Functional study of protective immunity following DNA vaccination

against viral haemorrhagic septicaemia in rainbow trout

Ph.D. thesis by

Dagoberto Sepúlveda Araneda

November 2015

Supervisor: Professor Niels Lorenzen

Table of Contents

1. Preface ................................................................................................. 1

Acknowledgements .................................................................................. 2

List of original manuscripts ...................................................................... 3

List of abbreviations ................................................................................. 4

Summary .................................................................................................. 6

Sammendrag (Danish Summary) .............................................................. 8

2. Part I: Background ............................................................................... 10

2.1 Aquaculture ........................................................................................... 10

2.2 Aquaculture and infectious diseases ......................................................... 10

2.3 Viral diseases in finfish aquaculture .......................................................... 12

2.4 Viral haemorrhagic septicaemia (VHS) ...................................................... 13

2.5 Viral haemorrhagic septicaemia virus (VHSV) ............................................ 14

2.5.1 VHSV- Infectious cycle .................................................................... 15

2.5.2 VHSV: distribution, diversity, and adaptability ................................... 16

2.6 Control of viral diseases in aquaculture ..................................................... 18

2.7 Fish immune system .............................................................................. 19

2.7.1 Innate immunity ............................................................................ 19

2.7.2 Adaptive immunity ......................................................................... 21

2.8 Vaccines against VHSV ........................................................................... 24

2.9 DNA vaccine against VHSV ...................................................................... 26

2.10 In this thesis ....................................................................................... 32

3. Part II: Hypotheses and objectives .............................................. 34

4. Part III: Publications .................................................................... 36

4.1 Manuscript I .......................................................................................... 36

4.2 Manuscript II ......................................................................................... 65

4.3 Manuscript III ........................................................................................ 85

5. Part IV: General discussion and perspectives ............................... 105

5.1 Section I

5.1 Manuscript I ..................................................................................... 106

5.2 Section II

5.2.1 Manuscript II ................................................................................. 107

5.2.2 Manuscript III ................................................................................ 109

6. Concluding remarks..................................................................... 112

7. References .................................................................................. 115

1

1. Preface

This thesis describes the work carried out at the Section of Fish Health, Department of

Animal Science, Aarhus University, under the supervision of the Professor Niels

Lorenzen.

This Ph.D. project was funded by the Chilean National Scholarship Program for

Graduate Studies (Conicyt) and by the European Commission under the 7th

Framework Programme for Research and Technological Development contract

FP7311993, TargetFish.

This thesis is divided into four parts. Background information is described in Part I.

Hypothesis and objectives are presented in Part II. The major studies are described in

Part III, and are presented in three manuscripts. The Discussion and perspectives are

described in Part IV.

2

Acknowledgment

I would like to thank all the people who contributed in some way to the work described

in this thesis.

First of all, I would like to express my sincere gratitude to my supervisor, Professor

Niels Lorenzen, for his guidance, motivation, and fruitful discussions.

I would like to thank Jesper Skou Rasmussen, Helle Kristiansen, Helle Frank Skall,

Ellen Lorenzen, and those who worked in the lab at the beginning of my project, Katja

Einer-Jensen, Brian Dall Schyth, for the useful feedback and insightful comments on

my work.

My sincere thanks to Inge Marie Jepsen, Hanne Møller Purup, Torben Egil Kjær, Lene

Nørskov, Tommy Hejl, Hanne Bucholtz, Lisbeth Kjær Troels for the skillful support and

helpful assistance, without their effort my job would have undoubtedly been more

difficult.

I thank my fellow lab mates Dennis Bela-ong, Sekar Larashati and Niccoló Vendramin,

for the stimulating discussions at work, but also for sharing social activities out of the

lab that made my stay in Denmark a pleasant time.

Last but not least, I thank my parents, sister, and friends for the love, care and

constant encouragement during this thesis and my life in general.

3

List of original manuscripts

Manuscript I

Dagoberto Sepúlveda, Niels Lorenzen

“Can VHS virus bypass the protective immunity induced by DNA vaccination in

rainbow trout?”

Submitted to PLOs One

Manuscript II

Dagoberto Sepúlveda, Ellen Lorenzen, Jesper Skou Rasmussen, Katja Einer-Jensen,

Bertrand Collet, Chris Secombes, Niels Lorenzen

“Time-course study of the immune protection induced by an interferon-

inducible DNA vaccine against viral haemorrhagic septicaemia virus in

rainbow trout”

Intended for submission to Fish and Shell Fish Immunology

Manuscript III

Dagoberto Sepúlveda, Jesper Skou Rasmussen, David Parra, Niels Lorenzen

“Attempt to mimicking antibody-antigen complexes by DNA vaccination in a

fish virus model”

Intended for submission to Fish and Shell Fish Immunology

4

List of abbreviations

ADCC Antibody-dependent cell-mediated cytotoxicity

APC Antigen presenting cells

Bp Base pair

BF2 Bluegill fry cells

cDNA Complementary deoxyribonucleic

CPE Cytopathic effect

CMV Cytomegalovirus

CTL Cytotoxic T lymphocytes

DNA Deoxyribonucleic acid

EAVR Early antiviral responses

ELISA Enzyme-linked immunosorbent assay

EPC Epithelioma papulosum cyprinid cells

Fc Fragment crystallizable region of Immunoglobulins

G Glycoprotein IFN Interferon

GALT Gut-associated lymphoid tissue

ILT Interbranchial lymphoid tissue

I.M Intramuscular

I.P Intraperitoneal

ISG Interferon stimulated genes

ISRE Interferon-stimulated response element

L Large protein, RNA-dependent RNA polymerase

LAVR Long-term antiviral responses

M Matrix protein

MAb Monoclonal Antibody

MEM Eagle’s minimum essential medium

5

MHC Major Histocompatibility complex

mRNA Messenger RNA

N Nucleoprotein

NV Non-virion protein

P Phosphoprotein

PAMP Pathogen-associated molecular pattern

PRR Pattern recognition receptors

RNA Ribonucleic acid

RT-qPCR Quantitative reverse transcription polymerase chain reaction

SAVR Specific antiviral responses

ssRNA Single-stranded RNA

TCID50 Tissue culture infective dose

VHS Viral haemorrhagic septicaemia

Viruses

IHNV Infectious haematopoietic necrosis virus

ISAV Infectious salmon anaemia virus

HIRRV Hirame rhabdovirus

SAV Salmon alphavirus

VHSV Viral haemorrhagic septicaemia virus

6

Summary

Functional studies of protective immunity related to DNA vaccination of rainbow trout

against viral haemorrhagic septicemia (VHS)

This paper addresses some functional and safety aspects associated with DNA

vaccination of farmed fish. Vaccination against infectious diseases using DNA is a

relatively new approach for disease prevention in husbandry animals and humans. The

concept is based on insertion of the gene encoding a protective antigen of a disease-

causing agent into a eukaryotic expression vector, typically, a plasmid with a strong

cytomegalovirus (CMV) promoter. Upon intramuscular injection of purified vaccine

DNA, some cells in the vaccinated animal will take up the plasmid and express the

vaccine antigen. This will activate the animal's immune system. The principle does not

work for all diseases. In this context, it is interesting that DNA vaccination against the

diseases caused by rhabdoviruses in salmonids has been found to be particularly

efficient. This has led to the commercialization of a DNA vaccine for the Atlantic salmon

in Canada.

In Europe, VHS is one of the most important viral diseases in farmed rainbow trout,

and an experimental DNA vaccine based on the genre of the viral surface glycoprotein

(G) has given promising results under laboratory conditions.

The work in this thesis has aimed to clarify the following questions, all of which are

related to the potential use of DNA vaccination to prevent disease in aquaculture:

I: Would VHS virus be able to bypass the protective immunity induced by DNA

vaccination after a few passages in vaccinated fish?

II: Would a DNA vaccine, in which the CMV promoter is replaced by an interferon-

inducible Mx promoter provide improved safety and practical benefits?

III: Could the combination of the Fc domain of the immunoglobulin molecule with the

vaccine antigen in a fusion protein promote the immunogenicity of the antigen?

Despite the fact that RNA viruses such as VHS virus is known to have a high mutation

rate and thus the ability to adapt to new host conditions, the results showed that the

DNA vaccine induced a robust protection which was not bypassed by the virus during

repeated passaging in vaccinated animals. However, although vaccination protected

against the disease, some of the vaccinated fish still got infected following exposure to

7

VHS virus and were capable of transmitting the infection to non-vaccinated cohabitant

fish.

Use of an interferon-inducible trout Mx promoter in the DNA vaccine gave interesting

results both in terms of consumer safety and understanding of the development of

protective immunity. In contrast to the CMV promoter, the trout Mx promoter was not

active in human cells. On the other hand, a variable protective effect was observed in

fish over time, and the results thereby emphasized the importance of examining the

effect of new vaccines over a prolonged period of time.

A range of DNA vaccine constructs were made, encoding Fc-fusion proteins by

combining the constant Fc domain of the trout antibody molecules with the soluble part

of the viral G protein. None of these induced protection, but since the fusion proteins

were also not secreted from transfected cells, it was not possible to finally answer

question III.

The results are presented in the thesis in the form of three manuscripts for publication

in scientific journals.

8

Sammendrag

Funktionelle undersøgelser af beskyttende immunitet i forbindelse med DNA

vaccination af regnbueørreder mod viral haemorrhagisk septikæmi (VHS).

Denne afhandling adresserer nogle af de fundamentale funktionelle og

sikkerhedsmæssige aspekter i forbindelse med DNA vaccination af opdrætsfisk.

Vaccination mod smitsomme sygdomme ved hjælp af DNA er en relativt ny strategi for

sygdomsforebyggelse i husdyr og mennesker. Konceptet bygger på at genet for et

beskyttende antigen fra det sygdomsfremkaldende agens indsættes i en eukaryot

expressions vektor, typisk i form af et plasmid med en stærk promoter fra

cytomegalovirus (CMV). Ved intramuskulær injektion af oprenset DNA vil plasmidet

optages i nogle få af det vaccinerede dyrs celler. Disse vil udtrykke vaccine-antigenet,

som igen vil aktivere dyrets immunsystem. Princippet fungerer dog ikke lige godt for

alle sygdomme. I den sammenhæng er det interessant, at DNA vaccination mod

sygdomme forsaget af rhabdovirus i laksefisk har vist sig at være særdeles effektiv.

Dette har ført til kommercialisering af en DNA vaccine til Atlantisk laks i Canada. I

Europa er VHS en af de vigtigste virussygdomme i opdræt af regnbueørred og en

eksperimentel DNA vaccine baseret på genet for det virale overflade glykoprotein (G)

har givet lovende resultater under laboratorieforhold.

