death. Within the past
decade (2003-2013), a novel sublineage of the North American genotype IV, designated
type IVb, emerged in the Great Lakes region of the US and Canada. Large-scale fish
kills marked the virus entry into new water systems, with a few known exceptions (e.g.
Lake Superior, Baseline Lake). VHSV-IVb was found to be capable of naturally infecting
many ecologically and economically important fish species and was quickly regarded as
a threat to Great Lakes commercial and recreational fisheries. The many unknowns
regarding potential effects of this disease on the health of wild and cultured Great Lakes
fish populations prompted a large investigation, in which this dissertation is a part. The
investigation aimed to a) compare the susceptibility of different fish species, b) identify
potential species involved in virus transmission (carrier fish hosts, invertebrates), c)
develop supplementary and non-lethal testing methods, and d) investigate host

5

immunity. In addition to providing information on the ability to predict what species of
fish may be most affected by this disease in the short and long-term, this research will
allow for more targeted surveillance efforts and provide a better understanding of how
the virus is maintained in water systems from year to year. The collection of studies in
this dissertation in particular focused on host immune responses to VHSV-IVb. Many
studies focused on muskellunge, Esox masquinongy (Order Esociformes, family
Esocidae), an important Great Lakes top predator, and also a prized sport fish, that was
involved in initial mortality events and was later confirmed experimentally to be highly
susceptible to the disease.

VIRAL HEMORRHAGIC SEPTICEMIA VIRUS

Structure and infection cycle. Viral hemorrhagic septicemia virus (VHSV) is a
member of the family Rhabdoviridae (order Mononegavirales) and the genus
Novirhabdovirus (Dietzgen et al. 2012). Rhabdoviruses are a diverse group of viruses
that include important pathogens of vertebrates, invertebrates and plants (Dietzgen et
al. 2012). All rhabdoviruses have a single molecule of linear, negative sense singlestranded RNA (ssRNA) that encodes five structural proteins. These proteins are
designated L (RNA-dependent RNA polymerase), G (glycoprotein), N (nucleoprotein), P
(phosphoprotein) and M (matrix protein) (Dietzgen et al. 2012). An additional nonstructural (or non-virion) protein (NV) characterizes viruses of the genus
Novirhabdovirus (Kurath & Leong 1985, Schütze et al. 1999). The genus

6

Novirhabdovirus contains viruses of fish including VHSV, infectious hematopoietic
necrosis virus (IHNV) of salmonids, Hirame rhabdovirus of Japanese olive flounder
(Paralichthys olivaceus) and Snakehead virus of warm-water species of Southeast Asia
(Dietzgen et al. 2012). Both IHNV and VHSV are reportable pathogens to the World
Organization for Animal Health (OIE).

Genes are arranged in the order 3ʼ-N-P-M-G-NV-L-5ʼ and non-coding leader and trailer
regions are present at the 3ʼ and 5ʼ ends respectively (Dietzgen et al. 2012) (Figure 1.1).
The N, P and L proteins together with the viral RNA form the nucleocapsid (Dietzgen et
al. 2012). G proteins are spike-like homotrimeric proteins that span the lipid bilayer and
are the only surface protein (Gaudin et al. 1992). Like other rhabdoviruses, the G
proteins of VHSV are a key protein involved in virulence and tropism because of its role
in initiating the infection cycle (Bearzotti et al. 1995). G proteins mediate virus
attachment to host cellular membrane receptors during the first step of the infection
cycle (Smail & Snow 2011). While the exact receptors used by VHSV and other fish
novirhabdoviruses is not known, evidence suggests that the receptor complex contains
fibronectin as a component (Bearzotti et al. 1999, Liu & Collodi 2002). The expression of
the host cell receptors used by a particular virus thus influence host range as well as
tissue tropism (Bearzotti et al. 1999). Adsorption is followed by receptor-mediated
endocytosis, fusion with primary lysosomes, and release of the nucleocapsid into the
host cell cytoplasm (Granzow et al. 1997). The virus uses its own RNA polymerase (L)
to replicate in the cytoplasm and proteins are produced using the host cellʼs translational

7

machinery (Dietzgen et al. 2012). The virion is assembled at the inner surface of the cell
membrane and the glycoprotein-studded envelope is acquired during budding from the
cell. The function of the NV protein is not fully defined. The protein is expressed in
infected cells but not present in mature virions of IHNV (Kurath & Leong 1985). The NV
gene product of IHNV was found to be essential for viral replication and it believed to
have a role in pathogenicity (Thoulouze et al. 2004).

Transmission and VHS disease. Transmission of VHSV is horizontal and spread by
direct contact with infected fish or water. VHSV is shed in urine and reproductive fluids
from infected fish (Neukirch 1985, OIE 2012). Transmission of the virus by ingestion of
infected fish has also been demonstrated experimentally in rainbow trout
(Oncorhynchus mykiss) (Schönherz et al. 2012). Unlike IHNV, there is no evidence that
VHSV can be vertically transmitted. Reservoirs of VHSV include clinically infected fish
as well as inapparently infected fish (covert carriers) (OIE 2012). VHSV-IVb has recently
been isolated from several invertebrate species including amphipods [Diporeia spp.
(Faisal & Winters 2011)] and leeches [Myzobdella lugubris (Faisal & Schulz 2009)].
Their role as potential reservoirs or vehicles for virus transmission is unclear at this
time.

Epithelial tissues of gill and skin, particularly near the base of fins (Harmache et al.
2006), are important sites of early viral replication and considered primary routes of
entry (Neukirch 1984, Yamamoto et al. 1992). Leukocytes and endothelial cells are

8

considered important cell types for virus replication (Wolf 1988, Yamamoto & Clermont
1990). The virus spreads in blood to early target organs including spleen and kidney
where it causes marked necrosis of hematopoietic tissue of the head kidney (reviewed
in Kim & Faisal 2011b, Smail & Snow 2011). As the kidney and spleen are main
immune organs, damage likely impairs host response to this virus. The cause of death
is blood extravasation leading to deprivation of vital tissues of oxygen. VHSV can cause
disease in all life stages of susceptible species; however generally mortality is highest in
younger fish and naïve adults (OIE 2012). Mortality caused by VHS (genotype I) in
freshwater rainbow trout for example often reaches 80-100% in fry and fingerling
rainbow trout (Olesen 1998) and typically less (30-70%) in adults (Skall et al. 2005).
The temperature range of VHSV-I and VHSV-IVb are similar; the virus has an optimum
of 9-12°C and an upper limit of 18-20°C (Goodwin & Merry 2011, OIE 2012).

Geographic distribution and genotypes. Prior to the 1980ʼs, VHSV was thought to be
strictly a disease of cultured rainbow trout in Western Europe, where it caused extensive
economic losses for the aquaculture industry (Wolf 1988, Olesen 1998, Snow et al.
2004). This was challenged when VHSV was isolated from coho (O. kisutch) and
chinook (O. tshawytsha) salmon returning to hatcheries in the Pacific Norwest region of
the U.S. in 1988 (Brunson et al. 1989, Hopper 1989). Subsequent investigations
revealed that the virus exists endemically among an extensive reservoir of free-ranging
marine and anadromous fish in the North Pacific Ocean off the western coast North
America (Meyers et al. 1992, Meyers & Winton 1995, Hedrick et al. 2003), the North

9

Atlantic Ocean, North and Baltic Seas around Europe (Dixon et al. 1997, Mortensen et
al. 1999, Smail 2000), as well as coastal Japan (Takano et al. 2000). Based on
historical evidence and genetic analysis of isolates, VHSV is considered to have
originated from a marine source and adapted several times to freshwater, where it
proved to be highly virulent to cultured rainbow trout in Europe (Einer-Jensen et al.
2004) and naïve populations of wild freshwater fish in the Great Lakes region of North
America.

Based on phylogenetic analyses, four main genotypes (I-IV) are recognized that cluster
based on geographic location (Snow et al. 1999, Einer-Jensen et al. 2004, Snow et al.
2004, Einer-Jensen et al. 2005, Pierce & Stepien 2012). Some genotypes are further
divided into multiple sublineages (Ia-Ie, IVa-IVc). Genotypes I-III are predominantly
isolates endemic among marine species around Western Europe, whereas genotype IV
consists of North American isolates, as well as a few from East Asia. Genotype Ia
contains the highly virulent freshwater isolates from cultured rainbow trout in western
Europe. Genotype III has caused losses of cultured turbot (Scophthalmus maximus) in
the British Isles. In North America, type IVa exists among marine species in the
northeastern Pacific Ocean, particularly in herrings and sardines (Meyers & Winton
1995, Hedrick et al. 2003). Natural epizootics due to VHSV-IVa have been documented
in small Pacific herring (Clupea pallasi) and several other marine fish species (Meyers
et al. 1999, Traxler et al. 1999). VHSV-IVa proved to be highly virulent in juvenile Pacific
herring under experimental conditions (Kocan et al. 1997). In older herring, the virus

10

establishes a chronic infection with mortality associated with stress (Meyers et al. 1994).
Isolates of IVa are found in wild and farmed marine fish species in Japan (Takano et al.
2000, Isshik et al. 2001, Nishizawa et al. 2002) and Korea (Kim et al. 2003). Type IVb
consists of freshwater isolates from the Great Lakes region of the U.S. and Canada.
The VHSV strain isolated from fish off the Atlantic coast of Canada in 2000 (Gagne et
al. 2007), was recently designated type IVc (Pierce & Stepien 2012).

VHSV-IVb in the Great Lakes. VHSV-type IVb emerged in the Great Lakes region of
North America within the last decade. The earliest known occurrence of this sublineage
is 2003 from muskellunge collected from Lake St. Clair, Michigan (Elsayed et al. 2006).
In 2006-2007, the virus was isolated from large-scale die-offs of free-ranging fish in
Lake Ontario (Groocock et al. 2007, Lumsden et al. 2007), the St. Lawrence River,
(Groocock et al. 2007), Lake St. Clair and Lake Huron (Faisal et al. 2012). The virus
then spread to all of the Great Lakes, to several inland lakes in Michigan, Wisconsin,
and New York, as well as to at least one site in the Mississipi watershed in Ohio
(reviewed in Kim & Faisal 2011a,b, Faisal et al. 2012). Analysis of partial G gene
sequences of over 100 Great Lakes isolates from 2003-2009 by Thompson et al. (2011)
revealed a low genetic diversity (maximum 1.05%), consistent with a recent introduction.

Species involved in wild fish die-offs, distribution and history of spread, pathogenicity,
and host range has been discussed previously (Kim 2010, Kim & Faisal 2010b, Kim &
Faisal 2011, Faisal et al. 2012). Currently, 28 fish are regulated under the VHS Federal

11

Order aimed at preventing spread of VHSV-IVb into aquaculture facilities (United States
Department of Agriculture, Animal and Plant Health Inspection Service 2008). These
fish were naturally infected with the virus, though some fish did not show clinical signs of
disease. Experimental studies by Kim & Faisal (2010a, b, c) have identified two
important determinants in the outcome of infection, which include the infective dose and
fish species. The virus is highly virulent in naïve juvenile fish of the following species:
Great Lakes muskellunge (family Esocidae) (Kim & Faisal 2010c), lake herring/cisco
(family Salmonidae; Coregonus artedii) (Weeks et al. 2011), and largemouth bass
(family Centrarchidae, Micropterus salmoides). The lethal dose 50 (LD50) by the
intraperitoneal (IP) injection route of infection was 2.2 plaque forming units (pfu) for
2

musky and 1.5 x 10 pfu for largemouth bass. Yellow perch (family Percidae, Perca
5

flavescens) are of moderate susceptibility (IP LD50 2.5 x 10 pfu). Most salmonids
(family Salmonidae), including rainbow trout, brook trout (Salvelinus fontinalis), brown
trout (Salmo trutta), chinook salmon, and coho salmon were relatively resistant (IP
6

7

LD50: 1.4 x 10 to > 7 x 10 pfu) (Kim & Faisal 2010a). Survivors may play a role in
maintaining the virus in the ecosystem (Kim and Faisal 2012, Faisal et al. 2012).
Preliminary studies in our laboratory suggested that salmonids shed the virus for an
extended period after infection, despite never displaying signs of disease (Shavalier &
Faisal, pers. com.). Studies by Kim and Faisal (2012) demonstrated that muskellunge
can indeed shed the virus for up to 15 weeks post-infection and after that, may resume
shedding upon exposure to stressors (e.g. handling).

12

IMMUNE RESPONSE OF FISH

Lymphoid tissues. Fish have both innate (non-specific) and adaptive (specific)
immune systems that are similar in form and in some case function to mammals, with a
few key differences. The immune tissues of teleost fish differ from mammals in that they
lack bone marrow, lymph nodes and germinal centers (Zapata & Amemiya 2000). The
anterior (head) kidney is the primary lymphoid tissue, a main site of hematopoiesis, and
the site of B lymphocyte development, whereas the thymus is the primary T lymphocyte
tissue (Fange 1986, Rombout et al. 2005, Zapata et al. 2006, Todo et al. 2011, Roberts
2012). The spleen and kidney both have a reticuloendothelium system for filtering and
trapping pathogens, and are both sites of lymphocyte activation of B and T cells (Zwollo
et al. 2005, Solem & Stenvik 2006, Ye et al. 2011). Melanomacrophage centers
throughout the reticuloendothelial tissue house many macrophages that clear material
from circulation, including immune complexes. Teleosts have mucosa-associated
lymphoid tissue (MALT) in the skin, gill, and gut, although organized germinal centers
(in skin, or similar to Peyerʼs patches in mammalian gut) have not been identified
(reviewed in Salinas et al. 2011). The gills contain phagocytic cells along brachial
capillaries and lymphocytes (Roberts 2012). Mucus coating epithelial tissue of skin, gill,
and gastrointestinal tract is a protective barrier for physically trapping and sloughing
organisms to prevent colonization (Roberts 2012).

13

Innate immunity. The innate immune system is an ancient form of host defense
against infection consisting of barriers, soluble (humoral) substances, and cells. Innate
immunity in fish has been the subject of several thorough reviews (Ellis 2001,
Magnadottir 2006, Whyte 2007, Gomez & Balcazar 2008). The innate response is
based on a system of evolutionarily conserved host receptors expressed on cell
surfaces, inside cellular compartments and on soluble molecules that distinguish
microbes from host products (Medzhitov & Janeway 1997). Pattern recognition
receptors (PRRs) bind molecules called pathogen-associated molecular patterns
(PAMPs) present only on microbes, thus giving the host the ability to distinguish an
invading pathogen from self molecules and cells. This system of recognition allows for a
rapid “non-specific” response in the absence of prior exposure. Engaged PRRs activate
innate pathways such as phagocytosis, complement, coagulation, inflammation and
apoptosis (Medzhitov & Janeway 1997).

The skin and mucus layer contain numerous protective humoral substances that inhibit
pathogen entry and replication (e.g. lysozyme, lectins, proteases, complement) and
many of these substances are also present in serum and tissue fluids (Ellis 2001, Fast
et al. 2002, Esteban 2012, Najafian & Babji 2012). In some cases, the repertoire of
innate humoral substances of fish is more diverse than that of mammals (Vasta et al.
2011, Nakao et al. 2011, Roberts 2012). It is thought that such unique structural and
functional diversity of the teleost innate humoral immunity could allow a broader
recognition of pathogens during the innate response. In some cases, innate humoral

14

molecules may cause non-specific neutralization or binding to viruses in assays.
Lectins are humoral molecules characterized by the presence of a carbohydrate
recognition domain that binds sugar and glycoprotein PAMPS (Vasta et al. 2011).
Ladderlectin, isolated from plasma of rainbow trout, was shown to bind directly to
VHSV-IVb in vitro (Reid et al. 2011). Another lectin, the 6S inhibitor of infectious
pancreatic necrosis virus (IPNV) is present in sera of some salmonids without prior
exposure to IPNV (Jørgensen 1973, Dorson & de Kinkelin 1974, Park & Reno 2005),
and this substance can result in neutralization of IPNV in cell culture by preventing viral
adsorption (Kelly & Nielsen 1985, Park & Reno 2005).

Cells of innate immunity of teleosts include macrophages, granulocytes (neutrophils,
basophils, eosinophils, eosinophilic granular cells), nonspecific cytotoxic cells, and
dendritic-like cells. Macrophages and neutrophils are specialized phagocytic cells that
kill pathogens directly, and are also important in initiating inflammation including
releasing cytokines to increase mobilization of additional phagocytes (Roberts 2012).
Cytokines involved in fish immunity have been recently reviewed (Alejo & Tafalla 2011).
Macrophages can be found throughout the body but are concentrated in kidney, spleen
and in some species, atrium of heart, as reticuloendothelial cells (Roberts 2012).
Interestingly, some B cells of fish also have potent phagocytic activity suggesting their
role in innate as well as adaptive responses (Li et al. 2006, Zhang et al. 2010). In
mammals, natural killer (NK) cells are the effector cells of innate anti-viral immunity. NKlike cells that kill virus-infected and tumor cells have been found in several fish species

15

(Fischer et al. 2006, Nakanishi et al. 2011). Several other types of cells with non-specific
cytotoxic activities have been identified in fish (reviewed in Nakanishi et al. 2011).

Adaptive (acquired) immunity. Fish are the most primitive vertebrates to have an
adaptive immune system. The major components of adaptive immunity include
lymphocytes (B and T cells), major histocompatibility complex (MHC) molecules, T and
B cell receptors, and immunoglobulins (Iwana & Nakanishi 1996, Roberts 2012). Upon
first encounter with an antigen, the adaptive response takes time to develop as antigenspecific cells are selected, proliferate, and undergo changes in affinity. In mammals,
dendritic cells are professional antigen presenting cells (APCs) that have a primary role
in antigen presentation to T cells, thereby linking innate and adaptive responses
(Banchereau et al. 2000). T cells can be divided into two main functional groups.
Cytotoxic (killer) T cells express CD8 co-receptor, interact with endogenous antigens
(generated inside cells) presented on MHC Class I molecules, and kill abnormal cells by
inducing apoptosis or releasing molecules. Helper T cells express co-receptor CD4,
recognize exogenous antigens presented in the context of MHC Class II by APCs, and
function to activate and coordinate other immune cells, including B cells, through
cytokine production. B cells secrete antibodies (plasma B cells) or differentiate into longterm memory B cells. Antibodies function in direct neutralization of pathogens, or
opsonize pathogens for removal by phagocytes. T cells can be further divided into
several effector subsets based on cytokine production and cell surface markers.
Functional and gene expression evidence to date suggest the presence of helper and

16

cytotoxic T cells in teleosts and similar activation pathways (Fischer et al. 2006, Toda et
al. 2011). However, physical characterization of cells involved in the adaptive immune
response of fish has been hindered compared to mammals due to lack of specific
reagents (Nakanishi et al. 2002, Fischer et al. 2006, Nakanishi et al. 2011, Esteban et
al. 2012, Verrier et al. 2011). Functionally analogous dendritic cells have been recently
defined in teleosts (tDC) (Bassity & Clark 2012). In addition to being morphologically
similar to mammalian DCs, this cell type induced a potent, proliferative mixed leukocyte
response and was phagocytic and migratory suggesting its role as a specialized APC
(Bassity & Clark 2012). Recent advances have characterized many T-cell associated
proteins in multiple species of fish including surface co-receptors (CD8, CD4), proteins
associated with signal transduction (e.g. CD3 complex), and many secreted cytokines
(Laing & Hansen 2011). One study recently showed proliferation of CD4+ T cells in
ginbuna crucian carp (Carassius auratus langsdorfii) after allogenic and antigen-specific
stimulation (Toda et al. 2011), supporting that CD4+ T cells in fish are analogous to
mammalian CD4 + helper T cells (Laing & Hansen 2011). The finding of sequence
homologues for MHC Class I, TCR and CD8 coreceptor in several fish species suggests
that antigen presentation to cytotoxic T cells is also similar to what occurs in higher
vertebrates (Utke et al. 2007). More studies that link protein expression with function are
needed.

The main humoral component of the adaptive immune response is immunoglobulin,
which is secreted by B cells. In teleosts, immunoglobulin M (IgM) is the predominant

17

immunoglobulin type in plasma and the main one associated with systemic responses
(Kaattari & Piganelli 1996). IgM has multiple effector functions including neutralization,
precipitation and agglutination, opsonization and complement-mediated functions
(Roberts 2012). Two additional immunoglobulin classes including IgD (Wilson et al.
1997, Edholm et al. 2010) and IgT (also called IgZ, and analogous to mammalian IgA)
(Danilova et al. 2005, Hansen et al. 2005) have more recently been identified in several
species of teleosts. Studies have shown IgT to be an important immunoglobulin in the
gut mucosa (Zhang et al. 2010) and skin mucus of rainbow trout (Bordon 2013). The
role of IgD is not yet elucidated (Castro et al. 2013). Multiple B cell subsets exist in fish
including B cells expressing only IgM, IgD or IgT and one subset that expresses both
IgM and IgD (Salinas et al. 2011). Recent studies using pyrosequencing and CDR3length spectratyping have shown that VHSV infection significantly changes profiles of B
cells in spleen, which is evidence of clonal expansion of B cells (Castro et al. 2013).
Interestingly, both immunoglobulin M (IgM)+ and IgT+ B cells responded. This evidence
suggests an initial relationship between IgT and the anti-viral response (Castro et al.
2013).

Immune memory occurs in teleosts but does not share all the features associated with
immune memory in mammals (Kaattari 1992, Ma et al. 2013). Characteristics of fish
memory response associated with subsequent exposures to an antigen include a more
rapid antibody response, higher antibody titers, and increased sensitivity to antigens
(reviewed in Kaattari 1992). The overall increase in magnitude of antibody titers

18

between the primary and secondary response is not as profound in teleosts as it is in
mammals, though this can vary based on temperature (Roberts 2012). In teleosts,
isotype-switching does not occur during affinity maturation as it does in mammals
(Stavnezer et al. 2008). Instead, distinct B cell subsets produce the different isotypes
(e.g. IgM, IgT) as discussed above. Affinity maturation of IgM antibody in rainbow trout
involves a gradual, and slower switch from lower to higher affinity antibodies, which
corresponds to a higher degree of disulfide polymerization of IgM (Ye et al. 2011). Ye et
al. (2011) showed that low affinity, low titer antibody subpopulations are progressively
replaced by higher affinity subpopulations that have higher titers and appear later. The
highest affinity antibodies appeared 15 weeks after exposure to TNP-keyhole limpet
hemocyanin and remained elevated through 27 weeks when the study period was
terminated.

IMMUNE RESPONSE TO VIRAL HEMORRHAGIC SEPTICEMIA VIRUS

Innate immune response of fish to VHSV. VHSV activates a robust innate antiviral
immune response followed by an adaptive response and induction of immunological
memory. The ability of the host to prevent and limit virus infection initially by physical
and chemical barriers is an important determinant in the outcome of disease (reviewed
in Esteban 2012). The essential role of external barriers is evidenced by the fact that
much higher doses of VHSV-IVb by immersion are required to produce disease and
mortality compared to the intraperitoneal route (Encinas 2010, Kim & Faisal 2010 a,b,c).

19

Even between individuals of a susceptible species, huge variation exists in their ability
to prevent the development of clinical infection. The genetic component of host
resistance was recently studied using a collection of rainbow trout clones with
differential susceptibility to VHSV by immersion (Verrier et al. 2012). The study revealed
that mechanisms determining host resistance were indeed associated with
innate/intrinsic mechanisms and not specific host immunity. In particular, the induction
of interferon by some clones prevented early viral replication whereas this response was
absent in cells from susceptible fish. Induction of a systemic antiviral state by cytokine
release may occur following early VHSV replication in epidermal cells of fin tissues
(Quillet et al. 2001, 2007).

The innate antiviral response of fish to VHSV is mediated primarily by cytokines,
notably, the type I (α/β) interferon (IFN) system and its inducible proteins (Verrier et al.
2011, Purcell et al. 2011). VHSV stimulates a rapid and robust type I IFN response by a
variety of host cells (Tafalla et al. 2008, Hansen et al. 2012, Lovy et al. 2013). The IFN
system of fish has been recently reviewed (Zou & Secombes 2011). Double-stranded
RNA intermediates produced during early viral replication (Smail & Snow 2011) and viral
surface glycoproteins of VHSV and IHNV are both potent inducers of the IFN response
in fish (Acosta et al. 2006, Verjan et al. 2008, Smail & Snow 2011). It is host Toll-like
receptors or other PRRs that recognize these conserved viral patterns (Purcell et al.
2011). IFNs induce an antiviral state in neighboring cells through regulating the

20

transcription of numerous genes including proteins that inhibit viral replication (Ellis
2001, Murphy et al. 2008).

The IFN response and associated changes in host gene expression are studied by
quantitiative PCR or subtractive suppressive hybridization (Verrier et al. 2011). Studies
have demonstrated a “core” set of interferon-stimulated genes (ISGs) involved in the
antiviral response to rhabdoviral infection or DNA vaccination, many of which are
conserved among vertebrates (reviewed in (Verrier et al. 2011). VHSV is a potent
inducer of the intracytoplasmic “VHSV and interferon-induced genes” vig1 (aka viperin)
vig2 and ISG15/vig3 as well as Mx 1-3 genes. Mx increases in the liver after infection of
yellow perch with VHSV-IVb and upregulation was correlated with increasing viral load
(Olson et al. 2013). In mammals, Mx proteins have broad range anti-viral activities
against RNA viruses (Leong et al. 1998) and direct antiviral activity of teleost Mx
proteins have been confirmed by transfection experiments (Caipang et al. 2003). Virusinfected cells also produce secretable antiviral factors (e.g. chemoattractants Vig7, Vig8,
and galectin 9). In response to DNA vaccination, the upregulation of these factors is
similar, and timing of peak levels corresponds to the timing of the non-specific/crossprotective phase that occurs within a week after vaccination (Purcell et al. 2004, 2012).

The antibody response to VHSV. The specific antibody response to VHSV is best
characterized in terms of the neutralizing antibody response. Early studies showed that
rainbow trout exposed to low-virulence strain of VHS [F1, Denmark (Jensen 1965);

21

genotype Ia] showed enhanced resistance against subsequent challenge with a virulent
strain (Jørgensen 1976). Early studies also demonstrated the presence of virusneutralizing antibodies in serum of rainbow trout after experimental and natural infection
(Jørgensen 1971, de Kinkelin et al. 1977, Dorson & Torchy 1979, Jørgensen 1982,
Olesen & Jørgensen 1986). Neutralization is mediated by IgM antibodies based on
studies that demonstrated rabbit antiserum against trout IgM inhibits the neutralization
reaction (Olesen & Jørgensen 1986). The reaction of neutralizing antibodies against
neutralizing epitopes of virus G proteins is measured by the 50% plaque neutralization
test (PNT) and serum neutralization test (SNT). The in vitro neutralization reaction
between VHSV-neutralizing antibodies and the virus is enhanced by the addition of
complement, a heat-labile molecule, which is supplied by adding unheated, naïve
rainbow trout sera into the reaction (Dorson & Torchy 1979). Serum being tested for
antibodies is first heated to 45°C for 20-30 minutes in order to destroy any pre-existing
complement that may be bound to antibodies in the serum (Dorson & Torchy 1979,
Sakai 1981). The specific mechanism by which complement aids in neutralization still
remains unclear (Purcell et al. 2012) but some results suggest that the classical
complement activation pathway is involved (Lorenzen et al. 1999). The complement
system is a collection of serum proteins that is a powerful innate defense (reviewed in
Boshra et al. 2006). Complement may facilitate direct lysis of enveloped viruses
(Lorenzen et al. 1999), aid neutralizing antibodies in the formation of immune
complexes or inhibit virus attachment to cell receptors (Pier et al. 2004).

22

The protective nature of specific antibodies against VHSV has been demonstrated by
passive transfer experiments. Rainbow trout immunized with neutralizing serum, or the
purified immunoglobulin portion of serum from challenge survivors (Bernard et al. 1983),
were protected against VHSV challenge (de Kinkelin et al. 1977, Lorenzen et al. 1999).
Subsequent studies demonstrated that antibodies against the G proteins mediated
protection (Lorenzen et al. 1990). Rainbow trout injected with monoclonal antibodies
against N, M1 (now phosphoprotein), and M2 (now matrix) proteins in contrast, were not
protected (Lorenzen et al. 1990). It is know well established that the G proteins of VHSV
and IHNV are the major surface antigen that elicits the production of neutralizing
antibodies (Lorenzen et al. 1990, Bearzotti et al. 1995). The time course of
development of neutralizing antibodies after experimental infection has been studied
primarily in rainbow trout after infection with VHSV-genotype I. Neutralizing antibodies
begin to appear by 2-4 weeks after infection with peak responses occurring by 6-10
weeks after infection at 10°C (reviewed in Lorenzen & LaPatra 1999). Because the
antibody response does take several weeks, the role of neutralizing antibodies in
protecting naïve fish from a primary, acute infection is unlikely (Lorenzen & LaPatra
1999, Purcell et al. 2012). Neutralizing antibodies after VHSV infection have been
reported to last at least a year in serum of survivors (Olesen & Jørgensen 1986).

Non-neutralizing antibodies, or binding antibodies, are also induced after rhabdoviral
infection. Various assays have been developed for measuring non-neutralizing
antibodies against VHSV including ELISA, western blot (Lorenzen et al. 1993),

23

immunofluorescence (Olesen et al. 1991, Jørgensen et al. 1991) and counter-current
immunoelectrophoresis (Enzmann et al. 1993, Enzmann & Konrad 1990, 1993). ELISA
assays used immunoglobulin-captured virus as the coating phase (Jørgensen et al.
1991, Olesen et al. 1991, Fregeneda-Grandes & Olesen 2007), viral-infected cell culture
supernatants (Kim et al. 2008), or recombinant G protein fragments (Encinas et al.
2011). Detection of rainbow trout serum antibodies against VHSV were detected using
anti-rainbow trout antibodies or antisera.

In several studies where PNT and ELISA were performed in parallel, more survivors
were detected by ELISA-based techniques (Jørgensen et al. 1991, Olesen et al. 1991,
Encinas et al. 2011). It was hypothesized by Olesen et al. (1991) that this is because
with ELISA, more types of antibodies can be detected. PNT detects only neutralizing
antibodies, or those reacting against neutralizing epitopes on the G proteins. Olesen et
al. (1991) hypothesized that ELISAs (and immunofluorescence techniques) would
measure additional populations of antibodies such as those against non-neutralizing
areas of the G proteins, as well as against other antigens exposed on the plate: L (RNAdependent RNA polymerase), G (glycoprotein), N (nucleoprotein), P (phosphoprotein,
formerly M1) and M (matrix protein, formerly M2), however, this was not characterized in
that study. From studies with IHNV, it seems that the most prevalent antibodies after
rhabdovirus infection are against G and P proteins, in which G is embedded. Using
western blot, Ristow (1993) characterized antibodies against all five structural proteins
of IHNV in sera of rainbow trout after IHNV infection. Antibodies were detected against L

24

(RNA-dependent RNA polymerase), G (glycoprotein), N (nucleoprotein), M1 (now P,
phosphoprotein) and M2 (now M, matrix protein). In fish that had only been exposed
once to IHNV, the M1 (now P) protein was the most commonly detected protein by
western blot, while in fish that had repeated exposures to IHNV, antibodies against the
G protein were the most frequently produced.

Neutralizing antibodies appear to decline sooner in rainbow trout after VHSV (genotype
I) infection than do non-neutralizing antibodies. For example, Olesen et al. (1991)
experimentally infected rainbow trout with VHSV-I at 0 and 14 weeks. The highest titers
of both PNT and ELISA (using Ig-captured VHSV infected-cell culture supernatants)
were detected by 10 weeks post-infection. Between 18-23 weeks (the last sampling
point), there was a decline in PNT titers (to 80) while ELISA titers increased. However,
the fact that fish in this study received a second exposure at 14 weeks post-infection
should be taken into account. Encinas et al. (2011) found a similar trend of PNT
antibodies being shorter-lived using ELISA that employed purified fragment of
recombinant G proteins of VHSV-genotype I as coating phase (Encinas et al. 2011).
Between 4 and 10 weeks after infection, the percentage of fish detected as survivors
based on ELISA increased to 100%. In contrast, in that same time period, the
percentage positive decreased from 65% to 35%. Of sera testing positive by only one
assay, 40% of sera were PNT negative, ELISA positive and only 3% were PNT positive,
ELISA negative.

25

In general, less is known about protection induced by non-neutralizing antibodies
against fish rhabdoviral infections. Some studies have shown that non-neutralizing
antibodies can also be protective (Lorenzen et al. 1990). The immune response of other
species of fish besides rainbow trout against VHSV is also in general less studied.
Some species do not mount a neutralizing antibody response, at least not in levels
detectable by standard PNT assays. Pacific herring surviving infection with VHSV-IVa
are protected upon subsequent exposure, and passive transfer of sera produces
protection, yet neutralizing antibodies are not detected by standard neutralization
assays (Kocan et al. 2001, Hershberger et al. 2007, Hershberger et al. 2011). It is
possible that the neutralizing antibody response in some species could be reduced or
absent.

VHSV immune response and temperature. Immune responses of fish to infectious
organisms are influenced by host factors (e.g. age, health status, genetic variability) and
external factors such stress, light level, and temperature (Bly et al. 1997, Le Morvan et
al. 1998, Kollner et al. 2002). Temperature in particular has been shown to affect the
host response and disease course following VHSV infection. VHSV and other fish
rhabdoviruses are generally considered “hit and run” viruses; that is, they cause severe,
acute disease resulting in high mortality of their host (Purcell et al. 2012). However,
VHSV can also establish a long-term, persistent disease state in asymptomatic, carrier
fish, which is usually associated with low temperatures and virus infection of neural
tissues (Wolf 1988, Smail & Snow 2011, Purcell et al. 2012). Persistent infections with

26

VHSV have been demonstrated up to 400 days in rainbow trout held at low water
temperatures of 4°C (Jørgensen 1982, Neukirch 1984, 1985) or up to 200 days in
Pacific herring infected with VHSV-IVa (Hershberger et al. 2010). The ability of VHSV to
cause recurring outbreaks of disease may be associated with persistence of the virus in
carrier fish (Purcell et al. 2011). At higher temperatures, the virus is cleared more rapidly
(Jørgensen 1982, Goodwin & Merry 2011).