Arbejdet i denne afhandling har søgt at afklare nedenstående spørgsmål, som alle er

relateret til mulighederne for anvendelse af DNA vaccination til sygdomsforebyggelse i

akvakultur:

I: Vil VHS virus være i stand til at omgå den beskyttende immunitet induceret af DNA

vaccination efter få passager i vaccinerede fisk?

II: Vil en DNA vaccine, hvor CMV promotoren er udskiftet med en interferon inducibel

Mx promoter give sikkerheds- og anvendelsesmæssige fordele?

III: Kan kombination af Fc-delen fra immunglobulin molekylet med vaccineantigenet i

et fusionsprotein fremme immunogeniciteten?

På trods af at RNA virus som VHS virus er kendt for at have en høj mutationsfrekvens

og dermed evne til at tilpasse sig nye vilkår, viste resultaterne at DNA vaccination

inducerede en robust beskyttelse, som virus ikke omgik i løbet af få passager i

vaccinerede dyr. Det var dog også klart at selvom vaccinationen beskyttede mod

9

sygdom, blev en del af fiskene alligevel inficerede og dermed i stand til at overføre

infektionen til ikke vaccinerede dyr.

Anvendelse af en interferon-inducibel ørred Mx-promoter i DNA vaccinen gav

interessante resultater både i forhold til forbrugersikkerhed og forståelse af udvikling

af beskyttende immunitet. I modsætning til CMV-promoteren var ørred Mx-promoteren

ikke aktiv i humane celler. Til gengæld varierede den beskyttende effekt i fisk over tid

og resultaterne understregede dermed vigtigheden i at undersøge effekten af nye

vacciner over et længere tidsrum.

Det viste sig at være vanskeligt at få transfekterede celler til at udskille

fusionsproteiner bestående af den konstante Fc-del af antistofmolekyler og G

proteinet. Det var derfor ikke muligt endegyldigt at besvare spørgsmål III.

Phd-projektets resultater er præsenteret i afhandlingen i form af 3 manuskripter til

publikation i videnskabelige tidsskrifter.

10

2. Part I: Background

2.1 Aquaculture

The aquaculture industry has progressively grown for the last two decades, playing an

increasingly important role in a food supply. This expansion is the result of both,

depleting wild fish stocks by overfishing, and increasing demand for seafood by an

expanding world population (FAO 2014). In the 1990’s, the aquaculture industry

provided only 25% of the total fish consumed. In 2012, the aquaculture industry

represented 42% of the total fish consumed, corresponding to 66.6 million tons fish

production (FAO 2014). Projections indicate that aquaculture will keep developing to

become the largest supplier of consumed fish and fish products (Hall 2011).

Developing a sustainable and fast-growing aquaculture industry has required facing

different types of challenges. One is the technical limitations of upgrading fish farms

from extensive low scale production to intensive high scale production(Brudeseth,

Wiulsrød et al. 2013). Another is the environmental impact of introducing exotic fish

species into new farming areas. However, the major challenge that has constantly

threatened the sustainability of the aquaculture industry is infectious diseases (Leung

and Bates 2013).

2.2 Aquaculture and infectious diseases

Infectious diseases are one of the major causes of economic losses in the aquaculture

industry. The negative effects on fish farming can reduce meat quality or cause high

mortality rates, which in some cases reach up to 100% of the production. The impact

of the infectious diseases in some countries has almost shut down the industry within a

short time, generating economic losses and social problems such as the loss of labor

11

because of reduced production (Grischkowsky and Amend 1976; Olesen 1998; Skall,

Olesen et al. 2005; Kibenge, Godoy et al. 2012).

One study that monitored Norwegian farms for over 14 years showed that the annual

average fish loss was 8.6% of the production, of which 85% (7.31% of total

production) was due to fish deaths, most of them caused by pathogens (Dixon 2012).

In 2007 the global market for finfish aquaculture had a value of US$ 20 billion and the

7.31% loss by pathogens was valued at 1.4 billion dollars (Dixon 2012). Therefore,

reducing losses, which includes controlling the outbreaks of infectious diseases, is a

priority for the aquaculture industry.

Aquacultured fish are highly susceptible to infectious diseases because of the

characteristics of fish farms and aquatic environments, which often provide ideal

conditions for the spread of infectious diseases. First, the aquatic environment often

lacks physical barriers to avoid pathogen transmission between farmed fish and wild

fish reservoirs. Moreover, the pathogens are able to survive a long time outside the

host in this environment, which facilitates the spread of pathogens over long distance

by migratory wild fish or sea currents (Meyer 1991; McCallum, Harvell et al. 2003;

Kurath and Winton 2011; Kibenge, Godoy et al. 2012). Second, fish farming has

intrinsic characteristics that promote pathogen outbreaks, and are not present under

wild conditions. Some of these characteristics are: high-density fish stocks of a single

fish species, stress associated with crowding, handling, continuous production. On top

of this geographical movement of cultured fish promotes the introduction of exotic

pathogens (Kurath and Winton 2011; Kibenge, Godoy et al. 2012). Not only do these

conditions promote disease outbreaks in aquaculture, but they may also favor selection

of pathogens with higher virulence, or pathogens with the capacity to infect new hosts.

12

2.3 Viral diseases in finfish aquaculture

Infectious diseases in finfish aquaculture are caused by different types of

microorganisms. Bacteria are the most prevalent causative disease agent in

aquaculture, responsible for 54.4% of all cases, followed by 22.6% for viruses 19.4%

for parasites and 3.1% for fungi (Kibenge, Godoy et al. 2012). Although bacterial

diseases are the most common, it is possible to control most of them effectively by

using antibiotics and vaccines.

On the contrary, viral diseases have been harder to control, mainly due to the high

susceptibility of aquatic animals at early growth stages, lack of antiviral treatments,

low-efficacy of commercial vaccines, and fast viral spreading (Kibenge, Godoy et al.

2012).

Understanding aquaculture sensitivity and the risk to an expanding global trade in

farmed fish and their products, The World Organization of Animal Health (OIE) has

delineated standards and guidelines to control the risks of spreading diseases. One of

these guidelines is to define a list of aquatic animal diseases, which must be reported

to the authorities in order to take proper contingency measures. The selection of the

diseases in the list follows three main criteria;

If the disease can cause significant production losses at a national or

multinational level,

If there is scientific evidence that the disease causes significant morbidity or

mortality in wild aquatic animal populations

If the agent is of public health concern.

Out of the 10 diseases listed, 8 are caused by a viral etiological agent, an indication of

the international concern about the damages that viral diseases can cause. The listed

viral diseases are: the Epizootic haematopoietic necrosis virus (EHNV), the infectious

13

salmon anaemia virus (ISAV), the salmonid alphavirus (SAV), the infectious

haematopoietic necrosis virus (IHNV), the koi herpes virus disease (KHV), the red sea

bream iridoviral disease (RSIV), the spring viremia of carp (SVCV), and the viral

haemorrhagic septicaemia (VHSV) (OIE 2012).

This work will focus on VHSV, one of the most important viral diseases in the world,

and the one of greatest concern in rainbow trout fish farming in Europe (EURL-FISH

2015).

2.4 Viral haemorrhagic septicaemia (VHS)

Viral haemorrhagic septicaemia was confirmed to be caused by a viral etiological agent

in 1963. At that time, the virus was named Egtved virus, taking the name of the

Danish village close to the isolation area. Currently, it is named viral haemorrhagic

septicaemia virus (VHSV) and has been isolated from over 80 fish species in Asia,

North America, and Europe (Skall, Olesen et al. 2005).

VHS has a high impact on aquaculture, affecting economically important fish species,

such as the rainbow trout and turbot in Europe (Ross, McCarthy et al. 1995; Smail and

Snow 2011), and the Japanese flounder and olive flounder in Japan and Korea,

respectively (Schlotfeldt, Ahne et al. 1991; Isshik, Nishizawa et al. 2001; OIE 2012).

The mortalities in an outbreak can reach up to 100% in rainbow trout fry and 30-70%

in adults. Moreover, massive mortalities by VHS have been registered in marine and

freshwater wild fish species in North America, causing environmental disasters

(Lumsden, Morrison et al. 2007; OIE 2012).

VHSV is transmitted horizontally by direct contact with infected fish or with water

containing virions from infected fish. The major entry portals of the virus are the gills,

skin and fin base (Harmache, LeBerre et al. 2006; Brudeseth, Skall et al. 2008; OIE

2012).

14

The VHS has different variants. The acute form is the most common form of the

disease, characterized by clinical signs such as skin darkening, exophthalmia, anaemia,

haemorrhage, and abnormal swimming behavior, often leading to high mortality. The

chronic form is usually a persistent infection in the absence of external clinical signs.

Finally, the nervous form is characterized by infection mainly in the brain tissue,

generating changes in the swimming behavior.

The form of the disease, the clinical signs, and the severity of mortality rates caused

by VHSV depend on several factors such as pathogen-related factors (strain of VHSV,

virus loads), host-related factors (species, growth stage, stress level) and

environmental-related factors (water temperature, oxygen concentration, water

salinity) (Smail and Snow 2011; OIE 2012).

2.5 Viral haemorrhagic septicaemia virus (VHSV)

The viral haemorrhagic septicaemia virus (VHSV), a member of the genus

Novirhabdovirus, belonging to the Rhabdoviridae family. VHSV is an enveloped virus

with the characteristic bullet shape of the rhabdovirus, 100 nm in length and 60 nm

width (Figure 1C). The genome is a single molecule of linear, negative-sense, single-

stranded RNA (ssRNA) of 11,4 kb in size (Walker, Benmansour et al. 2000).

The genome encodes six proteins, five of which are structural proteins, the

nucleoprotein (N), the phosphoprotein (P), the matrix protein (M), the glycoprotein (G)

and the RNA-dependent RNA polymerase (L). VHSV and other Novirhabdovirus

genomes further encode a non-structural protein denominated NV (Figure 1A-B). The

function of NV protein is not completely understood, but it has been associated with

apoptosis control in infected cells (Ammayappan and Vakharia 2011). The NV gene is

absent in the classical mammalian rhabdoviruses like vesicular stomatitis virus (VSV)

15

and rabies virus (RV) (Fig. 1) and it might be speculated that the NV protein somehow

reflects adaptation to poikilothermic animal hosts.