The immune system of teleosts functions through a range of permissible temperatures,
with a speciesʼ optimal immune response occurring at that speciesʼ normal summer
temperature (Ellis 1988, Bly & Clem 1992, Ellis 2001, Magnodittir 2006). Temperatures
below a speciesʼ permissible range tend to have an immunosuppressive effect on the
immune system by inhibiting or delaying responses, particularly, the development of
adaptive immunity. The priming phase of the adaptive response is perhaps the most
sensitive to non-permissible temperatures, due to slower primary T helper and CTL
response, however once memory cells are generated, they are not impaired (reviewed
in Iwama & Nakanishi 1996). At colder temperatures, innate responses are prolonged to
compensate for the suppression of the adaptive response, however, antibodies are
eventually produced even at colder temperatures (Bly & Clem 1992, Le Morvan et al.
1998). For example, in a study by Lorenzen et al. (2009), rainbow trout were vaccinated
with the VHSV or IHNV DNA vaccine and held at 5°C, 10°C or 15°C for 40 days prior to
VHSV challenge. Neutralizing antibodies were produced by a higher proportion of fish,
and in higher titers, when fish were kept at their optimal temperature (15°C). In contrast

27

antibodies were completely absent by this time in fish held at 5°C. In this same study,
fish vaccinated with IHNV- DNA vaccine were only cross-protected against VHSV
challenge if they were acclimated to the colder temperatures (5-10°C). Since crossprotection is mediated by innate anti-viral immunity, these results suggest that at colder
temperatures, innate responses lasted longer.

It has been suggested that at colder temperatures, the maintenance of the virus or
antigen for a longer period of time, actually prolongs the induction of the adaptive
response during the priming phase, leading to an even more robust response when it
does develop (Jørgensen 1982, Lorenzen et al. 2009). Fish acclimated at 10°C to
VHSV, followed by transferring of fish to a range of temperatures, found that antibody
titers and protection were higher by 9 week PV in fish that were actually moved to colder
temperatures. These studies suggest that vaccination against rhabdoviruses seems
optimal in terms of long-term protection if fish are vaccinated and kept at the low end of
permissible temperatures in order to prolong the period of time in which antigen is
presented to the immune system (Lorenzen et al. 2009).

DNA VACCINATION AGAINST VHSV AND THE ANTIBODY RESPONSE

DNA vaccination technology. DNA vaccination is defined as the intentional transfer of
genetic material to somatic cells for the purposes of influencing the immune system
(Tonheim et al. 2008). Genetic material from a pathogen of interest is introduced by

28

means of recombinant DNA technology into an expression vector (plasmid). The
plasmid is used to deliver the gene of interest to host cells. The plasmids used in DNA
vaccines can be described as consisting of two units: the antigen expression unit and
the production unit (Schirmbeck & Reimann 2001; Figure 1.1). The antigen expression
unit consists of the promotor/enhancer, antigen sequence, and polyadenylation
sequences needed for expression of the protein by the vaccinated host. The production
unit consists of elements necessary for amplification of the plasmid in bacteria as well
as selectable markers (antibiotic resistance genes). The plasmid is replicated in
Escherichia coli (E. coli) to generate large amounts and then is subsequently purified
back out for use as a DNA vaccine.

DNA vaccination against fish rhabdoviruses. DNA vaccinations have been
investigated for many fish pathogenic viruses, bacteria, and parasites (reviewed in
Tonheim et al. 2008). The most successful DNA vaccines for aquatic animals, and
animals in general, are those against fish rhabdoviruses (Lorenzen & LaPatra 2005,
Kurath et al. 2007). Anderson et al. (1996) first reported the efficacy of a DNA vaccine in
fish. Rainbow trout vaccinated with a plasmid containing the glycoprotein (G) gene of
IHNV produced antibodies and were significantly protected from lethal challenge with
the virus. The basic vaccine construct consisted of the G gene of IHNV located
downstream of the cytomegalovirus (CMV) promoter in the eukaryotic expression vector
(pcDNA3, Invitrogen). Vaccines containing the same design but containing the G gene
of VHSV genotype I were soon developed and proved highly efficacious in protecting

29

rainbow trout from lethal challenge with VHSV (Lorenzen et al. 1998, Heppell et al.
1998). A number of DNA vaccines based on this same construct design have since
been developed for other strains of both IHNV and VHSV and their efficacy shown in
other fish species besides rainbow trout including chinook, sockeye (Oncorhynchus
nerka) and Atlantic (Salmo salar) salmon (reviewed in Kurath et al. 2007). Effective DNA
vaccines for other fish rhabdoviruses have also been developed. These include
vaccinations against the North American spring viremia of carp virus (SVCV) [isolate
SVCVnc (Goodwin 2002)] in koi (Cyprinus carpio koi) (Emmenegger & Kurath 2008)
and Hirame rhabdovirus in Japanese flounder (Paralichthys olivaceus) (Takano et al.
2004). Vaccine efficacy is commonly expressed in terms of relative percent survival
(RPS) which is calculated according to the following formula: [1-(average CPM of
vaccinates/average CPM of negative controls)] × 100 (Amend 1981).

In typical laboratory trials of the IHNV and VHSV DNA vaccines, duplicate or triplicate
groups of fish are given a single injection (intramuscularly in skeletal muscle) of a small
quantity of plasmid DNA (generally 0.1 to 1 µg) and challenged by immersion 4-10
weeks later. Virus doses used for challenge vary based on virus, virus strain, size and
species of fish, however, doses consistently produce high cumulative percent mortality
(75-100%) in control (empty plasmid)-vaccinated fish virus (Kurath et al. 2007). In
salmonids, DNA vaccines are highly efficacious even under these lethal challenge
circumstances, providing nearly 100% protection. In 2005, the first DNA vaccine,
Apex®-IHN (Novartis), was approved for use in Canada for the prevention of IHN in

30

farmed Atlantic salmon (Salonius 2007). In Europe and in US, the commercialization of
the Apex-IHN vaccine as well as other DNA vaccines in fish are restricted due to safety
concerns (Alonso & Leong 2013).

Following intramuscular injection of IHNV/VHSV DNA vaccines, some pDNA is injected
directly into cells (e.g. myocytes, antigen presenting cells) and some goes into the
extracellular compartment between cells. Extracellular pDNA is taken up by surrounding
cells and/or re-distributed from the site of administration. Myocytes in particular at the
site of immunization are important for initial plasmid uptake and subsequent antigen
expression (Boudinout et al. 1998, Garver et al. 2005, Tonheim et al. 2007). Host cellsʼ
use their own translational machinery to transcribe the pDNA into mRNA and then
translate the mRNA into protein (reviewed in Tonheim et al. 2008). This in vivo
expression of antigenic proteins that occurs following DNA vaccination ensures the
vaccinated animal “sees” the antigen in its native conformation with all proper posttranslational modifications (Schirmbeck & Reimann 2001). This aspect is important for
maintaining epitope integrity and for generating neutralizing (protective) antibodies
(Schirmbeck & Reimann 2001). Immediate and transitory detection of plasmid DNA in
blood following injection suggests that redistribution of pDNA to other sites is likely by
means of the circulatory system (Garver et al. 2005, Tonheim et al. 2007). Upregulation
of specific receptors used by dendritic cells at the site of VHSV G injection, suggest
recruitment of this cell type and their role as antigen-presenting cells of plasmid DNA
(Ortega-Villaizan et al. 2009).

31

Three phases of antiviral immunity. Expression of the G protein by host cells
induces early, non-specific immune responses that are later replaced by a long-lasting,
specific immunity. The immune response of salmonids to IHNV-G and VHSV-G DNA
vaccination has been described in terms of three sequential but interrelated phases of
immunity (Lorenzen & LaPatra 2005, Kurath 2005, and as reviewed in Kurath et al.
2007). These phases are the early antiviral response (EAVR), the specific antiviral
response (SAVR) and the long-term antiviral response (LAVR).

The EAVR phase is characterized by a rapid onset of protection and low specificity. In
rainbow trout immunized with DNA vaccines encoding the G protein of IHNV and VHSV,
protection is established within days after vaccination. Significant levels of protection
compared to control fish occur as early as 4 days (60 degree days, 15°C) postvaccination (PV) in rainbow trout immunized with the pIHN-G (1 µg) and subsequently
challenged with IHNV. Similarly, a high level of protection occurred in VHSV-G
vaccinated fish challenged only 8 days after vaccination (Lorenzen et al. 2000).
Numerous studies have confirmed the ability of IHNV and VHSV-G DNA vaccines to
also provide cross-protection at early time points against fish viruses other than those in
which they were vaccinated against. For example, this type of early cross-protection
was observed in rainbow trout vaccinated with pVHS-G and challenged with IHNV at 4,7
and 14 days (60, 105, 210 degree days) but not by 28 days PV (420 degree days)
(LaPatra et al. 2001).

32

Early phase protection is mediated by non-specific antiviral immune mechanisms such
as IFN production. The exact mechanisms of early protection and cross protection
against fish rhabdoviruses following DNA vaccination are still being investigated but a
characteristic transcriptional pattern is upregulated (Purcell et al. 2012). Transcriptional
upregulation of Mx and IFN regulatory factors 1 and 2 (IRF-1 and IRF-2) prove that like
VHS virus, IHN and or VHS DNA vaccination induce a type I interferon response
(Boudinot et al. 1998, Kim et al. 2000, McLauchlan et al. 2003, Purcell et al. 2004).
Recent analysis by cDNA microarray has demonstrated upregulation of other humoral
defense genes (e.g. complement C3, complement regulatory proteins, MHC class II
associated molecules) (Byon et al. 2006).

The specific antiviral response (SAVR) phase is characterized by high RPS against the
homologous virus against which trout were vaccinated and the presence of neutralizing
antibodies (Kurath et al. 2007). The shift from the EAVR to this specific phase of
protection occurs between 2-4 weeks PV (Lorenzen et al. 2002a, LaPatra et al. 2001).
Cross-protection with the IHNV and VHSV no longer occur in DNA-vaccinated rainbow
trout challenged after 4 weeks post-vaccination (Kurath et al. 2007). While some
neutralizing antibodies can be detected after rhabdovirus DNA vaccine as early as 23
days PV in a few fish (Boudinot et al. 1998), high seroprevalence can be expected
between 6-12 weeks PV in sera of rainbow trout, chinook salmon and Atlantic (reviewed
in Kurath et al. 2007). Challenged fish show enhanced neutralizing antibody response in

33

comparison to control fish in several studies suggesting that vaccination primes the
neutralizing antibody response (Kurath et al. 2007). Though most studies test
neutralizing antibodies, Byon et al. (2006) demonstrated ELISA binding IgM antibodies
in sera of Japanese flounder after VHSV-I (isolate KRRV9822) DNA vaccination.

The kinetics of neutralizing antibody development and protection are dependent on
temperature and vaccine dose. The dose-response to VHS DNA vaccine was reviewed
in Lorenzen et al. (2000). Fish given higher (but still low) doses of the vaccine (1 or 0.1
µg) were significantly protected compared to fish given only 0.01 µg. Similarly, LaPatra
et al. (2000) found that 1 µg of IHN DNA vaccine induced complete protection but a
lesser dose (0.1 µg) did not protect 150 g fish. Some studies have found that higher
doses of the vaccine result in higher seroprevalence and titers (Kurath et al. 2007).
Furthermore the timing of the virus challenge in fish given a constant dose affected
protection. Protection ranged from 80-97% relative percent survival (RPS) between 3
weeks and 4 months (at 11-14°C), but protection was reduced by 5.6 months PV (64%
RPS). Interestingly, the level of protection in many studies is higher than the proportion
of fish with detectable neutralizing antibodies (McLauchlan et al. 2003, reviewed in
Purcell et al. 2012) or ELISA-antibodies (Byon et al. 2006). These findings lead authors
to conclude that cell-mediated immunity is likely involved in protection as well, or that
neutralizing/non-neutralizing antibodies below assay detection limits are still sufficient to
provide protection. Indeed, passive transfer of neutralizing sera diluted to the point

34

where it is no longer neutralizing in vitro, still provides protection after challenge
(LaPatra et al. 1994).

The long-term antiviral response (LAVR) is used to refer to a phase months to years
after fish rhabdoviral vaccination. It is thought that DNA vaccines induce long-term
immunological memory in fish and protection mediated during the LAVR is due to
effector cells of adaptive immunity (Kurath et al. 2007). This phase is the least well
characterized. Studies have reported that after VHSV DNA vaccination, neutralizing
antibody levels begin to decline 3-6 months PV and are completely absent by 9 months
PV (McLaughlin et al. 2003). Kurath et al. (2006) demonstrated significant protection
(66% RPS) of rainbow trout as late as 2 years after IHNV vaccination. Similarly, rainbow
trout vaccinated with the VHSV DNA vaccine were significantly protected (88% RPS)
after challenge 9 months PV (McLauchlan et al. 2003). In both studies antibodies were
no longer detectable at the time of challenge. After IHNV vaccinated fish were
challenged at 2 years PV, a strong neutralizing antibody response was present by 28
days PC, which suggests an anamnestic neutralizing response was mounted.

RESEARCH QUESTIONS AND OBJECTIVES

The overall project goal was to better understand the host immune response to the
Great Lakes strain of VHSV (type IVb). Four main research questions were addressed
in this study. The research questions and related objectives are as follows:

35

1. Do muskellunge surviving VHSV-IVb mount a neutralizing antibody response
that can be demonstrated by an in vitro assay (Chapter 2)? The objective of this
study was to investigate the kinetics of neutralizing antibody production and virus
levels in serum following experimental immersion exposure. An experimental
infection of muskellunge using three different virus doses was carried out and blood
was collected from fish at various time points over the course of 20 weeks. Sera
were tested for neutralizing antibodies using a PNT assay. Virus levels in sera were
also quantified, so that the relationship between the development of antibodies and
viremia could be assessed. The PNT assay and its reagents were established for
use in our laboratory based on previously published studies on VHSV- genotype I
and IHNV.
2. Can evidence of immunity to VHSV-IVb be detected among wild fish residing
in an enzootic lake (Chapter 3, 5)? The objectives of this study were to determine
if fish from Lake St. Clair, Michigan have detectable levels of neutralizing antibodies
in their sera, and if so, whether differences exist in terms of prevalence and titers
between species. Sera from past fish mortality events and virus surveillance efforts
(2004 to 2009) were tested, as well as samples collected for the purposes of this
study in May 2010 and 2011. Testing of sera from multiple years was helpful to
detect patterns of antibody levels in relationship to known outbreaks of the virus.
3. Do fish infected with VHSV-IVb also produce binding antibodies (Chapter 4,
5)? To explore this question, a competitive ELISA was developed for measuring
binding antibodies to VHSV-IVb. Rabbit antiserum to VHSV-IVb was produced for

36

use as the competition antibody. As part of the assay development, a positivenegative threshold was established for muskellunge, and comparisons were made
between antibody results obtained by cELISA and PNT for several groups of fish.
Using muskellunge as an indicator species, I evaluated the use of the cELISA as a
surveillance tool for detecting virus presence in five water systems in the Great
Lakes (Chapter 5). Muskellunge were sampled between 2005 and 2012 from two
enzootic areas (Lake St. Clair and Lower Fox River/Green Bay, Wisconsin) and from
three inland lakes that were considered negative for VHSV-IVb (Lake Hudson and
Thornapple Lake, MI, and Butternut Lake, WI). Virus exposure status, based on
seroconversion, was compared to virus isolation results from samples collected from
the same individuals or from fish during parallel sampling periods. In Chapter 4, I
preliminarily investigated the use of the cELISA for detecting antibodies in other
species sampled from Lake St. Clair.
4. What is the efficacy of a DNA vaccine encoding the VHSV-IVb glycoprotein
(Chapter 6)? The objectives of this study were to 1) evaluate efficacy of a VHSVIVb DNA vaccine in muskellunge based on induction of an adaptive immune
response, and protection and virus clearance following lethal challenge with VHSVIVb and 2) evaluate ability of vaccine to elicit an innate immune response as
determined by cross-protection challenge against IHNV in rainbow trout. The vaccine
containing the VHSV-IVb glycoprotein gene was developed as part of this study. The
design was based on previously published VHSV-genotype I and IHNV constructs.
The virus challenge time-point post-vaccination (7 weeks) was selected based on

37

experimental studies in Chapter 2.

38

APPENDIX

39

Figure 1.1. A) Schematic of the typical Novirhabdovirus virion (70 nm x 180 nm) and its
linear, negative-sense single-stranded RNA molecule of approximately 11,000
nucleotides. The genome encodes 5 structural proteins [L (RNA-dependent RNA
polymerase), G (glycoprotein), N (nucleoprotein), P (phosphoprotein) and M (matrix
protein)] and one non-virion protein (NV). B) Diagram showing the general design of a
DNA vaccine containing the G gene of viral hemorrhagic septicemia virus (VHSV)
(modified from Schirmbeck and Reimann (2001) and Kurath et al. (2007)). For
interpretation of the references to color in this and all other figures, the reader is referred
to the electronic version of this dissertation.

A

B

Production unit!

Expression unit!
CMV Promoter!

Antibiotic
resistance!

VHSV glycoprotein
(G) gene!
Origin of replication!
Poly A tail!

40

CHAPTER 2
DEVELOPMENT OF NEUTRALIZING ANTIBODY RESPONSES IN MUSKELLUNGE,
ESOX MASQUINONGY (MITCHILL), EXPERIMENTALLY EXPOSED TO VIRAL
HEMORRHAGIC SEPTICEMIA VIRUS (GENOTYPE IVB)

Millard EV, Faisal M (2012) Development of neutralizing antibody responses in
muskellunge, Esox masquinongy (Mitchill), experimentally exposed to viral
haemorrhagic septicaemia virus (genotype IVb). J Fish Dis 35:11-18

41

ABSTRACT

A complement-dependent 50% plaque neutralization test was used to assess the
neutralizing antibody response of muskellunge, Esox masquinongy, experimentally
infected with viral hemorrhagic septicemia virus (VHSV, genotype IVb) by immersion.
Groups of muskellunge were challenged with varying concentrations of VHSV-IVb:
2

-1

3

-1

Group 1 with 10 plaque forming units (pfu) mL , Group 2 with 4 × 10 pfu mL , Group
5

-1

-1

3 with 10 pfu mL , and Group 4 with 0 pfu mL . The fish were held at a temperature
of 11 ± 1°C and were sampled over a 20-week period. Neutralizing antibodies were not
detected in sera of any of the negative control fish throughout the study. Low
neutralizing titers were detected in Groups 1-3 by 6 days post-infection (PI). Neutralizing
titers of ≥ 80 were not detected again until 3, 4 and 7 weeks PI for Groups 2, 3 and 1
respectively. Peak titers for those groups occurred 16, 11 and 17 weeks PI. VHSV-IVb
was detected in sera up to 11 weeks PI. Results of this study show that survivors can be
detected by a serological technique, despite being virus negative. This may benefit the
investigation of VHSV-IVb distribution in the Great Lakes and the study of host immune
responses to this emerging sublineage.

INTRODUCTION

Viral hemorrhagic septicemia virus (VHSV) is an internationally reportable pathogen to
the World Organisation for Animal Health (OIE). This fish rhabdovirus is a member of

42

the genus Novirhabdovirus (Walker et al. 2000). Since its initial isolation in Denmark in
the 1960ʼs (Jensen 1963), the virus has resulted in devastating losses of farmed
rainbow trout, Oncorhynchus mykiss (Walbaum), throughout Europe and is now known
to occur endemically among marine fish of the Pacific and Atlantic Coast of North
America, Europe and Japan (reviewed in Skall et al. 2005). A novel freshwater VHSV
strain, genotype IVb, has been present in the Great Lakes ecosystem since at least
2003 (Elsayed et al. 2006) and has caused multiple wild fish die-off events particularly in
2005 and 2006 (reviewed in Bain et al. 2010, Kim & Faisal 2011a). To restrict the
spread of VHSV-IVb into new geographic areas in North America, federal and state
agencies limited the movement of live fish across the states, implemented strict
biosecurity measures in gamete-collection facilities and hatcheries and began wild fish
surveillance programs to trace the virusʼ distribution.

To date, the gold standard diagnostic test for VHSV in fish, as outlined by the U.S. Fish
and Wildlife Service and American Fisheries Society-Fish Health Section (USFWS and
AFS-FHS 2010) and the OIE (2009), is virus isolation in cell culture from tissue samples
and subsequent confirmation by the reverse-transcriptase polymerase chain reaction
(RT-PCR). Although ideal for clinically infected fish, this testing method has some
disadvantages in that it requires fish to be sampled lethally. It also relies on the virus
being present at levels that are high enough at the time of tissue collection for isolation
on susceptible cell lines, a condition that does not often exist in recovered, sub-clinically
infected, and carrier fish. As a result, there is a need to use additional methods to

43

determine VHSV presence within a population or a waterbody.

Experimental studies have unraveled some aspects of the biology of this emerging
VHSV sublineage and have focused mainly on differential host susceptibilities, disease
course, and tissue changes (Al-Hussinee et al. 2010, Kim & Faisal 2010a, Al-Hussinee
et a. 2011). When Kim and Faisal (2010c) infected juvenile muskellunge, the fish were
found to be particularly susceptible to VHSV-IVb. Some individuals in the same tanks
with fish experiencing high mortality however were able to survive. Furthermore,
muskellunge that survived one virus challenge seemed resistant to re-infection following
a second challenge suggesting that some protective benefit had been obtained (Kim &
Faisal 2012). Observations such as these highlight how little is actually known about
VHSV-IVb host interactions, particularly the host immune response.

From studies on other genotypes of VHSV and infectious hematopoietic necrosis virus
(IHNV), it has been demonstrated that the viral surface glycoproteins (G) elicit the
development of immune responses including antibody formation. For example, viral G
proteins induce antiviral interferons (IFNs) (Acosta et al. 2006) and expression of
immune related genes (Novoa et al. 2006, Tafalla et al. 2007). G proteins also induce
the formation of neutralizing antibodies, which are an important part of the protective
immune response (Engelking & Leong 1989, Lorenzen et al. 1990).

Unlike the relatively short-term presence of virus in infected fish tissues, neutralizing

44

antibodies last longer after infection and as such their presence has been used
successfully to indicate past exposure of fish to rhabdoviruses (reviewed by LaPatra
1996). In a recent serological survey of a VHSV-IVb endemic waterbody, Lake St. Clair
(Michigan, USA), neutralizing antibodies against the virus were detected in sera of
several fish species, with muskellunge expressing the highest antibody titers (Millard &
Faisal 2012b). The current study was designed to characterize the development of
neutralizing antibodies in muskellunge following experimental infection.

MATERIALS AND METHODS

Cell culture and virus isolate. The epithelioma papulosum cyprini (EPC; Fijan et al.
o

1983) cell line was used throughout this study. The cell line was maintained at 25 C in
2

150cm tissue culture flasks (Corning) in minimum essential medium with Earleʼs salts
(EMEM, Invitrogen) supplemented with 10% fetal bovine serum (FBS; Hyclone), 10%
tryptose phosphate broth (TPB; BD Biosciences), 2.0 mM L-glutamine (Invitrogen) and
sodium bicarbonate (12 mM; Sigma-Aldrich).

The index VHSV-IVb strain, MI03, originally isolated from muskellunge from Lake St.
Clair (Elsayed et al. 2006) was propagated on the EPC cell line at 15°C in EMEM with
2% FBS, 10% TPB, 2.0 mM L-glutamine, 12 mM sodium bicarbonate, penicillin
−1

−1

−1

(100 IU mL ), streptomycin (100 μg mL ) and amphotericin B (2.5 μg mL )

45

(Invitrogen). When approximately 90% cytopathic effects (CPE) were present, media
o

from infected cell cultures was clarified by centrifugation (2900 × g for 30 min at 4 C).
o

Supernatant was divided into aliquots and stored at -80 C until immediately before use.
The stock titer was determined by plaque assay on 7% polyethylene glycol (PEG)treated EPC cells as described in Batts & Winton (1989).

Muskellunge. Certified juvenile muskellunge were obtained from the Rathbun National
Fish Hatchery (Iowa Department of Natural Resources, Moravia, Iowa, USA). Fish were
acclimatized for 3 weeks in holding tanks in a continuous flow-through system at the
Michigan State University Research Containment Facility (URCF; East Lansing,
Michigan, USA). Muskellunge were fed live fathead minnows, Pimephales promelas
(Rafinesque), obtained from Anderson Minnow Farm (Lonoke, Arkansas, USA). Fish
were maintained according to protocols approved by Michigan State University's
Institutional Animal Care and Use Committee.

Experimental infection. Acclimatized juvenile muskellunge (4 months old;
17.0 ± 5.0 g; total length 16.4 ± 1.2 cm) were divided into four groups for the immersion
challenge. Groups 1-3 consisted of approximately 80 fish each. Fish were exposed to
2

-1

the following concentrations of VHSV: Group 1 with 10 plaque forming units (pfu) mL ,
3

-1

Group 2 with 4 × 10 pfu mL

5

-1

and Group 3 with 10 pfu mL . A fourth group (Group 4;

70 fish) served as the negative control. The doses were chosen to produce acute,

46

subacute, and chronic courses of infection as determined for juveniles of this species
(Kim & Faisal 2010c). The experimental infection was performed in aquaria that
contained static, aerated water to which the virus was added and the immersion
challenge continued for 90 min. The negative control group received sterile cell culture
media. Each challenge group was then distributed between two randomly assigned
replicate tanks. Water temperature was maintained at 11 ± 1°C throughout the duration
of the study. Fish were monitored on a daily basis and tissues including kidney, spleen
and heart were collected from any fish that died.

Four fish per group (two per replicate; 16 total) were sampled at the following time
points post-infection (PI): 0, 12, 24, and 36 hr; 2, 4, 6 and 8 days and then once weekly
up to 9 weeks PI. The number of sampled fish was then reduced to two fish per
challenge group in order to allow for a longer observation period. Two fish per group
were sampled 11, 12, 13, 15, 16, 17, 18 and 20 weeks PI until all fish had been
sampled. Moribund fish were selected for sampling first, and if no morbidity was
observed, fish were selected randomly. Fish were euthanized with an overdose of
–1

tricaine methanesulfonate (MS-222; Argent Chemical Laboratories; 0.25 mg mL ) and
blood samples were aseptically collected by puncture of the caudal vein. The blood
o

samples were allowed to coagulate overnight at 4 C, centrifuged at 4500 × g for 20 min
o

o

at 4 C, and sera were collected. Sera aliquots were stored at -80 C until assayed.

47

Assessment of VHSV in infected muskellunge serum samples. The EPC cell line
was used for isolation of VHSV from sampled muskellunge serum samples and tissues
from mortalities in accordance with suggested procedures (USFWS & AFS-FHS 2010).
The amount of infectious virus in sera was quantified by the plaque assay. RT-PCR,
using VHSV specific primers, was used to confirm the re-isolation of VHSV from
samples that exhibited CPE after two successive passes. Cell culture negative sera
were also tested by RT-PCR to detect levels of virus that may have been below the
detection limit of the cell culture assays.

50% plaque neutralization test (50%PNT). Sera from disease-free lake trout,
Salvelinus namaycush (Walbaum), was used as a source of fish complement. Lake trout
–1

had been reared in isolation at URCF. Fish were anesthetized using 0.1mg mL

MS-

222 buffered with sodium bicarbonate and blood was collected non-lethally. Sera were
separated as previously described and stored in single-use aliquots in liquid nitrogen
until immediately before use. The naïve sera used as a source of complement the assay
(1:30 dilution, unheated) was tested to be free of background neutralization activity prior
to use.

The 50%PNT was performed based on procedures previously described for detecting
neutralizing antibodies in rainbow trout sera against VHSV genotype I and IHNV
(Olesen & Jørgensen 1986, LaPatra et al. 1993) with some modifications. Fish sera
o

were heat inactivated for 30 min at 45 C. Serial two fold dilutions of sera, starting at a

48

3

dilution of 1:20, were then mixed with an equal volume of virus containing 1.5 x 10 pfu
o

VHSV-IVb, and incubated at 18 C for 60 min with constant gentle agitation. Half way
through the incubation, an equal volume of naïve lake trout complement was added to
the reaction. Serum-virus-complement mixtures were then inoculated onto replicate
wells of PEG pretreated EPC cells grown in 24 well tissue culture plates (Corning). Virus
o

adsorption was allowed to take place for one hour at 15 C before the inoculum was
removed and cells overlaid with 1% methylcellulose. After 5-6 days of incubation at
o

15 C, plates were fixed and stained with 0.5% crystal violet in 50% formalin and
plaques enumerated. The neutralizing antibody titer is reported as the reciprocal value
of the highest serum dilution causing a 50% reduction in the average number of plaques
counted for the negative control. The negative control was naive muskellunge sera with
complement and virus. Both a negative and positive control muskellunge sera were
included each time the assay was performed.

RESULTS

Disease course and virus re-isolation. The experimental challenge caused morbidity
and mortality in infected muskellunge that varied among the three challenge groups.
Moribund and dead fish exhibited classical signs and lesions of VHSV such as severely
anemic gills, external, and internal hemorrhages. Other than fish euthanized for
sampling, Groups 1, 2 and 3 experienced 2.4%, 7.6% and 12.5% mortality respectively.

49

Most mortality occurred between 6 days and 4 weeks PI. VHSV was re-isolated from the
internal organs of dead fish from Groups 1-3 and its identity confirmed by RT-PCR. No
VHSV-related mortalities were observed in the negative control group.

VHSV was isolated from the sera of some of the sampled fish; however, the number of
viremic fish varied between challenge groups (Table 2.1). In Group 1, none of the
sampled fish tested positive by cell culture. In Group 2, 12.9% of sampled fish were
positive. VHSV was first detected in sera 6 days p.i and last detected 6 weeks PI. In
Group 3, 23.2% of sera were positive. The virus was detected as early as 2 days PI and
as late as 11 week PI. For both groups, peak viral titers occurred 6 days PI. All virus
positive cell culture samples were confirmed positive for VHSV-IVb by RT-PCR.
Additionally, RT-PCR performed on samples not exhibiting cytopathic effects after the
second passage revealed the presence of the virus in one fish from Group 1 sampled
15 days PI and one additional fish from Groups 2 and 3 sampled 6 and 4 weeks PI,
respectively.

Neutralizing antibody response. Negative control muskellunge sampled between 0
and 9 weeks PI did not have any detectable neutralizing activity (titer <20). Some, but
not all, of the serum samples from virus-challenged muskellunge were positive for
neutralizing antibodies. Low titers (20-80) were detected in Groups 1-3 by 6 days PI
(Figure 2.1). Titers of ≥ 80 were detected again 3, 4 and 7 weeks PI for Groups 2, 3,
and 1, respectively. Overall, 9.0% (7/78) of samples were positive for neutralizing

50

2

-1

activities from fish challenged with the lowest concentration of VHSV (10 pfu mL ). In
3

-1

the group challenged with 4 × 10 pfu mL

(Group 2), 29.0% (20/69) of fish developed

neutralizing antibodies. A peak titer of 5120 for this Group was detected in a sample
taken 16 weeks PI, which was the last sampling point. For Group 3 fish, that were
5

-1

challenged with the highest concentration of virus (10 pfu mL ), 20.0% (13/65) of the
serum samples had neutralizing antibodies. The highest titers were detected between 9
weeks PI and the end of the sampling period (13 weeks PI). The highest titer (20,480)
was detected 11 weeks PI.

DISCUSSION

When naïve muskellunge were exposed to VHSV-IVb, virus-neutralizing antibodies
developed in sera of some individuals. This finding clearly demonstrates the ability of
this susceptible species to mount a humoral immune response against the virus. The
immune response, including the development of neutralizing antibodies, of rainbow trout
to other genotypes of VHSV has been reviewed (Lorenzen & LaPatra 1999). Evidence
of an adaptive immune response to VHSV, although not neutralizing antibodies, has
been reported for Pacific herring, Clupea pallasii (Valenciennes), (Kocan et al. 2001,
Hershberger et al. 2007) as well as wild brown trout (Salmo trutta L.), pike (Esox lucius
L.) and whitefish (Coregonus sp.) (Enzmann et al.1993). This report is the first to
characterize the development of neutralizing antibodies following experimental exposure

51

to VHSV-IVb; as well as the first to examine the immune response of experimentally
infected muskellunge to a viral pathogen to the best of our knowledge.

In the present study, neutralizing antibodies were not detected in any of the negative
control fish at the starting dilution of 1:20. Therefore a titer of <20 is defined as being
negative for neutralizing antibodies to VHSV-IVb. Relatively low neutralizing titers up to
80 were detected in some experimentally infected muskellunge in the first 6 days after
immersion challenge; with titers of 20 and 40 occurring as early as 36 hr PI. Similarly,
low neutralization titers occurring the first week after infection were reported by LaPatra
et al. (1993). The nature of this early neutralizing activity remains to be elucidated and
could, partially or totally, be related to innate humoral factors that were expressed upon
VHSV exposure. Specific antibody development takes 3-4 weeks in rhabdovirus
immunized rainbow trout (LaPatra et al. 2001, Boudinot et al. 1998) and this response is
influenced by many factors such as temperature (Le Morvan et al. 1998). In this context,
anti-viral factors such as IFN-like substances (Pakingking et al. 2004) and Mx protein
(Cuesta & Tafalla 2009) that are induced early following infection with VHSV could limit
virus infectivity of cells.

Neutralizing antibody titers of ≥80 were not detected again until 3 weeks PI. In samples
taken 6 weeks PI, 62.5% (5/8) of Groups 2 and 3 muskellunge had neutralizing
antibodies. Titers increased throughout the sampling period with the maximum titers for
the groups occurring 17 (1280), 16 (5120) and 11 weeks (20480) for Groups 1, 2 and 3

52

respectively. Such titers are comparable to those obtained from serum samples of
rainbow trout experimentally infected with VHSV genotype I (Olesen & Jørgensen 1986,
Olesen et al. 1991). In both studies, neutralizing antibodies were detected by 4 weeks
PI and maximum titers of 5120-10,240 occurred 6-10 weeks PI. In this study, it is
interesting to note that out of all the antibody positive fish, only 4 of these individuals
also tested positive for the virus. Moreover, all of these double positive fish were
sampled 6 weeks PI and earlier, meaning that later during the disease course, after
mortalities have ceased, it seems that survivors are more likely to be detected based on
the presence of virus-specific antibodies in their sera, versus the virus itself.

The total duration of neutralizing antibodies in sera of survivors of VHSV infection is
currently unknown. Olesen et al. (1991) reported a decrease to a mean titer of 80
between 18 and 23 weeks PI, even following a second virus challenge 14 weeks PI. On
the other hand, Olesen and Jørgensen (1986) reported some trout having VHSV
neutralizing antibodies for over a year. In the present study, antibodies were detected as
long as 17 weeks PI; however, the total duration was not determined due to the
sampling design. In a concurrent study in our laboratory (data not shown), we found that
neutralizing antibodies are still detectable in sera of several muskellunge that survived a
double exposure to VHSV-IVb 90 weeks after their original exposure. Frequent nonlethal blood collection from this group is underway in order to determine the maximum
duration of circulating neutralizing antibodies.

53

This study has also demonstrated that a number of infected muskellunge have not
mounted a VHSV neutralizing antibody response. Fish exposed to the lowest dose
2

−1

(10 pfu mL ) had considerably more non-responders than Groups 2 and 3. While this
finding may suggest an optimum virus concentration for neutralizing antibody
development, the frequent sampling in the experiment does not allow confirmation of
this observation. This variation in antibody responses and titers reflects the individual
heterogeneity among experimental fish. It is also indicative of multiple pathways of
defense that the muskellunge immune system utilizes toward the same inducing
antigen.

From the present study, it was determined that muskellunge, an important Great Lakes
species, can carry evidence of past exposures to the virus, even when negative for the
virus itself. These results will have important implications in investigating the distribution
and spread of VHSV in the Great Lakes as well as provide the background information
necessary to assess the efficacy of potential vaccine preparations.

ACKNOWLEDGMENTS

The authors thank the Great Lakes Fishery Trust (Grant #08WRGR0006) for supporting
this study.

54

APPENDIX

55

-1

Table 2.1. Virus isolation and titer range in plaque forming units (pfu) mL of serum of
sampled muskellunge, Esox masquinongy, experimentally infected with viral
hemorrhagic septicemia virus (VHSV, genotype IVb). “VHSV Pos.” column indicates
number of VHSV-positive serum samples/ number of samples tested. Groups were
2

-1

challenged by immersion in the following concentrations of virus: 10 pfu mL
3

-1

1), 4 × 10 pfu mL

5

-1

(Group 2), and 10 pfu mL

Group 1
VHSV
Pos.

Group 2
VHSV Titer range
-1
Pos.
(pfu mL )

0

0/4

0/3

0.5
1
1.5
2

0/4
0/4
0/4
0/4

4

Days PI

(Group

(Group 3).
Group 3
Titer range
VHSV
-1
Pos.
(pfu mL )

a

-

0/4

0/3
0/4
0/4
0/4

a

-

0/3
0/4
0/4
1/4

-

0/4

0/4

-

2/4

5.0 × 10 -7.0 × 10

6

0/4

2/4

2/4

9.5 × 10

8
Weeks PI
2

0/4

0/3

5.0 × 10 -7.6 × 10
-

2/4

8.5 × 10

0/4

1/4

2.3 × 10

3/4

3

0/4

1/4

5.0 × 10

0/4

5.0 × 10 -3.6 × 10
-

1/4
0/4
0/4

6.0 × 10
-

1/4

5.0 × 10

a

a

2

6

5
2
4

6

3

4

-

5.0 × 10

2
2
5
4
2

0/4
0/4
0/4

3/4
0/4
2/4

7

0/4

0/4

1.0 × 10 -8.0 × 10
-

8

0/4

0/4

-

1/4

1.0 × 10

9

0/4

0/4

-

2/4

1.0 × 10 -3.6 × 10

11
12
13
15
16
17
18
20

0/2
0/2
0/2
0/2
0/2
0/2
0/2
0/2

0/2
0/2
0/2
0/2
0/1
-

-

1/2
0/2
0/2
-

3.0 × 10
-

a

2
3
4
3

Although four fish were sampled, there was not enough serum from one fish for
testing.

56

5

3

4
5
6

2.3 × 10 -4.2 × 10
-

3

4

Figure 2.1. Development of neutralizing antibodies measured by the 50% plaque
neutralization test (50%PNT) in three groups of muskellunge, Esox masquinongy,
challenged by immersion with viral hemorrhagic septicemia virus (VHSV-IVb). An
2

3

antibody titer of ≥20 is positive. Groups 1, 2 and 3 were exposed to 10 , 4 × 10 and
5

-1

50% plaque neutralization test titer

10 pfu mL respectively. The last sampling points were 20, 16 and 13 weeks for
Groups 1, 2 and 3, respectively.

20480
10240
5120
2560
1280
640
320
160
80
40
20
<20
20480
10240
5120
2560
1280
640
320
160
80
40
20
<20
20480
10240
5120
2560
1280
640
320
160
80
40
20
<20

Group 1

Group 2

Group 3

012

4

6

8

2

Days post-infection

4

6

8

10 12 14 16 18 20

Weeks post-infection

57

Figure 2.1. (contʼd)
a

Several samples did not yield enough serum to be included for the 50%PNT test:
Group 1 (one fish each at 24 hr and 5 weeks PI); Group 2 (two fish at 0 hr PI, one fish
each at 12 hr and 8 days PI); and Group 3 (two fish at 0 hr PI, and one fish each at 12
hr, 2 days and 4 weeks PI).

58

CHAPTER 3

HETEROGENEITY IN LEVELS OF SERUM NEUTRALIZING ANTIBODIES AGAINST
VIRAL HEMORRHAGIC SEPTICEMIA VIRUS GENOTYPE IVB AMONG FISH
SPECIES IN LAKE ST. CLAIR, MICHIGAN, USA

Millard EV, Faisal M (2012) Heterogeneity in levels of serum neutralizing antibodies
against viral hemorrhagic septicemia virus genotype IVb among fish species in
Lake St. Clair, Michigan, USA. J Wildl Dis 48:405-415

59

ABSTRACT

The presence of neutralizing antibodies against viral hemorrhagic septicemia virus
(VHSV-IVb) was investigated in sera of 13 fish species collected from Lake St. Clair,
Michigan, USA, a VHSV-endemic water body. We tested 297 sera collected May 2004–
June of 2010, using a complement-dependent 50% plaque neutralization test (50%
PNT). Neutralizing antibodies were detected in 23% (67/297) of the samples. The
highest overall antibody prevalence (85%, 34/40) and mean positive antibody titer
(12,113; SD = 11,699) were detected in muskellunge (Esox masquinongy). Antibodies
were also detected in 50% (15/30) of sampled northern pike (E. lucius), 25% (15/61) of
freshwater drum (Aplodinotus grunniens), and 7% (3/41) of smallmouth bass
(Micropterus dolomieu). All sera from channel catfish (Ictalurus punctatus), lake
sturgeon (Acipenser fulvescens), quillback (Carpiodes cyprinus), rock bass (Ambloplites
rupestris), shorthead redhorse (Moxostoma macrolepidotum), silver redhorse (M.
anisurum), walleye (Sander vitreus), white perch (Morone americana), and yellow perch
(Perca flavescens) were negative. Antibodies in one or more fish species were detected
in all sampling years (2004, 2006, 2007, 2009, and 2010), whereas in parallel sampling
periods, VHS virus was only detected in 2006 and 2009. Our results suggest the
continued presence of VHSV-IVb in the Lake St. Clair ecosystem, and underscore the
importance of assessing immune responses of fish populations to determine prior virus
exposure.

60

INTRODUCTION

Viral hemorrhagic septicemia virus (VHSV) is a pathogenic fish virus belonging to the
family Rhabdoviridae, genus Novirhabdovirus. In spring-summer of 2005 and 2006, a
novel North American VHSV sublineage (designated IVb) was implicated in fish die-off
events in Lake St. Clair, Lake Erie, and Lake Ontario. Mortality episodes involved
numerous fish species including freshwater drum (Aplodinotus grunniens; Lumsden et
al. 2007), round gobies (Neogobius melanostomus; Groocock et al. 2007), muskellunge
(Esox masquinongy), gizzard shad (Dorosoma cepedianum), yellow perch (Perca
flavescens), and others (Winton et al. 2008, Kim & Faisal 2011). Archived samples date
the presence of VHSV-IVb in the Lake St. Clair ecosystem to at least 2003 (Elsayed et
al. 2006). The discovery of this reportable virus in the Great Lakes initiated fish
movement restrictions and surveillance testing for purposes of zoning and monitoring
the spread of the virus. Within the next 5 yr, VHSV-IVb was isolated from all five Great
Lakes, the St. Lawrence River, and several inland lakes in Michigan, New York, Ohio,
and Wisconsin, USA (Kim & Faisal 2011).

Fish are tested for VHSV following guidelines of the American Fisheries Society (Batts &
Winton 2010) and the World Organization for Animal Health (OIE 2012). Laboratory
detection of VHSV requires isolation in cell culture followed by confirmation of the
isolated virus, most commonly by reverse transcriptase polymerase chain reaction (RTPCR). Although this approach is currently considered the gold standard for VHS

61

diagnosis, populations that exhibit low infection prevalences or those that have
recovered and cleared the virus, can be missed.

Fish surviving infection with VHSV (Olesen & Jørgensen 1986) and other pathogenic
rhabdoviruses, including infectious hematopoietic necrosis virus (IHNV; Amend & Smith
1974, Jørgensen et al. 1991) and spring viremia of carp virus (SVCV; Ahne 1986),
mount an adaptive immune response that includes the production of neutralizing
antibodies. Survivors demonstrate enhanced resistance to disease upon subsequent
exposures (LaPatra et al. 1993, Kocan et al. 2001, Ahne et al. 2002). Antibodies to
VHSV remain in sera for an extended period after the virus is no longer detectable, and
have been used in epizootiologic studies to identify fish populations that previously have
been infected (LaPatra 1996).

In Lake St. Clair, following the initial large-scale mortality events in spring-summer 2006,
and despite surveillance testing, the virus was not isolated again until a smallmouth
bass (Micropterus dolomieu) die-off occurred in June 2009. One logical explanation for
the absence of VHSV outbreaks in Lake St. Clair in 2007, 2008, and 2010, is that fish
residing in this infected zone might have developed some immunity to the virus. We
sought to determine if fish in Lake St. Clair have neutralizing antibodies against VHSVIVb and compare the levels of these antibodies among resident fish species.

MATERIALS AND METHODS

62

Field serum samples. Most fish were captured using trap nets set in Anchor Bay,
Lake St. Clair (Figure 3.1) by personnel of the Michigan Department of Natural
Resources (MDNR) and the Aquatic Animal Health Laboratory (AAHL, Michigan State
University, East Lansing, Michigan, USA). In May 2010, blood samples were collected
by caudal venip

uncture from channel catfish (Ictalurus punctatus, n = 25),

freshwater drum (n = 10), muskellunge (n = 21), northern pike (Esox lucius, n = 16),
quillback (Carpiodes cyprinus, n = 7), rock bass (Ambloplites rupestris, n = 24),
shorthead redhorse (Moxostoma macrolepidotum, n = 6), silver redhorse (M. anisurum,
n = 13), smallmouth bass (n = 25), walleye (Sander vitreus, n = 10), white perch
(Morone americana, n = 12) and yellow perch (n = 3). Lake sturgeon (Acipenser
fulvescens, n = 15) were sampled by the MDNR the first week of June 2010. Blood was
sampled nonlethally from all 13 species and fish were released, with the exception that
11 of the 21 muskellunge were euthanized following blood collection to obtain tissues
(kidney, spleen, and heart) for VHSV testing. Gametes were also collected from all
mature muskellunge for VHSV testing. Water temperature in May 2010 was 9–15°C.

Archived sera from muskellunge (n = 6) and freshwater drum (n = 40) in May 2007 and
muskellunge (n = 10) in 2009, originated from apparently healthy fish collected for
VHSV surveillance. Serum samples from northern pike in May 2004 (n = 10), all species
in May 2006 (n = 28), and smallmouth bass in June 2009 (n = 16), were collected as
part of targeted sampling efforts during, or immediately following, mortality events in

63

Lake St. Clair. Most fish were captured alive and immediately transported on ice to the
AAHL. Upon arrival, blood was collected and tissues (kidney and spleen) and
reproductive fluids (from ripe fish) were sampled for VHSV testing.

For all captured muskellunge, weight and total length were recorded and a dorsal fin ray
clip was taken for age estimation (Clark et al. 2004) by MDNR researchers. Blood
samples from all fish were kept at 4°C overnight and then centrifuged (2,500 × g, 10
min, 4°C). Sera were aliquoted and stored at -80°C until testing.

Cell line and virus isolate. The epithelioma papulosum cyprini (EPC) cell line (Fijan et
al. 1983) was cultured at 25°C in Earleʼs minimum essential medium (EMEM; Life
Technologies) supplemented with 10% tryptose phosphate broth (TPB; Becton,
Dickinson and Company), 10% fetal bovine serum (FBS; Hyclone), 2 mM L-glutamine
(Life Technologies), and buffered with 7.5% sodium bicarbonate. The Great Lakes strain
of VHSV-IVb (Elsayed et al. 2006) was propagated on the EPC cell line and aliquots of
supernatant were stored at -80°C for use in the 50% plaque neutralization test (PNT).
The virus titer was determined by plaque assay as described in Batts & Winton (1989).

Virus isolation. All serum samples, tissues, and reproductive fluids collected were
tested for VHSV in accordance with suggested procedures (Batts and Winton, 2010).
Serum samples were diluted 1:20 (v/v) in EMEM with 10% TPB, 14 mM tris buffer,
-1

-1

penicillin (100 U mL ), streptomycin (100 µg mL ; Life Technologies), gentamicin

64

-1

-1

sulfate (100 µg mL ; Sigma-Aldrich), and amphotericin B (2.5 µg mL ; Lonza). Tissues
and reproductive fluids were diluted 1:4 (w/v), homogenized, and centrifuged (2,500 × g,
30 min, 4°C). Homogenate supernatants and serum samples were inoculated onto EPC
cells grown to confluency in cell culture medium (EMEM with 10% TPB, 5% FBS, 2 mM
L-glutamine, 14 mM tris, and 100 U penicillin, 100 µg streptomycin, 100 µg gentamicin
-1

sulfate, and 2.5 µg amphotericin B mL ). Plates were incubated at 15°C for two
successive passages and any samples exhibiting cytopathic effects (CPE) were subject
to confirmation using RT-PCR.

Neutralizing antibody detection. A complement-dependent 50% PNT (LaPatra et al.
1993) was used with some modifications to detect VHSV-IVb neutralizing antibodies in
fish sera. Samples were heat inactivated for 30 min at 45°C and serial twofold dilutions,
at a starting dilution of 1:20, were mixed with an equal volume of VHSV-IVb (2.0×10

4

-1

plaque-forming units mL ). Serum-virus mixtures were incubated for 30 min at 18°C
with constant gentle motion. An equal volume of unheated naïve sera from lake trout
(Salvelinus namaycush), diluted 1:30, was added as a source of complement and the
incubation period was repeated. Mixtures were adsorbed for 1 hr at 15°C to 7%
polyethylene glycol pretreated EPC cells in duplicate wells of 24-well tissue culture
plates. Fluid was then removed and cells overlaid with 1% methylcellulose in cell culture
medium (with 2X concentrated EMEM). After incubation for 5 days at 15°C, the cells
were fixed and stained with 0.5% crystal violet in formalin. Titers are reported as the

65

reciprocal of the highest serum dilution that reduced the number of virus plaques by
50% compared to the negative control. Each time the assay was performed, positive
and negative control sera were included.

Complement source. Naïve sera used in the 50% PNT were collected from adult lake
trout that had been raised at the Michigan State University Research Containment
-1

Facility (MSU UCRF). Prior to blood collection, fish were anesthetized using 0.1mg mL
tricaine methanesulphonate (MS-222, Argent Chemical Laboratories) buffered with
sodium bicarbonate. Blood was processed as previously described, and aliquots of
pooled sera were stored in liquid nitrogen until immediately before use.

Naïve sera examination. Serum samples from three groups of naïve muskellunge,
originating from hatcheries with no history of VHSV exposure, were tested with the 50%
PNT to determine if innate neutralization activities against VHSV-IVb exist in sera of
unexposed fish. Fish were maintained at 11±2°C in a flow-through system at the MSU
UCRF and fed fathead minnows (Pimephales promelas) raised in a farm with no prior
history of VHSV. Blood was collected weekly from Groups I (n = 30; ~4 mo old) and II (n
= 46; ~6 mo old) for 9 wk. Fish from these groups were euthanized at the time of blood
-1

collection with an overdose of MS-222 (0.25 mg mL ) and organs tested in cell culture
to confirm the absence of VHSV. Blood samples were collected nonlethally from Group
III (n = 7, ~22 mo old) muskellunge every 6 wk for six sampling periods (total of 42
sera).

66

Specificity of VHSV-IVb neutralizing antibodies. To test the specificity of the serum
neutralizing activities against VHSV-IVb, six Lake St. Clair muskellunge sera (titers of
20–20,480) and a single freshwater drum serum (titer of 40) were tested with IHNV
(strain 220-90; LaPatra et al. 1994). Serum samples were tested in parallel with VHSVIVb and IHNV using the 50% PNT conditions as described previously.

Statistical analysis. Differences in the occurrence of antibodies between species, and
between mature versus immature muskellunge, were compared using a two-tailed
Fisherʼs exact test. Pearsonʼs correlation was used to determine if antibody titers in sera
of muskellunge were associated with weight or length. Significant differences were
accepted at P < 0.05.

RESULTS

Virus isolation. All serum samples from Lake St. Clair fish were negative for VHSV.
Tissues collected in 2004, 2007, and 2010, and all reproductive fluids from muskellunge
also were negative. VHSV was isolated in cell culture from tissues of 19 of 28 fish
sampled in 2006. All muskellunge, northern pike, rock bass, and shorthead redhorse
were VHSV-positive, as were 55% of freshwater drum and 20% of silver redhorse. Most
of these fish exhibited external and internal petechial hemorrhages, exophthalmia, and
congestion of internal organs. VHSV was also isolated from tissues of smallmouth bass

67

(25%; 4/16) collected during a mortality event in June 2009. The identity of VHSV was
confirmed in all cell culture-positive samples by RT-PCR.

Innate neutralization of VHSV by sera of naïve muskellunge. Most (89%) naïve
muskellunge sera did not neutralize VHSV-IVb at the starting serum dilution of 1:20 (titer
<20; Table 3.1). Titers of 20–80 were observed in sera of 13 of 118 muskellunge (11%).
Therefore, only neutralization titers >80 are considered antibody-positive for wild
muskellunge and its congener, northern pike. A specific cutoff value for other fish
species could not be ascertained because of the absence of samples collected from
naïve fish. Results for these species are reported at titers of ≥20.

Neutralization specificity of fish sera. Neutralization titers ranging from 20 to 20,480
in muskellunge sera and 40 in a freshwater drum serum did not cross-react with IHNV at
titers of ≥40 (Table 3.2).

Neutralizing antibodies against VHSV-IVb. Blood samples were collected from a
total of 297 Lake St. Clair fish representing 13 species (Table 3.3). Neutralizing
antibodies against VHSV-IVb were detected in 23% (67/297) of sera, and in all study
years (2004, 2006, 2007, 2009, 2010). Four species, muskellunge, northern pike,
freshwater drum, and smallmouth bass, were antibody-positive. The overall antibody
prevalence in muskellunge (85%; 34/ 40) was significantly higher than northern pike
(50%; 15/30; P = 0.0032), freshwater drum (25%; 15/61; P < 0.0001), and smallmouth

68

bass (7%; 3/41; P < 0.0001). Excluding archived sera, fish sampled in 2010 followed a
similar trend; muskellunge exhibited the highest antibody prevalence (86%; 18/21),
followed by northern pike (63%; 10/16), freshwater drum (20%; 2/10), and smallmouth
bass (12%; 3/25). All sera from channel catfish, lake sturgeon, quillback, rock bass,
shorthead and silver redhorses, walleye, white perch, and yellow perch were negative
(titers <20).

Neutralization titers reached the highest levels in muskellunge; 91% (31/34) of antibodypositive individuals had titers of 2,560–40,960 (Figure 3.2). In comparison, maximum
titers for other antibody-positive species did not exceed 1,280 (northern pike and
freshwater drum) and 320 (smallmouth bass). Mean positive titers were higher for
muskellunge (mean = 12,113; SD = 11,699) compared to northern pike (mean = 608;
SD = 449), freshwater drum (mean = 201; SD = 339), and smallmouth bass (mean =
213; SD = 92). Among muskellunge, the prevalence of neutralizing antibodies was
significantly higher in reproductively mature fish than immature fish (Table 3.4; P <
0.0001). Antibody titer in this species was not correlated with length (n = 40, r = 0.24, P
= 0.14), or weight (n = 39, r = 0.18, P = 0.27).

For some species, lower antibody prevalences were detected during VHSV-positive
sampling events. The lowest yearly antibody prevalence for muskellunge (33%; 1/3)
was detected in 2006 when this species was VHSV-positive. In 2006, anti-VHSV
antibody prevalence in freshwater drum (0%; 0/11) was also lower compared to 2007

69

(33%; 13/40) and 2010 (20%; 2/10). Similarly, none of the smallmouth bass from a
VHSV-positive mortality event in 2009 were antibody-positive, whereas the following
year, 12% (3/25) had antibodies.

DISCUSSION

Neutralizing antibodies are targeted against viral glycoproteins and are an essential
component of the protective immune response of fish against rhabdoviruses (Engelking
& Leong 1989, Lorenzen et al. 1990). Muskellunge that survive infection with VHSV-IVb
develop neutralizing antibodies in their sera (Millard & Faisal 2012), and are partially
resistant to disease, despite high levels of shed virus in surrounding water (Kim & Faisal
2012). In this context, fish residing in a VHSV-endemic water body, such as Lake St.
Clair, could be intermittently re-exposed to VHSV, without experiencing disease
outbreaks. In this study, we detected evidence of an acquired immune response in sera
of four of the 13 fish species sampled, albeit at different prevalences and titers.

The majority of Lake St. Clair muskellunge (85%) and northern pike (50%) were positive
for VHSV-IVb antibodies. Neutralizing antibodies were detected in fewer sera from
freshwater drum (25%) and smallmouth bass (7%), and were completely absent in sera
from channel catfish, lake sturgeon, quillback, rock bass, shorthead redhorse, silver
redhorse, walleye, white perch, and yellow perch. Interspecific variation in antibody
prevalence could be related to differences in speciesʼ ecologic niches, which would

70

influence the frequency of exposure to the virus. Esocids (muskellunge and northern
pike) are top predators in the Lake St. Clair ecosystem. It is possible that their high
trophic position provides more opportunities for virus exposure through ingestion of
VHSV-infected prey fish. Neutralizing antibody prevalence also might be influenced by
speciesʼ susceptibility to the virus, or differences in the immune mechanisms employed
against VHSV-IVb. Of the 13 species sampled, seven (muskellunge, northern pike,
freshwater drum, smallmouth bass, rock bass, and shorthead and silver redhorses)
were positive for VHSV in this investigation, and the remaining species, with the
exception of lake sturgeon and quillback, are also known to be naturally susceptible
(USDA–APHIS 2008). The dose of VHSV-IVb necessary for infection and disease in
these species, however, is variable. It is possible that a relatively low virus
concentration in the water could infect and induce a neutralizing antibody response by
species of high susceptibility (e.g., muskellunge; Kim & Faisal 2010c); whereas species
that are less susceptible or resistant (e.g., lake sturgeon; M. Faisal, unpubl. data) might
not become infected. However, because the experimental susceptibility among species
included in this study is known only for muskellunge, yellow perch, and smallmouth bass
(Kim & Faisal 2010a, b), a correlation between occurrence of antibodies and
susceptibility cannot be made at the present time. The absence of antibodies detectable
by the 50% PNT in the nine aforementioned antibody-negative species does not
necessarily imply the lack of an immune response. VHSV induces both cell-mediated
and humoral defenses, including non-neutralizing antibodies that would not be
detectable by the 50% PNT (Lorenzen et al. 1999).

71

Antibody prevalence was lower in immature muskellunge (ages 2–3) than reproductively
mature fish (ages 6–20). Juvenile muskellunge (4 mo old) produce neutralizing
antibodies to VHSV-IVb under laboratory conditions (Millard & Faisal 2012). Thus, the
absence of antibodies in immature Lake St. Clair muskellunge is not due to an inability
to mount a neutralization response. Rather, it could be due to having fewer opportunities
for virus exposure. Spawning congregations, for example, have been identified as a
likely route for VHSV transmission, due to increased fish densities and increased virus
concentration in the water from infected individuals (Hershberger et al. 2010, VHSV
Expert Panel and Working Group 2010). The absence of serum antibodies in immature
muskellunge, as well as the small sample size, could also suggest an increased
mortality rate upon VHSV exposure of young muskellunge. Intraspecific variation in
neutralization titers could also be attributed to exposure dose, as well as time postexposure.

Antibody-positive status for muskellunge and northern pike was defined as having a
neutralization titer of >80. This is because when we tested several groups of naïve
juvenile muskellunge, with no history of VHSV-IVb exposure, a few individuals exhibited
low levels of innate virus neutralization. The nature of the low level neutralization
remains to be elucidated; however, some virus inhibitors have been reported in sera of
unexposed rainbow trout (Oncorhynchus mykiss; Dorson & de Kinkelin 1974, Park &
Reno 2005). The absence of neutralization activities in the majority of non-esocid fish

72

sera tested in this study led us to believe that such VHSV inhibitors might not be present
in sera of these species. Thus, titers of ≥20 were considered positive. By setting the
definition of antibody-positive slightly higher for esocid fish, the number of antibodypositive fish might be underestimated by seven individuals.

Evidence for the presence of immunoglobulins in several antibody-positive Lake St.
Clair muskellunge was confirmed using an indirect ELISA that utilizes a monoclonal
anti-muskellunge IgM antibody (Millard, Kaattari, and Faisal, data not shown). In our
analysis, we also confirmed that virus neutralization activity in muskellunge sera was
specific to VHSV, because no cross-reactivity with IHNV was observed at titers >20.
Rainbow trout sera with antibodies against VHSV-I also do not cross-react with IHNV or
SVCV (Olesen & Jørgensen 1986).

Virology results from this study, as well as additional testing of tissue samples of Lake
St. Clair fish by the AAHL, confirm the inability to isolate VHSV during the summers of
2007 and 2010. Antibodies, however, were detected in all sampling years. The duration
of virus in tissues, induction and duration of the immune response, and outcome of
rhabdovirus infection are influenced by several factors, including virus dose, water
temperature, and various properties of the host (LaPatra 1998, Lorenzen & LaPatra
1999). When Kim (2010) infected muskellunge held at 11°C with VHSV-IVb by
immersion, virus titers peaked within the first 2 wk postinfection (pi). Most fish sampled
later in the disease course had no, or substantially lower, virus titers in tissues,

73

however, virus was detectable in some fish up to 9 wk pi. At higher temperatures, VHSV
is cleared from tissues more rapidly (Jørgensen 1982, Goodwin & Merry 2011).
Antibody titers peak later in the disease course, by approximately 6 wk in rainbow trout
infected with VHSV-I at 13°C (Olesen & Jørgensen 1986), and by 11–16 wk in
muskellunge infected with VHSV-IVb at 11°C (Millard & Faisal 2012). The duration of
VHSV-neutralizing antibodies is not fully understood, but antibodies remain detectable
in some fish 1 yr pi (Olesen & Jørgensen 1986).

In 2006, when fish were collected immediately following a VHSV outbreak, the virus was
readily isolated in cell culture. Antibodies were only detected in 14% of the serum
samples collected from these fish, and all antibody-positive individuals were also viruspositive, indicating an early convalescent stage of infection. Neutralizing antibodies
were detected in >85% of muskellunge sampled in May of 2007, 2009 and 2010,
indicating that the majority of adult muskellunge have survived encounters with the
virus. Given the high proportion of antibody-positive individuals, it seems likely that adult
muskellunge have established immunity to the virus.

In the absence of acute disease outbreaks, serologic testing would aid in the
identification of fish populations that previously have been infected with VHSV. Blood
samples can be taken non-lethally, which avoids the unnecessary sacrifice of valuable
and endangered fish stocks. Our results suggest that muskellunge is a good species to
sample non-lethally as an indicator of past VHSV-IVb exposure. Assessing immunity

74

among Great Lakes fish stocks would provide a more accurate understanding of VHSV
distribution, and could potentially be used to predict the vulnerability of a given
population to VHSV outbreaks, both of which would provide valuable information for
fisheries managers.

ACKNOWLEDGMENTS

We thank the Michigan Department of Natural Resources for assistance with sample
collection, Scott LaPatra (Clear Springs Foods, Inc.) for valuable insight with the
neutralization assay and for providing the IHNV isolate, and Stephen Kaattari (Virginia
Institute of Marine Science, College of William and Mary) for the monoclonal antimuskellunge IgM antibody. This study was supported with funding from the U.S. Fish
and Wildlife Service (Grant USDI US 30181AG013 FWS).

75

APPENDIX

76

Table 3.1. Distribution of neutralizing titers against viral hemorrhagic septicemia virus
in 118 serum samples from naïve juvenile muskellunge (Esox masquinongy). Group I (~
4 mo.; 17 cm, 1 SD; 19 g, 4 SD), Group II (~ 6 mo.; 16 cm, 2 SD; 17 g, 5 SD), Group III
(~ 22 mo.; 32 cm, 1 SD; 148 g, 21 SD).

a

Titer

Group I (n = 30)

Group II (n = 46)

<20
20
40
80

30
0
0
0

40
1
4
1

Group III (n = 7)
35
5
2
0

a

Overall
105 (89.0%)
6 (5.1%)
6 (5.1%)
1 (0.9%)

Blood was collected non-lethally at 6-wk intervals (6 time points; 42 total sera).

77

Table 3.2. Serum neutralization specificity of Lake St. Clair muskellunge (1-6; Esox
masquinongy) and freshwater drum (7; Aplodinotus grunniens). Sera were tested in
parallel with viral hemorrhagic septicemia virus (VHSV-IVb) and infectious
hematopoietic necrosis virus (IHNV) using a 50% plaque neutralization test (50% PNT).

50% PNT titers to:
Serum
no.
1
2
3
4
5
6
7

VHSV-IVb

IHNV

20
320
320
1,280
10,240
20,480
40

<20
<20
<20
20
<20
<20
<20

78

Table 3.3. Prevalence (%) of viral hemorrhagic septicemia virus (VHSV-IVb) neutralizing antibodies among Lake St. Clair
fish species.

Species

No. antibody-positive/No. tested (prevalence [%])

a

Muskellunge
Northern pike

b

2004

2006

2007

2009

2010

Overall

—

1/3 (33)

6/6 (100)

9/10 (90)

18/21 (86)

34/40 (85)

2/10 (20)

3/4 (75)

—

—

10/16 (63)

15/30 (50)

Mean
positive
titer

SD

c

d

12,113

11,699

e

608

449

f

201

339

Freshwater drum

—

0/11

13/40 (33)

—

2/10 (20)

15/61 (25)

Smallmouth bass

—

—

—

0/16

3/25 (12)

3/41 (7)

213

92

Channel catfish

—

—

—

—

0/25

0/25

NA

NA

Lake sturgeon

—

—

—

—

0/15

0/15

NA

NA

Quillback

—

—

—

—

0/7

0/7

NA

NA

Rock bass
Shorthead
redhorse
Silver redhorse

—

0/3

—

—

0/24

0/27

NA

NA

—

0/2

—

—

0/6

0/8

NA

NA

—

0/5

—

—

0/13

0/18

NA

NA

Walleye

—

—

—

—

0/10

0/10

NA

NA

White perch

—

—

—

—

0/12

0/12

NA

NA

Yellow perch

—

—

—

—

0/3

0/3

NA

NA

Overall

2/10 (20)

4/28 (14)

19/46 (41)

9/26 (35)

33/187 (18)

67/297 (23)

a
b

Scientific names are provided in text.
Dashes indicate no sera were collected.

79

c

NA

NA

Table 3.3 (contʼd)
c

NA = Not applicable.

d

Significantly higher antibody prevalence compared to northern pike (P = 0.0032), smallmouth bass, and freshwater drum
(P < 0.0001).
e
f

Significantly higher antibody prevalence compared to smallmouth bass (P < 0.0001) and freshwater drum (P = 0.0192).
Significantly higher antibody prevalence compared to smallmouth bass (P = 0.0332).

80

a

Table 3.4. Maturity, sex, age, and size of Lake St. Clair muskellunge (Esox masquinongy) included in this study .
Sex

Immature
Mature

a

b

Range (mean, SD)

M

F

ND

Age (years)

3

0

1

2–3 (3, 1)

23

11

0

6–20 (11, 3)

Length
(cm)
45–73
(61, 12)
88–127
(107, 11)

Weight (g)
452–2200
(1370, 717)
4000–13,800
(8285, 3079)

No. antibody positive
(>80)/ No. sampled
c

0/4

34/34

Table excludes two muskellunge (titers of <20) captured in 2006 due to incomplete measurements.

b

Classification of immature fish was based upon the absence of ripe gametes and was confirmed by examination of
reproductive organs. In one immature muskellunge, sex could not be determined (ND).
c

Antibody prevalence of immature muskellunge was significantly less than that of mature muskellunge (P < 0.0001).