Figure 1. Genomic organization of (A) VHSV, and (B) Rabies virus. (C) VHSV

particle. Images from viral zone web page (www.viralzone.expasy.org)

2.5.1 VHSV infection cycle

When VHSV enters the host, the glycoprotein of the virion recognizes and attaches to a

host receptor on the plasma membrane (Bearzotti, Delmas et al. 1999), and enters

into the host cell by receptor-mediated endocytosis (Liu, Liu et al. 2011). Later, a pH-

dependent fusion of the viral envelope with the vesicle membrane releases the viral

ribonucleocapside (RNP), consisting of the genomic (-)ssRNA, N, L, and P into the

cytoplasm, where transcription and replication take place. The transcription of the

messenger RNAs from the genomic viral (-)ssRNA is carried out by a complex of the

RNA-dependent RNA polymerase (L) and the phosphoprotein (P). The same complex is

A

B

C

16

responsible for the post-transcriptional processing of the mRNA (capping and

polyadenylation) (Maclachlan and Dubovi 2010).

The VHVS genomic RNA only has one promoter, which is located at the 3’ end of the

genome. The viral RNA polymerase attaches to this promoter and moves along the

genome in the 3’ to 5’ direction. The six mRNAs are generated by a mechanism called

Attenuated Transcription or Stop-Start Transcription. Briefly, after transcribing the first

gene, the polymerase reaches the stop/start signal in the intergenic region, and a

fraction of the polymerases moves to the next gene to continue the transcription, while

another fraction of the polymerases fall off and attaches to the promoter to start again.

The result of this process is a higher transcription of the genes located at the 3’ end

(close to the promoter), reducing the transcription level gradually, according to the

position in the genome N>P>M>G>NV>L.

The next step in the viral cycle is the translation of the viral mRNAs by the host

machinery. When high levels of N and P are accumulated in the cytoplasm, the RNA

polymerase switches from synthesizing mRNA to synthesizing the antigenome, which is

later used as a template for the synthesis of the new genomic RNA. The nucleoprotein

attaches to the newly formed genomic RNA to form RNP, which, along with M protein,

binds the transmembrane glycoproteins inserted in the plasma membrane. The last

step of the cycle is the shedding that occurs when the virions are formed by the

budding of new virus particles from the plasma membrane (Maclachlan and Dubovi

2010).

2.5.2 VHSV: distribution, diversity, and adaptability

RNA viruses, and thus VHSV, have a characteristic high genetic variability, which is an

advantage for rapid adaptation to environmental or host changes. The capacity for

high variability is due to high replication rates, a large population size, and elevated

17

mutation rates due to the absence in most RNA viruses of a proofreading activity in

RNA polymerases (Manrubia and Lázaro 2006). To compare, while a vertebrate host

genome has a mutation rate of 1x10-9 substitution per site per year (sub/site/year),

the VHSV genome has a mutation rate of 2,58 x 10-4 sub/site/year (Holmes 2009;

Pierce and Stepien 2012). According to the quasispecies theory, the variability of RNA

viruses rapidly generates a heterogeneous virus population in the infected organisms.

This population consists of a dominant genome and a very diverse spectrum of low-

frequency viral genomes. The frequency of one of these viral genomes can change in

response to a shift in the environmental conditions or a shift to a new host (Manrubia

and Lázaro 2006; Domingo, Sheldon et al. 2012).

A demonstration of this principle is the capacity of certain viruses to generate escape

mutants under selective pressure, such as the immune protection induced by a vaccine

or the inhibition of the viral replication by antiviral drugs. Influenza A virus and human

respiratory syncytial virus are able to generate escape mutants to neutralizing

antibodies in embryonated chicken eggs and cell culture, respectively (Lambkin,

McLain et al. 1994; Tomé, Frabasile et al. 2012). However, retroviruses, such as the

human and simian immunodeficiency virus (HIV and SIV), have shown a high capacity

to escape the protection induced by vaccination and the effect of antiviral drugs

(Barouch, Kunstman et al. 2002; Barouch, Kunstman et al. 2003; Adetokunboh,

Atibioke et al. 2015). Therefore, the variability of RNA viruses must be taken into

account in the design, testing, and application of either vaccines or antiviral drugs.

Evidence for the high adaptation capacity of VHSV is shown by its distribution in

diverse environmental conditions (marine water, freshwater, and different

temperatures) and its ability to infect over 80 fish species. Phylogenetic analyses have

divided VHSV in four genotypes (I-IV). Genotype I is subdivided into Ia – Ie, and

together with genotype II and III were found in European freshwater and northern

18

European marine environments. Genotype IV is subdivided into IVa – IVc, which were

found in the Great Lakes in North America, Japan, Korea, and the west coast of

Canada and USA (Pierce and Stepien 2012). VHSV was recently declared eradicated

from the freshwater environment in Denmark (Olesen, Skall et al.). However, the virus

is still present in the marine environment, and although marine isolates of VHSV are

not- or low-virulent to rainbow trout, recent outbreaks have occurred in sea reared

rainbow trout in other Scandinavian countries (Dale, Orpetveit et al. 2009). The

potential risk of the emergence of a new high virulent variant is therefore constant and

requires disease control programs to have to be strictly applied and updated

continually.

2.6 Control of viral diseases in aquaculture

The constant growth of the aquaculture industry requires establishing of programs to

control the impact of fish diseases and to reduce the associated economic losses of the

industry. These programs include surveillance measures to detect the pathogens at

early stages of infection, avoiding spread of the diseases, management of infected fish

(quarantine, killing infected fish, fallowing periods) and prophylactic measures

(vaccination, probiotic, immunostimulant) (Kibenge, Godoy et al. 2012).

Even though all measures play a significant role in the disease control program, the

application of vaccines was a turning point in the development and expansion of the

aquaculture industry. The first vaccines commercially available were against the

bacterial enteric redmouth disease (ERM, yersiniosis) and vibriosis in the USA in 1976,

while the first anti-viral vaccine was against spring viremia carp in 1982 (Evelyn

1996). Although vaccines since then have been successful tools for controlling many

bacterial diseases in aquaculture and reducing the needs for antibiotic treatments

(Midtlyng, Grave et al. 2011), the outcome of attempts to develop and implement

19

vaccines against viral diseases has not been similarly promising. On major challenge is

that traditional vaccines based on killed virus usually are too expensive to produce to

be cost-efficient for use in aquacultured fish. And while many veterinary vaccines

against viral diseases in husbandry animals are based on attenuated virus, authorities

have been hesitating to allow use of live viral vaccines in the aquatic environment,

where environmental safety issues are often hard to predict. The way forward thus

seems to be development of recombinant subunit vaccines. However, the design of

such vaccines has appeared to require a better understanding of how the antiviral

immune mechanisms in fish can be efficiently activated and long lasting immunity

established.

2.7 Fish Immune system

Fish, as well as mammals, possess the two classical arms of the high vertebrate

immune system, the innate and the adaptive immunity. However, because of the

evolutionary distance between the fish and mammal immune system, there are some

structural and functional differences.

2.7.1 Innate immunity

Innate immunity in all vertebrates and invertebrates has the same fundamental

characteristics: quickly induced after infection, not specific to an antigen, and the

molecules involved in this immunity are encoded by germ-line genes. Some of the

components of the innate immunity that play an important role in the defense against

viral infection are outlined below:

The physical barriers are the first line of defense against pathogens, blocking

their entry into the organism. Skin, gut, and gill in combination with the

secretion of mucus are the main components of the physical barriers (Ángeles

20

Esteban 2012). The mucus contains lysozymes, proteases, cathepsin B, alkaline

phosphatase, and antimicrobial peptides, providing different strategies to

defend the organisms from a broad range of pathogens (Collet 2014).

Complement System. In higher vertebrates, the classical, mannan-binding

lectin (MBL), and the alternative complement pathway have been associated

with antiviral responses. In salmonid, the complement system has also been

involved in the fish defense against viruses. For instance, after salmon

alphavirus (SAV) infection, components of the classical pathway were up-

regulated (Desvignes, Quentel et al. 2002). Another example involves the VHSV

neutralization by fish serum in vitro, which is a complement-dependent process

(Olesen and Jørgensen 1986).

Cellular components such as Macrophages, Neutrophils, and Natural Killer-like

cells, among other immune cells, play two roles in the immune responses. First,

destroying directly pathogens or infected cells. Second, these cells can trigger

the activation of the adaptive immunity by processing the antigens, along with

producing cytokines and other soluble factors (Collet 2014).

Pattern Recognition Receptors (PRRs) are cellular receptors specialized in

recognizing the pathogen-associated molecular pattern (PAMPs), such as

lipopolysaccharides (LPS), flagellins and peptidoglycan in bacteria, or DNA CpG

motif, dsRNA, and glycoprotein in viruses. Toll-like receptors and RIG-like

helicases are essential PRRs for detection of viral components such as

intermediates of the virus replication in fish. The activation of these receptors

leads to the activation or synthesis of antiviral protein, inflammatory cytokines,

and chemokines (Workenhe, Rise et al. 2010).

Interferons (IFNs) are cytokines induced by the recognition of viral components

by PRRs. Fish and mammals IFN type I appears to induce similar activation

pathways. IFNs play an important role activating an antiviral status in infected

21

and in neighboring cells, inducing the up-regulation of the IFN-stimulated genes

(ISG), which are effector antiviral proteins such as protein kinase R (PKR), Vig-

1, ISG15, and Myxovirus resistance protein (Mx) (Workenhe, Rise et al. 2010;

Collet 2014).

Even though most components of the innate immunity in fish have homologous in

mammals, fish often have multiple isoforms of innate immune molecules such as

complement proteins and C-reactive proteins, and thereby showing a higher diversity

in the innate immune response than the one found in mammals (Magnadottir 2010).

This possibly reflects that innate protective mechanisms play a more important role in

fish, particularly at low temperatures, at which activation of adaptive mechanisms

takes longer time.

2.7.2 Adaptive immunity

The second arm of the vertebrate immune system is the adaptive immunity. The

adaptive immunity in both mammals and fish is characterized by: acting on a specific

antigen, requiring a longer activation time than the innate immunity, inducing long-

lasting protection, and eliciting an enhanced response to a second exposure to the

pathogen (immunological memory). It is important to take into account that the

temperature affects physiological processes in fish such as metabolism and the

immune responses in general, however, the adaptive immunity appears to be more

sensitive to this factor, compared to the innate immunity. A study in channel catfish

showed that at low temperature, the antigen processing and presentation was

impaired, causing a down-regulation of responses of helper T cells, and cytotoxic T

cells, generating a low antibody production (Vallejo, Miller et al. 1991). Therefore,

considering that the fish farms are located in different geographic areas, the

22

temperature might have a huge impact on the adaptive responses induced by

vaccination.

One major difference between mammals and fish is the organs involved in the

production and maturation of immune cells. In mammals, the production of red blood

cells, platelets and white blood cells (B and T lymphocytes, macrophages) occurs in the

bone marrow, while the antigen presentation and lymphocyte maturation take place in

the lymph nodes. Fish lack bone marrow, lymph nodes, germinal centers and mucosal-

associated lymphoid tissue (MALT) (Tort, Balasch et al. 2003). Instead, the major

lymphoid organs in fish are the head kidney, the thymus, and the spleen. The head

kidney is a primary lymphoid organ analogous to the bone marrow. The head kidney is

where all stages of maturation of B cell occurs, and the production antibodies take

place. Other regions of the kidney also possess some immunological functions (K2-K5

in figure 2). The thymus is another primary organ and has been associated with the

maturation of T cells. The spleen, a secondary lymphoid organ, is where the antigen

presentation occurs and thereby the place where the adaptive responses are initiated.

Additionally, two secondary lymphoid organs have been identified in fish, the

interbranchial lymphoid tissue (ILT) (Haugarvoll, Bjerkås et al. 2008; Koppang, Fischer

et al. 2010), and the gut-associated lymphoid tissue (GALT), the latter contains

intraepithelial T and B lymphocytes within the gut lamina propia (Figure 2) (Zapata,

Diez et al. 2006; Salinas, Zhang et al. 2011).

Figure 2. Location of lymphoid organs in teleost. Taken from (Kibenge, Godoy et al.

2012).

23

The adaptive immunity in fish, as well in mammals, can be divided into cellular and

humoral responses:

Cellular immunity. Fish immune system involves different types of cells that can

detect, internalize, process and present dangerous molecules on the Major

Histocompatibility Complex (MHC), activating multiple immune mechanisms. Due to

the blurry borders between the innate and adaptive immunity, some of these cells play

an essential role in both, the innate and adaptive immunity. Cells involved in the

adaptive immunity are (i) Cytotoxic T lymphocytes (CTL), which through cell-mediated

cytotoxicity kill cells infected by intracellular pathogens, which present antigens on

MHC I. (ii) Helper T cells, which activation is mediated by the internalization and

presentation of the exogenous antigen on the MHC II of macrophages, dendritic cells

or B-cells. Helper T cells are important for further activation of the humoral immunity.

Although, the main cellular components of the adaptive immunity in mammals are also

present in fish, the full characterization of these cell populations in fish is still limited

compared mammals (Fischer, Koppang et al. 2013).

Humoral immunity. As in mammals, the B-lymphocytes in fish or rather their

developed stage called plasma cells are responsible for the production of antibodies. It

has been identified that fish produce three immunoglobulin isotypes IgM, IgD, and IgT,

while mammals produce IgM, IgE, IgA, IgD, and IgG.

- IgM is the most prevalent antibody isotype in fish, and can be found as a

membrane-bound form and as a tetrameric secreted form. IgM is associated

with systemic functions, involving complement fixation, agglutination, binding

mannose, binding lectin, signaling cellular cytotoxicity (Kibenge, Godoy et al.

2012).

- IgD functions in both, fish and mammals, are not clear yet. In mammals, IgD

has been associated with some ancient function of surveillance and

24

inflammation, linking the innate and the adaptive immunities (Ma, Ye et al.

2013; Ye, Kaattari et al. 2013).

- IgT has been associated with the mucosal defense, which involves the combat

of parasitic infections. An analysis of surviving fish following a parasitic infection

showed a high titer of IgT parasite-specific in the gut, while very low

concentration in the serum. On the contrary, the same fish had a low titer of

IgM parasite-specific in the gut and high titer in the serum (Zhang, Salinas et

al. 2010).

Affinity maturation and immunological memory are two mechanisms that are better

developed in mammals than in fish. In mammals, low-affinity IgM changes to high-

affinity IgG in a process designated as class switching, which is absent in fish.

However, fish antibodies do pass through an affinity maturation process, but the

magnitude is lower than in mammals. The immunological memory in mammals is

defined as the secondary response to the same antigen, which triggers the production

of antibodies with higher affinity and titers. Memory B cells have also been

characterized in teleost, but with differences in the antibodies titer and affinity in the

secondary response than in mammals (Ma, Ye et al. 2013; Ye, Kaattari et al. 2013).

From the immunological point of view, an ideal vaccine has to induce the adaptive

immunity and particularly long-lasting protective mechanisms. However, taking into

account that the adaptive and innate immunity are connected, a vaccine also has to be

able to induce the innate immunity effectively.

2.8 Vaccines against VHSV

The losses caused by VHSV in economically important fish species such as the rainbow

trout, Japanese flounder, and turbot have driven the development and testing of all

kinds of vaccination strategies: live-attenuated vaccines, whole inactivated killed

25

vaccine, purified subunit proteins of the pathogen, purified proteins produced from

cloned genes, and DNA vaccines (Leong, Anderson et al. 1996). However, currently no

commercial vaccine against VHSV is available. Any commercial vaccine for aquaculture

should possess some features: be safe for the fish, the consumer, and the

environment, have a low cost per dose and be able to induce a high protection against

the pathogen.

In 1995, De Kinkelin and collaborators compared different vaccination strategies

against VHSV and showed that the inactivated virus vaccines induced protection by

intraperitoneal injection, but the efficiency was low when the delivery was performed

by immersion. Moreover, considering the need to use cell culture for its production, the

cost per dose makes this type of vaccine too expensive to be applied. Another type of

vaccine tested was the live-attenuated virus vaccine, which showed higher protection

along with lower cost production than inactivated virus vaccines, and the possibility of

immunizing fish through immersion, a convenient strategy considering the laborious

work to perform the injection on a fish farm with several thousand fish per cage.

Despite the advantages, the live-attenuated virus vaccine against VHSV showed some

residual virulence, which holds a risk of reemergence of the disease considering the

wild fish population surrounding the fish farms that can become reservoirs (De

Kinkelin, Bearzotti et al. 1995; Adelmann, Köllner et al. 2008).

Recombinant subunit vaccines are based on the expression of the VHSV glycoprotein,

which is the target of neutralizing and protective antibodies (Lorenzen, Olesen et al.

1990). The first recombinant subunit vaccine was the expression of the glycoprotein in

E.coli, but it showed a low immunogenicity (Lorenzen, Olesen et al. 1993). The second

involved the expression of the glycoprotein in baculovirus, which provides immune

protection, although part of this protection was a non-specific protection induced by

the insect cell (Lecocq-Xhonneux, Thiry et al. 1994).

26

A more recent, and so far the most promising type of recombinant vaccine tested

against VHSV is the DNA vaccine. This vaccine induces both short and long-term

protection along with a cheaper and simpler production than the other types of

vaccines.

2.9 DNA vaccine against VHSV

A DNA vaccine consists of a naked plasmid encoding an antigenic protein, controlled by

a eukaryotic promoter. After injection into the muscle, the antigen encoded in the DNA

vaccine is translated by the host machinery, inducing a specific protective response

against the antigen (Figure 3).

The first step in developing a DNA vaccine started in 1990, when Wolf and

collaborators showed that it was possible that the host cell machinery expresses a

reporter gene in mice muscle, after intramuscular injection of an expression vector

(Wolff, Malone et al. 1990). Later, the same principle was used, but, in this case,

injecting a DNA vaccine encoding the hemagglutinin protein from influenza, proving

that the intracellular expression of a viral antigen was able to induce immune

protection against this virus in mice. (Ulmer, Donnelly et al. 1993).

In 1996, Anderson and collaborators tested the same principle in rainbow trout. First,

they were able to detect expression of luciferase in the fish muscle after intramuscular

injecting the expression vector (Anderson, Mourich et al. 1996). Second, this

experiment was followed by the intramuscular injection of a plasmid encoding the

glycoprotein of IHNV in rainbow trout, which was able to induce strong immune

protection against this virus (Anderson, Mourich et al. 1996; Gomez-Chiarri, Livingston

et al. 1996). A few years later, similar results were obtained using a DNA vaccine

against VHSV (Heppell, Lorenzen et al. 1998).

27

Figure 3. Development of the DNA vaccine against VHS.

The DNA vaccine concept was later tested against different viral pathogens that affect

the aquaculture. So far, DNA vaccines have been successful against Novirhabdovirus

such as IHNV, VHSV and Hirame rhabdovirus (HIRRV) (Takano, Iwahori et al. 2004),

but has shown low efficacy against other virus, like IPNV or ISAV (Mikalsen, Torgersen

et al. 2004; Mikalsen, Sindre et al. 2005) or no protection against salmon alphavirus

(SAV) and Atlantic halibut nodavirus (Sommerset, Skern et al. 2005; Mutoloki and

Evensen 2011).

The total immune protection induced by the DNA vaccine against VHSV in rainbow

trout consists of two major mechanisms (Figure 4): first, shortly after intramuscular

injection of the DNA vaccine, the VHSV glycoprotein is expressed in the host cells,

inducing a non-specific and short-term protection that can be detected around 4 days

28

post-vaccination (dpv) and fades away after 3-10 weeks depending on the water

temperature. This protection was able to protect against a heterologous virus such as

IHNV, which is another Novirhabdovirus, and against the Atlantic halibut nodavirus, a

non-enveloped virus that belongs to Betanodavirus family. However, the DNA vaccine

was not able to induce protection against bacterial infection such as Yersinia ruckeri

and Aeromonas salmonicida, suggesting the mechanisms to be exclusively of an

antiviral nature (Lorenzen, Lorenzen et al. 2002).

Figure 4. Different immune protective mechanism induced by the DNA vaccine against

VHSV. (Lorenzen, Lorenzen et al. 2002)

Gene expression analyses have shown that early protection correlated with the

upregulation of Mx in the muscle, one of the antiviral proteins induced by IFN type I

(Boudinot, Blanco et al. 1998; Acosta, Petrie et al. 2005). Similar studies using the

DNA vaccine against IHNV have shown an upregulation of IFN type I, and thereby

several interferon-stimulated genes (ISGs), such as IRF-3, Mx-1, Vig-1, and Vig-8.

This expression analysis was carried out in the anterior kidney and spleen,

demonstrating that the antiviral mechanisms were not only induced at the injection

29

site but systemically (Purcell, Nichols et al. 2006). Additionally, some evidence of the

presence of Natural killer-like cells has been identified at short time after vaccination

(Utke, Kock et al. 2008), which agree with the upregulation of IFN-γ, usually

expressed by natural killers cells in mammals (Purcell, Nichols et al. 2006).

The non-specific and short protection induced by the DNA vaccine is followed by the

specific and long lasting protection of the adaptive immune response, which can last

for over two years, according to a time-course study using the DNA vaccine against

IHNV (Kurath, Garver et al. 2006). An additional characteristic of the DNA vaccine is

the capacity to induce both arms of the adaptive immunity, the humoral and the

cellular immune response (Liu 2011).