81

Figure 3.1. Lake St. Clair (Lake Erie watershed, USA) showing sampling sites. Most
fish were captured using trap nets (!) positioned between 42° 39'N, 82° 46'W and 42°
37'N, 82° 46'W in Anchor Bay. Lake sturgeon (Acipenser fulvescens) were captured
using survey setlines (!: 42° 37'N, 82° 37'W). Smallmouth bass (Micropterus dolomieu)
from a mortality event in June of 2009 (!: 42° 29'N, 82° 53'W), and a single
muskellunge (Esox masquinongy) collected May 2006 (": 42° 21'N, 82° 54'W), were
caught using scoop nets.

82

Figure 3.2. Antibody titers measured by a 50% plaque neutralization test (50% PNT)
against viral hemorrhagic septicemia virus (IVb) from Lake St. Clair fish species that
were antibody-positive: muskellunge (! Esox masquinongy), northern pike (! E.
lucius), smallmouth bass (! Micropterus dolomieu), freshwater drum (" Aplodinotus
grunniens). Neutralization titers of >80 are considered positive for muskellunge and
northern pike. The parenthesis denote the number of sera with titers of <20/number
sampled.

a

All freshwater drum in 2006 and smallmouth bass in 2009 had titers of <20.

83

CHAPTER 4
APPLICATION OF A COMPETITIVE ELISA FOR DETECTING CIRCULATING
ANTIBODIES AGAINST VIRAL HEMORRHAGIC SEPTICEMIA VIRUS (GENOTYPE
IVB) IN REPRESENTATIVE FISH SPECIES IN LAKE ST. CLAIR, MICHIGAN, USA

84

ABSTRACT

Viral hemorrhagic septicemia virus genotype IVb (VHSV-IVb) is a serious
novirhabdovirus of freshwater fish that has invaded the Laurentian Great Lakes region
or North America within the last decade. We tested sera (n = 113) from nine fish species
for the presence of antibodies against VHSV-IVb using a competitive enzyme-linked
immunosorbent assay (cELISA). Fish were sampled in May 2010 from Lake St. Clair, a
VHSV-IVb endemic lake in Michigan, USA. Inhibition of the virus-specific rabbit
antiserum by sera of freshwater drum (Aplodinotus grunniens), smallmouth bass
(Micropterus dolomieu), and channel catfish (Ictalurus punctatus) strongly suggests the
presence of VHSV-IVb antibodies in sera of these species. This cELISA does not
require species-specific reagents and is a rapid, efficient and non-lethal test for
determining indirectly the presence of VHSV-IVb in fish populations.

BODY

Viral hemorrhagic septicemia virus (VHSV) (family Rhabdoviridae, genus
Novirhabdovirus) is a serious viral pathogen of fishes worldwide (Wolf, 1988). The
introduction of VHSV-genotype IVb (VHSV-IVb) in the Laurentian Great Lakes of North
America has led to mass mortality events among naïve populations of wild, freshwater
fish (Elsayed et al. 2006, Groocock et al. 2007, Lumsden et al. 2007, Faisal et al. 2012).
VHSV-IVb has been isolated from at least 28 freshwater fish species (United States

85

Department of Agriculture, Animal and Plant Health Inspection Service, 2008). Existing
surveillance programs are hampered by the inability to detect asymptomatically infected
and recovered fishes using the standard method of virus isolation in cell culture.
Additionally, for threatened and endangered fish species, a non-lethal test is desired.
Serological assays, that detect virus-specific antibodies, are a viable alternative or
supplementary tool to determine VHSV exposure in a fish population (HattenbergerBaudouy et al. 1995, Millard & Faisal, 2012b, Schyth et al. 2012).

Recently, we used a serum neutralization assay [50% plaque neutralization test (PNT)]
to assess the presence of neutralizing antibodies in wild and experimentally exposed
muskellunge (Esox masquinongy) (Millard & Faisal 2012a, 2012b). Over a period of
several years, antibodies were detected in a much greater proportion of wild-caught
muskellunge from Lake St. Clair, Michigan than was the virus. While the PNT remains a
valuable tool for measuring protective antibody levels, not all circulating antibodies have
functional VHSV-neutralizing activity. Furthermore, the assay requires 5-6 days to
perform and is not cost effective for large-scale serological surveys. For these reasons,
indirect enzyme-linked immunosorbent assays (ELISA) have been developed and used
to measure VHSV-genotype I antibodies in rainbow trout (Oncorhynchus mykiss)
(Jørgensen et al. 1991, Olesen et al. 1991, Fernandez-Alonso et al. 1998, Encinas et al.
2011). The absence of commercially available anti-immunoglobulins for many Great
Lakes species makes assessing VHSV-IVb antibodies by indirect ELISA not feasible.
Conversely, competitive ELISAs (cELISAs) are well suited as antibody screening

86

assays for pathogens with wide host ranges. Antibody levels are measured according to
the degree to which sera inhibit the binding of a virus-specific hyperimmune antiserum.
cELISAs have previously been developed for several rhabdoviruses including spring
viremia of carp virus (SVCV; Dixon et al. 1994), vesicular stomatitis virus (Alvarado et
al. 2002) and rabies virus (Zhang et al. 2009).

In this study, we applied a cELISA to assess the presence of VHSV-IVb antibodies in
113 serum samples from nine fish species representing a range of trophic levels. Fish
were caught on May 5 and 7, 2010 from the Anchor Bay area of Lake St. Clair,
Michigan, a lake that is known to be enzootic for VHSV-IVb. The previous summer,
VHSV-IVb was isolated from smallmouth bass (Micropterus dolomieu) that were a part
of a mortality event that occurred near to the sampling site. Fish were caught using trap
nets positioned between 42° 39'N, 82° 46'W and 42° 37'N, 82° 46'W. Tested species
were channel catfish (Ictalurus punctatus), rock bass (Ambloplites rupestris), northern
pike (Esox lucius), silver redhorse (Moxostoma anisurum), smallmouth bass, freshwater
drum (Aplodinotus grunniens), white perch (Morone Americana), shorthead redhorse
(Moxostoma macrolepidotum), and quillback (Carpiodes cyprinus). The number of fish
tested ranged from 5 to 24 per species (Table 4.1). Blood samples were collected nonlethally. Sera were separated after centrifugation (2500 ! g, 10 min, 4°C) and stored at 80°C.

87

Fish sera were heat-treated (45°C for 30 min) and incubated overnight in phosphate
buffered saline (PBS) with 1% nonfat dried milk (PBS-1%NFDM, Sigma-Aldrich, St.
Louis, Missouri, USA) prior to running the cELISA to reduce non-specific binding. Plates
were washed five times with PBS containing 0.05% Tween 20 (Sigma-Aldrich) in an
automated microplate washer following each step unless noted. Briefly, 96-well
®

polystyrene microplates (Microlon 600, Greiner Bio-One, Monroe, North Carolina, USA)
-1

were coated with purified VHSV-IVb (1 µg·mL ) overnight at 4°C. Unbound sites were
blocked with 5% nonfat dried milk (PBS-5%NFDM, Sigma-Aldrich). Fish test sera and
control sera, diluted 1:10 in PBS-1%NFDM, were added to duplicate wells and
incubated prior to addition of rabbit hyperimmune serum to VHSV-IVb (1:512,000). A
commercial anti-rabbit IgG horseradish peroxidase conjugate (Sigma-Aldrich) was used
as the detection antibody and a colorimetric substrate was used to develop the reaction.
After 30 min and without washing, color development was stopped with 3 M sulfuric
acid. Results were interpreted as the percent inhibition of the optical density (OD) of the
rabbit hyperimmune serum according to the following formula: [1-(Average OD405/490
fish serum/ Average OD405/490 rabbit serum)] x 100

The rabbit hyperimmune serum to VHSV-IVb used in the cELISA as the competition
antibody was produced by immunization of New Zealand white rabbits with purified
VHSV-IVb (data not shown). Negative inhibition values were treated as zeros in all
analyses. Data were analyzed using the MIXED procedure in SAS Version 9.2 (SAS

88

Institute, Inc. 2010). Mean percent inhibition for each species was compared to the
overall mean percent inhibition. Differences in mean percent inhibition among species
were tested using pair-wise comparisons of least-squares means. The type-I error rate
was set at 0.05.

Some fish sera were able to successfully block the rabbit hyperimmune serum from
binding to purified VHSV-IVb during the assay (Figure 4.1). The overall mean percent
inhibition for all 113 fish tested was 8.8% (SD = 16.1%) and the overall median was
3.6%. Inhibition levels were the highest in freshwater drum. Percent inhibition values for
this species ranged from 0.0 to 88.3% (mean = 29.4%, SD = 33.3%) (Table 4.1).
Smallmouth bass percent inhibition ranged from 0.0 to 58.9% (mean = 20.0%, SD =
21.1%). The mean of both freshwater drum [t = 4.62, degrees of freedom (df) = 104, p <
0.0001] and smallmouth bass (t = 2.76, df = 104, p = 0.0034) was significantly greater
than the overall mean percent inhibition calculated for all 113 sera. Inhibition was also
detected in several sera of channel catfish (range = 0.0 to 63.9%; mean = 10.9%, SD =
14.7%). The mean of this species was, however, not significantly different than the
overall mean (t = 0.74, df = 104, p = 0.2311). Mean percent inhibition levels of rock bass
(t = 1.95, df = 104, p = 0.9732), northern pike (t = 1.69, df = 104, p = 0.9533), silver
redhorse (t = 1.51, df = 104, p = 0.9327), white perch, (t = 1.15, df = 104, p = 0.8732),
shorthead redhorse (t = 1.18, df = 104, p = 0.8795) and quillback (t = 0.43, df = 104, p =
0.6670) also did not differ from the overall mean. Percent inhibition of these species
ranged from 0.0 to 18.4%.

89

Mean percent inhibition levels differed among the resident fish species, F8,104 = 5.17, p
< 0.0001. Freshwater drum had significantly greater mean percent inhibition than
channel catfish (t = 3.48, df = 104, p = 0.0007), rock bass (t = 4.91, df = 104, p <
0.0001), northern pike (t = 4.62, df = 104, p < 0.0001), silver redhorse (t = 4.50, df =
104, p < 0.0001), white perch (t = 3.93, df = 104, p = 0.0002), shorthead redhorse (t =
3.63, df = 104, p = 0.0004), and quillback (t = 3.02, df = 104, p = 0.0032). Smallmouth
bass had a greater mean percent inhibition compared to rock bass (t = 3.38, df = 104, p
< 0.0010), northern pike (t = 3.18, df = 104, p = 0.0020), silver redhorse (t = 3.05, df =
104, p = 0.0029), white perch (t = 2.63, df = 104, p = 0.0097) and shorthead redhorse (t
= 2.49, df = 104, p = 0.0145). The mean percent inhibition of freshwater drum was not
different than smallmouth bass (t = 1.55, df = 104, p = 0.1243), nor were any of the
other pairwise comparisons between species (Table 4.2).

Without established positive-negative cELISA thresholds for the species tested in this
research we were unable to estimate seroprevalence for each species, which hinders
our ability to conclude what species were producing VHSV-IVb antibodies.
However, the wide range in inhibition values for species such as smallmouth bass,
freshwater drum, and channel catfish is suggestive that at least some individuals of
these species were producing VHSV-IVb antibodies. For other species, such as silver
redhorse, white perch, and quillback where maximum inhibition values ranged from 9.5
to 18.4%, VHSV-IVb antibody production remains less certain. Current studies are

90

directed toward determining positive-negative threshold(s) for the cELISA. These
studies are challenged by the absence of reliable sources of fish for some species that
we can be certain have not been exposed to VHSV-IVb.

The presence of high inhibitions in several fish species residing in Lake St. Clair is
suggestive of current or previous infection with VHSV-IVb. Both smallmouth bass and
freshwater drum have been involved in fish kills from which VHSV-IVb was isolated
during the four-year period preceding collection (Faisal et al. 2012). Prior to collection in
June 2009, VHSV-IVb was isolated from tissues of smallmouth bass involved in a
mortality episode. Differences in the occurrence and levels of antibodies among species
could be related to differences in susceptibility to VHSV-IVb, immune response to the
virus or life history. A longer life span, piscivorous diet or engaging in certain behaviors
such as spawning congregations would likely increase exposures to VHSV-IVb, and
thereby the presence of antibodies in surviving fish.

Some sera of smallmouth bass, freshwater drum, and channel catfish produced
considerable inhibition levels by cELISA, but were negative for neutralizing antibodies
by PNT (titers <20) in our previous study (Millard & Faisal 2012b). This suggests the
presence of non-neutralizing antibodies against the glycoprotein, or antibodies against
other viral proteins exposed in the virus preparation (Olesen et al. 1991). Detection of
antibodies in non-neutralizing sera by ELISA-based assays has been reported
previously for rainbow trout (Olesen et al. 1991, Encinas et al. 2011) and common carp

91

(Cyprinus carpio) against VHSV-genotype I and SVCV, respectively (Dixon et al. 1994).
Some studies have found that non-neutralizing antibodies last longer after infection
compared to neutralizing antibodies (Olesen et al. 1991, Encinas et al. 2011). It is also
possible that some species have a reduced or absent neutralizing antibody response to
VHSV. Pacific herring (Clupea pallasii) for example do not mount a detectable
neutralizing antibody response against VHSV-IVa, despite survivors being protected
against subsequent viral challenges (Kocan et al. 2001, Hershberger et al. 2007).

Despite the limitations, the results of this study suggest that the developed cELISA can
be applied to many different species without the need for custom immunoglobulins
against each species tested. Furthermore, while not the intent of the present
investigation, it appears that the cELISA is capable of detecting a broader range of
VHSV-IVb antibodies compared to the PNT, and as such, may be more sensitive in
detecting prior virus exposure for some species. Our results suggest that smallmouth
bass and freshwater drum would be good species to target for future VHSV-IVb
serological surveys. The VHSV-IVb cELISA is a promising tool for cost-effective and
timely VHSV-IVb serological surveys.

ACKNOWLEDGEMENTS

We thank T.O. Brenden (Quantitative Fisheries Center, Michigan State University) for
his help with statistical analyses and critical review of this manuscript. This study was

92

supported with funding from the United States Fish and Wildlife Service (USFWS
F11AP00569/F11AP00105).

93

APPENDIX

94

Table 4.1. Viral hemorrhagic septicemia virus-genotype IVb competitive ELISA results
of 113 fish of 9 species sampled from Lake St. Clair, MI, May 2010. Bold font indicates
species for which the mean inhibition was significantly different than the overall mean.
SD = standard deviation

Percent inhibition
Species

Scientific name

Channel catfish
Rock bass
Northern pike
Silver redhorse
Smallmouth bass
Freshwater drum
White perch
Shorthead redhorse

Ictalurus punctatus
Ambloplites rupestris
Esox lucius
Moxostoma anisurum
Micropterus dolomieu
Aplodinotus grunniens
Morone Americana
Moxostoma
macrolepidotum
Carpiodes cyprinus

Quillback
Overall

95

No.
fish
24
21
14
14
12
10
8
5
5
113

Mean

SD

Median

Range

10.9
2.8
2.4
3.1
20.0
29.4
3.1
1.4

14.7
2.5
2.3
4.1
21.1
33.3
4.0
1.7

5.3
1.9
1.4
1.7
7.4
9.4
0.6
0.2

0 – 63.9
0 – 6.9
0 – 8.1
0 – 14.5
0 – 58.9
0 – 88.3
0 – 9.5
0 – 3.6

6.1
8.8

8.6
16.1

0.1
3.6

0 – 18.4
0 – 88.3

Table 4.2. Pairwise differences in least-squares means for percent inhibition values
from 13 fish species from Lake St. Clair, Michigan. Blood samples were collected in May
of 2010 and tested for VHSV-IVb antibodies by cELISA. Differences in least-squares
means (Diff.), t and p values for comparisons between species are shown. Degrees of
freedom = 104. The type-I error rate was set at 0.05. Bolded p values indicate
statistically significant comparisons. Species abbreviations are as follows: CCF =
channel catfish, ROB = rock bass, NOP = northern pike, SIR = silver redhorse, SMB =
smallmouth bass, FRD = freshwater drum, WHP = white perch, SHR = shorthead
redhorse, QUI = quillback.

Species Species
CCF
ROB
CCF
NOP
CCF
SIR
CCF
SMB
CCF
FRD
CCF
WHP
CCF
SHR
CCF
QUI
ROB
NOP
ROB
SIR
ROB
SMB
ROB
FRD
ROB
WHP
ROB
SHR
ROB
QUI
NOP
SIR
NOP
SMB
NOP
FRD
NOP
WHP
NOP
SHR
NOP
QUI
SIR
SMB
SIR
FRD
SIR
WHP
SIR
SHR
SIR
QUI
SMB
FRD

Mean percent inhibition
Diff.
t
p
8.12
1.93
0.0564
8.50
1.79
0.0758
7.80
1.65
0.1027
-9.10
-1.83
0.0706
-18.44
-3.48
0.0007
7.84
1.36
0.1758
9.55
1.38
0.1707
4.85
0.70
0.4854
0.37
0.08
0.9388
-0.32
-0.07
0.9471
-17.22
-3.38
0.0010
-26.57
-4.91
<.0001
-0.28
-0.05
0.9614
1.43
0.20
0.8387
-3.27
-0.47
0.6415
-0.70
-0.13
0.8961
-17.60
-3.18
0.0020
-26.94
-4.62
<.0001
-0.66
-0.11
0.9163
1.06
0.14
0.8859
-3.65
-0.50
0.6202
-16.90
-3.05
0.0029
-26.24
-4.50
<.0001
0.04
0.01
0.9950
1.75
0.24
0.8117
-2.95
-0.40
0.6885
-9.35
-1.55
0.1243
96

Table 4.2 (contʼd)
SMB
SMB
SMB
FRD
FRD
FRD
WHP
WHP
QUI

WHP
SHR
QUI
WHP
SHR
QUI
SHR
QUI
QUI

16.94
18.65
13.95
26.28
28.00
23.29
1.71
-2.99
-4.70

2.63
2.49
1.86
3.93
3.63
3.02
0.21
-0.37
-0.53

0.0097
0.0145
0.0657
0.0002
0.0004
0.0032
0.8314
0.7104
0.5986

97

Figure 4.1. Distribution of percent inhibition values of sera from 113 fish of nine
species from Lake St. Clair, Michigan, May 2010. Sera were tested for antibodies
against viral hemorrhagic septicemia virus-genotype IVb using a competitive ELISA.
The mean percent inhibition value of each species is indicated by a red dash. Dashed
and dotted horizontal lines indicate the overall mean (8.8) and median (3.6) percent
inhibition values.

98

CHAPTER 5
DETECTION OF ANTIBODIES TO VIRAL HEMORRHAGIC SEPTICEMIA VIRUS-IVB
IN SERA OF MUSKELLUNGE (ESOX MASQUINONGY) USING A COMPETITIVE
ELISA

99

ABSTRACT

A competitive enzyme-linked immunosorbent assay (cELISA) was developed for the
detection of antibodies to viral hemorrhagic septicemia virus-genotype IVb (VHSV-IVb)
in fish sera. Assay conditions were standardized using known negative and positive
muskellunge (Esox masquinongy) sera. A positive-negative threshold of 14.6%
inhibition was established based on analysis of sera of 60 muskellunge with no previous
exposure to VHSV-IVb. The cELISA was then used to investigate immune responses of
wild muskellunge sampled from five water bodies in Michigan and Wisconsin, USA
between the years of 2005 and 2012. Antibodies were detected in fish from Lake St.
Clair, Michigan and Lower Fox River/Green Bay, Wisconsin. Both water systems were
considered enzootic for VHSV-IVb. Additionally, antibodies were detected in
muskellunge from Thornapple Lake, a Michigan inland lake previously considered
negative for VHSV-IVb based on virus isolation methods. Muskellunge populations from
Lake Hudson, MI and Butternut Lake, Wisconsin lacked evidence of an immune
response to VHSV-IVb. When results of the cELISA were compared to the 50% plaque
neutralization test for several groups of fish, there was 78.4% agreement between the
tests for antibody presence. The cELISA is a rapid and efficient test for the detection of
binding antibodies to VHSV-IVb and will be a useful non-lethal tool for monitoring the
spread of this serious pathogen.

INTRODUCTION

100

Viral hemorrhagic septicemia virus (VHSV) (family Rhabdoviridae, genus
Novirhabdovirus) (Dietzgen 2012) is the causative agent of viral hemorrhagic septicemia
(VHS). VHS is a World Organization for Animal Health (OIE) reportable disease of
freshwater and marine fish. In the Great Lakes region of North America, a new
sublineage of the virus, designated VHSV-IVb, has emerged within the last decade and
caused large-scale mortality events in wild freshwater fish populations (Elsayed et al.
2006, Groocock et al. 2007, Lumsden et al. 2007, Faisal et al. 2012). VHSV-IVb has
since been isolated from at least 28 fish species (United States Department of
Agriculture Animal and Plant Health Inspection Service 2008) and concerns exist
regarding its continued spread to naïve fish populations and potential introduction into
aquaculture systems (VHSV Expert Panel and Working Group 2010). Ongoing VHSVIVb surveillance efforts aim to determine virus distribution and delineate VHSV-IVb
positive and negative zones in the Great Lakes basin. Testing for VHSV-IVb is
commonly conducted according to the international standard, which is virus isolation
from fish tissues in cell culture followed by confirmation of the isolated virus by reversetranscriptase polymerase chain reaction (OIE 2012, United States Fish and Wildlife
Service and American Fisheries Society-Fish Health Section 2010). During and
immediately following VHS outbreaks, VHSV can be readily isolated in this manner (OIE
2012, Schyth et al. 2012). Detection of the virus from clinically healthy fish (carriers) is
more problematic and dependent on many factors such as environmental temperature
and time since exposure (OIE 2012).

101

In the absence of clinical VHS disease outbreaks, such as for surveillance purposes or
surveying enzootic populations, the use of antibody-based detection assays can
increase the capacity to screen populations for exposure to VHSV (HattenbergerBaudouy 1995, LaPatra et al. 1996, Millard and Faisal 2012b, Schyth et al. 2012). After
the onset of antibody production, circulating antibodies have been detected in the range
of months to years after VHS infection in sera of surviving fish (Olesen & Jørgensen
1986, Olesen et al. 1991, Enzmann & Konrad 1993, Millard and Faisal 2012a). Another
benefit of serological testing is that it can be done non-lethally, an attribute that is
especially ideal for the study of threatened, valuable and long-lived species. It is
important to consider that the presence of virus-specific antibody indicates
seroconversion but does not necessarily indicate fish ever developed clinical VHS
disease and does not distinguish between subclinically infected and recovered fish.

Neutralizing antibodies are directed against viral surface glycoproteins (G) and are an
important component of the protective immune response of rainbow trout
(Oncorhynchus mykiss) to VHSV (Lorenzen et al. 1990, Bearzotti et al. 1995, Lorenzen
& LaPatra 1999). This set of antibodies is measured by neutralization tests such as the
50% plaque neutralization tests (PNT) that rely on an exogenous source of fish
complement (Dorson & Torchy 1979). Rainbow trout surviving infection with VHS also
produce binding antibodies and several indirect enzyme-linked immunosorbent assays
(ELISA) have been developed for their detection (Jørgensen et al. 1991, Olesen et al.

102

1991, Fernandez-Alonso et al. 1998, Encinas et al. 2011). In some of these studies,
binding antibodies were detected in a higher percentage of fish and lasted longer than
did neutralizing antibodies. Compared to the PNT, ELISA was considered a more
feasible test for surveillance studies given the capacity to screen a larger number of
samples simultaneously, quicker test results, and reduced processing costs (Olesen et
al. 1991, LaPatra 1996).

In this study we developed a competitive ELISA (cELISA) to detect antibodies to VHSVIVb and applied its use to sera from wild muskellunge (Esox masquinongy) from
systems where the virus is enzootic and in systems where the virus has never been
detected. Competitive ELISAs, such as the one developed by Dixon et al. (1994) for
detection of spring viremia of carp virus (SVCV) antibodies, utilize a virus-specific rabbit
antiserum to compete with fish sera for virus binding sites. As such, the cELISA can be
used on multiple fish species without the need to develop specific anti-immunoglobulin
reagents for each species tested as would be necessary for indirect ELISAs. This is a
more feasible approach given the wide range of fish hosts for VHSV-IVb and the
absence of commercially available antisera to immunoglobulins for many of these
species. We focused on muskellunge for this initial study due to the known susceptibility
of this species to VHSV-IVb (Kim & Faisal 2010) and the availability of sera from
unexposed, experimentally infected and naturally infected populations. Furthermore,
muskellunge have been shown to be good indicators for VHSV-IVb presence in an

103

ecosystem based on the presence of neutralizing antibodies in a previous study (Millard
& Faisal 2012b).

MATERIALS AND METHODS

Preparation of virus stock and purified virus. The MI03 index strain of VHSV-IVb
(Elsayed et al. 2006) was used throughout this study. To produce purified virus for use
as the cELISA coating antigen and for immunization of rabbits, VHSV-IVb was
propagated in Epithelioma papulosum cyprini (EPC) cells (Winton et al. 2010) in
minimum essential medium with Earleʼs salts (EMEM; Life Technologies) supplemented
with 10% tryptose phosphate broth (BD Biosciences), 2.0 mM L-glutamine (Life
Technologies) and 12 mM sodium bicarbonate (Sigma-Aldrich). After cells had lifted,
flasks were briefly frozen at -20°C and cell debris were removed by centrifugation (5200
®

! g, 30 min, 4°C). Resulting supernatants were filtered using a Corning bottle-top
vacuum filter system and ultracentrifuged using a SW 32 Ti rotor (Beckman Coulter) for
80,000 ! g for 2.5 hr at 4°C (Olesen et al. 1999). Concentrated virus pellets were
resuspended in TE buffer (20 mM Tris-HCl, pH 7.5 and 1 mM EDTA, pH 8.0) and stored
at -80°C.

For purification, virus suspensions were overlaid on 10-60% discontinuous sucrose
gradient columns and ultracentrifuged (80,000 ! g, 2.5 hr, 4°C). The fraction containing
the purified virus was used immediately for immunization of rabbits. For use as the

104

cELISA coating antigen, this fraction was given a final wash by pelleting through TE
buffer (80,000 ! g, 150 min, 4°C). The resulting pellet of purified virus was resuspended
in phosphate buffered saline [PBS (0.138 M NaCl; 0.0027 M KCl); pH 7.4; SigmaAldrich), divided into aliquots, and stored at -80°C until use. The protein concentration of
®

®

the final virus preparation was determined using a Qubit Fluorometer with the Qubit
Protein Assay Kit (Life Technologies).

The virus used in experimental infections and the PNT was propagated in EPC cells as
previously described (Millard & Faisal 2012b). The titers of the virus stocks were
determined by a virus plaque assay on EPC cells. Cells were pretreated with a 7%
solution of polyethylene glycol (PEG; 20,000 MW; JT Baker) (Batts and Winton 1989).

Production of polyclonal antibody to VHSV-IVb. Three New Zealand white rabbits
were given a series of two immunizations of purified VHSV-IVb over a 5-week period.
Purified virus (0.5 mL, 10

10-12

TCID50) was emulsified 1:1 in Freundʼs complete

adjuvant (Sigma-Aldrich) for the first immunization and in Freundʼs incomplete adjuvant
(Sigma-Aldrich) for the second. The immunizations were administered subcutaneously
-1

over the dorsum (10 sites, 0.1 mL·site ). Serum was collected from each rabbit prior to
immunization (pre-immune sera) and 3 weeks after the second immunization
(hyperimmune serum). Sera were heat-treated for 45°C for 45 min, divided into aliquots,
and stored at -80°C. Rabbits and fish were cared for in accordance with guidelines

105

defined by Michigan State Universityʼs (MSU) Institutional Animal Care and Use
Committee (AUF 07/07-123-00, 09/10-140-00, 02/10-013-00).

Competitive ELISA procedure and control sera. The immunoassay was initially
established as an indirect ELISA system following general principles described by Voller
et al. (1979) and Crowther (2009). A variety of types of plates, blocking and diluent
reagents, and incubation times and temperatures were tested in preliminary trials and
titrations were carried out to determine several suitable reagent combinations. Purified
-1

virus (coating antigen) at a starting dilution of 16 µg·mL

and rabbit hyperimmune

serum to VHSV-IVb at a starting dilution of 1:2000 were tested in 2-fold serial dilutions.
The anti-rabbit IgG horseradish peroxidase conjugate (Sigma-Aldrich) was tested at
dilutions of 1:1000, 1:2500 and 1:5000. Final concentrations for the competitive
immunoassay were then optimized using three fish sera controls. The positive control
sera were from two adult muskellunge collected from a VHSV-IVb positive water body in
Michigan that were positive for neutralizing antibodies by PNT (titers of 20,480 and
1280). The negative control consisted of a pooled serum sample from naïve hatcheryreared muskellunge from outside the Great Lakes basin that were negative by PNT.
After standardization trials, fish sera controls were included on each assay plate and
consisted of the high positive control (PNT titer of 20,480), a low positive control (1:80
dilution of the high positive), and the negative control serum.

106

®

Polystyrene microplates (96-well, Microlon 600 with chimney wells; Greiner Bio-One)
were used as the solid phase for the cELISA. Plates were washed five times with PBS
containing 0.05% Tween 20 (PBS-T20; Sigma-Aldrich) in an automated microplate
washer (BioTek, ELx405™) following each step, and were sealed for all incubations
-1

unless otherwise stated. Briefly, assay plates were coated with 100 µl·well
-1

VHSV-IVb in PBS at a concentration of 1 µg·mL

of purified

and incubated overnight (16-18 hr) at

4°C in a humid chamber. Prior to running the assay, plates were washed and
-1

unoccupied sites were blocked with the addition of 420 µl·well

of PBS containing 5%

nonfat dried milk (PBS-5%NFDM; Sigma-Aldrich) for 1 hr at 37°C. Fish control and test
sera, diluted 1:10 in PBS-1%NFDM (hereafter referred to as sample diluent) and
pretreated as described in the following paragraph, were then added to duplicate wells
-1

(100 µl·well ). At this time, sample diluent alone was added to duplicate diluent control
wells and 12 wells reserved for the rabbit serum control. After incubating for 1 hr at
25°C, rabbit hyperimmune serum to VHSV-IVb (1:512,000 in sample diluent, 100
-1

µl·well ) was added to rabbit serum control wells and all wells that had received fish
serum during the previous step. Sample diluent alone was again added to diluent control
wells and plates were incubated for 1 hr at 25°C. A commercial anti-rabbit IgG
horseradish peroxidase conjugate (Sigma-Aldrich), 1:1000 in sample diluent, was then
-1

added to all wells (100 µl·well ) and the incubation period of 1 hr at 25°C was repeated.
Enzymatic development of the plates was started by the addition of substrate (0.4

107

-1

mg·mL

o-Phenylenediamine in phosphate citrate buffer containing 3 mM hydrogen
-1

peroxide; 100 µl·well ) and color development proceeded for 30 min at 25°C in the
dark. Without washing, the reaction was stopped with 3 M sulfuric acid. An absorbance
microplate reader (BioTek ELx405) was used to read dual wavelength optical density at
405 and 490 nm (OD405/490). The average OD405/490 of the diluent control wells was
subtracted from each well. Results were interpreted as the percent inhibition of the
average OD405/490 of the rabbit serum control. The following formula was used to
calculate percent inhibition: [1- (Average OD405/490 fish serum / Average OD405/490
rabbit serum control)] × 100. Negative inhibition values were treated as zeros in all
analyses.

All fish control and test sera were heat-treated for 30 min at 45°C. The night prior to
performing the assay, fish sera (22 µl each) were diluted 1:10 in sample diluent in 96well dilution plates that had been pre-blocked with PBS-5%NFDM. Rabbit serum
(1:1000) was also pre-incubated overnight prior to the assay in sample diluent.

Establishment of a positive-negative threshold for muskellunge. Muskellunge were
obtained from Rathbun National Fish Hatchery (Iowa Department of Natural Resources,
Moravia, Iowa), which is located outside the Great Lakes basin (hereafter referred to as
naïve muskellunge). Naïve muskellunge were kept in a flow-through system at the MSU
Research Containment Facility and fed fathead minnows (Pimephales promelas)

108

purchased from Anderson Minnow Farm (Lonoke, Arkansas). Naïve muskellunge and
minnows were certified to be free of VHSV and other reportable viruses. Blood was
drawn from the caudal vein of 60 naïve muskellunge (average weight ~80 g) using a
sterile syringe and needle. Blood samples were allowed to clot at 4°C overnight and
then centrifuged (2500 ! g, 10 min, 4°C). Sera were collected, divided into aliquots, and
stored at -80°C. These sera were tested by the cELISA in order to determine a baseline
level of inhibition present in sera of muskellunge that were not exposed to VHSV. The
positive-negative threshold for the cELISA test was established by calculating the upper
95% confidence limit to the 95

th

quantile of the inhibition values for the naïve

muskellunge. We chose to use this quantile approach for determining thresholds rather
than the more common “mean + 3 standard deviations (SD)” approach because its
interpretation is not dependent on the underlying distribution of percent inhibition values
of naïve fish sera. The approach is also more robust to decisions regarding how
negative inhibitions are calculated and to potential anomalous observations. The
quantiles and their upper 95% confidence limits were calculated in SAS 9.2 (SAS
Institute, Inc. 2010). Confidence limits for the quantiles were calculated using the
bootstrapping approach of He & Hu (2002). A total of 5,000 bootstrap samples were
used to calculate the confidence limits.