When the humoral response was analyzed, neutralizing antibodies could be detected

after 8 weeks post-vaccination, disappearing after 9 months post-vaccination, even

though the protection was intact. This result was an indication that the neutralization

by antibodies was not the only mechanism involved the protection. The cellular

responses showed that after intramuscular injection of the DNA vaccine in rainbow

trout, a higher number of pathogen-specific T cells were found in the blood in

comparison with non-injected fish, demonstrating that cellular responses play an

important role in the elimination of the cell that express the glycoprotein encoded in

the DNA vaccine (Utke, Kock et al. 2008).

The activation mechanisms of B cell, T helper cells and cytotoxic T cells by the DNA

vaccine are known for mammals (Figure 5). However, taking into account that

mammals and fish possess similar immune components, it is assumed that the

mechanisms involved in fish are comparable to the ones in mammals. Briefly, after

vaccination, the transfected cells, such as non-immune cells (myocytes) and antigen-

presenting cells (APC), express the antigen, which is processed into short peptides,

and presented for MHC class I (Figure 5B). The transfected APCs trigger the activation

30

of CD8+ cells (CTL) to further eliminate infected cells or in this case, transfected non-

immune cells with the vaccine. Additionally, the antigen secreted from the transfected

cells can be taken up by APCs, which can process it and present it on MHC class II,

activating CD4+, thereby enhancing B cell differentiation into specific antibody

producing cells (Figure 5C) (Rice, Dossett et al. 2008)

Figure 5. Immune mechanisms induced by the DNA vaccine against VHSV. (A)

Immunization or rainbow trout with a DNA vaccine. (B) Activation mechanisms in

different cell types by the expression of the VHSV glycoprotein. (C) Activation of

lymphocyte populations.

31

According to Figure 4, the protection induced by the DNA vaccines against VHSV and

IHNV can be divided into two major mechanisms, the non-specific and the specific

mechanisms. Nevertheless, it has been proposed that the protection induced by these

rhabdovirus DNA vaccines follows a model consisting of three sequential phases

instead of just two phases. The first one, the early antiviral responses (EAVR), is an

early, non-specific, and short-term protection associated with innate mechanisms. The

second phase is the specific antiviral responses (SAVR), which involves the humoral

responses in the presence of neutralizing antibodies and cellular immunity with the

presence of cytotoxic T cells, both associated with the adaptive immune response. The

third phase, the long-term antiviral responses (LAVR), is also associated with adaptive

immune responses, but without the presence of neutralizing antibody. In this third

phase the protection is slightly lower than in SAVR (Figure 6) (Kurath, Garver et al.

2006; Kurath, Purcell et al. 2007).

Figure 6. Model of the three phases of the host response to the IHNV and VHSV DNA

vaccines. Taken from (Kurath, Purcell et al. 2007). ISG: interferon-stimulated genes.

32

Besides inducing a strong protection against VHSV, DNA vaccines also possess a

technical advantage, such as low production costs, thermostability (it does not need

special storage conditions), short production time, and the manufacture of multivalent

vaccines is straightforward in comparison with other vaccine strategies.

In 2007, the first DNA vaccine was licensed and commercialized for aquaculture

against IHNV, in British Columbia, Canada, without outbreaks until now (Salonius,

Simard et al. 2007).

2.10 In this thesis

The experimental work in this thesis covers some aspects of the host-pathogen

interaction between rainbow trout immunized with the DNA vaccine and the fish

rhabdovirus VHSV. The contents are divided into two sections depending on the point

of view of this interaction.

The first section takes the VHSV point of view of this interaction and focuses on

whether VHSV, as a member of RNA viruses with high genetic variability capacity, is

able to escape from the protection induced by the DNA vaccine in rainbow trout,

thereby compromising the practical use of the vaccine.

The second section takes the vaccine point of view. The aim of this section was to

evaluate two innovative strategies for generic improvement of DNA vaccine vectors.

The first one analyzed the kinetics of the protection induced by a DNA vaccine with an

interferon-inducible rainbow trout Mx promoter controlling the expression of the VHSV

glycoprotein instead of the cytomegalovirus promoter used in the highly protective

current DNA vaccine. In terms of consumer safety, a fish-derived promoter would be

preferable to a mammalian virus promoter and an inducible promoter might improve

timing and duration of protective immunity in the fish. The second strategy aimed at

improvement of immunogenicity of a DNA vaccine encoding the secreted form of the

33

VHSV glycoprotein by linking the antigen to the constant region of the fish

Immunoglobulin (Fc). Such Fc-fusion proteins have been used in mammals to improve

the recognition of low immunogenic antigens by improving the Fc receptor mediated

take up by APCs and thereby promoting the antigen processing, presentation, and

further activation of the adaptive immune mechanisms. This approach could be a way

to improve the efficacy of DNA vaccines in fish in general.

34

3 Part II: Hypotheses and objectives

This thesis comprises the three manuscripts presented below.

3.1 Manuscript I

Title “Can VHS virus bypass the protective immunity induced by DNA

vaccination in rainbow trout?”

Hypotheses: Due to its intrinsic high variability capacity, VHSV is able to generate

mutants that can escape from the different protective mechanisms induced by the

glycoprotein gene DNA vaccine.

Objectives:

- Evaluate the ability of VHSV to escape from the innate and the adaptive antiviral

mechanism induced by the DNA vaccine by serial passaging of the virus in rainbow

trout immunized with the DNA vaccine.

-Evaluate the ability VHSV to escape from the neutralizing antibodies response by after

serial passaging or the virus in cell culture in the presence of serum from rainbow trout

immunized with the DNA vaccine.

3.2 Manuscript II

Title: “Time-course study of the immune protection induced by an interferon-

inducible DNA vaccine against viral haemorrhagic septicaemia virus in

rainbow trout”

Hypotheses: A DNA vaccine with a trout-derived IFN-inducible Mx promoter,

controlling the expression of VHSV glycoprotein is protective against VHS and a safer

alternative to the DNA vaccine with CMV promoter.

35

Objective:

-Evaluate the capacity of a DNA vaccine with the IFN-inducible Mx promoter to express

the antigen in fish and human cell lines.

-Evaluate the early, specific, and long-term antiviral responses induced by a DNA

vaccine with IFN-inducible Mx promoter.

3.3 Manuscript III

Title: “Attempt to mimic antibody-antigen complexes by DNA vaccination in a

fish virus model”

Hypotheses: The immunogenicity of a DNA vaccine encoding a secreted antigen can

be made more immunogenic by linking a secreted form of the VHSV glycoprotein to

the Fc domain of the fish immunoglobulin.

Objectives:

-Generate DNA vaccines encoding the secreted form of the VHSV glycoprotein linked to

the Fc domain of each immunoglobulin of rainbow trout.

- Analyze the secretion of the Fc-fusion proteins in cell culture.

- Evaluate the protection induced by DNA vaccines encoding the Fc-fusion proteins in

rainbow trout

36

4. Part III: Publications

4.1 Manuscript I

Can VHS virus bypass the protective immunity induced by DNA

vaccination in rainbow trout?

Dagoberto Sepúlveda, Niels Lorenzen

Submitted to PLOs One

37

Can VHS virus bypass the protective immunity induced by DNA vaccination in

rainbow trout?

Dagoberto Sepúlvedaa, Niels Lorenzen

a.

Fish Health Section, Department of Animal Science, Aarhus University, Aarhus, Denmarka

Corresponding author: Niels Lorenzen (niels.lorenzen@anis.au.dk)

38

ABSTRACT

DNA vaccines encoding viral glycoproteins have been very successful for induction of

protective immunity against diseases caused by rhabdovirus in cultured fish. However, the

vaccine is based on a single viral gene and since RNA viruses are known to possess high

variability and adaptation capacity, this work aimed at evaluating whether viral

haemorrhagic septicaemia virus (VHSV) was able to evade the protective immune response

induced by the DNA vaccination.

VHSV is a negative strand RNA virus, a member of the Rhabdoviridae family. The virus

causes lethal disease in rainbow trout and other economically important cultured fish

species.

The experiments comprised repeated serial passages of a highly pathogenic VHSV isolate

in fish cells in the presence of neutralizing fish serum (in vitro approach), and in rainbow

trout injected with the DNA vaccine (in vivo approach).

For the in vitro approach, the passaged virus was as sensitive as the parental virus to serum

neutralization. For the in vivo approach, the passaged viruses did not show increased

virulence nor increased persistence in vaccinated fish. However, a few vaccinated fish still

carried virus 4 weeks after challenge with passaged or parental virus and were able to

spread the infection to cohabitant naïve fish. The results demonstrated that the DNA

vaccine induces a robust protection, but also that the immunity is non-sterile and that it is

important not to consider vaccinated fish as virus free in veterinary terms.

39

INTRODUCTION

Viral haemorrhagic septicaemia virus (VHSV) is a negative-sense, single-stranded RNA

virus, which belongs to the Novirhabdovirus genus within the Rhabdoviridae family

(Walker, Benmansour et al. 2000). VHSV is the causative agent of the viral haemorrhagic

septicaemia (VHS), a serious and economically important disease of farmed rainbow trout

(Oncorhynchus mykiss) in Europe, causing high mortalities in all fish stages (Skall, Olesen

et al. 2005).

Currently, no commercial vaccine against VHS is available. Several vaccination strategies

have been tested to control this disease, among them live attenuated vaccines, inactivated

vaccines, and recombinant protein vaccines, but with limited efficiency or compromised

safety aspect (De Kinkelin, Bearzotti et al. 1995; Lorenzen and Olesen 1996). By contrast,

DNA vaccines have shown promising results by consistently protecting fish against

rhabdoviruses such as VHSV or Infectious hematopoietic necrosis virus (IHNV)

(Anderson, Mourich et al. 1996; Heppell, Lorenzen et al. 1998). This led to the licensing

and use of a DNA vaccine against IHNV in Atlantic salmon in Canada since 2005

(Salonius, Simard et al. 2007), with no outbreaks reported since.

The traditional DNA vaccine against VHSV consists of a plasmid designed for expression

of the viral surface glycoprotein (G) in eukaryotic cells. After injecting this vaccine

intramuscularly into the fish, an early non-specific interferon associated antiviral protection

is triggered (Lorenzen, Lorenzen et al. 2002; Sommerset, Lorenzen et al. 2003). The

temporary non-specific protection is followed by a specific and long lasting immunity,

40

including both the cellular and the humoral immune arms of adaptive immunity (Lorenzen,

Lorenzen et al. 1999; Utke, Kock et al. 2008)

Although the high efficacy of the DNA vaccine against VHSV has been consistent under

experimental conditions, its protective effect might be threatened following repeated use

under field conditions, due to the high variability of RNA viruses. The genetic variability of

VHSV is reflected by its ability to adapt to different environments and to a huge number of

different fish host species (Skall, Olesen et al. 2005; Pierce and Stepien 2012; Schönherz,

Lorenzen et al. 2015). Whether this variability might also promote generation and selection

of VHSV mutants, capable of evading the immunological protection induced by the DNA

vaccine remain to be addressed.