Use of sera from fish surviving experimental VHSV-IVb infection to test assay.
The cELISA was first tested using sera from two groups of muskellunge that were
experimentally challenged with VHSV-IVb (MI03 strain) by immersion in previous

109

studies. Fish (n = 60-90 per dose group) were challenged by immersion in aerated
3

5

-1

water containing 4 ! 10 or 10 plaque-forming units (pfu)·mL

VHSV-IVb

(experimentally exposed fish) or an equal volume of tissue culture media (mockchallenged fish). Fish were sampled between 1 and 15 weeks (11 and 165 degree days)
post-infection (PI). The first group of sera tested by the cELISA represents a subset of
these samples and consisted of 27 sera from individual experimentally exposed fish and
16 mock-challenged fish (2-6 fish per time-point). Further details of this trial can be
found in Millard and Faisal (2012a).

The second group of sera consisted of 42 sera collected from 22 individual muskellunge
(4 months of age, 13 g) that were survivors of two previous VHSV-IVb immersion trials
(Kim & Faisal 2012). Fish (n = 234) were initially exposed to a low dose of the virus (1.4
3

-1

! 10 pfu·mL ) and survivors were re-challenged 22 weeks (1700 degree days) later
6

-1

with doses ranging from 10 to 10 pfu·mL . The 22 fish that survived both challenges
were later tagged with passive integrated transponder (PIT) tags and serial blood
samples were collected from the same individuals 67 and 79 weeks (5200 and 6000
degree days) PI (45 and 57 weeks post-second exposure). Water temperature was
maintained at approximately 11 ± 1°C for both studies.

Application of the assay to field-collected sera of muskellunge. A total of 200
blood samples were obtained from wild muskellunge sampled from five water bodies in

110

Michigan (MI) and Wisconsin (WI) (Table 5.1). Fish were sampled prior to or during
spawning between the years of 2005 and 2012. Most fish were of reproductively mature
age. Four locations were within the Great Lakes basin and included two areas
considered enzootic for VHSV-IVb (Lake St. Clair, MI and Lower Fox River/Green Bay,
WI) and two inland lakes (Lake Hudson and Thornapple Lake, MI) considered negative
for VHSV-IVb based on past viral testing. The fifth water body (Butternut Lake, WI) was
a VHSV-negative inland lake located outside the Great Lakes basin. Muskellunge from
Lake St. Clair were sampled in spring 2010 and spring 2011 near Anchor Bay and the
Detroit River (Figure 5.1). Lower Fox River muskellunge were collected in spring 2010,
2011 and 2012 near the mouth of the Fox River at Green Bay. Lower Fox River
muskellunge are primarily residents of Green Bay waters of Lake Michigan except
during spawning. The last isolations of VHSV-IVb from fish in Lake St. Clair and Green
Bay (and connected waters) were in spring 2009. As part of routine monitoring efforts in
the states of MI and WI, sera were collected from muskellunge from Butternut Lake in
spring 2005 and 2007 and from Lake Hudson and Thornapple Lake in spring 2008 and
2010.

No fish were euthanized for the purposes of this study. Blood samples were collected
non-lethally, or from fish euthanized for other purposes (health inspections or VHSV-IVb
surveillance for state regulatory agencies). All samples from fish tested in this study
were negative for VHSV and other viruses using virus isolation in cell culture as
described in Millard and Faisal (2012b), Faisal et al. (2012) and WI DNR (unpublished).

111

Samples tested by virus isolation included internal tissues from all Butternut Lake fish
(kidney, spleen) and most Lake St. Clair fish (kidney, spleen, heart), and non-lethal
samples (reproductive fluids and/or sera) from all fish from Lakes St. Clair and Hudson,
Lower Fox River and Thornapple Lake.

Comparison of cELISA and PNT results. Results obtained by cELISA and PNT were
compared for several groups of muskellunge to assess the agreement between the two
methods. The first group consisted of fish from the first experimental immersion trial (n =
27, sampled between 1 and 15 weeks PI). The second group consisted of the wild
muskellunge from Lake St. Clair (n = 38) and Lower Fox River (n = 23). Muskellunge
from Lower Fox River (all years) and muskellunge from Lake St. Clair (2011) were
tested by PNT for this study. PNT results of remaining sera as well as details of the
assay are as reported in Millard and Faisal (2012a, 2012b). Briefly, two-fold dilutions of
heat-treated test sera and positive and negative controls were combined with an equal
volume of pre-titrated VHSV-IVb. Mixtures were incubated for 30 min at 18°C. Sera from
naïve lake trout (Salvelinus namaycush) was then added as a source of complement
and the incubation was repeated. Samples were plaque assayed on EPC monolayers
and fixed and stained with 0.5% crystal violet in 50% formalin after 6 days. The
neutralization titer of each test sera was defined as being the reciprocal of the highest
serum dilution that prevents the formation of 50% of the viral plaques in relation to the
negative control sera. Cohenʼs kappa was used to assess agreement of the assays for
classifying muskellunge as positive or negative for antibodies based on the respective

112

thresholds (PNT ≥160, cELISA ≥14.6%). Spearman rank correlation was used to assess
correlation between PNT and cELISA titers.

RESULTS

Establishment of the cELISA. The individual rabbit hyperimmune serum that was
selected for use in the assay neutralized VHSV-IVb in vitro and had a strong reactivity to
the virus in preliminary indirect ELISA trials. Pre-immune serum from this rabbit was
non-neutralizing and did not bind to purified virus in the ELISA. In preliminary trials, we
determined that blocking and diluent reagents containing nonfat milk reduced
background levels better than did reagents with bovine serum albumin or goat serum.
Incubation periods were carried out at 25°C since temperature could be maintained
consistently in an incubator compared to room temperature. Sufficient washing after
each step with PBS-T20, as opposed to PBS alone, was critical for inhibition detection
and assay reproducibility.

In a series of competitive ELISAs, we determined that inhibition was best detected at
any given dilution of fish control sera when a) the virus was coated to plates under nonsaturating conditions, b) the maximum absorbance of the rabbit serum control wells was
approximately 1.0 ± 0.2 OD, and c) the conjugate was used at higher compared to lower
concentrations. At the minimum conjugate dilution tested (1:1000), negligible
background was present in diluent control wells and the reagent was still economical, so

113

this dilution was used for subsequent trials. The final concentrations of purified virus and
rabbit hyperimmune serum for use in the cELISA were determined by titrations in the
presence of 1:10 dilutions of negative and positive fish sera controls. The optimal
-1

concentration of virus was determined to be 1.0 µg·mL

-1

VHSV-IVb (0.1 µg·well ) and

the optimal dilution of rabbit hyperimmune serum was 1:512,000. Using these
conditions, rabbit serum control wells resulted in an OD of approximately 1.0 and the
difference between percent inhibitions of positive and negative fish sera controls was
maximized. Positive fish sera controls effectively inhibited rabbit serum from binding and
resulted in inhibition of color development by 86% and 84% compared to rabbit control
wells. The negative control serum resulted in only 3-6% inhibition. Using these
concentrations, results indicated that the cELISA was capable of making a clear
distinction between muskellunge sera with and without antibodies to VHSV-IVb. These
reagent concentrations were used for all subsequent assays.

Several treatments of the fish control sera were tested during preliminary trials. Heattreatment of the serum resulted in lower non-specific binding of negative fish sera.
Incubating overnight in diluent containing 1%NFDM also reduced non-specific binding.
Using the final cELISA protocol and reagent concentrations, a small trial was done to
determine whether fish sera should be screened at a dilution higher than 1:10. A panel
of fish sera with and without VHSV-IVb antibodies was tested at dilutions of 1:10, 1:20,
1:40 and 1:80 (data not shown). At a serum dilution of 1:10, optimal positive/negative
ratios for fish sera were achieved, so all test sera for this study were screened at the

114

single, fixed dilution of 1:10. This method of testing was selected over an end-point
dilution technique so that large numbers of sera could be screened at once.

Establishment of a positive-negative threshold for muskellunge. The distribution of
inhibition values of sera from naïve fish used to establish the positive-negative threshold
is shown in Figure 5.2. Of the 60 naïve muskellunge, sera from 50 fish (83.3%)
exhibited less than 5% inhibition of the rabbit serum, and 58 fish (96.7%) showed less
than 10% inhibition. The observed range of the inhibition values in naïve muskellunge
was 0% to 18.8% (mean = 3.0%; SD = 3.5%). A positive-negative threshold of 14.6%
inhibition was obtained using the 95

th

quantile. For the purpose of this study, inhibition

values ≥14.6% were considered positive for VHSV-IVb antibodies. Using the traditional
“mean + 3 SD” approach would have resulted in a very similar threshold (13.5%).

Use of sera from fish surviving experimental VHSV-IVb infection to test assay.
Inhibition values from fish that survived the single VHSV-IVb challenge are shown in
Figure 5.3. Inhibition was first detected 7 weeks PI (33.3% of fish). The highest
proportions of cELISA positive fish were detected 11 weeks (66.7%) and 15 weeks
(100%) PI. Of the 18 fish sampled after the onset of antibody production (between 7 and
15 weeks PI), eight fish (44.4%) blocked the rabbit hyperimmune serum from binding to
VHSV-IVb during the assay. Inhibition values of seropositive fish ranged from 31.6% to
69.7% (mean = 51.0%; SD = 11.9%). The range in inhibition for the 16 mock-challenged
fish in contrast was 0.0% to 3.1% (mean = 1.3%; SD = 1.4%).

115

Of the 22 muskellunge that survived two VHSV-IVb exposures, inhibition levels above
the threshold were detected in sera of 7 fish (31.8%) sampled 67 weeks PI (45 weeks
post-second exposure) and in 10 fish sampled 79 weeks PI (57 weeks post-second
exposure) (Table 5.2). Inhibition values of seropositive fish ranged from 14.7% to 46.1%
(mean = 25.2%; SD = 8.2%).

Application of the assay to field-collected sera of muskellunge. Inhibition values of
all wild fish are shown in Figure 5.4 by location and year and data are summarized in
Table 5.1. Seroprevalence is given as apparent (versus true) seroprevalence by
definition since sensitivity and specificity of this assay has not been determined. Of the
38 muskellunge collected from Lake St. Clair, 33 fish (86.8%) had antibodies to VHSVIVb based on the threshold of 14.6 % inhibition. Seroprevalence was 83.3% in 2010 and
90.0% in 2011. Inhibition values for all Lake St. Clair muskellunge ranged from 0.0% to
86.8% (mean = 55.6%; SD = 26.9%). Inhibition values of seropositive fish from Lake St.
Clair ranged from 22.2% to 86.8 % (mean = 63.1%; SD = 19.7%). Mean positive
antibody titer was slightly greater in May 2010 (mean = 68.2; SD = 18.3) compared to
May 2011 (mean = 58.9; SD = 20.3).

Of the 24 muskellunge from Lower Fox River tested by cELISA, nine fish (37.5%) had
antibodies to VHSV-IVb. Antibodies were detected in 60.0% of fish in April 2010 and
85.7% of fish in May 2012. No antibodies were detected in fish sampled in May 2011.

116

Inhibition values of all seropositive fish ranged from 17.1% to 53.7 % (mean = 32.8%;
SD = 12.0%). The mean antibody titer was higher in fish sampled in April 2010 (mean =
41.5; SD = 13.2) compared to May 2012 (mean = 28.5; SD = 9.7).

All 78 muskellunge collected from Butternut Lake (n = 18) and Lake Hudson (n = 60)
were negative for VHSV-IVb antibodies. For Butternut Lake fish, inhibition values
ranged from 0.1% to 8.1% (mean = 2.5%; SD = 2.0%). For Lake Hudson fish, inhibition
values ranged from 0.0% to 11.3% (mean = 2.9%; SD = 2.9%).

Thornapple Lake, like Butternut Lake and Lake Hudson, was considered to be VHSVfree based on past viral testing. Muskellunge in Thornapple Lake, however, showed
evidence of an immune response against VHSV-IVb. Of the 60 fish collected in 2008
and 2010, 25 fish (41.7%) had inhibition levels greater than the threshold.
Seroprevalence and mean positive antibody titers were greater for fish sampled in April
2008 compared to 2010. In 2008, 16 of 30 muskellunge (53.3%) were seropositive.
Inhibition values for cELISA-positive fish in 2008 ranged from 18.3% to 80.5% (mean =
51.8%; SD = 16.7%). In 2010, 9 of 30 muskellunge (30.0%) had antibodies to VHSVIVb. Inhibition values of seropositive fish in 2010 ranged from 19.2% to 60.8 % (mean =
39.2%; SD = 13.6%). Inhibition values for all Thornapple muskellunge ranged from 0.0%
to 80.5% (mean = 21.2%; SD = 24.7).

117

Comparison of cELISA with PNT results. Results by cELISA and PNT agreed in
classifying fish as positive or negative for antibodies to VHSV-IVb for the majority of
sera tested (Figure 5.5). Overall, results were in agreement for 69 of 88 fish tested by
both assays (78.4%, Cohenʼs kappa coefficient = 0.55). For the experimentally exposed
fish alone, 29.6% were positive by cELISA and 33.3% were positive by PNT. Agreement
between the tests was 88.9% (24 of 27 fish) (Cohenʼs kappa coefficient = 0.74). Of the
observations that were in agreement, 7 of 24 fish (29.2%) were positive by both assays
and 17 of 24 fish (70.8%) were negative by both assays. For the wild muskellunge from
Lake St. Clair and Lower Fox River, 41 fish (67.2%) were positive by cELISA and 53 fish
(86.9%) were positive by PNT. Agreement between the tests for wild muskellunge was
73.8% (45 of 61)(Cohenʼs kappa coefficient = 0.3). For the samples that agreed, 39 of
45 fish (86.7%) were positive by both assays and 6 of 45 fish (13.3%) were negative by
both assays. There were several sera from experimentally challenged and wild
muskellunge that were positive by only one assay. Of results that disagreed, 16 fish
were PNT positive/cELISA negative and 3 fish were PNT negative/cELISA positive.
Over half (56.3%) of the sera that were PNT positive/cELISA negative were from
muskellunge sampled from the Lower Fox River in May 2011. There was some
correlation between neutralizing titers and cELISA percent inhibition values. The
Spearman correlation coefficients were 0.80 (P < 0.0001) and 0.6 (P < 0.0001) for
experimentally exposed and wild fish, respectively.

DISCUSSION

118

In the current study, we developed a cELISA for the detection of antibodies in sera of
fish previously exposed to VHSV-IVb under experimental and natural (field) conditions.
Competition and inhibition ELISAs are used to quantify a substance (in our case, serum
antibodies to VHSV-IVb) by measuring its ability to interfere with an established pretitrated direct or indirect ELISA system (Crowther 2000). In our assay, microplates
coated with purified VHSV-IVb were used as the solid phase. Fish test sera and rabbit
serum (as a source of polyclonal antibody to VHSV-IVb) were added sequentially and
bound rabbit immunoglobulin was then detected with a commercially available enzymelabeled conjugate. When optimizing reagent concentrations, we found that inhibition of
the rabbit antiserum was most apparent when the purified virus was used at a non-1

saturating concentration (1.0 µg·mL ) and when the maximum absorbance in the
absence of inhibition was approximately 1.0 OD. In wells that previously received fish
serum containing antibodies to VHSV-IVb, the polyclonal antibody was effectively
outcompeted (or blocked) from binding, thus inhibiting enzymatic color change in those
wells. This variation of the competitive assay is sometimes referred to as a blocking
ELISA (Crowther 2000, Jordan 2005). Expressed on a scale of 0 to 100 percent
inhibition of the competing antibody, results are proportional to antibody activity in the
test sample (Wright et al. 1993). The presence of immunoglobulins in several sera that
produced inhibition was confirmed in preliminary trials using an indirect ELISA system
with a monoclonal antibody produced against muskellunge immunoglobulin M (Kaattari
et al., unpublished). The main reason for developing a competitive ELISA was however
to provide a test that could be used on multiple Great Lakes fish species, even on the

119

same plate, without the need for developing species-specific immunoglobulin reagents
as would be required for indirect ELISAs. Application of the assay for testing antibodies
in other species was beyond the scope of this initial study.

Fish sera were heat-treated (Dixon et al. 1994, Kibenge et al. 2002) and pre-incubated
with nonfat milk, as suggested by Kim et al. (2007), and purified virus was used to coat
plates to minimize the possibility for false positive reactions. These measures were
undertaken since some untreated sera from normal fish can exhibit high levels of nonspecific binding (Olesen et al. 1991, Kibenge et al. 2002, Kim et al. 2007). Using the
aforementioned conditions, sera from 60 juvenile muskellunge reared in a hatchery
setting and considered to be naïve to VHSV-IVb (e.g. never exposed and never
infected) were tested by the cELISA. Most sera produced inhibition values less then 5%
indicating only a low level of non-specific binding. With these data, we established a
threshold (14.6% inhibition) above which, inhibition would be unlikely to be due to nonspecific binding. For the purposes of this study, we considered sera with inhibition
≥14.6% to be positive for antibodies to VHSV-IVb. While the threshold provided a
means for us to interpret our results, we recognize that serological thresholds are often
dynamic and can be biased based on extrinsic (e.g. diet, presence of other pathogens)
and intrinsic factors (e.g. species, age, sex) of the host, as well as by laboratory testing
methods. A large number of sera of known antibody-negative and antibody-positive
status would be needed in order to refine the threshold in the future; however, this is

120

difficult in the absence of a true gold standard reference test for determining VHSV-IVb
seropositivity.

Using sera from experimental trials, we determined that the cELISA could detect
antibodies to VHSV-IVb in some fish that had been previously exposed to the virus. The
threshold provided a clear distinction between seropositive and seronegative fish.
Antibodies were first detected in fish sampled 7 weeks post-immersion exposure and
the highest percentage of seropositive fish were detected later, between 11 and 15
weeks PI. Low levels of inhibition could still be distinguished in sera from another group
of muskellunge that was experimentally challenged approximately one year prior to
sampling. It is important to consider that not all fish exposed to the virus by immersion
necessarily became infected, and furthermore, not all infected fish necessarily
seroconvert.

When the cELISA was applied to wild populations of muskellunge, VHSV-IVb antibodies
were detected in three of five populations tested. Muskellunge from two populations
(Butternut Lake, WI and Lake Hudson, MI) were negative for antibodies based on the
threshold. Muskellunge from both locations were expected to be seronegative as VHSV
has never been detected from fish from these locations. Antibodies were detected in 60
to 90% of muskellunge from Lower Fox River sampled in 2010 and 2012 and in 80 to
90% of fish from Lake St. Clair sampled in 2010 and 2011. These two locations were
considered enzootic for VHSV-IVb. The virus has been present in Lake St. Clair since at

121

least 2003 and in Green Bay and its connected waters since 2007 (reviewed in Faisal et
al. 2012). Prior to sampling, VHSV-IVb was last confirmed from fish in these systems in
the spring and early summer 2009. Seroconversion may be due to recent exposure(s),
indicating the persistence of the virus in these populations. Alternatively, given the
longevity the muskellunge, it is possible that fish were survivors from the last known
occurrence of VHSV in these areas (2009). At the time of sampling, fish may have been
fully recovered or may have been asymptomatic carriers of the virus, perhaps in tissues
not sampled or at levels below detection limits of cell culture isolation. The high
seroprevalence detected in fish from Lake St. Clair and Lower Fox River likely indicates
that some immunity has been established to VHSV-IVb in these populations. While the
association between virus neutralizing antibodies and protective immunity against fish
rhabdoviruses is well established (reviewed in Lorenzen & LaPatra 1999, Purcell et al.
2012), less is known about the role of binding antibodies. Binding antibodies in general
function to opsonize pathogens for removal by phagocytes.

Antibodies were also detected in wild muskellunge from Thornapple Lake.
Approximately 50% of muskellunge were seropositive in 2008 and 30% of fish were
seropositive in 2010. Thornapple Lake is an inland lake in Michigan that was not
considered to be positive for VHSV-IVb based on previous testing by virus isolation of
reproductive fluids and/or sera from 60 muskellunge from this population collected
yearly since 2006. The finding that fish in Thornapple Lake had prior exposure to VHSVIVb was not surprising given this lakeʼs high angler use and proximity to VHSV-IVb

122

positive regions of the Lake Michigan watershed. Although the presence of crossreactive antibodies could be an alternative explanation for this finding, this is unlikely
given that no other fish rhabdoviruses are known to occur in Michigan. Our results
suggest that the use of serological tests for detecting VHSV-IVb antibodies in fish sera
can increase the likelihood of detecting, indirectly, the presence of the virus in a
population (or water system) in the absence of clinical disease outbreaks. Given the
longevity of antibodies compared to the time in which the virus can be isolated, the
likelihood of detecting VHSV exposure based on antibodies could be expected to be
higher than the likelihood of detecting the virus itself on most sampling occasions, the
exception being when fish are sampled during outbreaks of clinical disease. This was
shown recently for cultured rainbow trout naturally infected with VHSV-I in Denmark
(Schyth et al. 2012). This has especially important implications when clinically healthy
fish are being tested, as is done for surveillance purposes. In fish that have high levels
of circulating antibodies, virus may be cleared even more rapidly from tissues following
infection, further decreasing the time-frame in which VHSV could be isolated.

In the absence of providing preliminary estimates of diagnostic sensitivity and
specificity, due to the fact that we do not have a gold standard reference test for
determining true antibody status, Cohenʼs kappa was used to compare agreement with
the PNT. An overall kappa value of 0.55 indicated there was a moderate level of
agreement beyond chance in the assaysʼ ability to differentiate seropositive and
seronegative muskellunge (Landis & Koch 1977). Agreement was not necessarily

123

expected due to differences in the types of antibodies each assay measures and this
has been discussed in the context of several fish rhabdoviruses (Olesen et al. 1991,
Ristow et al. 1993, Smail & Snow 2011). While the PNT measures antibodies directed
against neutralizing epitopes of viral glycoproteins, immunoassays measure binding
antibodies that could react with any exposed virus proteins (e.g. nucleoprotein, matrix
protein). Given the greater diversity of antibodies theoretically detectable by the cELISA,
it is interesting that the majority of discrepant results in this study came from sera that
were actually PNT positive/ cELISA negative and this requires further study. Binding
antibodies have been found to persist longer than neutralizing antibodies in sera of
rainbow trout following experimental infection with VHSV-I (Olesen et al. 1991, Encinas
et al. 2011). This was not evident in the current study for muskellunge following
experimental infection with VHSV-IVb for the time period tested (1-15 weeks PI) but it is
possible that an extended sampling period or sample size would have revealed less
agreement and correlation. Less agreement and correlation was detected for wild fish
and this could reflect differences in sampling time in relation to exposure time and
number of exposures. Further studies are needed to characterize the types of
antibodies detected by the cELISA and the kinetics of production and duration of
neutralizing and binding antibodies against VHSV-IVb in Great Lakes fish.

The cELISA has the potential to be a useful, non-lethal research and/or surveillance tool
to detect fish that have survived infection with VHSV-IVb. Compared to the PNT, the
cELISA assay has a faster turn-around time for results and is easier to perform and less

124

costly in terms of both technician time and materials. We suspect the further use of the
cELISA as a supplement to VHSV-IVb surveillance efforts in the Great Lakes will reveal
a wider distribution of the virus than is currently known.

ACKNOWLEDGEMENTS

We would like to thank the staff of the Fisheries Division of the Michigan and Wisconsin
Departments of Natural Resources (DNR) for their valuable help with site selection and
fish collection. We would also like to thank Bo Norby (MSU College of Veterinary
Medicine) and the anonymous reviewers of Diseases of Aquatic Organisms for critical
review of this manuscript. This study was possible through funding provided by the U.S.
Fish and Wildlife Service (USDI US 30181AG013 FWS and USFWS
F11AP00569/F11AP00105) and the Great Lakes Fishery Trust/U.S. Geological Survey
(USDI 08WRGR0006 USGS). Support for T. Brenden was provided by contributing
partners of the MSU Quantitative Fisheries Center, which includes Council of Lake
Committee Agencies, the Michigan DNR, the Great Lakes Fishery Commission, MSUʼs
College of Agriculture and Natural Resources, MSU Extension, and MSU
AgBioResearch..

125

APPENDIX

126

Figure 5.1. Map of Michigan and Wisconsin, USA showing sites where sera were
collected from muskellunge (Esox masquinongy) during health surveys of wild fish, 2005
to 2012. Fish sampled from the Anchor Bay and Detroit River sites are collectively
referred to as Lake St. Clair, MI muskellunge in this study. Muskellunge sampled from
the Lower Fox River, WI are primarily residents of Green Bay waters of Lake Michigan
except during spawning.

127

Figure 5.2. Distribution of cELISA percent inhibition values from sera of unexposed
muskellunge (n = 60) that were used to establish a positive-negative threshold of 14.6
% inhibition (dashed line).

128

Figure 5.3. Percent inhibition values by cELISA of muskellunge (n = 27) sampled
between 1 and 15 weeks after immersion challenge with VHSV-IVb. Samples with
percent inhibition values ≥14.6 (dashed line) are positive for antibodies to VHSV-IVb.
Horizontal lines represent the mean percent inhibition.

129

Figure 5.4. Distribution of percent inhibition values by cELISA of sera from 200 wild
muskellunge by sampling location and year. Samples with percent inhibition values
≥14.6 (dashed line) are positive for antibodies to VHSV-IVb.

130

Figure 5.5. Comparison between cELISA and 50% plaque neutralization test (PNT)
using sera from 38 wild muskellunge from a Lake St. Clair, Michigan (circles), 23
muskellunge from Lower Fox River, Wisconsin (diamond) and 27 experimentally
exposed muskellunge sampled 1-15 weeks post-exposure (triangles). Titers ≥14.6
percent inhibition (cELISA) and ≥160 (PNT) are considered positive for VHSV-IVb
antibodies. In order to display all titers, PNT titers were arbitrarily transformed by adding
two and resulting values plotted on a logarithmic scale.

100"

cELISA+/PNT-"

cELISA+/PNT+"

90"
80"

Percent inhibition (cELISA)"

70"
60"
50"
40"
30"
20"
10"
0"
cELISA-/PNT-"

-10"
1"

10"

cELISA-/PNT+"

100"

1000"
PNT "

10000" 100000"

131

Table 5.1. Competitive ELISA results for 200 free-ranging muskellunge shown by location and sampling event. Fish were
sampled between 2005-2012 from water bodies in Michigan (MI) and Wisconsin (WI). Sera with percent inhibition values
≥14.6 by cELISA are positive for antibodies to VHSV-IVb. Tissues and/or non-lethal samples (ovarian fluid, sera) from all
fish were negative for VHS virus by cell culture isolation. A positive VHSV-IVb status indicates that the virus had been
isolated from fish from that water system prior to the first sampling event. Confidence intervals (CI) were calculated using
R (R Core Team 2012). SD = standard deviation

0
0

Percent inhibition
Mean
Overall
positive
mean (SD)
(SD)
2.8 (2.5)
2.2 (1.6)

30
30

0
0

-

3.3 (2.9)
2.4 (2.9)

0.0 - 11.3
0.0 - 9.8

2008, Apr 11
2010, Mar 31

30
30

53.3 (35.8 - 70.1)
30.0 (16.4 - 48.3)

51.8 (16.7)
39.2 (13.6)

28.8 (27.9)
13.6 (18.6)

0.0 - 80.5
0.0 - 60.8

Pos.

2010, Apr 28
2011, May 10
2012, May 1

5
12
7

60.0 (20.4 - 90.0)
0
85.7 (41.9 - 98.0)

41.5 (13.2)
28.5 (9.7)

27.3 (21.7)
4.0 (4.3)
24.5 (13.8)

2.7 - 53.7
0.0 - 14.5
0.4 - 40.9

Pos.

2010, May 5-19 18
2011, May 2620
Jun 3

83.3 (59.1 - 94.5)

68.2 (18.3)

57.7 (29.3)

0.0 - 86.2

90.0 (67.6 - 97.5)

58.9 (20.3)

53.6 (25.1)

5.1 - 86.8

Butternut Lake, WI

VHSVIVb
status
Neg.

2005, Apr 2
2007, Apr 24

No.
of
fish
8
10

Lake Hudson, MI

Neg.

2008, Apr 11
2010, Apr 7

Thornapple Lake, MI

Neg.

Lower Fox River, WI

Lake St. Clair, MI

Location

Sampling event

% cELISA positive
(95% CI)

132

Range
0.3 - 8.1
0.1 - 5.2

Table 5.2. Percent inhibitions of muskellunge testing positive (≥14.6% inhibition) for
VHSV-IVb antibodies by cELISA on at least one time point sampled. Fish were
challenged at 0 and 22 weeks with VHSV-IVb by immersion and blood was sampled
from 22 survivors 67 and 79 weeks post-infection (45 and 57 weeks post-second
exposure). SD = standard deviation
Percent inhibition

1
2
3
4
5
6
7
8
9
10
No. seropositive (%)

67 weeks
(45 weeks)
14.7
16.3
23.4
33.2
20.6
31.5
9.7
28.4
11.2
11.4
7 (31.8 %)

79 weeks
(57 weeks)
17.9
18.2
29.2
33.4
23.6
29.8
21.5
46.1
25.1
16.3
10 (50.0 %)

No. fish total
Mean positive (SD)

22
24.0 (7.3)

20
26.1 (9.0)

Fish number (no.)

a

a

No blood could be collected from 2 fish

133

CHAPTER 6

VACCINATION WITH A PLASMID CONTAINING THE VHSV-IVB GLYCOPROTEIN
GENE PROTECTS MUSKELLUNGE (ESOX MASQUINONGY) AGAINST
CHALLENGE

134

INTRODUCTION

The objective of this study was to evaluate the immune response of fish following
vaccination with a new DNA vaccine against the Great Lakes strain of viral hemorrhagic
septicemia virus (VHSV). The Great Lakes VHSV (genotype IV, sublineage b) is a
reportable fish virus that has emerged in the Great Lakes region of North America within
the past decade (reviewed in Kim and Faisal 2011a,b). The virus has been responsible
for large-scale mortality events of wild fish. A number of subsequent laboratory trials
have confirmed the pathogenicity of this isolate in a broad range of freshwater fish
hosts, many of which are important from both an ecological and economical standpoint
to Great Lakes fisheries (Al-Hussinee et al. 2010, Kim & Faisal 2010b, Goodwin & Merry
2011). Strict biosecurity measures and a limited distribution of VHSV in North America
have restricted the virus to wild populations. Concerns regarding effects of VHSV-IVb on
wild fish populations as well as the potential introduction of VHSV-IVb into Great Lakes
hatcheries have prompted this investigation. Muskellunge have a high susceptibility to
VHSV-IVb and have an important role as a top predator in Great Lakes ecosystems
(Kim & Faisal 2010c). Vaccination could provide a level of immunity in stocked fish.

VHSV is a member of the genus Novirhabdovirus (family Rhabdoviridae) along with
several other fish pathogenic viruses such as infectious hematopoietic necrosis virus
(IHNV), Hirame rhabdovirus, and Snakehead rhabdovirus (Dietzgen et al. 2012). DNA
vaccines encoding the immunogenic glycoprotein (G) of VHSV (genotype I) and IHNV

135

are highly efficacious in conferring protection to salmonids following severe virus
challenge (Anderson et al. 1996, Lorenzen et al. 1998, Corbeil et al. 2000, Lorenzen et
al. 2000, Lorenzen et al. 2002, Lorenzen & LaPatra 2005, Kurath et al. 2007). The
surface G protein is one of six proteins encoded by the single-stranded RNA genome of
fish novirhabdoviruses (Dietzgen et al. 2012). The surface G proteins of VHSV and
IHNV initiate host cell infection and induce production of protective and neutralizing
antibodies (Engelking & Leong 1989, Lorenzen et al. 1990, Bearzotti et al. 1995,
reviewed by Coll 1995). The basic VHSV and IHNV DNA vaccine construct consists of
the G gene inserted downstream of the CMV promoter in the eukaryotic expression
vector pCDNA3.1 (Invitrogen).

DNA vaccines in salmonids induce cellular and humoral immunity (reviewed in Kurath et
al. 2007, Purcell et al. 2012). A robust innate immune response is evident within days
after vaccination that is cross-protective against other fish viruses. Specific immunity
develops within several weeks and is characterized by high efficacy, loss of crossprotection, and development of neutralizing antibodies. Protection 1-2 years after
vaccination after neutralizing antibodies have declined suggests establishment of longterm immunological memory. In the current study, the efficacy of a VHSV-IVb DNA
vaccine was tested in muskellunge. We also measured neutralizing and binding
antibodies to assess the adaptive, humoral response to vaccination. In a follow up
study, the ability of the construct to induce an innate, antiviral response was tested in
rainbow trout (Oncorhynchus mykiss) following an early challenge with IHN virus.

136

MATERIALS AND METHODS

Ethics statement. Michigan State University (MSU) Institutional Biosafety Committee
approved protocols involving recombinant nucleic acid molecules according to
guidelines required by the National Institute of Health (registration #3280). Live fish
experiments at MSU were designed and carried out with oversight by MSUʼs
Institutional Animal Care and Use Committee (AUF # 09/10-140-00 and 02/10-013-00).

Construction of eukaryotic expression plasmid pVHSivb-G. A plasmid encoding
the glycoprotein (G) gene (GenBank # GQ385941) of the VHSV-IVb isolate MI03
(Elsayed et al. 2006) was used. The design of the construct was based on DNA
vaccines against VHSV genotype I and IHNV (Anderson et al. 1996, Heppell et al. 1998,
Lorenzen et al. 1998). Plasmid construction, replication, and purification were
outsourced as a custom project to Life Technologies. Briefly, the full-length open
reading frame of the VHSV-IVb (G) gene (1524 bp) was assembled from synthetic
oligonucleotides and/or PCR products. An EcoRI restriction site sequence followed by a
kozak consensus sequence ending in the initial Met (start) codon of the G gene was
added to the 5ʼ end. An XbaI restriction site sequence was added following the
termination codon of VHSV-G. The fragment was cloned into the eukaryotic expression
vector pcDNA_3.1+ (Invitrogen) using EcoRI and XbaI restriction sites. The resulting
plasmid is hereafter referred to as pVHSivb-G following the naming scheme of other fish

137

rhabdoviral DNA vaccines. The plasmid was replicated in Escherichia coli K12 cells and
subsequently purified out for use as the vaccine. The pcDNA_3.1+ vector without the G
gene insert was replicated in E. coli cells for use as a plasmid control (or mock) vaccine.
The pDNA purity and concentration for both plasmid preparations were determined by
UV spectroscopy. DNA sequencing of the plasmid confirmed the correct sequence and
-1

orientation of the insert. The plasmid DNA was diluted to a concentration of 1 mg mL
in sterile phosphate buffered saline (PBS) and stored at -80°C until use.

Muskellunge vaccination and challenge. Juvenile muskellunge were obtained
outside the Great Lakes basin from the Rathbun National Fish Hatchery (Iowa
Department of Natural Resources, Moravia, IA) and reared to experimental size at the
University Research Containment Facility, Michigan State University, East Lansing, MI.
Fish were fed live fathead minnows (Pimephales promelas) obtained outside the basin
and tested to be free of VHSV and other reportable viruses. This same cohort of
muskellunge was used for the pilot virus challenge and vaccination trials to assess
protection and development of antibodies (Figure 6.1). Tanks received single-pass
water and water temperature was maintained at 11 ± 1°C.

To analyze the effect of vaccination on survival following lethal VHSV-IVb challenge, a
protection trial was carried out. Two treatments (vaccination with pVHSivb-G or the
empty plasmid) were assigned to 3 tanks each. Muskellunge (n = 180) were randomly
assigned to tanks so that each tank had 30 fish. Fish ranged in weight from 40 to 70 g

138

(mean = 56 g, SD = 9) but each tank had an approximately equal total biomass. Fish
were given several days to acclimate to the experimental tanks prior to vaccination. On
-1

day 0, fish were anesthetized by immersion in water containing 0.1 g L
-1

methanesulfonate (MS-222) and 0.3 g L

tricaine

sodium bicarbonate. Each fish was given a

single intramuscular injection of 10 µg of the pVHSivb-G vaccine or 10 µg of the empty
plasmid in a volume of 50 µl sterile PBS. The vaccine was administered in the right
dorsal epaxial muscle, anterior to the dorsal fin.

5

Seven weeks later (539 degree days), fish were challenged by immersion with 10 pfu
-1

mL

VHSV-IVb. The MI03 isolate from muskellunge (Elsayed et al. 2006) was used.

The virus was grown in the epithelioma papulosum cyprini (EPC) cell line from the
®

fathead minnow (ATCC CRL-2872) and titer determined by plaque assay (Batts &
Winton 1989). Fish from each tank were challenged in separate glass aquaria
containing 15 L of chilled, aerated water to which the virus had been added. The density
of fish in each challenge aquaria was approximately 112 g or 2 fish L

-1

. The immersion

challenge lasted 90 min and fish were returned back to their respective tank. Fish were
monitored for a period of 28 days during which morbidity was selected as an end-point
whenever possible. Mean cumulative percent mortality (CPM) of triplicate tanks of each
vaccine treatment were used to calculate relative percent survival (RPS) according to

139

the following formula: RPS = [1-(average CPM of pVHSivb-G tanks/average CPM of
plasmid control tanks)] × 100 (Amend 1981).

After 28 days post-challenge (PC; 11 weeks PV) surviving fish were euthanized using
-1

an overdose of MS-222 (0.25 g L ) buffered with sodium bicarbonate. Tissues were
collected from all fish for virus titration. A blood sample was also collected from all fish
and a subset, representing 85% of survivors, was tested for antibodies to VHSV-IVb.
The subset tested included all plasmid control fish (n = 17) and an approximately equal
number of pVHSivb-G survivors (n = 18; 6 fish selected at random from each triplicate
tank).

5

-1

The dose that was used for challenge of vaccinated fish (10 pfu mL ) was determined
in a pilot challenge (Figure 6.1) to account for the possibility that older and/or larger
muskellunge might have a lower VHSV-IVb susceptibility. Groups of 5 fish each were
7

-1

challenged by immersion with doses ranging from 0 to 10 pfu mL

and mortality

recorded for 28 days (Table 6.1).

Quantification of VHSV-IVb in tissues by plaque assay. The mixed sample of
kidney, spleen, heart and liver tissue collected from mortalities and survivors [28 days
post-challenge (PC)] was used for virus titration by plaque assay. Tissues were stored
at -80°C until testing. Tissues were diluted 1:10 (weight: volume) in minimum essential

140

medium (MEM) with Earleʼs salts supplemented with 10% tryptose phosphate broth
-1

(TPB), 100 IU mL

-1

penicillin, 100 µg mL

-1

streptomycin, 100 µg mL

gentamicin

-1

sulfate, and 2.5 µg mL amphotericin and adjusted to pH 7.5 with 1 M tris buffer.
Tissues were homogenized in a stomacher on high speed for 3 min. Homogenate was
transferred into a tube and centrifuged (2,500 × g, 30 min, 4°C). Tissue supernatant was
transferred to a 1.5 mL microcentrifuge tube and a 10-fold serial dilution series was
performed in tissue culture media. Tissue culture media, hereafter referred to as MEM5-T, was Earleʼs salt-based MEM with 10% TPB, 5% fetal bovine serum, 2 mM L-1

glutamine, 100 IU mL

-1

penicillin, 100 µg mL

-1

streptomycin, 100 µg mL

gentamicin

-1

sulfate, 2.5 µg mL amphotericin and 12 mM tris buffer. Tissue supernatant and three
dilutions of each sample were plaque-assayed on confluent EPC cells grown in flat
bottom 24-well plates. Briefly, media was removed from cell monolayers and cells were
pre-treated for 10 min with 200 µL well

-1

of 7% polyethylene glycol (PEG; 20,000 MW;

Batts and Winton 1989). The PEG was a 7% solution in tissue culture media (MEM-5T). Samples were inoculated in a volume of 100 µL well
room temperature. Finally, 1 mL well

-1

-1

and incubated for 30 min at

methylcellulose overlay (0.75% in 2X

concentrated MEM-5-T) was added to restrict the spread of the virus. Plaque assays
were incubated at 15°C for 6 days and then fixed and stained with 1% crystal violet in
50% formalin. Virus titers are calculated as pfu gram
detection thresholds were 100 pfu gram

-1

-1

of tissue. The lower and upper
7

and 3.0 × 10 pfu gram
141

-1

tissue, respectively.

Detection of an adaptive humoral immune response. A serological analysis of
vaccinated, unchallenged muskellunge was conducted in parallel in order to determine if
protection observed in the challenge study might be due to induction of an adaptive,
humoral immune response (Figure 6.1, antibody response trial). Ten muskellunge per
treatment (pVHSivb-G or plasmid control) were vaccinated on the same day and same
manner as described above and fish were held in replicate tanks of 5 fish each. Blood
samples were collected non-lethally from each fish on week 0 (prior to vaccination) and
at 7 and 11 weeks post-vaccination (PV). We selected larger fish (75 g) for this group
intentionally to increase the likelihood we could collect blood non-lethally. Another small
group of vaccinated muskellunge was held in a divided tank to serve as supplementary
or replacement blood donors in the case that we were unable to obtain sufficient
quantity of blood from an individual. On average, blood was drawn from 2 of these
additional fish per treatment at each time-point. Blood samples were kept at 4°C
overnight prior to centrifugation (2500 ! g, 20 min, 4°C). Sera were stored at 80°C until
testing.

Antibody testing. Samples were tested for VHSV-IVb neutralizing G antibodies by
50% PNT on EPC cells as previously described (Millard & Faisal 2012b). Pooled serum
from naïve lake trout (Salvelinus namaycush) was used as a source of complement in
the assay. Positive and negative fish sera controls were included each day the assay
was performed. Neutralizing antibody titers are reported as the reciprocal of the highest

142

serum dilution causing at least a 50% reduction of virus plaques compared to negative
control sera. Sera were tested in 2-fold serial dilutions from 1:20 to 1:640. Neutralizing
titers ≥160 are considered to be positive for neutralizing antibodies (Millard & Faisal
2012b).

Sera were also tested for binding antibodies using a VHSV-IVb competitive enzymelinked immunosorbent assay (cELISA) as described in Millard et al. (in press). The
assay uses a polyclonal rabbit antiserum to compete with fish sera for purified VHSVIVb binding sites. Antibody results are reported as percent inhibition according to the
formula: [1- (Average OD405/490 fish serum / Average OD405/490 rabbit serum control)]
× 100. Percent inhibition ≥ 14.6 are positive for antibodies to VHSV-IVb.

Study 2: Rainbow trout vaccination and cross-protection challenge. To assess the
efficacy of the pVHSivb-G DNA vaccine to confer early antiviral immunity, a challenge
study was carried out in S. LaPatraʼs laboratory. Rainbow trout (4 g) were vaccinated
®

with 1 µg of the pVHSivb-G vaccine, empty plasmid (plasmid control), or the Apex-IHN
vaccine (Aqua Health Ltd). Another group of the same size fish were vaccinated with
only PBS (negative control). Water temperature was maintained at approximately
14.5°C. On 7 days PV (100 degree days), duplicate groups of 25 fish each from each
5

-1

treatment were challenged by immersion with a lethal dose (10 pfu mL ) of IHNV
(strain 220-90; LaPatra et al. 1994). The PBS-injected control group was not challenged

143

at this time. Mortalities were recorded for a period of 28 days at which point survivors
were euthanized. For comparison, the same challenge trial and 28 day observation
period was repeated with different fish at a later time-point (28 days, 400 degree days
PV).

Data analysis. Survival, prevalence percentages, and virus titers were analyzed as a
completely randomized design with subsampling. Even though we administered the
vaccine to individual fish, by virtue of fish being pooled into tanks by treatment, we could
not treat fish as independent units. Instead, for the purpose of analyses, tanks were
considered experimental units and fish were observational units. Tank survival rates and
prevalence percentages were arcsine square root transformed and virus titer levels
were loge + 1 transformed prior to analysis. Data were analyzed using the MIXED
procedure in SAS Version 9.2 (SAS Institute, Inc. 2010). For the cross-protection
challenges, differences in survival among the different treatments were tested using
pair-wise comparisons of least-squares means. For all tests, the type-I error rate was
set at 0.05.

RESULTS

Efficacy of pVHSivb-G in conferring protection to muskellunge. Muskellunge were
vaccinated with 10 µg of pVHSivb-G or the plasmid without VHSV-G gene insert
5

-1

(plasmid control) and challenged 7 weeks PV with a lethal dose (10 pfu mL ) of

144

VHSV-IVb by immersion. Kinetics of mortality development are shown in Figure 6.2. The
virus dose used was the lowest dose that resulted in ≥ 60% mortality in a pilot VHSVIVb challenge study conducted several months prior with muskellunge of the same age
and size (Table 6.1). There was a significant effect of vaccination on survival, F1,4 =
22.71, p = 0.0089. Plasmid control tanks experienced percent mortality of 83.3%, 80.0%
and 80.0% (mean = 81.1%, SE = 4.2%). Fish died between day 6 and 28 PC with a
steep incline in mortality between 7 and 11 days PI that accounted for 80% of the
overall mortality. The mean days until death was 10.6 days (SE = 0.5). Fish vaccinated
with pVHSivb-G experienced mortality of 60.0%, 33.3% and 40.0% (mean = 44.4%, SE
= 5.3%). The mean days until death was 12.4 days (SE = 1.0). The RPS of pVHSivb-G
vaccinated muskellunge was 45.2% compared to plasmid-vaccinated fish.

External gross pathology of fish that died from either treatment included severely pale
(white) gills, areas of diffuse and petechial hemorrhage near and inside the mouth and
fins, and diffuse and ecchymotic hemorrhage on cranium and along dorsal surface.
Gross pathology of internal organs included swollen and pale spleens, pale heart,
hyperemia of renal, gastrointestinal and liver vessels, congested and swollen kidney
tissue, and focal areas of hemorrhage of swim bladder, liver, and visceral adipose
tissue.

Quantification of VHSV-IVb in tissues of mortalities and survivors. A mixed tissue
sample from all mortalities and all surviving vaccinated fish (28 days PC) were titrated

145

for VHSV-IVb by virus plaque assay (Table 6.2). The majority of fish that died had high
virus titers regardless of vaccination treatment as expected. VHSV-IVb was detected by
virus plaque assay in tissues from 100% of plasmid control fish and 94.7% (SE = 3.6%)
of pVHSivb-G vaccinated fish. Titers of most fish from both groups (81.8%, 90 of 110
7

fish) were greater than 3.0 × 10 pfu gram

-1

which was the maximum detection limit of
5

the assay. Of the plasmid vaccinated group, titers ranged from 4.1 × 10 to >3.0 × 10
-1

7

7

-1

pfu gram . Titers of pVHSivb-G vaccinated fish ranged from 0 to >3.0 × 10 pfu gram .

Surviving fish from the pVHSivb-G vaccinated group had a significantly lower infection
prevalence than plasmid control fish, F1,4 = 21.99, p = 0.0094. Tissues of 82.4% (SE =
9.5%) of the plasmid-control vaccinated fish were virus positive. In contrast, only 4.0%
(SE = 2.8%) of pVHSivb-G vaccinated fish were positive. There was also a significant
difference in mean virus titer between vaccinated groups, F1,4 = 28.22, p = 0.0060
(Figure 6.3). The overall mean titer for pVHSivb-G vaccinated survivors was 4.2 × 10
pfu gram

-1

2

-1

(SE = 2.9 × 10 pfu gram ). Virus titers in tissues of surviving plasmid
6

control fish ranged from 0 to 3.0 × 10 pfu gram
5

2

-1

-1

5

-1

(mean = 3.3 × 10 pfu gram , SE =

1.9 × 10 pfu gram ). The mean virus titer was 100 fold higher in one of the plasmid
control tanks compared to the others (Figure 6.3).

146

Analysis of the adaptive humoral immune response to pVHSivb-G. Sera from
vaccinated, unchallenged fish and challenge survivors were tested by 50% PNT and
cELISA to determine if VHSV-IVb vaccination induced an adaptive humoral immune
response (Table 6.3). By 7 weeks PV, that corresponded to the timing of virus challenge
in the parallel study, only one of 12 pVHSivb-G vaccinated fish were seropositive
(neutralizing antibody titer of 320). By 11 weeks PV, neutralizing antibodies were
produced by 60.0% (6 of 10) of pVHSivb-vaccinated fish. Titers of seropositive fish were
320 (2 of 6 fish) and ≥640 (4 of 6 fish). Two fish (one per treatment) had neutralizing
titers prior to vaccination, and two additional sera from the plasmid control group were
neutralizing at later time points. Unfortunately there was not enough serum left to retest
these four samples.

Sera from a subset of challenge survivors collected 28 days PC (11 weeks PV) was also
tested for neutralizing antibodies (Table 6.4). Neutralizing antibodies were detected in
100.0% of pVHSivb-G vaccinated fish but only 11.8% of plasmid controls. Titers of
pVHSivb-G vaccinated fish ranged from 160 to ≥640. Ten of the 18 seropositive fish
(55.6%) had neutralizing titers ≥640. Titers of plasmid control fish ranged from 0 to 320,
and the majority of fish had titers of <20 (58.8%, 10 of 17 fish).

Binding antibodies against VHSV-G protein were not detected by cELISA in any of the
vaccinated, unchallenged fish at week 0, 7 or 11 PV based on the threshold of 14.6%
inhibition (Table 6.3). To follow up on this result, sera were collected again from these

147

fish at 15 and 20 weeks PV. All sera were still negative for antibodies by cELISA (data
not shown). Binding antibodies were produced by a few of the challenge survivors.
Antibodies were detected in two VHSivb-G vaccinated fish and one plasmid control fish.
For VHSivb-G vaccinated fish, titers ranged from 0.0 to 23.0% inhibition (mean = 4.7%,
SD = 6.2%). For the plasmid vaccinated survivors, titers ranged from 0 to 21.4%
inhibition (mean = 5.3%, SD = 5.4%).

Saprolegnia infection post-challenge. An outbreak of Saprolegnia sp. occurred
following virus challenge of muskellunge. The fungus was found on some moribund and
dead fish between 8 and 28 days PC from 5 of 6 tanks. The number of dead fish with
the observed fungus was 8 plasmid-vaccinated fish and 14 pVHSivb-G vaccinated fish.
Virus titers in internal organs of most fish with external fungus were high (between 2.9 ×
4

7

10 and >3.0 × 10 ) with the exception of two fish that were negative. The pVHSivb-G
tank with the lowest CPM (33.3%) had the highest amount of mortalities with
Saprolegnia (9 of 10 fish). For these reasons, the fungus was not likely to be the cause
of most mortality observed in this study though its involvement cannot be ruled out. Fish
were treated with formalin several times in an attempt to control the infection. The
fungus was not found on any vaccinated, unchallenged fish.

Study 2: Rainbow trout vaccination and cross-protection challenge. In a follow up
study, juvenile rainbow trout were immunized with 1 µg of the pVHSivb-G vaccine,
®

Apex-IHN vaccine, or an empty plasmid and protection was evaluated following

148

challenge with IHN virus at several time points PV. Pair-wise comparisons of leastsquares means between treatments indicated that pVHSivb-G (t = 4.88, df = 3, p =
®

0.0165) and Apex-IHN vaccination (t = 6.86, df = 3, p = 0.0063) resulted in significantly
lower mortality after the 7 day challenge compared to the plasmid control. Differences in
survival of pVHSivb-G and IHN vaccinated groups were not statistically significant (t =
1.98, df = 3, p = 0.1416). Percent mortality of replicate tanks with each group is shown
in Table 6.5. The mean CPM of plasmid control tanks was 62.0% (SE = 6.9%). The
mean CPM of the pVHSivb-G vaccinated fish was 24.0% (SE = 6.1%) and the mean
®

CPM of Apex-IHN vaccinated fish was 12% (SE = 4.6%). The RPS of pVHSivb-G and
®

Apex-IHN treatment groups was 61.3% and 80.6%, respectively.

Following the later challenge (28 days PV) vaccination protected fish only against
challenge with the homologous virus as expected. The mean CPM of pVHSivb-G
vaccinated fish was 58.0% (SE = 7.1%), which was similar to the mean CPM of the
plasmid control fish (52.9%, SE = 7.1%). As such, the RPS of pVHSivb-G fish was <0.
As expected, pair-wise comparisons of least-squares means indicated rainbow trout
®

vaccinated with Apex-IHN were significantly protected against IHN challenge
compared to plasmid-vaccinated (t = 3.47, df = 3, p = 0.0402) and pVHSivb-G
vaccinated (t = 3.76, df = 3, p = 0.0328) fish. Differences in survival between plasmid—
vaccinated and pVHSivb-G treatments were not statistically significant (t = 0.29, df = 3,
®

p = 0.7909). The mean CPM of Apex-IHN vaccinated fish was 8.0% (SE = 3.9%). The

149

®

RPS of Apex-IHN vaccinated fish was 84.9%. Only a few mortalities (< 4% CPM)
occurred in unchallenged, PBS-injected fish during either trial.

DISCUSSION

DNA vaccines against fish novirhabdoviruses for salmonids are among the most
efficacious DNA vaccines developed to date (Lorenzen & LaPatra 2005, Kurath et al.
2007). In typical laboratory trials of the IHNV and VHSV DNA vaccines, duplicate or
triplicate groups of fish are given a single IM injection of 0.1 to 1 µg of vaccine and
challenged by immersion 4-10 weeks later (reviewed in Kurath et al. 2007). A virus dose
producing ≥ 60% CPM is used for challenge as this increases the validity of the RPS
calculation (Johnson et al. 1982). In salmonids, DNA vaccines are highly efficacious
under these circumstances, providing nearly 100% protection. In this study, we
developed a similar DNA vaccine construct but containing the G gene of the emerging
VHSV-IVb Great Lakes genotype. Results of challenge trials in both muskellunge and
rainbow trout indicate successful uptake and/or transfection of host cells with the
pVHSivb-G DNA plasmid and subsequent expression of the VHSV-IVb G protein. In
muskellunge, immune responses mounted against the expressed G protein protected
fish from a severe VHSV-IVb immersion challenge 7 weeks (539 degree days) later.
Fish vaccinated with a 10 µg dose of pVHSivb-G had a significant, albeit moderate, level
of protection (45% RPS) compared to fish vaccinated with the empty plasmid. The
challenge dose used in this study resulted in 81% mortality in plasmid control fish and

150

this level of mortality is consistent with typical post-vaccination challenge trials of IHNV
and VHSV salmonids (Kurath et al. 2007). Nevertheless it is important to note that this is
a much higher dose than fish would likely encounter in the wild and as such, we expect
that muskellunge would be protected to an even greater degree under natural
circumstances.

The VHSV-IVb DNA vaccine primed the adaptive immune response of muskellunge for
virus challenge as evident from the analysis of sera from survivors sampled 4 weeks PC
(or 11 weeks PV). Neutralizing antibodies were produced by 100% of the pVHSivb-G
vaccinated survivors tested and most sera had titers of 320 to ≥640. In contrast, low
levels of neutralizing antibodies were detected in sera from only 12% of the plasmid
control survivors, and this is consistent with an unprimed response expected by 4
weeks after VHSV-IVb challenge in this species (Millard & Faisal 2012a). The
seroprevalence of pVHSivb-G vaccinated survivors was higher also compared to
pVHSivb-G vaccinated, unchallenged fish at this same time point (11 weeks PV),
indicating an anamnestic immune response to challenge and not just a primary adaptive
response to vaccination.

Sera collected from vaccinated, unchallenged fish showed that neutralizing antibodies
against the G protein were only beginning to develop by 7 weeks PV, corresponding to
the timepoint in which fish from the protection trial were challenged. The fact that only a
single fish had seroconverted based on neutralizing titers ≥ 160 may be the reason for

151

the moderate level of protection. However, numerous studies have found that detectable
neutralizing antibodies is not necessary for protection of DNA-vaccinated salmonids
following rhabdovirus challenge (reviewed in Kurath et al. 2007). Sera with low
neutralizing titers (20-40), even non-neutralizing sera, from IHNV-challenged and DNAvaccinated fish can still provide protection in vivo following passive transfer (Traxler et
al. 1999, LaPatra et al. 1994). It is possible then that neutralizing titers <160, and even
below the PNT detection limit (<20), are protective against VHSV-IVb in muskellunge as
well. We consider PNT titers 160 and above to be positive for neutralizing antibodies in
muskellunge sera and this is conservative compared to some other research groups.
However, non-specific reactions of normal sera and/or fluctuations in titers from normal
day-to-day variability can occur with this assay (Olesen & Jørgensen 1986, Millard &
Faisal 2012b) making low levels of neutralization difficult to interpret. We suspect this
was the cause for the neutralizing titers detected in a few of the pre-vaccination and
plasmid-control sera in this study.

Although the neutralizing antibody response is often used as an assessment of
protective immunity induced by DNA vaccination in salmonids, non-neutralizing
(binding) antibodies and specific cell-mediated cytotoxicity are expected to be involved
in protection as well (Lorenzen et al. 1998, Lorenzen et al. 2000, Kurath et al. 2007). In
this study, binding antibodies were not induced by vaccination and as such were
unlikely to have contributed to survival of muskellunge against challenge. Low levels of
antibodies were detectable by cELISA by 4 weeks PC in sera of a few challenge

152

survivors from both treatment groups. We can conclude that it was VHSV-IVb infection,
and not vaccination, that induced binding antibodies in these survivors. Muskellunge
surviving the pilot virus challenge also produced cELISA-binding antibodies when sera
were tested between 6 and 12 weeks PC (data not shown) and binding antibodies after
VHSV-IVb immersion challenge has been reported previously for this species as well
(Millard et al. in press). The specificity of these antibodies detectable after VHSV-IVb
infection to the G protein though is unknown since whole, purified VHSV is used as the
coating phase in cELISA. It is possible that there are differences in the native VHSV-IVb
G protein conformation, the G protein expressed by muskellunge cells, and the G
proteins on the plate after virus purification and this requires further investigation. Hightiter neutralizing sera from both survivors and vaccinated, unchallenged fish were
negative by cELISA confirming that at least some types of neutralizing G antibodies are
not detectable by VHSV-IVb cELISA, which was unclear based on results in a previous
study (Millard et al. in press).

The fact that pVHSivb-G significantly reduced viral prevalence and titers after challenge
is important to consider from a fisheries management aspect. Infected muskellunge
surviving VHS shed high titers of virus for several months and may reassume shedding
following a stressful circumstance (Kim & Faisal 2012). The role of virus shedders in
amplifying infection was clear in this study. All tanks were challenged with the same
virus load, yet 4 weeks later viral tissue loads were considerably higher in fish kept
together in one of three plasmid control tanks. Vaccinating fish prior to stocking in the

153

Great Lakes could be done in an effort to establish herd immunity among wild
populations. Furthermore, it could help reduce virus transmission by decreasing viral
load, and presumably the amount of shed virus, in fish that do become infected.
Neutralizing antibodies likely played a role in clearing infection in pVHSivb-G vaccinated
survivors. A reduced persistence of virus in tissues of VHSV DNA vaccinated fish after
challenge has been reported in rainbow trout (Lorenzen et al. 2000, 2009) and Pacific
herring (Clupea pallasii) (Hart et al. 2012).

In a follow-up study, a low dose of the pVHSivb-G vaccine protected rainbow trout
against virulent IHN virus challenge at an early time point (7 days PV) suggesting
induction of a robust innate immune response by this species. This result also suggests
that the pVHSivb-G plasmid is functional, and mediates appropriate expression of the
protein upon injection. Early phase protection after VHSV DNA vaccination in salmonids
is mediated by non-specific antiviral immune mechanisms (e.g. interferon system) that
can provide cross-protection against other fish rhabdoviruses within days PV (reviewed
in Kurath et al. 2007, Purcell et al. 2012, Lorenzen et al. 1998, LaPatra et al. 2001). This
cross-protection does not seem to occur in all species though based on a recent study
in Pacific herring (Hart et al. 2012). It will be interesting to follow up on this aspect of the
immune response in muskellunge and other species. Following the later challenge at 28
days PV, only fish vaccinated with Apex®-IHN were protected against challenge with
the homologous virus. This was expected based on the transition to specific immunity

154

that occurs within several weeks after rhabdoviral DNA vaccination in rainbow trout
(Lorenzen et al. 2002a, LaPatra et al. 2001).

The protection conferred by pVHSivb-G DNA vaccination in muskellunge was less then
that conferred to salmonids by VHSV-I DNA vaccination despite a similar construct and
study design. Muskellunge are evolutionarily different compared to salmonids and
immune responses probably differ in some ways. Furthermore, stress may have a
stronger immunosuppressive effect on species such as muskellunge that are less
domesticated to aquaculture conditions compared to rainbow trout. This idea is
supported by results from a recent study by Hart et al. (2012) with Pacific herring
vaccinated with VHSV-genotype I DNA vaccine and challenged with VHSV-IVa, to which
they are highly susceptible. Although a higher level of protection may have been
observed in herring with a homologous DNA vaccine, combined with our study, it seems
as if the typical response of salmonids to DNA vaccines does not necessarily apply to
other fish species. It is also possible that the vaccine dose used overwhelmed the
immune system leading to a suppressive effect, though other studies using similar
doses on a per weight basis still observed a high level of protection (Lorenzen et al.
2000).

Live minnows fed to muskellunge were likely the source of the external Saprolegnia
infection that occurred PC. With the exception of two fish, viral titers of fish that died with
fungus were sufficient to cause mortality, suggesting viral infection was the cause of

155

death. We cannot rule out an effect on our results, however, since compromise of the
skin barrier may have lead to re-infection from virus shed into the water. The fact that
the fungal outbreak occurred post-challenge is suggestive of reduced immune function
that could be due to destruction of hematopoietic and immune tissues by the virus.

Rhabdoviral pathogens represent a significant threat to the aquaculture industry in North
America including the Great Lakes region. This study provides an important starting
point for VHSV-IVb vaccine development and provides useful information about the
antiviral immune response to DNA vaccination in a non-domesticated fish species.
While no DNA vaccines have been approved to date in the U.S. for aquacultured
species, the Apex®-IHN DNA vaccine was approved for commercial use as a
preventative strategy against IHNV infection for cultured Atlantic salmon (Salmo salar)
in Canada in 2005 (reviewed in Salonius et al. 2007). DNA vaccines with inducible
promoters of fish origin and mechanisms capable of limiting the duration of the vaccine
in tissues have been developed (Alonso et al. 2011, Alonso et al. 2003). Should the
development of DNA vaccination against VHSV-IVb continue in the future, incorporation
of some of these safety mechanisms would seem appropriate.

ACKNOWLEDGEMENTS

We would like to thank Niels Lorenzen and Katja Einer-Jensen for critical review of the
VHSV-IVb DNA vaccine construct and sequence. We are grateful to the Great Lakes

156

Fishery Trust for funding (#2012.1257).

157

APPENDIX

158

Figure 6.1. Schematic. A pilot virus challenge was first carried out with a group of 28
unvaccinated muskellunge (Esox masquinongy) in order to determine the dose to be
used for the protection challenge (Table 6.1). On day 0, two trials were started. For the
protection trial, 90 fish (3 tanks of 30 fish each) were vaccinated with the pVHSivb-G
DNA vaccine and 90 fish (3 tanks of 30 fish each) were vaccinated with the empty
5

-1

plasmid vaccine. Fish were then challenged with 10 plaque-forming units mL VHSVIVb by immersion 7 weeks post-vaccination (PV). After 28 days, blood and tissues were
collected from surviving fish for virus titration (Table 6.2) and serology (Table 6.4). For
the antibody response trial, another group of fish was used. Blood was drawn nonlethally from 10-12 fish from each vaccine treatment group prior to vaccination, and on
week 7 and 11 PV for serological testing (Table 6.3). Fish were not challenged at any
point.

159

Figure 6.2. Development of cumulative percent mortality (CPM) in triplicate tanks of
muskellunge (Esox masquinongy) vaccinated with pVHSivb-G DNA vaccine (squares)
or the empty plasmid (triangles) and challenged with VHSV-IVb 7 weeks (539 degree
days) post-vaccination. The day of virus challenge is considered day 0. The pVHSivb-G
significantly reduced mortality compared to the control (p = 0.0089). The relative percent
a

survival (RPS) of pVHSivb-G vaccinated muskellunge was 45.2%.

Cummulative mortality (%)

100
80
60
40
20
0
0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

Days post VHSV-IVb challenge
a

RPS = [1-(average CPM of pVHSivb-G groups/average CPM of plasmid control
groups)] × 100

160

Figure 6.3. Virus titers in tissues of vaccinated muskellunge (Esox masquinongy) 28
days after virus challenge with VHSV-IVb. A-C designations represent triplicate tanks
within each treatment. Titers were transformed by adding one in order to depict titers of
-1

0 plaque forming units gram .

161

Table 6.1. A pilot study was conducted in order to determine the minimum VHSV-IVb
dose that would be lethal to ≥60% of muskellunge (Esox masquinongy) by immersion.
Five fish were infected at each dose, except for control fish (n = 3). A dose of 10
-1

plaque-forming units (pfu) mL

was chosen for challenge of the vaccinated fish.

-1

Challenge dose (pfu mL )
0
10
10
10
10
10

Percent mortality
0.0

3

20.0

4

40.0

5

100.0

6

80.0

7

100.0

162

5

Table 6.2. VHSV-IVb infection prevalence and tissue titers of mortalities and survivors. Fish were vaccinated with an
empty plasmid or the pVHSVivb-G DNA vaccine and challenged with VHSV-IVb 7 weeks (539 degree days) postvaccination. Data from survivors is from fish sampled 28 days post virus-challenge. Virus titers are in plaque-forming units
gram

-1

of tissue.

a, b

Infection prevalence
Standard error (%)
Mean virus titer

Mortalities
plasmid control
72/72 (100.0%)
0
>2.7 × 10

7

>9.2 × 10

Virus titer (range)

4.1 × 10 - >3.0 × 10

a
b

>2.1 × 10

5

Standard error

5

pVHSivb-G
36/38 (94.7%)
3.6
>2.1 × 10
7

7

3.3 × 10

6

0 - >3.0 × 10

Survivors
plasmid control
14/17 (82.4%)
9.5
1.9 × 10

7

5

2/50 (4.0%)
2.8
4.2 × 10

5

0 - 3.0 × 10

pVHSivb-G

2.9 × 10
6

c

2c
2

0 - 1.1 × 10

4

Excludes data for 3 mortalities due to improper storage.
7

7

Virus titers greater than 3.0 × 10 were treated as 3.0 × 10 for calculations.

c

pVHSivb-G vaccinated survivors had a lower infection prevalence (p = 0.0094) and mean tissue viral titer (p = 0.0060)
compared to plasmid control survivors.

163

Table 6.3. Antibody titers of muskellunge (Esox masquinongy) vaccinated with pVHSivb-G or the plasmid control. Blood
was drawn non-lethally from 10-12 fish from each treatment prior to vaccination (0d) and 7 and 11 weeks later. Fish were
not challenged at any point with VHSV-IVb. Sera were tested for neutralizing G antibodies by 50% plaque neutralization
test (PNT) and for binding G antibodies by VHSV-IVb cELISA. Parentheses indicate number of sera with the specified
titer.

Treatment

No. PNT
positive/no.
tested

pVHSivb-G
0d
7 weeks
11 weeks

1/12
1/12
6/10

plasmid control
0d
1/12
7 weeks
1/12
11 weeks
1/10
a

a

No. cELISA
positive/ no.
tested

Neutralizing antibody titers

cELISA percent
inhibition range

Mean
cELISA
percent
inhibition
(SD)

<20 (10), 40 (1), 160 (1)
<20 (9), 40 (2), 320 (1)
20 (2), 40 (1), 80 (1), 320 (2), ≥640 (4)

0/10
0/10
0/10

0.2 - 6.7
0 - 8.4
1.2 - 6.2

3.2 (2.1)
2.7 (2.7)
3.3 (1.6)

<20 (10), 40 (1), ≥640 (1)
<20 (9), 40 (1), 80 (1), 160 (1)
<20 (8), 20 (1), 320 (1)

0/10
0/10
0/10

0 - 6.4
0 - 8.4
1.2 - 6.2

3.3 (2.1)
2.0 (1.4)
1.3 (1.3)

50% PNT titers of ≥160 are positive for VHSV-IVb neutralizing antibodies

164

Table 6.4. Neutralizing antibody titers in vaccinated muskellunge (Esox masquinongy)
surviving challenge with VHSV-IVb. Fish were sampled 28 days post-challenge.