Some evidence about how the genetic variability of VHSV induces the generation of escape

mutants under selective conditions was shown when rainbow trouts were injected with a

plasmid encoding a neutralizing single chain antibody (scAb) against the G protein of

VHSV. In this case, a neutralization escape mutant was isolated from the survivors to the

infection (Lorenzen, Cupit et al. 2000). Similarly, selective conditions provided by

neutralizing monoclonal antibodies in vitro promoted the growth of neutralization-resistant

virus variants (Bearzotti, Monnier et al. 1995). These escape mutant had only one or few

mutations in the glycoprotein, demonstrating that specific selective conditions can quickly

promote a genetic change in the virus population both in vivo and in vitro. Previous reports

have shown that the DNA vaccine induced a slightly reduced immune protection against a

serotype heterologous to the one used in the vaccine (Lorenzen, Lorenzen et al. 1999). This

suggests that the difference of relatively few amino acids in G protein of VHSV can reduce

41

protection induced by a DNA vaccine. Similar observations have been made for IHNV

(Garver, LaPatra et al. 2005).

While mutations in the G protein can affect the efficacy of the adaptive protection,

mutations in e.g. the non-structural-protein (NS) may potentially affect the ability of the

virus to bypass the innate protection induced by the DNA vaccine. This viral protein has

been suggested to inhibit the apoptotic signal in virus infected cells at an early stage of

virus infection, thus affecting the virulence of the VHSV (Ammayappan and Vakharia

2011).

The aim of this work was to determine whether VHSV within a few generations, under the

selective pressure of DNA vaccine-induced immunity, will be able to develop mutants that

can escape from the innate or the adaptive protective mechanisms induced by the vaccine.

This information will allow us to evaluate the robustness of the current DNA vaccines and

potentially design a safer vaccination strategy.

MATERIALS AND METHODS

Cells. The fish cell lines used in this study were EPC (epithelioma papulosum cyprini)

(Fijan, Sulimanović et al. 1983) and BF2 (bluegill fry fibroblast) (Wolf, Gravell et al.

1966). The cells were maintained in minimum essential media (MEM) supplemented with

10% fetal bovine serum (FBS), 100 U/mL of Penicillin and 100 µg/mL of Streptomycin.

EPC and BF2 were grown for 24 hrs at 24 °C and 21 °C, respectively, and then maintained

at 15 °C.

Virus. To propagate VHSV, BF2 cell cultures were inoculated with low MOI of the virus

and maintained at 15°C until a complete cytopathic effect (CPE) was observed. The

42

supernatant was collected and centrifuged at 4500 x g for 15 min at 4 °C to eliminate

cellular debris. The virus was stored at - 80 °C. The titer was determined using the method

of 50% tissue culture infective doses (TCID50) per mL, in BF2 cells (Reed and Muench

1938).

Virological examination. Supernatants from cell culture or tissue homogenates were used

to prepare 10-fold serial dilutions, which were inoculated onto BF2 monolayer cultured at

60-80% confluence in a 24-well plate. The cultures were maintained at 15 °C for 7 days

when each well was examined. The wells with CPE were considered positive. The identity

of the CPE-causing virus was confirmed by PCR as outlined below.

Vaccination. For in vivo experiment, outbreed all female rainbow trout hatched and reared

under pathogen-free laboratory conditions and with a weight of 2-8 g were used. For the

vaccination, the fish were anesthetized in 0,01% benzocaine and injected intramuscularly

(IM) in the left epaxial muscle below the dorsal fin with 25 µL of purified DNA plasmid in

saline solution (0.9% NaCl), as described earlier (Lorenzen, Lorenzen et al. 1999). This

study included two groups, one vaccinated with 0,1 µg, and another vaccinated with 1,0 µg

of the plasmid pcDNA3-VHSV-G encoding the glycoprotein of VHSV DK3592b (Heppell,

Lorenzen et al. 1998). Non-vaccinated fish were used as controls. All fish were maintained

in pathogen-free laboratory facilities in 120 L aerated aquaria supplied with recirculated

water at 8-10°C. One day before inoculation with virus (challenge), the fish were

transferred to aerated aquaria of 8 L supplied with running tap water in a contained

experimental facility.

43

Passaging of VHSV in vaccinated fish. Infection trials included into two treatment

groups; one group challenged 1 week post-vaccination, and another group challenged 6

weeks post-vaccination (Figure 1).

Figure 1

The experiment included successive passages of the VHSV isolate DK3592b (parental

virus) in fish immunized with the DNA vaccine. In the first passage, each group was

subdivided into 3 subgroups; I) non-vaccinated fish, II) fish vaccinated with 0,1 µg of

plasmid and III) fish vaccinated with a dose of 1,0 µg of plasmid. Every subgroup was kept

in 2 aquaria with 25 fish in each. The infection was carried out by immersion in static water

with a virus concentration of 1 x 105 TCID50 mL

-1, in 8 L water for 3 hrs. After this, water

flow was restored. The experiment was monitored 3 times per day, and moribund fish were

euthanized with an overdose of benzocaine and stored at - 20 °C until further analysis.

At 21 days post infection, the surviving fish were euthanized with an overdose of

benzocaine. The moribund fish sampled were dissected, and spleen, liver, heart, head

kidney, and brain were collected and pooled per fish in MEM. Organs were homogenized

in a TissueLyser (Qiagen) for 2 min at 20 Hz. The homogenate was centrifuged at 4500 x g

for 15 min, and the supernatant was collected to be treated with gentamicin overnight at 4

°C. After the antibiotic treatment, the virus content was titrated and on BF2 cells and the

samples stored at -80 °C (Reed and Muench 1938). This homogenate was used for the next

passage.

The next passages had the same groups and subgroups as the previous passage, but in this

case, 10 fish were used per aquaria. The infection was performed by intraperitoneal

44

injection of the tissue homogenate supernatant collected from the previous passage.

Homogenates included organs from both survivors and dead fish. The procedure for

monitoring and sampling was performed as explained in the first passage.

Figure 2

Examination for VHSV escape mutants to the vaccine-induced immunity. A challenge

trial with vaccinated fish was performed to compare the performance of the parental virus

with that of the passaged virus, under the selective pressure induced by DNA vaccination.

The comparison was based on the accumulated mortality, the virus carrier state of the

vaccinated fish, as well as the level of virus transfer to co-habitant naïve fish.

To obtain sufficient virus for immersion challenge, the virus collected after successive

passages in vaccinated fish (passaged virus), was passaged once in BF2 cells and titrated as

outlined above. An RT-PCR assay with primers specific for the VHSV N-gene was used to

confirm the identity of the passaged virus. Total RNA was isolated from 100µL of each

supernatant using the RNeasy mini kit (Qiagen). Subsequently, the cDNA was synthesed

using the iScript kit (BioRad) following the manufacturer's recommendation. The qPCR

reaction contained 1 µL of the cDNA, 7,5 µL of SuperMix, 0,03 µL of Rox, 0.3 µL of each

primer VHSV-N-For 5’-AGG TCT CAG ATG TCA TCA AGG AG-3’ and VHSV-N-Rev

5’-CGG TGG AGC TCC TGA AGT T-3’, and 5.87 µL of free-nuclease water. The PCR

amplification program involved an initial step at 50°C for 2 min, and then a denaturation

95°C for 2 min. These steps were followed by 40 cycles 95°C for 15 sec and 60°C for 1

min. Amplification was performed in MX Pro-Mx3005P thermocycler (Stratagene)

Table 1

45

The fish involved were vaccinated with 1,0 µg of the DNA vaccine and challenged at 1 or 6

weeks post vaccination, using virus passaged at similar time points in vaccinated fish. Fish

challenged with the parental DK3592b virus served as controls. As an example, when the

fish were challenged 1 week post-vaccination the passaged virus used was VHSV-1W0,1

and VHSV-1W1,0. With each virus, the infection was performed in duplicates with 20 fish

in each aquarium. The infection was carried out by immersion in static water in the 8 L

aquaria for 3hrs with a final concentration of the virus at 1x104 TCID50 mL

-1. After this, the

water flow was restored.

Two weeks post-infection, 10 non-vaccinated (naïve) fish were transferred into each

aquarium to cohabitate with the vaccinated fish. The naïve fish were tagged by cutting a

part of the tail fin. After another 2 weeks, all surviving fish were euthanized with an

overdose of anesthetics. The spleen, heart, liver, kidney and brain were collected and

pooled from each fish individually. The tissues were homogenated and used for virological

examination.

In vitro passaging of VHSV in the presence of neutralizing trout serum. Serial 10-fold

dilutions of VHSV DK3592b (30 µL) were mixed in wells of 96-well plates with 15 µL of

a 1/40 dilution of the heat treated (30 min at 45°C) neutralizing trout serum obtained after

repeated immunization with the DNA vaccine encoding the VHSV glycoprotein (pcDNA3-

VHSV-G) (Utke, Kock et al. 2008). Following incubation for 1 h at 15°C, 15 µL of 1/40

dilution of normal trout serum was added per well as complement source (Olesen and

Jørgensen 1986). After incubation overnight at 15°C, 10 µL of the treated virus was added

46

to the wells of a 96-well plate with EPC cell culture, seeded the previous day, and

incubated for 30 min at 15 °C. Finally, to maintain the selective environment, 50 µl of

MEM 5% supplemented with a dilution of 1/640 of the immune serum and a dilution of

1/640 of the complement was added.

Five replicates well were used for each virus dilution. The cells were incubated at 15 °C for

7 days. After that, we collected the supernatant from 3 wells given the lowest virus

inoculum still causing CPE. Serial dilutions were made with each of the 3 collected

supernatants. The dilutions were mixed with antiserum for the second passage, following

the same procedure as for the first passage, except that only 3 replicate cell culture wells

were inoculated with diluted antiserum/virus mixtures. This passaging under selective

(antiserum) pressure was repeated 11 times.

Evaluation of virus´ susceptibility to serum neutralization. The susceptibility of the

passaged and the parental virus to the neutralizing effect of the immune serum was

compared using the plaque neutralization assay (PNT) (Olesen and Jørgensen 1986), in

which the titer of the serum is defined as the reciprocal value of the highest serum dilution

reducing the number of plaques by 50% compared to a normal trout serum control. To

evaluate the presence of mutations, the full-length glycoprotein gene of each passaged virus

was amplified by PCR and sequenced. The total RNA isolation and the cDNA synthesis

was performed following the same procedures outlined above. The PCR amplification was

performed with primers flanking the G gene using the high fidelity DNA polymerase

Herculase II Fusion (Agilent Technology) according to the procedure recommended by the

47

manufacturer. The alignment step in the PCR was performed at 57oC, using the forward

primer 5´ TAC AAT CGT GCC GTC GAA G 3´and the reverse primer 5´ AGG TCA CAG

TTG AGG TAG TTG 3´.