PNT Titer
<20
20
40
80
160
320
≥640
Number of fish tested
% positive ( ≥160)

pVHSivb-G
Number sera (%)
0
0
0
0
2 (11.1%)
6 (33.3%)
10 (55.6%)
18
100.0%

165

plasmid control
Number sera (%)
10 (58.8%)
3 (17.6%)
0
2 (11.8%)
1 (5.9%)
1 (5.9%)
0
17
11.8%

Table 6.5. Cumulative percent mortality (CPM) of rainbow trout (Oncorhynchus mykiss)
challenged by immersion 7 days (100 degree days) and 28 days (400 degree days)
5

-1

post-vaccination with 10 pfu mL IHN virus. Number designations 1 and 2 indicate
replicate tanks. Relative percent survival (RPS) = [1-(average CPM of vaccinated
fish/average CPM of plasmid control fish)] × 100

Mean
CPM

No. fish

No. dead

CPM

1

2

1

2

1

2

7d IHNV challenge
plasmid control
Apex®-IHN

25
25

25
25

14
4

17
2

56.0
16.0

68.0
8.0

62.0

pVHSivb-G

25

25

7

5

28.0

20.0

24.0

28d IHNV
challenge
plasmid control
Apex®-IHN
pVHSivb-G

27
25
25

24
25
25

14
0
13

13
4
16

51.9
0.0
52.0

54.2
16.0
64.0

52.9

12.0

b

8.0
58.0

a
a

RPS
0
80.6
61.3

0
84.9
<0

a

Significantly less than plasmid control: Apex®-IHN (p = 0.0063) and pVHSivb-G (p =
0.0165).
b

Significantly less than plasmid control (p = 0.0402) and pVHSivb-G (p = 0.0328).

166

CHAPTER 7
CONCLUSIONS AND FUTURE RESEARCH

167

The studies performed in fulfillment of this dissertation have improved our
understanding of the immune response of fish against a rhabdoviral pathogen, viral
hemorrhagic septicemia virus (VHSV). Specifically, studies elucidated some kinetics of
the adaptive, humoral immune response of a susceptible host, the muskellunge, to
VHSV-type IVb. The combined experimental and field studies performed herein shed
light on virus trafficking and how wild populations have adapted to the presence of
VHSV-IVb over time. Two serological assays have been applied for the first time to the
study of Great Lakes fish and VHSV-IVb. These assays, the 50% plaque neutralization
test (PNT) and competitive ELISA (cELISA), will open a new horizon for future
serological studies in the Great Lakes and elsewhere in North America, where the
VHSV genotype IV is endemic.

CONCLUSIONS

From the experimental study detailed in Chapter 2, we concluded that muskellunge
mount a pronounced neutralizing antibody (NAb) response as part of their adaptive
immune defense against VHSV-IVb, and that this response can be detected by an in
vitro method, the PNT. The immersion dose of VHSV-IVb influenced the NAb response,
as well as the onset and duration of viremia. Medium and high immersion doses of virus
3

5

-1

[4 x 10 and 10 plaque forming units (pfu) mL ] produced a state of viremia, indicating
2

-1

a systemic infection, while the lowest dose of virus (10 pfu mL ) did not. An earlier

168

onset [2 days post-infection (PI) vs. 6 days] and longer duration (11 weeks vs. 6 weeks)
of viremia was apparent in fish infected with high vs. medium doses. Muskellunge
produced NAbs (≥160) by 5-7 weeks PI, levels peaked between 9-16 weeks and were
detectable at least until 17 weeks PI. The onset of NAb production corresponded to a
rapid clearance of the virus from systemic circulation in fish infected with a medium
dose of the virus indicating that antibodies were critical in limiting systemic infection.
The lowest immersion dose resulted in a reduced proportion of fish producing antibodies
suggesting that a certain threshold of infection is required for induction of the NAb
response in this species. In fish infected with a highest dose of the virus, less NAb were
detected compared to the medium dose group and this could be indicative of a)
extensive damage of kidney and spleen that led to a reduced ability to mount an
immune response or b) NAb were not detectable because they were saturated with
virus and subsequently cleared from circulation (this is a possibility especially since high
dose fish remained viremic for a longer period of time). The findings of this study were
indispensible for the better understanding of the host immune response against this
emerging pathogen.

One objective of this study initially was to compare the relationship of isolatable virus
from tissues and serum antibody levels throughout the disease course. By combining
serological results of this study for the medium dose group, with virus tissue data from
these same fish from a parallel study (Kim 2010), it is clear that a trend exists (Figure
7.1). During acute to chronic disease (~1-6 weeks PI), virus isolation remains superior in

169

detecting exposed fish. However, from the onset of antibody production through the end
of the study period, there is clear trend in the proportion of fish being virus positive (in
tissues) going down and the proportion of antibody-positive fish going up. Furthermore,
antibodies were more consistently detected across multiple sampling periods later in the
disease course compared to virus isolation. Combined with results from wild fish
studies, we can conclude that combined virus isolation and serological testing would
increase the diagnostic ability to detect VHSV-IVb in a given population—by being able
to not only detect infected fish, but also those that are recovering from past infection,
and those with established immunity.

A DNA vaccine encoding the glycoprotein (G) of VHSV-IVb was developed and proved
to be a useful tool for studying adaptive immunity in muskellunge (Chapter 6). Exposure
to the VHS DNA vaccine (pVHSivb-G) followed by VHSV-IVb challenge increased
seroprevalence and titers of survivors compared to challenge or vaccination alone. By
only four weeks after challenge, neutralizing antibodies were produced in elevated
levels (320 to ≥640) in 100% of pVHSivb-G vaccinated survivors. This suggests that the
G protein was expressed by host cells and presented to lymphocytes during the primary
phase of the adaptive immune response. Vaccination conferred significant, albeit
moderate, protection against a lethal VHSV-IVb challenge 7 weeks (539 degree days)
post-vaccination. Considering that the dose used in challenge was far higher than what
could be environmentally realistic during disease outbreaks, fish vaccinated with the
pVHSivb-G will, most probably, be able to combat the disease under natural conditions.

170

Protection was mediated, at least in part by, neutralizing antibodies. The fact that
vaccinated, unchallenged controls did not have a robust nAB response by 7 weeks
suggests NAb levels below the PNT detection threshold (<20) and/or cell-mediated
immunity also played a role in protection. Some studies have suggested the role of nonneutralizing (binding) antibodies against the G protein as a protective mechanism
particularly in vaccinated survivors that do not have NAb. For muskellunge, nonneutralizing antibodies are not likely to play a role in protection following vaccination, as
the pVHSivb-G vaccine did not induce production of any cELISA detectable antibodies.
Finally, pVHSivb-G vaccination, and its induced immune mechanisms, drastically limited
viral replication and/or increased viral clearance as evidenced by significantly reduced
viral loads in vaccinated survivors compared to plasmid control survivors. Like VHSV
genotype I DNA vaccines, the pVHSivb-G vaccine also protected rainbow trout against
an early challenge with a different rhabdovirus, proving that the expressed VHSV-IVb G
protein also induced innate antiviral mechanisms.

As concluded in Chapters 3- 5, fish residing in Lake St. Clair, a VHSV-IVb enzootic lake,
show evidence of virus exposure based on the presence of circulating antibodies. Prior
to this investigation, one hypothesis for the absence of die-offs in this ecosystem after
initial fish kills of 2005-2006 was that some fish had established immunity to the virus.
We detected a high seroprevalence in muskellunge over multiple years after the 2006
die-off events. Given the protective nature of NAbs against rhabdoviruses, we can be

171

confident that a significant percentage of sub-adult to adult fish of this species have
established immunity and are not at immediate risk of a VHSV-IVb outbreak.

Parallel testing by virus isolation and serological assays over a multiple year period
allowed a better understanding of disease dynamics. In general, the presence of VHSVIVb in the Lake St. Clair ecosystem was more likely to be indicated based on serological
vs. virus isolation evidence. The sampling period following 2006 mortality events
revealed a number of fish positive for both antibodies and virus, indicating an early
convalescent stage of disease. In contrast, recovered populations sampled a year later
in 2007 and in 2010 (a year after a 2009 mortality event) were antibody positive and
virus isolation negative.

Heterogeneity in antibody prevalence and titers existed between resident Lake St. Clair
fish species when tested by PNT (chapter 3) and cELISA (chapter 4). Muskellunge,
smallmouth bass, freshwater drum, and northern pike were the only 4 of 15 species that
had NAb. In chapter 4, binding antibodies by cELISA were detected in these species
(except northern pike) and also channel catfish. Muskellunge had the greatest
seroprevalence and titers of both neutralizing (and binding) antibodies compared to
other species. Based on this, muskellunge are a good serological indicator species for
VHSV-IVb presence in an ecosystem. This is likely related to aspects of this species life
history (e.g. longevity, top predator, spawning congregation behaviors) and VHSV-IVb
susceptibility, all of which favor increased virus exposures. Our results support previous

172

studies that suggest that neutralization and ELISA-based binding assays are detecting
different types of antibodies. Samples with high titers of antibodies by cELISA had high,
low or negative PNT results, and vice versa.

The cELISA was also used to screen other populations of muskellunge for virus
exposure (Chapter 5). Antibodies were detected in muskellunge from Lower Fox
River/Green Bay, Wisconsin and Thornapple Lake, MI. Results suggest that the cELISA
can be used to monitor the persistence of the virus in enzootic areas and detect virus
exposure in areas previously considered negative for the virus. Experimental results
showed that the antibody response by cELISA in some muskellunge lasts over a year
after exposure. Thus, the detection of antibodies in a population indicates exposure
could have occurred at least up to one year prior to the sampling event. The cELISA is a
valuable and efficient test for screening fish sera for antibodies against VHSV-IVb.
Results from Chapter 4 suggest that the developed assay can be applied to many
different species without the need for species-specific reagents.

In summary, the detection of a robust neutralizing response in muskellunge in the wild
and after experimental infection and vaccination proves their susceptibility to VHSV-IVb
is not due to an inability to mount an antibody response to this virus. This piece of
knowledge is critical to understanding the long-term effects that VHSV-IVb will have on
populations of this important predator and game species in the Great Lakes, whose

173

sport and commercial fisheries are estimated in billions of dollars of revenue to the state
of Michigan.

FUTURE RESEARCH DIRECTIONS AND APPLICATIONS

Like with every scientific study, one ends up with more questions than answers. Future
studies could build on the novel findings of studies detailed in this dissertation,
including, but not limited to the following:

Variation in presence, type, and titers of antibodies that we observed in these studies
are likely due in part to differences in host immune response mounted against VHSVIVb. It appears there are definitely differences both between individuals of the same
species as well in between species. As such, additional studies on innate and acquired,
humoral and cell-mediated defenses against VHSV-IVb are warranted. Fish are
estimated to consist of ~27,000 species, most of which have not been studied and their
basic physiological mechanisms unknown, not to mention their host defense
mechanisms. For example, it is currently unknown whether binding antibodies are
preferentially induced over NAb by some fish species in the defense against VHSV. Do
some species lack any adaptive humoral mechanisms and rely instead on innate and/or
adaptive cell-mediated immunity? These questions are important from both basic and
applied sciences perspectives. Answers to these questions would help characterize
immune responses of different fish species, help determine what types of tests could

174

best be used to detect immune responses in a population, and what type of vaccines
(and adjuvants) may be best suited for different species.

However, it is unclear at this time whether vaccination against VHSV-IVb in hatcheries
within the Great Lakes is desired or feasible. If DNA vaccines are not likely to be
approved in the near future in the U.S. for aquacultured species, it is reasonable to
reconsider at this time whether DNA vaccine development for VHSV-IVb is a priority. If
so, the reduced protection in muskellunge following vaccination compared to what is
reported for salmonids would be interesting to investigate further. Future studies could
investigate the use of multiple vaccine and virus challenge doses and timeframes of
challenge post-vaccination. Tissues collected during the DNA vaccine trial could be
used to assess induction of innate or adaptive immune genes.

In terms of VHSV-IVb ecology and population health, several future studies are needed.
Detection of high seroprevalence of NAb in a wild population likely suggests the risk of
VHSV-IVb outbreak in that species is unlikely. However, what level of herd immunity is
needed to prevent an outbreak and how does the level of herd immunity in one fish
species influence the likelihood of virus outbreaks in other fish species? It would be
interesting to further study the roles of carrier fish in VHSV-IVb transmission and
disease dynamics. For example, can muskellunge, or other species with high levels of
antibodies, still shed the virus or does shedding only resume if antibodies decline past a
certain threshold? On that note, what is the role of salmonids and other fish species that

175

become infected but do not develop clinical disease? Perhaps these fish act as carriers
and have role in viral amplification and transmission by shedding the virus.
The unique relationship between the MSU-Aquatic Animal Health Laboratory and the
Michigan DNR allows for research findings to almost immediately become useful from a
management standpoint. The following are reasonable suggestions for how information
from this dissertation can be applied to future VHSV-IVb surveillance and regulatory
testing in Michigan. Modification and implementation of the suggestions below require
conversations and planning with Michigan regulatory agencies.

•

For VHSV-positive management (enzootic) areas: the persistence of the virus
could be monitored from now on using a non-lethal serological assay alone. This
would reduce the number of fish sacrificed for purposes of screening for virus
presence. Long-lived, susceptible species such as muskellunge (or northern pike,
smallmouth bass) would be good species to use for this based on our studies. Acute
disease outbreaks in VHS positive areas would still be submitted as diagnostic
cases and presence of the virus confirmed by traditional tests.

•

For VHSV-surveillance and VHSV- free zones: we recommend the addition of
serological testing to traditional virus isolation testing. Choice of fish for each assay
should definitely differ; for example a short-lived, VHSV-IVb susceptible species that
is quick to reproduce like bluegill, or better yet round gobies, an invasive species in
the Great Lakes, would be ideal to target for virus isolation. A long-lived, susceptible,

176

larger species that can be sampled non-lethally would be ideal for serological
analysis. Finances and staff time will be a limiting factor; it seems reasonable that
this could be implemented on a limited number of sites for several summers and
then results assessed to determine future testing priorities. It is important to
determine immediately what the finding of antibodies in the absence of virus means
from a management standpoint. For example, we are confident that the detection of
antibodies means that Thornapple Lake muskellunge population has been exposed.
However, this has not led to reclassification of this lake as a VHS-positive lake to
date.

•

Using the most sensitive techniques for understanding the distribution of VHSV-IVb
is critical from a management standpoint. It is therefore important to consider that
sensitive quantitative RT-PCR assays combined with serological testing could be
considered as an alternative to the conventional virus isolation tests.

Parallel results of VHS isolation and serology may reveal a variety of testing patterns
that would allow for a better understanding of disease dynamics on a population level.
For example, virus isolation and antibody positive status could indicate a recent VHSVIVb exposure. Virus negative status with concurrent high seroprevalence would indicate
a recovered population that has likely established immunity. Absence of virus and
antibodies may suggest a naïve population that is at immediate risk if the virus enters. It
would be interesting to study further what factors influence the relationship between

177

presence/absence of neutralizing and binding antibodies. Some studies have suggested
that neutralizing antibodies decline sooner than do binding antibodies. It has been found
for infectious hematopoietic necrosis virus, for example, that increased frequency of
exposures lead to a more diverse antibody repertoire (Ristow et al. 1993). Therefore it
would be interesting to sample a population throughout the course of a year including
before, during, and after outbreak(s) of VHSV-IVb. These will allow a better estimation
of when fish were exposed based on the presence of antibodies and a better
understanding of how immunity plays a role in protecting wild fish from virus outbreaks.

Finally, while not the sole intent of the present investigation, it appears that the cELISA
is a valuable and efficient test for screening fish sera of many species for antibodies
against VHSV-IVb. As this is the main reason for developing the cELISA, further
research should aim to examine sera from naïve/unexposed individuals of numerous
fish species in order to estimate negative/positive thresholds. This is necessary as we
expect that innate or intrinsic levels of background inhibition exist in some or all fish
species, and that levels likely vary based on species. Alternatively, one threshold could
be used; this would simplify the assay but likely result in a reduced sensitivity for some
species and a reduced specificity for other species. The likelihood of false negatives
and false positives, respectively, then would need to be taken into consideration in light
of the results. In that respect, the same is true regarding thresholds for the PNT. In
terms of further assay testing, it would be useful to measure intra- and inter-assay
variability on the PNT and cELISA and compare sensitivity and specificity for detecting

178

antibodies throughout the disease course in several species. While NAb are known to
be protective, more studies are needed to determine the types and protective nature of
the antibodies detectable by cELISA.

179

APPENDIX

180

Figure 7.1. Percent of sampled fish with virus in sera, tissues (kidney, spleen and/or
heart), and neutralizing antibodies over the course of VHSV-IVb infection and recovery.
3

-

Muskellunge were infected by immersion in a medium dose of the virus (4 × 10 pfu mL
1

). Four to five fish are represented at each time-point, except early in the disease
course (0 week = 3 fish, 0.5 week = 20 fish, 1 week = 8 fish). Mortality rate of fish dying
with clinical signs consistent with VHSV-IVb infection peaked between day 6 and 16
post-infection. This graph was added to emphasize that the combination of both
conventional virus isolation assays along with antibody testing allowed more consistent
detection of exposed fish throughout the disease course.
Virus positive (sera)"

Virus positive (tissues)"

80%"
70%"

Postitive"

60%"
50%"
40%"
30%"
20%"
10%"
0%"
Weeks post-infection "

181

Antibody positive"

REFERENCES

182

REFERENCES

Acosta F, Collet B, Lorenzen N, Ellis AE (2006) Expression of the glycoprotein of viral
haemorrhagic septicaemia virus (VHSV) on the surface of the fish cell line RTGP1 induces type 1 interferon expression in neighbouring cells. Fish Shellfish
Immunol 21:272-278
Ahne W (1986) The influence of environmental temperature and infection route on the
immune response of carp (Cyprinus carpio) to spring viremia of carp virus
(SVCV). Vet Immunol Immunop 12:383-386
Ahne W, Bjorklund HV, Essbauer S, Fijan N, Kurath G, Winton JR (2002) Spring viremia
of carp (SVC). Dis Aquat Organ 52:261-272
Alejo A, Tafalla C (2011) Chemokines in teleost fish species. Dev Comp Immunol
35:1215-1222
Al-Hussinee L, Huber P, Russell S, LePage V and others (2010) Viral haemorrhagic
septicaemia virus IVb experimental infection of rainbow trout, Oncorhynchus
mykiss (Walbaum), and fathead minnow, Pimphales promelas (Rafinesque). J
Fish Dis 33:347-360
Al-Hussinee L, Lord S, Stevenson R, Casey R and others (2011) Immunohistochemistry
and pathology of multiple Great Lakes fish from mortality events associated with
viral hemorrhagic septicemia virus type IVb. Dis Aquat Organ 93:117-127
Alonso M, Chiou PP, Leong JA (2011) Development of a suicidal DNA vaccine for
infectious hematopoietic necrosis virus (IHNV). Fish Shellfish Immunol 30:815823
Alonso M, Johnson M, Simon B, Leong JA (2003) A fish specific expression vector
containing the interferon regulatory factor 1A (IRF1A) promoter for genetic
immunization of fish. Vaccine 21:1591-1600
Alonso M, Leong, JC (2013) Licensed DNA vaccines against infectious hematopoietic
necrosis virus. Recent patents on DNA & gene sequences 7:62-65
Alvarado JF, Dolz G, Herrero MV, McCluskey B, Salman M (2002) Comparison of the
serum neutralization test and a competitive enzyme-linked immunosorbent assay
for the detection of antibodies to vesicular stomatitis virus New Jersey and
vesicular stomatitis virus Indiana. J Vet Diagn Invest 14:240-242.

183

Amend DF, Smith L (1974). Pathophysiology of infectious hematopoietic necrosis virus
disease in rainbow trout (Salmo gairdneri): Early changes in blood and aspects of
the immune response after injection of IHN virus. J Fish Res Board Can
21:1371–1378
Amend DF (1981) Potency testing of fish vaccines. Dev Biol Stand 49:447-454
Anderson ED, Mourich DV, Fahrenkrug SC, LaPatra S, Shepherd J, Leong JA (1996)
Genetic immunization of rainbow trout (Oncorhynchus mykiss) against infectious
hematopoietic necrosis virus. Mol Mar Biol Biotechnol 5:114-122
Bain MB, Cornwell ER, Hope KM, Eckerlin GE and others (2010) Distribution of an
invasive aquatic pathogen (viral hemorrhagic septicemia virus) in the Great
Lakes and its relationship to shipping. PLOS ONE 5:e10156
Banchereau J, Briere F, Caux C, Davoust J and others (2000) Immunobiology of
dendritic cells. Annu Rev of Immunol 18:767-811
Bassity E, Clark TG (2012) Functional identification of dendritic cells in the teleost
model, rainbow trout (Oncorhynchus mykiss). Plos One 7:e33196
Battige K (2011) Great Lakes spotted muskellunge restoration:Evaluating natural
recruitment and modeling spawning habitat in Green Bay, Lake Michigan.
University of Michigan
Batts WN, Winton JR (1989) Enhanced detection of infectious hematopoietic necrosis
virus and other fish viruses by pretreatment of cell monolayers with polyethylene
glycol. J Aquat Anim Health 1:284-290
Batts WN, Winton JR (2010). Viral hemorrhagic septicemia. In AFS–FHS (American
Fisheries Society– Fish Health Section). 2010. FHS blue book: Suggested
procedures for the detection and identification of certain finfish and shellfish
pathogens, 2010 edition. AFS–FHS, Bethesda, Maryland, 11 pp.
Bearzotti M, Delmas B, Lamoureux A, Loustau AM, Chilmonczyk S, Bremont M (1999)
Fish rhabdovirus cell entry is mediated by fibronectin. J Virol 73:7703-7709
Bearzotti M, Monnier AF, Vende P, Grosclaude J, de Kinkelin P, Benmansour A (1995)
The glycoprotein of viral hemorrhagic septicemia virus (VHSV): Antigenicity and
role in virulence. Vet Res 26:413-422
Bernard J, de Kinkelin P, Bearzotti-Le Berre M (1983) Viral hemorrhagic septicemia of
rainbow trout:Relation between the G polypeptide and antibody production in

184

protection of the fish after infection with the F25 attenuated variant. Infect Immun
39:7-14
Bly JE, Clem W (1992) Temperature and teleost immune functions. Fish Shellfish
Immunol 2:159-71
Bly JE, Quiniou SM, Clem LW (1997) Environmental effects on fish immune
mechanisms. Dev Biol Stand 90:33-43
Bordon Y (2013) Evolution: A gutsy defence of the skin. Nat Rev Immunol 13:616-617
Boshra H, Li J, Sunyer JO (2006) Recent advances on the complement system of
teleost fish. Fish Shellfish Immunol 20:239-262
Boudinot P, Blanco M, de Kinkelin P, Benmansour A (1998) Combined DNA
immunization with the glycoprotein gene of viral hemorrhagic septicemia virus
and infectious hematopoietic necrosis virus induces double-specific protective
immunity and nonspecific response in rainbow trout. Virology 249:297-306
Bozek MA, Burri TM, Frie RV (1999) Diets of muskellunge in northern Wisconsin lakes.
North American Journal of Fisheries Management 19:258-270
Brunson R, True K, Yancey J (1989) VHS virus isolated at Makah National Fish
Hatchery. Am Fish Soc Fish Health Sec Newslett 17:2837-2846
Byon JY, Ohira T, Hirono I, Aoki T (2006) Comparative immune responses in japanese
flounder, paralichthys olivaceus after vaccination with viral hemorrhagic
septicemia virus (VHSV) recombinant glycoprotein and DNA vaccine using a
microarray analysis. Vaccine 24:921-930
Cai Z, Cai L, Jiang J, Chang K-S, van der Westhuyzen DR, Luo G (2007) Human serum
amyloid a protein inhibits hepatitis c virus entry into cells. J Virol 81:6128-6133
Caipang CM, Hirono I, Aoki T (2003) In vitro inhibition of fish rhabdoviruses by
Japanese flounder, Paralichthys olivaceus Mx. Virology 317:373-382
Castro R, Jouneau L, Pham H-P, Bouchez O and others (2013) Teleost fish mount
complex clonal IgM and IgT responses in spleen upon systemic viral infection.
PLoS Pathog 9(1): e1003098. doi:10.1371/journal.ppat.1003098
Crowther JR (2000) The ELISA guidebook. Humana Press, Totowa, NJ
Clark RD, Jr., Hanchin PA, Lockwood RN (2004) The fish community and fishery of
Houghton Lake, Roscommon County, Michigan with emphasis on walleyes and

185

northern pike. Michigan Department of Natural Resources, Fisheries Special
Report 30, Ann Arbor, Michigan, 60 pp.
Coll JM (1995) The glycoprotein-g of rhabdoviruses. Arch Virol 140:827-851
Corbeil S, Kurath G, Lapatra SE (2000) Fish DNA vaccine against infectious
hematopoietic necrosis virus:Efficacy of various routes of immunisation. Fish
Shellfish Immunol 10:711-723
Crossman EJ (1986) The noble muskellunge: a review. In: Hall (ed) Managing muskies.
American Fisheries Society, special publication 15. Bethesda, Maryland Pages
1–13
Cuesta A, Tafalla C (2009) Transcription of immune genes upon challenge with viral
hemorrhagic septicemia virus (VHSV) in DNA vaccinated rainbow trout
(Oncorhynchus mykiss). Vaccine 27:280-289
Danilova N, Bussmann J, Jekosch K, Steiner LA (2005) The immunoglobulin heavychain locus in zebrafish: Identification and expression of a previously unknown
isotype, immunoglobulin Z. Nat Immunol 6:295-302
de Kinkelin P, Gerard JP, Dorson M, Le Berre M (1977) Viral hemorrhagic
septicemia:Demonstration of a protective immune response following natural
infection. Fish Health News 6:3-4
Dixon PF, Hattenberger-Baudouy AM, Way K (1994) Detection of carp antibodies to
spring viraemia of carp virus by competitive immunoassay. Dis Aquat Organ
19:181-186
Dietzgen RG, Calisher CH, Kurath G, Kuzmin IV and others (2012) Family rhabdoviridae. In: King AMQ, Adams MJ, Carstens EB, Lefkowitz E (eds) Virus
taxonomy. Elsevier, San Diego, p 686-713
Dorson M, de Kinkelin P (1974) Infectious pancreatic necrosis in Salmonidae:Existence
in the serum of undamaged trout of a 6 S molecule specifically neutralizing the
virus. C R Acad Sci Hebd Seances Acad Sci D 278:785-788
Dorson M, Torchy C (1979) Complement dependent neutralization of Egtved virus by
trout antibodies. J Fish Dis 2:345-347
Edholm E-S, Bengtén E, Stafford JL, Sahoo M, Taylor EB, Miller NW, Wilson M (2010)
Identification of two IgD+ B cell populations in channel catfish, Ictalurus
punctatus. J Immunol 185:4082-4094

186

Einer-Jensen K, Ahrens P, Forsberg R, Lorenzen N (2004) Evolution of the fish
rhabdovirus viral haemorrhagic septicaemia virus. J Gen Virol 85:1167-1179
Einer-Jensen K, Ahrens P, Lorenzen N (2005) Parallel phylogenetic analyses using the
N, G or NV gene from a fixed group of VHSV isolates reveal the same overall
genetic typing. Dis Aquat Organ 67:39-45
Ellis AE (1988) Optimizing factors for fish vaccination. In: Ellis AE (ed) Fish vaccination.
Academic Press, London, p 32–46
Ellis AE (2001) Innate host defense mechanisms of fish against viruses and bacteria.
Dev Comp Immunol 25:827-839
Elsayed E, Faisal M, Thomas M, Whelan G, Batts W, Winton J (2006) Isolation of viral
haemorrhagic septicaemia virus from muskellunge, Esox masquinongy (Mitchill),
in Lake St. Clair, Michigan, USA reveals a new sublineage of the North American
genotype. J Fish Dis 29:611-619
Emmenegger EJ, Kurath G (2008) DNA vaccine protects ornamental koi (Cyprinus
carpio koi) against North American spring viremia of carp virus. Vaccine 26:64156421
Encinas P, Gomez-Casado E, Fregeneda G, Olesen NJ, Lorenzen N, Estepa A, Coll JM
(2011) Rainbow trout surviving infections of viral haemorrhagic septicemia virus
(VHSV) show lasting antibodies to recombinant G protein fragments. Fish
Shellfish Immunol 30:929-935
Engelking HM, Leong JC (1989) The glycoprotein of infectious hematopoietic necrosis
virus elicits neutralizing antibody and protective responses. Virus Research
13:213-230
Enzmann PJ, Konrad M (1990) Antibodies against vhs in whitefish of the Lake of
Constance, West Germany. Bull Eur Assoc Fish Pathol 10:24
Enzmann PJ, Konrad M (1993) Longevity of antibodies in brown trout and rainbow trout
following experimental infection with VHS virus. Bull Eur Assoc Fish Pathol
13:193–194
Enzmann PJ, Konrad M, Parey K (1993) VHS in wild living fish and experimental
transmission of the virus. Fisheries Research 17:153-161
Esteban MÅ (2012) An overview of the immunological defenses in fish skin. ISRN
Immunology 2012:29

187

Esteban MÅ, Meseguer J, Tafalla C, Cuesta A (2008) Nk-like and oxidative burst
activities are the main early cellular innate immune responses activated after
virus inoculation in reservoir fish. Fish Shellfish Immunol 25:433-438
Faisal M, Schulz CA (2009) Detection of viral hemorrhagic septicemia virus (VHSV)
from the leech Myzobdella lugubris Leidy, 1851. Parasites & Vectors 2:45
Faisal M, Shavalier M, Kim RK, Millard EV, and others (2012) Spread of the emerging
viral hemorrhagic septicemia virus strain, genotype IVb, in Michigan, USA.
Viruses 4:734-760
Faisal M, Winters AD (2011) Detection of viral hemorrhagic septicemia virus (VHSV)
from Diporeia spp. (Pontoporeiidae, Amphipoda) in the Laurentian Great Lakes,
USA. Parasites & Vectors 4:2
Fange R (1986) Lymphoid organs in sturgeons (Acipenseridae). Vet Immunol
Immunopathol, 12 153–161
Fast MD, Sims DE, Burka JF, Mustafa A, Ross NW (2002) Skin morphology and
humoral non-specific defence parameters of mucus and plasma in rainbow trout,
coho and Atlantic salmon. Comparative Biochemistry and Physiology Part
A:Molecular & Integrative Physiology 132:645-657
Fernandez-Alonso M, Lorenzo G, Perez L, Bullido R, Estepa A, Lorenzen N, Coll JM
(1998). Mapping of linear antibody epitopes of the glycoprotein of VHSV, a
salmonid rhabdovirus. Dis Aquat Organ 34:167-176
Fijan N, Sulimanović D, Bearzotti M, Muzinic D, Zwillenberg LO, Chilmonczyk S,
Vautherot JF, de Kinkelin P (1983) Some properties of the Epithelioma
papulosum cyprini (EPC) cell line from carp Cyprinus carpio. Annales de lʼInstitut
Pasteur/Virologie 134E:207-220
Fischer U, Koppang EO, Nakanishi T (2013) Teleost T and NK cell immunity. Fish
Shellfish Immunol 35:197-206
Fischer U, Utke K, Somamoto T, Köllner B, Ototake M, Nakanishi T (2006) Cytotoxic
activities of fish leucocytes. Fish Shellfish Immunol 20:209-226
Fregeneda-Grandes JM, Olesen NJ (2007) Detection of rainbow trout antibodies against
viral haemorrhagic septicaemia virus (VHSV) by neutralisation test is highly
dependent on the virus isolate used. Dis Aquat Organ 74:151-158
Gagne N, Mackinnon AM, Boston L, Souter B, Cook-Versloot M, Griffiths S, Olivier G
(2007) Isolation of viral haemorrhagic septicaemia virus from mummichog,

188

stickleback, striped bass and brown trout in eastern Canada. J Fish Dis 30:213223
Garver K, Conway C, Elliott D, Kurath G (2005) Analysis of DNA-vaccinated fish reveals
viral antigen in muscle, kidney and thymus, and transient histopathologic
changes. Mar Biotechnol 7:540-553
Gaudin Y, Ruigrok RW, Tuffereau C, Knossow M, Flamand A (1992) Rabies virus
glycoprotein is a trimer. Virology 187:627-632
Gomez GD, Balcazar JL (2008) A review on the interactions between gut microbiota
and innate immunity of fish. FEMS Immunol Med Microbiol 52:145-154
Goodwin AE (2002) First report of spring viremia of carp virus (svcv) in North America. J
Aquat Anim Health 14:161-164
Goodwin AE, Merry GE (2011) Mortality and carrier status of bluegills exposed to viral
hemorrhagic septicemia virus genotype IVb at different temperatures. J Aquat
Anim Health 23:85–91
Granzow H, Weiland F, Fichtner D, Enzmann PJ (1997) Studies of the ultrastructure and
morphogenesis of fish pathogenic viruses grown in cell culture. J Fish Dis 20:110
Groocock GH, Getchell RG, Wooster GA, Britt KL and others (2007) Detection of viral
hemorrhagic septicemia in round gobies in New York State (USA) waters of Lake
Ontario and the St. Lawrence River. Dis Aquat Organ 76:187-192
Hansen JD, Landis ED, Phillips RB (2005) Discovery of a unique ig heavy-chain isotype
(IgT) in rainbow trout:Implications for a distinctive B cell developmental pathway
in teleost fish. Proc Natl Acad Sci U S A 102:6919-6924
Hansen JD, Woodson JC, Hershberger PK, Grady C, Gregg JL, Purcell MK (2012)
Induction of anti-viral genes during acute infection with viral hemorrhagic
septicemia virus (VHSV) genogroup IVa in Pacific herring (Clupea pallasii). Fish
Shellfish Immunol 32:259-267
Harmache A, LeBerre M, Droineau Sp, Giovannini M, Bremont M (2006)
Bioluminescence imaging of live infected salmonids reveals that the fin bases are
the major portal of entry for Novirhabdovirus. J Virol 80:3655-3659
Hart LM, Lorenzen N, LaPatra SE, Grady CA and others (2012) Efficacy of a
glycoprotein DNA vaccine against viral haemorrhagic septicaemia (vhs) in Pacific
herring, Clupea pallasii Valenciennes. J Fish Dis 35:775-779