RESULTS

In vivo approach: Evaluation of the ability of passaged VHSV to evade the innate and

adaptive protection induced by the DNA vaccine

Passaging in vaccinated rainbow trout. To test the ability of the virus to bypass the

innate immunity induced by the DNA vaccine, we passaged VHSV in rainbow trout

vaccinated one week before inoculation with virus. When the fish used for passaging had

been vaccinated with a dose of 0,1 µg of the DNA vaccine, we were able to isolate virus

from dead and vaccinated surviving fish in each of the successive 5 passages. When fish

given 1,0 µg of the DNA vaccine were used, we were able to isolate virus only until the 2nd

passage. This suggested that the protection induced by 1,0 µg of the DNA vaccine was able

to clear the virus in the immunized fish more efficiently than in fish vaccinated with 0,1 µg

DNA, although the latter dose was still highly protective (data not shown).

To test the ability of the virus to bypass the adaptive immune response induced by the DNA

vaccine, the passaging was performed in rainbow trout vaccinated 6 weeks before

inoculation. In this setup, VHSV was re-isolated from dead and vaccinated survivor fish in

all 4 passages using both vaccine doses (data not shown).

48

Comparison between passaged virus and parental virus. After successive passages, the

virus obtained from the last passages under selective conditions (passaged virus), was

amplified by 1 passage on BF2 cells and later compared with the parental VHSV isolate

DK3592b.

The comparison in the in vivo approach included three parameters; mortalities induced in

vaccinated and non-vaccinated fish, the capacity of the virus to infect and persist in

vaccinated fish, and the ability of the virus to spread the infection from vaccinated carriers

to cohabitant naïve fish.

The mortality rates caused by the passaged virus and the parental virus in vaccinated fish

were both low, ranging between 0-12 %. Only one of the duplicate aquaria with fish

inoculated with the VHSV-1W1,0, and one of the duplicates inoculated with the VHSV-

6W0,1, showed higher mortalities (about 20%) than the aquaria inoculated with the parental

virus (Figure 3). These results were considered to be due to an intergroup variability rather

than increased virulence of the passaged viruses. In non-vaccinated fish, all viruses induced

high mortality rates, indicating that the passaging had not affected the virulence.

Figure 3

The evaluation of the ability of the virus to infect vaccinated fish showed that both the

parental and passaged viruses were able to infect and persist in some of the vaccinated fish

for at least 4 weeks post-challenge (Table 1). The highest frequencies of carriers in

individual aquaria were obtained with the parental virus, with 36% and 15,8% of virus-

positive fish, when the evaluation was performed at 1 and 6 weeks post-vaccination,

respectively. Among the fish inoculated with passaged virus, the highest carrier frequencies

49

reached 13.6% and 10.5 % respectively. Therefore, the passaging had not improved the

ability of the passaged virus to persist in vaccinated fish.

The analysis of the capacity of the virus to spread from vaccinated carriers to naïve fish

showed that even though only a few fish could be detected as carriers, vaccinated fish were

able to transmit the infection to naïve cohabitants.

Table 2

In vitro approach: Evaluation of the ability of passaged VHSV escape from

neutralization by serum from fish immunized with the DNA vaccine

After 11 successive passages in cell culture in the presence of neutralizing trout serum, we

compared the ability of the passaged virus and the parental virus to escape from the

neutralization by a trout immune serum. As shown in Table 3, no significant difference in

susceptibility to neutralization was found between the passaged viruses and the parental

virus.

The glycoprotein gene of the 3 passaged viruses and the parental virus was sequenced and

compared to determine the presence of mutations. The 4 sequences had 100% nucleotide

identity (not shown).

Table 3

DISCUSSION

This work focused on analyzing whether the fish rhabdovirus VHSV was able to mutate

and escape from the immune protection induced by a DNA vaccine. Such mutants would

represent a potential risk of reemergence of the disease in vaccinated fish populations,

50

reducing the applied potential of the DNA vaccine. To our knowledge, this important

aspect remains to be addressed for the otherwise extensively analyzed and highly protective

fish DNA vaccines. Our results suggest that the immune response triggered by the vaccine

is rather robust and not easily bypassed by the virus.

The high genetic variability of RNA viruses is due to the high replication rates, large

population size, and high mutation rates, which together generate a diverse spectrum of

virus variants in every replication cycle and result in a population of one dominant virus

variant along with multiple low-frequency virus variants. This phenomenon is known as the

quasispecies theory and allows the virus to adapt rapidly to new environments, and to new

hosts (Domingo, Biebricher et al. 2001). When RNA viruses are exposed to selective

exogenous conditions, the frequency of some virus variants in the population could change,

favoring those with a certain advantage to replicate in the new condition. The fact that

VHSV has been found in a wide range of host fish species suggest a high adaptation

capacity (Skall, Olesen et al. 2005; Schönherz, Lorenzen et al. 2015). We, therefore,

questioned whether the virus would also be able to adapt DNA-vaccinated fish.

The experimental design included two approaches. First, an in vivo approach was used to

evaluate the ability of VHSV to evade the early innate or the later adaptive immune

protection induced by the DNA vaccine in rainbow trout fingerlings. Second, an in vitro

approach was applied, to evaluate the ability of the virus to evade the neutralizing effect of

serum from rainbow trout immunized with the DNA vaccine.

The analysis of the in vivo approach took into account that an escape mutant could have

diverse strategies to bypass the immune protection induced by the DNA vaccine. Among

51

these strategies, escape mutants could have increased virulence in vaccinated fish, causing

higher mortality rates as a result of a more efficient viral replication. Alternatively,

decreased virulence might reduce clearance, thereby allowing the virus to persist in the host

and spread to cohabitant fish. However, the in vivo approach evaluation showed no

advantage of any of the passaged viruses in comparison with the parental virus, neither

considering virulence, persistence in vaccinated fish nor the capacity of vaccinated fish

carriers to infect naïve cohabitant fish.

Our results demonstrate that the virus is unlikely to escape from the protective immune

response within 4-5 passages in vaccinated fish. However, we also show that DNA

vaccinated fish can become infected and that the infection can spread to naïve co-habitants.

Therefore, as for many other veterinary vaccines not providing sterile immunity, DNA

vaccinated animals cannot be considered virus-free in terms of trade regulations. In our

setup, naïve cohabitants were stocked with vaccinated carriers already 2 weeks post

exposure of the latter to the virus. It may be anticipated that a longer time gap would have

given the, presumably few, carrier fish time to eliminate the infection and hereby have

reduced the chance of transmission. Further time-course studies are needed to address this

aspect.

According to Read et al. 2015, infected vaccinated chickens were associated with a higher

risk of spread of highly virulent virus than the infected non-vaccinated chickens. The

rationale behind this theory was that non-vaccinated chickens would die out rapidly and

thereby allow the infection to be kept under control, while the infected vaccinated chicken

would survive the infection and allow the virus to persist, replicate and spread in the host

population for an extended period. Furthermore, the persistence in vaccinated chickens

52

could promote the selection of hyperpathogenic virus strains that could cause a more severe

disease (Read, Baigent et al. 2015).

While this scenario might be true for some highly lethal viruses, our results suggest that it

does not count for VHSV infections in rainbow trout. Although the virus causes high

mortality, some individuals often survive and clearance from such fish is slower compared

DNA vaccinated individuals (Lorenzen, Einer-Jensen et al. 2000). Furthermore, despite

repeated passaging in such fish we were unable to isolate escape mutants. Repeated

stocking with vaccinated fish, therefore, seems to be a viable strategy for reducing the

prevalence of VHSV in endemic zones.

The in vitro aim was look at the humoral immune response to DNA vaccination alone and

evaluate whether serial passages of VHSV in the presence of serum from rainbow trout

immunized with the DNA vaccine, would favour propagation of neutralization escape

mutants.

Previous works showed that viruses like VHSV and IHNV were able to generate escape

mutants resistant to neutralization by monoclonal antibodies after a few passages in cell

culture (Bearzotti, Monnier et al. 1995; Huang, Chien et al. 1996). Another setup showed

that it was possible to isolate an escape mutant from rainbow trout injected with a plasmid

encoding a neutralizing recombinant single chain antibody (scAb) against the glycoprotein

of VHSV (Lorenzen, Cupit et al. 2000).

However, after 11 passages, there was no evidence of mutants escaping the neutralizing

effect of the trout immune serum. This was confirmed by the fact that the passaged and the

parental virus had a 100% identical glycoprotein gene sequence. In contrast to this, mutants

53

escaping neutralization by mouse monoclonal antibodies carried one or more amino acid

substitutions in the translated G gene (Bearzotti, Monnier et al. 1995). The explanation for

why the apparently rather adaptable virus could not escape from the neutralization by our

fish immune serum might be due to different neutralization mechanisms used by mouse

monoclonal antibodies (mIgG1) and by fish polyclonal antibodies IgM in serum.

Neutralizing monoclonal antibodies bind a single epitope and thereby most likely interfere

with the process of infection, e.g. by preventing recognition of the receptors on the cell

membrane or generating aggregates incapable of infecting host cells (Reading and

Dimmock 2007). One or a few amino acids changes could prevent antibody binding and

thereby generate a full escape mutant. In contrast, to produce an escape mutant of the

neutralizing serum antibodies the virus may have to mutate at multiple sites, assuming that

vaccination induces a polyclonal neutralizing response. Furthermore, fish IgM neutralizing

activity depends on the presence of complement, implying that the neutralization

mechanism is not just a steric blocking of the viral infectivity, but a more complex

mechanism potentially involving different sites of the protein, which could be more

difficult for VHSV to bypass without affecting the functional biology of the virus particle

(Lorenzen, Olesen et al. 1990). In summary, the results from the in vitro analysis supported

the in vivo data and suggested that VHSV cannot easily escape even from the humoral

immune response alone as induced by DNA vaccination. The trout immune serum used

here was obtained from a hyperimmunized fish, and all fish do not seroconvert following a

single vaccine injection (Lorenzen et al. 1998). However, our failure to isolate escape

mutants in vivo suggest that the broad nature of the immune response triggered by the

vaccine, involving a range of both innate and adaptive mechanisms, makes escape mutation

incompatible with maintaining the viability and the infectious capacity of the virus.

54

Our setup only included the most stringent selective condition by challenging vaccinated

fish with virus carrying a G gene identical to the vaccine gene. It can therefore not be

excluded that viral escape from DNA vaccine induced immunity might arise under less

selective conditions, such as when the vaccine G gene is heterologous to that of the

infecting virus. The safest strategy would thus be to sequence the prevalent VHSV variants

in the fish population to be vaccinated and then perform the vaccination with a homologous

or at least genetically closely related vaccine gene.