189

Hattenberger-Baudouy AM, Danton M, Merle G, de Kinkelin P (1995). Serum
neutralization test for epidemiological studies of salmonid rhabdoviroses in
France. Vet Res 26:512-520
Hedrick RP, Batts WN, Yun S, Traxler GS, Kaufman J, Winton JR (2003) Host and
geographic range extensions of the North American strain of viral hemorrhagic
septicemia virus. Dis Aquat Organ 55:211-220
Heppell J, Lorenzen N, Armstrong NK, Wu T and others (1998) Development of DNA
vaccines for fish:Vector design, intramuscular injection and antigen expression
using viral haemorrhagic septicaemia virus genes as model. Fish Shellfish
Immunol 8:271-286
Hershberger P, Gregg J, Grady C, Collins R, Winton J (2010) Kinetics of viral shedding
provide insights into the epidemiology of viral hemorrhagic septicemia in pacific
herring. Marine Ecology Progress Series 400:187-193
Hershberger PK, Gregg JL, Grady CA, LaPatra SE, Winton JR (2011) Passive
immunization of Pacific herring against viral hemorrhagic septicemia. J Aquat
Anim Health 23:140-147
Hershberger PK, Gregg JL, Grady CA, Taylor L, Winton JR (2010) Chronic and
persistent viral hemorrhagic septicemia virus infections in pacific herring. Dis
Aquat Organ 93:43-49
Hershberger PK, Gregg J, Pacheco C, Winton J, Richard J, Traxler G (2007) Larval
pacific herring, Clupea pallasii (Valenciennes), are highly susceptible to viral
haemorrhagic septicaemia and survivors are partially protected after their
metamorphosis to juveniles. J Fish Dis 30:445-458
He X, Hu F (2002) Markov chain marginal bootstrap. J Amer Statist Assoc 97:783-795
Hopper K (1989) The isolation of VHSV from chinook salmon at glenwood springs,
orcas island, Washington. American Fisheries Society Fish Health Section
Newsletter 17
Isshik T, Nishizawa T, Kobayashi T, Nagano T, Miyazaki T (2001) An outbreak of VHSV
(viral hemorrhagic septicemia virus) infection in farmed Japanese flounder
Paralichthys olivaceus in japan. Dis Aquat Organ 47:87-99
Iwama GK, Nakanishi T (1996) The fish immune system:organism, pathogen, and
environment. Academic Press, San Diego, CA

190

Murphy K, Travers P, Walport M, Janeway C (2008) Janewayʼs immunobiology:7
Edition. Garland Science, New York, NY.

th

Jensen MH (1963) Preparation of fish tissue cultures for virus research. Bull Off Int
Epizoot, 59:131-134
Jensen MH (1965) Research on the virus of egtved disease. Ann NY Acad Sci 126:422426
Johnson KA, Flynn JK, Amend DF (1982) Duration of immunity in salmonids vaccinated
by direct immersion with Yersinia ruckeri and Vibrio anguillarum bacterins. J Fish
Dis 5:207-213
Jordan W (2005) Competitive enzyme-linked immunosorbent assay. In:Burns R (ed)
Immunochemical protocols, Vol 295. Humana Press, p 215-225
Jørgensen PEV (1971) Egtved virus:Demonstration of neutralizing antibodies in serum
from artificially infected rainbow trout (Salmo gairdneri). J Fish Res Board Can
28:875-877
Jørgensen PEV (1973) The nature and biological activity of IPN virus neutralizing
antibodies in normal and immunized rainbow trout (Salmo gairdneri). Arch
Gesamte Virusforsch 42:9-20
Jørgensen PEV (1976) Partial resistance of rainbow trout (Salmo gairdneri) to viral
haemorrhagic septicaemia (VHS) following exposure to non-virulent egtved virus.
Nord Vet Med 28:570-571
Jørgensen PEV (1982) Egtved virus:Temperature- dependent immune response of trout
to infection with low-virulence virus. J Fish Dis 5:47–55
Jørgensen PEV, Olesen NJ, Lorenzen N, Winton JR, Ristow SS (1991) Infectious
hematopoietic necrosis (IHN) and viral hemorrhagic septicemia (VHS):Detection
of trout antibodies to the causative viruses by means of plaque neutralization,
immunofluorescence, and enzyme-linked immunosorbent assay. J Aquat Anim
Health 3:100-108
Kaattari SL (1992) Fish B lymphocytes:Defining their form and function. Annu Rev Fish
Dis 2:161-180
Kaattari SL, Piganelli JD (1996) The specific immune system: humoral defense. In:
Iwama G, Nakanishi T (eds) The fish immune system. Academic Press, San
Diego, CA, p 207–54

191

Kelly RK, Nielsen O (1985) Inhibition of infectious pancreatic necrosis virus by serum
from rainbow trout (Salmo gairdneri) in Canadian hatcheries. Fish Pathology
19:245-251
Kibenge MT, Opazo B, Rojas AH, Kibenge FS (2002) Serological evidence of infectious
salmon anaemia virus (ISAV) infection in farmed fishes,using an indirect enzymelinked immunosorbent assay (ELISA). Dis Aquat Organ 51:1-11
Kim R (2010) Identification of host range, susceptibility, and disease course of viral
hemorrhagic septicemia virus (VHSV) in Great Lakes fish species., Ph.D. Thesis,
Michigan State University, East Lansing, MI
Kim R, Faisal M (2010a) Comparative susceptibility of representative Great Lakes fish
species to the North American viral hemorrhagic septicemia virus sublineage IVb.
Dis Aquat Organ 91:23-34
Kim R, Faisal M (2010b) Experimental studies confirm the wide host range of the Great
Lakes viral haemorrhagic septicaemia virus genotype IVb. J Fish Dis 33:83-88
Kim R, Faisal M (2010c) The Laurentian Great Lakes strain (MI03) of the viral
haemorrhagic septicaemia virus is highly pathogenic for juvenile muskellunge,
Esox masquinongy (Mitchill). J Fish Dis 33:513-527
Kim R, Faisal M (2011a) Emergence of viral hemorrhagic septicemia virus in the
Laurentian Great Lakes. In: Cipriano RC, Bruckner AW, Shchelkunov IS (eds)
Bridging America and Russia with shared perspectives on aquatic animal health.
Michigan State University, East Lansing, Michigan, p 113-122
Kim R, Faisal M (2011b) Emergence and resurgence of the viral hemorrhagic
septicemia virus (Novirhabdovirus, Rhabdoviridae, Mononegavirales). Journal of
Advanced Research 2:9-23
Kim RK, Faisal M (2012) Shedding of viral hemorrhagic septicemia virus (genotype IVb)
by experimentally infected muskellunge (Esox masquinongy). J Microbiol 50:278284
Kim CH, Johnson MC, Drennan JD, Simon BE, Thomann E, Leong JA (2000) DNA
vaccines encoding viral glycoproteins induce nonspecific immunity and mx
protein synthesis in fish. J Virol 74:7048-7054

Kim S, Lee J, Hong M, Park H, Park S (2003) Genetic relationship of the VHSV (viral
hemorrhagic septicemia virus) isolated from cultured olive flounder, Paralichthys
olivaceus in korea. J Fish Pathol 16:1-12

192

Kim W, Nishizawa T, Yoshimizu M (2007) Non-specific adsorption of fish
immunoglobulin M (IgM) to blocking reagents on ELISA plate wells. Dis Aquat
Organ 78:55-59
Kocan R, Bradley M, Elder N, Meyers T, Batts W, Winton J (1997) North American
strain of viral hemorrhagic septicemia virus is highly pathogenic for laboratoryreared Pacific herring. J Aquat Anim Health 9:279-290
Kocan RM, Hershberger PK, Elder NE, Winton JR (2001) Epidemiology of viral
hemorrhagic septicemia among juvenile Pacific herring and Pacific sand lances
in Puget Sound, Washington. J Aquat Anim Health 13:77-85
Kollner B, Wasserrab B, Kotterba G, Fischer U (2002) Evaluation of immune functions
of rainbow trout (Oncorhynchus mykiss)--how can environmental influences be
detected? Toxicol Lett 131:83-95
Kurath G, Garver KA, Corbeil S, Elliott DG, Anderson ED, LaPatra SE (2006) Protective
immunity and lack of histopathological damage two years after DNA vaccination
against infectious hematopoietic necrosis virus in trout. Vaccine 24:345-354
Kurath G, Leong JC (1985) Characterization of infectious hematopoietic necrosis virus
mrna species reveals a nonvirion rhabdovirus protein. J Virol 53:462-468
Kurath G, Purcell MK, Garver KA (2007) Fish rhabdovirus models for understanding
host response to DNA vaccines. CAB Reviews:Perspectives in Agriculture,
Veterinary Science, Nutrition and Natural Resources 2
Landis JR, Koch GG (1977) The measurement of observer agreement for categorical
data. Biometrics 33:159-174
Laing KJ, Hansen JD (2011) Fish T cells:Recent advances through genomics. Dev
Comp Immunol 35:1282-1295
LaPatra SE (1996) The use of serological techniques for virus surveillance and
certification of finfish. Annu Rev Fish Dis 6:15-28
LaPatra SE (1998) Factors affecting pathogenicity of infectious hematopoietic necrosis
virus (IHNV) for salmonid Fish. J Aquat Anim Health 10:121–131
LaPatra SE, Corbeil S, Jones GR, Shewmaker WD, Lorenzen N, Anderson ED, Kurath
G (2001) Protection of rainbow trout against infectious hematopoietic necrosis
virus four days after specific or semi-specific DNA vaccination. Vaccine 19:40114019

193

LaPatra SE, Lauda KA, Jones, GR (1994) Antigenic variants of infectious hematopoietic
necrosis virus and implications for vaccine development. Dis Aquat Organ
20:119–126
LaPatra SE, Turner T, Lauda KA, Jones GR, Walker S (1993) Characterization of the
humoral response of rainbow trout to infectious hematopoietic necrosis virus. J
Aquat Anim Health 5:165-171
Le Morvan C, Troutaud D, Deschaux P (1998) Differential effects of temperature on
specific and nonspecific immune defences in fish. J Exp Biol 201:165-168
Li J, Barreda DR, Zhang YA, Boshra H and others (2006) B lymphocytes from early
vertebrates have potent phagocytic and microbicidal abilities. Nat Immunol
7:1116-1124
Liu X, Collodi P (2002) Novel form of fibronectin from zebrafish mediates infectious
hematopoietic necrosis virus infection. J Virol 76:492-498
Lorenzen E, Einer-Jensen K, Martinussen T, LaPatra SE, Lorenzen N (2000) DNA
vaccination of rainbow trout against viral hemorrhagic septicemia virus:A dose
response and time course study. J Aquat Anim Health 12:167-180
Lorenzen E, Einer-Jensen K, Rasmussen JS, Kjær TB, Collet B, Secombes CJ,
Lorenzen N (2009) The protective mechanisms induced by a fish rhabdovirus
DNA vaccine depend on temperature. Vaccine 27:3870-3880
Lorenzen N, LaPatra SE (1999) Immunity to rhabdoviruses in rainbow trout:The
antibody response. Fish Shellfish Immunol 9:345-360
Lorenzen N, LaPatra SE (2005) DNA vaccines for aquacultured fish. Rev Sci Tech
24:201-213
Lorenzen N, Lorenzen E, Einer-Jensen K, Heppell J, Wu T, Davis H (1998) Protective
immunity to vhs in rainbow trout (Oncorhynchus mykiss, Walbaum) following
DNA vaccination. Fish Shellfish Immunol 8:261-270
Lorenzen N, Lorenzen E, Einer-Jensen K, LaPatra SE (2002) DNA vaccines as a tool
for analysing the protective immune response against rhabdoviruses in rainbow
trout. Fish Shellfish Immunol 12:439-453
Lorenzen N, Olesen NJ, Jørgensen PEV (1990) Neutralization of Egtved virus
pathogenicity to cell cultures and fish by monoclonal antibodies to the viral G
protein. J Gen Virol 71:561-567

194

Lorenzen N, Olesen NJ, Jørgensen PEV (1993) Antibody-response to VHS virus
proteins in rainbow trout. Fish Shellfish Immunol 3:461-473
Lorenzen N, Olesen NJ, Koch C (1999) Immunity to VHS virus in rainbow trout.
Aquaculture 172:41-61
Lovy J, Piesik P, Hershberger PK, Garver KA (2013) Experimental infection studies
demonstrating Atlantic salmon as a host and reservoir of viral hemorrhagic
septicemia virus type IVa with insights into pathology and host immunity.
Veterinary Microbiology 166:91-101
Lumsden JS, Morrison B, Yason C, Russell S and others (2007) Mortality event in
freshwater drum Aplodinotus grunniens from Lake Ontario, Canada, associated
with viral haemorrhagic septicemia virus, type IV. Dis Aquat Organ 76:99-111
Ma C, Ye J, Kaattari SL (2013) Differential compartmentalization of memory b cells
versus plasma cells in salmonid fish. Eur J Immunol 43:360-370
Magnadottir B (2006) Innate immunity of fish (overview). Fish Shellfish Immunol 20:137151
McLauchlan PE, Collet B, Ingerslev E, Secombes CJ, Lorenzen N, Ellis AE (2003) DNA
vaccination against viral haemorrhagic septicaemia (VHS) in rainbow trout:Size,
dose, route of injection and duration of protection--early protection correlates with
mx expression. Fish Shellfish Immunol 15:39-50
Medzhitov R, Janeway C (1997) Innate immunity: Impact on the adaptive immune
response. Curr Opin Immunol 9:4
Meyers TR, Short S, Lipson K (1999) Isolation of the North American strain of viral
hemorrhagic septicemia virus (VHSV) associated with epizootic mortality in two
new host species of Alaskan marine fish. Dis Aquat Organ 38:81-86
Meyers TR, Short S, Lipson K, Batts WN, Winton JR, Wilcock H, Brown E (1994)
Association of viral hemorrhagic septicemia virus with epizootic hemorrhages of
the skin in Pacific herring Clupea harengus pallasi from Prince William Sound
and Kodiak Island, Alaska, USA. Dis Aquat Organ 1: 27-37
Meyers TR, Sullivan J, Emmenegger E, Follett J, Short S, Batts WN (1992) Identification
of viral hemorrhagic septicemia virus isolated from pacific cod Gadus
macrocephalus in Prince William Sound, Alaska, USA. Dis Aquat Organ 12:167175

195

Meyers TR, Winton JR (1995) Viral hemorrhagic septicemia virus in North America.
Annu Rev Fish Dis 5:3-24
Mortensen H, Heuer O, Lorenzen N, Otte L, Olesen N (1999) Isolation of viral
haemorrhagic septicaemia virus (VHSV) from wild marine fish species in the
Baltic sea, Kattegat, Skagerrak and the North sea. Virus Res 63:95-106
Medzhitov R, Janeway C (1997) Innate immunity:Impact on the adaptive immune
response. Curr Opin Immunol 9:4
Medzhitov R, Janeway CA (2002) Decoding the patterns of self and nonself by the
innate immune system. Science 296:298-300
Millard EV, Brenden TO, LaPatra SE, Marcquenski S, Faisal M (In press) Detection of
antibodies to viral hemorrhagic septicemia virus-IVb in sera of muskellunge (Esox
masquinongy) using a competitive ELISA. Dis Aquat Organ
Millard EV, Faisal M (2012a) Development of neutralizing antibody responses in
muskellunge, Esox masquinongy (Mitchill), experimentally exposed to viral
haemorrhagic septicaemia virus (genotype IVb). J Fish Dis 35:11-18
Millard EV, Faisal M (2012b) Heterogeneity in levels of serum neutralizing antibodies
against viral hemorrhagic septicemia virus genotype IVb among fish species in
Lake St. Clair, Michigan, USA. J Wildl Dis 48:405-415
Najafian L, Babji AS (2012) A review of fish-derived antioxidant and antimicrobial
peptides:Their production, assessment, and applications. Peptides 33:178-185
Nakanishi T, Toda H, Shibasaki Y, Somamoto T (2011) Cytotoxic T cells in teleost fish.
Dev Comp Immunol 35:1317-1323
Nakao M, Tsujikura M, Ichiki S, Vo TK, Somamoto T (2011) The complement system in
teleost fish:Progress of post-homolog-hunting researches. Dev Comp Immunol
35:1296-1308
Neukirch M (1984) An experimental study of the entry and multiplication of viral
haemorrhagic septicaemia virus in rainbow trout, Salmo gairdneri Richardson,
after water-borne infection. J Fish Dis 7:231-234
Nishizawa T, Iida H, Takano R, Isshiki T, Nakajima K, Muroga K (2002) Genetic
relatedness among Japanese, American and European isolates of viral
hemorrhagic septicemia virus (VHSV) based on partial G and P genes. Dis Aquat
Organ 48:143-148

196

Novoa B, Romero A, Mulero V, Rodríguez I, Fernández I, Figueras A (2006) Zebrafish
(Danio rerio) as a model for the study of vaccination against viral haemorrhagic
septicemia virus (VHSV). Vaccine 24:5806-5816
OIE (World Organisation for Animal Health) (2012) Manual of diagnostic tests for
aquatic animals. www.oie.int/manual-of-diagnostic-tests-for aquatic animals
(accessed 10 Oct 2013)
Olesen NJ (1998) Sanitation of viral haemorrhagic septicaemia (VHS). J Appl Ichthyol
14:173-177
Olesen NJ, Jørgensen PEV (1986) Detection of neutralizing antibody to Egtved virus in
rainbow trout (Salmo gairdneri) by plaque neutralization test with complement
addition. J Appl Ichthyol 2:33-41
Olesen NJ, Lorenzen N, Jørgensen PEV (1991) Detection of rainbow trout antibody to
Egtved virus by enzyme-linked immunosorbent assay (ELISA),
immunofluorescence (IF),and plaque neutralization tests (50% PNT). Dis Aquat
Organ 10:31-38
Olesen NJ, Lorenzen N, LaPatra SE (1999) Production of neutralizing antisera against
viral hemorrhagic septicemia (VHS) virus by intravenous injections of rabbits. J
Aquat Anim Health 11:10-16
Olson W, Emmenegger E, Glenn J, Simchick C, Winton J, Goetz F (2013) Expression
kinetics of key genes in the early innate immune response to Great Lakes viral
hemorrhagic septicemia virus IVb infection in yellow perch (Perca flavescens).
Dev Comp Immunol 41:11-19
Ortega-Villaizan M, Chico V, Falco A, Perez L, Coll JM, Estepa A (2009) The rainbow
trout TLR9 gene and its role in the immune responses elicited by a plasmid
encoding the glycoprotein G of the viral haemorrhagic septicaemia rhabdovirus
(VHSV). Mol Immunol 46:1710-1717
Pakingking R, Jr., Okinaka Y, Mori K, Arimoto M, Muroga K, Nakai T. (2004) In vivo and
in vitro analysis of the resistance against viral haemorrhagic septicaemia virus in
Japanese flounder (Paralichthys olivaceus) precedingly infected with
aquabirnavirus. Fish Shellfish Immunol 17:1-11
Park KC, Reno PW (2005) Characteristics of inhibition of infectious pancreatic necrosis
virus (IPNV) by normal rainbow trout Oncorhynchus mykiss serum. Dis Aquat
Organ 63:43-52

197

Pier GP, Lyczak JB, Wetzler LE (2004) Immunology, infection, and immunity, Vol. ASM
Press, Washington, D.C.
Pierce LR, Stepien CA (2012) Evolution and biogeography of an emerging
quasispecies:Diversity patterns of the fish viral hemorrhagic septicemia virus
(VHSV). Molecular Phylogenetics and Evolution 63:327-341
Purcell MK, Kurath G, Garver KA, Herwig RP, Winton JR (2004) Quantitative expression
profiling of immune response genes in rainbow trout following infectious
haematopoietic necrosis virus (IHNV) infection or DNA vaccination. Fish Shellfish
Immunol 17:447-462
Purcell MK, Laing KJ, Winton JR (2012) Immunity to fish rhabdoviruses. Viruses 4:140166
Quillet E, Dorson M, Aubard G, Torhy C (2001) In vitro viral haemorrhagic septicaemia
virus replication in excised fins of rainbow trout: correlation with resistance to
waterborne challenge and genetic variation. Dis Aquat Organ 45:171-182,
Quillet E, Dorson M, Aubard G, Torhy C (2007) In vitro assay to select rainbow trout
with variable resistance/susceptibility to viral haemorrhagic septicaemia virus. Dis
Aquat Organ 76:7-16
R Core Team (2012). R:A language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria.
Ristow SS, de Avila J, LaPatra SE, Lauda KA (1993) Detection and characterization of
rainbow trout antibody against infectious hematopoietic necrosis virus. Dis Aquat
Organ 15:109-114
Reid A, Young KM, Lumsden JS (2011) Rainbow trout Oncorhynchus mykiss
ladderlectin, but not intelectin, binds viral hemorrhagic septicemia virus IVb. Dis
Aquat Organ 95:137-143
Roberts RJ (2012) Fish pathology. Wiley-Blackwell, Hoboken, NJ.
Rombout JH, Huttenhuis HB, Picchietti S, Scapigliati G (2005) Phylogeny and ontogeny
of fish leucocytes. Fish Shellfish Immunol 19:441–455
Sakai DK (1981) Heat inactivation of complements and immune hemolysin reactions in
rainbow trout, masu trout, coho salmon, goldfish and tilapia. Bull Jpn Soc Sci
Fish 47:565-571

198

Salinas I, Zhang Y-A, Sunyer JO (2011) Mucosal immunoglobulins and B cells of teleost
fish. Dev Comp Immunol 35:1346-1365
Salonius K, Simard N, Harland R, Ulmer JB (2007) The road to licensure of a DNA
vaccine. Curr Opin Investig Drugs 8:635-641
®

SAS Institute Inc (2010) SAS/STAT 9.22 Userʼs Guide. Cary, NC.
Schirmbeck R, Reimann J (2001) Revealing the potential of DNA-based
vaccination:Lessons learned from the hepatitis b virus surface antigen. Biol
Chem 382:543-552
Schönherz AA, Hansen MHH, Jørgensen HBH, Berg P, Lorenzen N, Einer-Jensen K
(2012) Oral transmission as a route of infection for viral haemorrhagic
septicaemia virus in rainbow trout, Oncorhynchus mykiss (Walbaum). J Fish Dis
35:395-406
Schütze H, Mundt E, Mettenleiter TC (1999) Complete genomic sequence of viral
hemorrhagic septicemia virus, a fish rhabdovirus. Virus Genes 19:59-65
Schyth BD, Ariel E, Korsholm H, Olesen NJ (2012) Diagnostic capacity for viral
haemorrhagic septicaemia virus (VHSV) infection in rainbow trout (Oncorhynchus
mykiss) is greatly increased by combining viral isolation with specific antibody
detection. Fish Shellfish Immunol 32:593-597
Scott WB, Crossman EJ (1973) Freshwater Fishes of Canada. Bulletin of the Fisheries
Research Board of Canada, vol. 184, Fisheries Research Board of Canada,
Ottawa, Ontario, Canada
Skall HF, Olesen NJ, Mellergaard S (2005) Viral haemorrhagic septicaemia virus in
marine fish and its implications for fish farming-a review. J Fish Dis, 28, 509-529.
Smail DA (2000) Isolation and identification of viral haemorrhagic septicaemia (VHS)
viruses from cod Gadus morhua with the ulcus syndrome and from haddock
Melanogrammus aeglefinus having skin haemorrhages in the north sea. Dis
Aquat Organ 41:231-235
Smail DA, Snow M (2011) Viral haemorrhagic septicaemia. In:Woo PTK, Bruno DW
(eds) Fish diseases and disorders:Viral, bacterial and fungal, Vol 3. CAB
International, Oxfordshire, UK, p 110-142
Snow M, Bain N, Black J, Taupin V and others (2004) Genetic population structure of
marine viral haemorrhagic septicaemia virus (VHSV). Dis Aquat Organ 61:11-21

199

Snow M, Cunningham CO, Melvin WT, Kurath G (1999) Analysis of the nucleoprotein
gene identifies distinct lineages of viral haemorrhagic septicaemia virus within the
European marine environment. Virus Res 63:35-44
Solem ST, Stenvik J (2006) Antibody repertoire development in teleosts—a review with
emphasis on salmonids and Gadus morhua L Dev Comp Immunol 30:57–76
Stavnezer J, Guikema JE, Schrader CE (2008) Mechanism and regulation of of class
switch recombination. Annu Rev Immunol 26:261-292
Tafalla C, Chico V, Pèrez L, Coll JM, Estepa A (2007) In vitro and in vivo differential
expression of rainbow trout (Oncorhynchus mykiss) Mx isoforms in response to
viral haemorrhagic septicaemia virus (VHSV) G gene, poly I:C and VHSV. Fish
Shellfish Immunol 23:210-221
Tafalla C, Sanchez E, Lorenzen N, DeWitte-Orr SJ, Bols NC (2008) Effects of viral
hemorrhagic septicemia virus (VHSV) on the rainbow trout (Oncorhynchus
mykiss) monocyte cell line rts-11. Mol Immunol 45:1439-1448
Takano T, Iwahori A, Hirono I, Aoki T (2004) Development of a DNA vaccine against
hirame rhabdovirus and analysis of the expression of immune-related genes after
vaccination. Fish Shellfish Immunol 17:367-374
Takano R, Nishizawa T, Arimoto M, Muroga K (2000) Isolation of viral haemorrhagic
septicaemia virus (VHSV) from wild Japanese flounder, Paralichthys olivaceus.
Bull Eur Assn Fish P 20:186-192
Thompson T, Batts W, Faisal M, Bowser P and others (2011) Emergence of viral
hemorrhagic septicemia virus in the North American Great Lakes region is
associated with low viral genetic diversity. Dis Aquat Organ 96:29-43
Thoulouze MI, Bouguyon E, Carpentier C, Brémont M (2004) Essential role of the NV
protein of Novirhabdovirus for pathogenicity in rainbow trout. J Virol 78:40984107
Toda H, Araki K, Moritomo T, Nakanishi T (2011) Perforin-dependent cytotoxic
mechanism in killing by CD8 positive T cells in ginbuna crucian carp, Carassius
auratus langsdorfii. Dev Comp Immunol 35:88-93
Tonheim TC, Bøgwald J, Dalmo RA (2008) What happens to the DNA vaccine in fish? A
review of current knowledge. Fish Shellfish Immunol 25:1-18

200

Tonheim TC, Leirvik J, Løvoll M, Myhr AI, Bøgwald J, Dalmo RA (2007) Detection of
supercoiled plasmid DNA and luciferase expression in Atlantic salmon (Salmo
salar l.) 535 days after injection. Fish Shellfish Immunol 23:867-876
Traxler GS, Anderson E, LaPatra SE, Richard J, Shewmaker B, Kurath G (1999) Naked
DNA vaccination of Atlantic salmon Salmo salar against IHNV. Dis Aquat Organ
38:183-190
Traxler G, Kieser D, Richard J (1999) Mass mortality of pilchard and herring associated
with viral hemorrhagic septicemia virus in British Columbia, Canada. American
Fisheries Society, Fish Health Section Newsletter 27:4-5
United States Department of Agriculture, Animal and Plant Health Inspection Service.
2008. VHS Regulated Species,
www.aphis.usda.gov/animal_health/animal_dis_spec/aquaculture/. Accessed
June 2013.
United States Fish and Wildlife Service and American Fisheries Society-Fish Health
Section (2010). Standard procedures for aquatic animal health inspections. In:
AFS-FHS. FHS blue book:suggested procedures for the detection and
identification of certain finfish and shellfish pathogens, 2010 edition. AFS-FHS,
Bethesda, Maryland.
Utke K, Bergmann S, Lorenzen N, Köllner B, Ototake M, Fischer U (2007) Cellmediated cytotoxicity in rainbow trout, Oncorhynchus mykiss, infected with viral
haemorrhagic septicaemia virus. Fish Shellfish Immunol 22:182-196
Utke K, Kock H, Schuetze H, Bergmann SM and others (2008) Cell-mediated immune
responses in rainbow trout after DNA immunization against the viral hemorrhagic
septicemia virus. Dev Comp Immunol 32:239-252
Vasta GR, Nita-Lazar M, Giomarelli B, Ahmed H and others (2011) Structural and
functional diversity of the lectin repertoire in teleost fish:Relevance to innate and
adaptive immunity. Dev Comp Immunol 35:1388-1399
VHSV Expert Panel and Working Group (2010) Viral hemorrhagic septicemia virus
(VHSV IVb) risk factors and association measures derived by expert panel. Prev
Vet Med 94:128-139
Verjan N, Ooi EL, Nochi T, Kondo H and others (2008) A soluble nonglycosylated
recombinant infectious hematopoietic necrosis virus (IHNV) G-protein induces
ifns in rainbow trout (Oncorhynchus mykiss). Fish Shellfish Immunol 25:170-180

201

Verrier ER, Langevin C, Benmansour A, Boudinot P (2011) Early antiviral response and
virus-induced genes in fish. Dev Comp Immunol 35:1204-121
Verrier ER, Langevin C, Tohry C, Houel A and others (2012) Genetic resistance to
rhabdovirus infection in teleost fish is paralleled to the derived cell resistance
status. Plos One 7:e33935
Voller A, Binwell DE, Bartlett A (1979) The enzyme-linked immunosorbent assay
(ELISA). A guide with abstract of microplate applications. Dynatech Europe,
Guernsey.
Walker PJ, Benmansour A, Dietzgen R, Fang RX, Jackson AO, Kurath G, Leong JC,
Nadin-Davies S, Tesh RB, Tordo N (2000) Family Rhabdoviridae. In:Van
Regenmortel MHV, Fauquet CM, Bishop DHL, Carstens EB, Estes MK, Lemon
SM, J. Maniloff J, Mayo MA, McGeoch DJ, Pringle CR, Wickner RB (eds) Virus
taxonomy. Classification and nomenclature of viruses. Academic Press, San
Diego, CA, p. 563–583
Winton JR, Kurath G, Batts WN (2008) Molecular epidemiology of viral hemorrhagic
septicemia virus in the Great Lakes region. United States Geological Service,
Fact Sheet 2008–3003. http://wfrc.usgs.gov/products/fs20083003.pdf (accessed
May 2010)
Weeks C, Kim R, Wolgamod M, Whelan G, Faisal M (2011) Experimental infection
studies demonstrate the high susceptibility of the salmonid, lake herring,
Coregonus artedi (Le sueur), to the Great Lakes strain of viral haemorrhagic
septicaemia virus (genotype IVb). J Fish Dis 34:887-891
Whyte SK (2007) The innate immune response of finfish--a review of current
knowledge. Fish Shellfish Immunol 23:1127-1151
Wilson M, Bengtén E, Miller NW, Clem LW, Du Pasquier L, Warr GW (1997) A novel
chimeric Ig heavy chain from a teleost fish shares similarities to IgD. PNAS
94:4593-4597
Winton J, Batts W, deKinkelin P, LeBerre M, Bremont M, Fijan N (2010) Current
lineages of the epithelioma papulosum cyprini (EPC) cell line are contaminated
with fathead minnow, Pimephales promelas, cells. J Fish Dis 33:701-704
Wolf K (1988) Fish viruses and fish viral diseases. Cornell University Press, Ithaca, NY
Wright PF, Nilsson E, Van Rooij EM, Lelenta M, Jeggo MH (1993) Standardisation and
validation of enzyme-linked immunosorbent assay techniques for the detection of
antibody in infectious disease diagnosis. Rev Sci Tech 12:435-450

202

Yamamoto T, Batts WN, Winton JR (1992) In vitro infection of salmonid epidermal
tissues by infectious hematopoietic necrosis virus and viral hemorrhagic
septicemia virus. J Aquat Anim Health 4:231-239
Yamamoto T, Clermont TJ (1990) Multiplication of infectious hematopoietic necrosis
virus in rainbow trout following immersion infection:Organ assay and electron
microscopy. J Aquat Anim Health 2:261-270
Ye J, Kaattari IM, Kaattari SL (2011) The differential dynamics of antibody
subpopulation expression during affinity maturation in a teleost. Fish Shellfish
Immunol 30:372-377
Zapata A, Amemiya CT (2000) Phylogeny of lower vertebrates and their immunological
structures. Curr Top Microbiol Immunol 248:67-107
Zapata A, Diez B, Cejalvo T, Gutierrez-de Frias C, Cortes A (2006). Ontogeny of the
immune system of fish. Fish Shellfish Immunol 20:126–136
Zhang S, Liu Y, Zhang F, Hu R (2009) Competitive ELISA using a rabies glycoproteintransformed cell line to semi-quantify rabies neutralizing-related antibodies in
dogs. Vaccine 27:2108-2113.
Zhang YA, Salinas I, Li J, Parra D and others (2010) IgT, a primitive immunoglobulin
class specialized in mucosal immunity. Nat Immunol 11:827-835
Zou J, Secombes CJ (2011) Teleost fish interferons and their role in immunity. Dev
Comp Immunol 35:1376-1387
Zwollo P, Cole S, Bromage E, Kaattari S (2005) B cell heterogeneity in the teleost
kidney: Evidence for a maturation gradient from anterior to posterior kidney. J
Immunol 174:6608-6616

203