In conclusion, our results support the low probability to generate an escape mutant under

optimal DNA vaccination conditions, when a strict protective immunity is induced. The

difficulties to bypass the protection induced by the DNA vaccine are an additional

advantage of this vaccine against VHSV, making it a safe prophylactic tool. However, we

also observed that some of the vaccinated fish can get subclinically infected and that the

infection can be transmitted to naïve cohabitants if these are stocked with the vaccinated

fish shortly after their exposure to the virus. Vaccinated fish from endemically infected

zones should, therefore, be considered to be potential carriers in terms of trade regulations.

ACKNOWLEDGMENTS

The authors gratefully acknowledge the excellent assistance from the technical staff of the

Fish Health Section, Department of Animal Science, University of Aarhus.

55

REFERENCES

1. Walker P, Benmansour A, Dietzgen R, Fang R, Jackson A, et al. (2000) Virus taxonomy:

classification and nomenclature of viruses. Seventh report of the International Committee on

Taxonomy of Viruses.

2. Skall HF, Olesen NJ, Mellergaard S (2005) Viral haemorrhagic septicaemia virus in marine fish

and its implications for fish farming–a review. Journal of fish diseases 28: 509-529.

3. De Kinkelin P, Bearzotti M, Castric J, Nougayrede P, Lecocq-Xhonneux F, et al. (1995) Eighteen

years of vaccination against viral haemorrhagic septicaemia in France. Veterinary research 26: 379-

387.

4. Lorenzen N, Olesen N (1996) Immunization with viral antigens: viral haemorrhagic septicaemia.

Developments in biological standardization 90: 201-209.

5. Heppell J, Lorenzen N, Armstrong NK, Wu T, Lorenzen E, et al. (1998) Development of DNA

vaccines for fish: vector design, intramuscular injection and antigen expression using viral

haemorrhagic septicaemia virus genes as model. Fish & Shellfish Immunology 8: 271-286.

6. Anderson E, Mourich D, Fahrenkrug S, LaPatra S, Shepherd J, et al. (1996) Genetic immunization

of rainbow trout (Oncorhynchus mykiss) against infectious hematopoietic necrosis virus. Molecular

Marine Biology and Biotechnology 5: 114-122.

7. Salonius K, Simard N, Harland R, Ulmer JB (2007) The road to licensure of a DNA vaccine.

Current Opinion in Investigational Drugs 8: 635.

8. Sommerset I, Lorenzen E, Lorenzen N, Bleie H, Nerland AH (2003) A DNA vaccine directed

against a rainbow trout rhabdovirus induces early protection against a nodavirus challenge in

turbot. Vaccine 21: 4661-4667.

56

9. Lorenzen N, Lorenzen E, Einer-Jensen K, LaPatra SE (2002) Immunity induced shortly after DNA

vaccination of rainbow trout against rhabdoviruses protects against heterologous virus but not

against bacterial pathogens. Developmental & Comparative Immunology 26: 173-179.

10. Utke K, Kock H, Schuetze H, Bergmann SM, Lorenzen N, et al. (2008) Cell-mediated immune

responses in rainbow trout after DNA immunization against the viral hemorrhagic septicemia virus.

Developmental & Comparative Immunology 32: 239-252.

11. Lorenzen N, Lorenzen E, Einer-Jensen K, Heppell J, Davis H (1999) Genetic vaccination of

rainbow trout against viral haemorrhagic septicaemia virus: small amounts of plasmid DNA protect

against a heterologous serotype. Virus research 63: 19-25.

12. 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.

13. Schönherz AA, Lorenzen N, Guldbrandtsen B, Buitenhuis B, Einer-Jensen K (2015) Ultra-deep

sequencing of VHSV isolates contributes to understanding the role of viral quasispecies. Veterinary

research.

14. Lorenzen N, Cupit PM, Einer-Jensen K, Lorenzen E, Ahrens P, et al. (2000) Immunoprophylaxis

in fish by injection of mouse antibody genes. Nature biotechnology 18: 1177-1180.

15. Bearzotti M, Monnier A, Vende P, Grosclaude J, De Kinkelin P, et al. (1995) The glycoprotein of

viral hemorrhagic septicemia virus (VHSV): antigenicity and role in virulence. Veterinary research

26: 413-422.

16. Garver KA, LaPatra SE, Kurath G (2005) Efficacy of an infectious hematopoietic necrosis (IHN)

virus DNA vaccine in Chinook Oncorhynchus tshawytscha and sockeye O. nerka salmon. Diseases

of Aquatic organisms 64: 13-22.

57

17. Ammayappan A, Vakharia VN (2011) Nonvirion protein of novirhabdovirus suppresses

apoptosis at the early stage of virus infection. Journal of virology 85: 8393-8402.

18. Fijan N, Sulimanović D, Bearzotti M, Muzinić D, Zwillenberg L, et al. Some properties of the

epithelioma papulosum cyprini (EPC) cell line from carp Cyprinus carpio; 1983. Elsevier. pp. 207-

220.

19. Wolf K, Gravell M, Malsberger RG (1966) Lymphocystis virus: isolation and propagation in

centrarchid fish cell lines. Science 151: 1004-1005.

20. Reed LJ, Muench H (1938) A simple method of estimating fifty per cent endpoints. American

journal of epidemiology 27: 493-497.

21. Olesen NJ, Jørgensen P (1986) Detection of neutralizing antibody to Egtved virus in rainbow

trout (Salmo gairdneri) by plaque neutralization test with complement addition. Journal of Applied

Ichthyology 2: 33-41.

22. Domingo E, Biebricher CK, Eigen M, Holland JJ (2001) Quasispecies and RNA virus evolution:

principles and consequences: Landes Bioscience Austin.

23. Read AF, Baigent SJ, Powers C, Kgosana LB, Blackwell L, et al. (2015) Imperfect Vaccination Can

Enhance the Transmission of Highly Virulent Pathogens. PLoS Biol 13: e1002198.

24. 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.

Journal of Aquatic Animal Health 12: 167-180.

25. Huang C, Chien M-S, Landolt M, Batts W, Winton J (1996) Mapping the neutralizing epitopes on

the glycoprotein of infectious haematopoietic necrosis virus, a fish rhabdovirus. Journal of General

Virology 77: 3033-3040.

26. Reading S, Dimmock N (2007) Neutralization of animal virus infectivity by antibody. Archives of

virology 152: 1047-1059.

58

27. Lorenzen N, Olesen NJ, Jørgensen PV (1990) Neutralization of Egtved virus pathogenicity to cell

cultures and fish by monoclonal antibodies to the viral G protein. Journal of General Virology 71:

561-567.

59

Figure 1. Vaccination and challenges for the in vivo approach.

60

Figure 2. Scheme of the passaging of VHSV in vaccinated fish for the first two passages.

Passage 3 and 4 followed the procedure of passage 2.

61

Figure 3. Comparison of the total mortality of rainbow trout challenged with the parental

virus, and the passaged virus. The challenges were performed at 1 week post vaccination

(A) and at 6 weeks post vaccination (B). White bars correspond to non-vaccinated fish and

black bars correspond to fish vaccinated with the DNA vaccine.

62

Table 1. Passaged virus

Virus

Label

Status of fish used for passaging

Time post vaccination Vaccine dose

VHSV-1W*0,1** 1 week 0,1 µg

VHSV-1W1,0 1 week 1,0 µg

VHSV-6W0,1 6 weeks 0,1 µg

VHSV-6W1,0 6 weeks 1,0 µg

* Refers to number of weeks between the vaccination and challenge

** Refers to vaccine dose (µg)

63

Table 2. Comparison of the carrier status of vaccinated rainbow trout challenged with the

parental virus or with the virus passaged in vaccinated fish.

Challenge

time

Virus

Replicate

Aquaria

N°

Vaccinated fish

positive in virological

examination (%)

Cohabitant naïve fish

positive in virological

examination (%)

1 wpv*

DK3592b

1 36,4 100

2 5,6 0

VHSV-1w 0,1

1 5,6 0

2 13,6 0

VHSV-1w 1,0

1 0 0

2 5,0 100

6 wpv

DK3592b

1 0 0

2 15,8 80

VHSV-6w 0,1

1 10,5 60

2 5 60

VHSV-6w 1,0

1 0 0

2 10,0 0

* wpv: Weeks post vaccination

64

Table 3. Neutralizing titers of trout immune serum for in vitro passaged and parental virus.

Parental Virus

DK3592b

Passaged

Virus1

Passaged

Virus2

Passaged

Virus3

IS CS IS CS IS CS IS CS

50% PNT

titer

5120 <40 20480 <40 5120 <40 10240 <40

IS= Immune serum collected from rainbow trout repeatedly immunized with the DNA

vaccine

CS= Control serum collected from non-vaccinated rainbow trout

65

4.2 Manuscript II

Time-course study of the immune protection induced by an

interferon-inducible DNA vaccine against viral haemorrhagic

septicaemia virus in rainbow trout

Dagoberto Sepúlveda, Ellen Lorenzen, Jesper Skou Rasmussen, Katja

Einer-Jensen, Bertrand Collet, Chris Secombes, Niels Lorenzen

Intended for submission to Fish and Shell Fish Immunology

66

Time-course study of the immune protection induced by an interferon-

inducible DNA vaccine against viral haemorrhagic septicaemia virus in

rainbow trout

Dagoberto Sepúlvedaa, Ellen Lorenzena, Jesper Skou Rasmussena, Katja Einer-Jensenb,

Bertrand Colletc, Chris Secombesd, Niels Lorenzena

a Department of Animal Science, Aarhus University, Denmark

b Qiagen, Aarhus, Denmark

c Marine Scotland, Marine Laboratory, Aberdeen, United Kingdom

d Aberdeen University, Aberdeen, United Kingdom

67

ABSTRACT

Attempts to improve the efficacy and safety of DNA vaccines against viral diseases in

aquacultured fish have focused on elements of the plasmid vector such as the

regulatory elements driving the expression of vaccine antigen. Ideally, the expression

of the antigen should be controlled by a fish-derived promoter, with minimal or no

activity in human cells.

In the work presented here, we compared a DNA vaccine with the interferon-inducible

Mx promoter from rainbow trout, and a DNA vaccine with a cytomegalovirus promoter

(CMV), both encoding the viral haemorrhagic septicaemia virus (VHSV) glycoprotein G.

The in vitro analysis showed that while the DNA vaccine with the CMV promoter

constitutively induced expression of the G protein in both fish and human cell lines, the

DNA vaccine with the Mx promoter inducibly promoted expression of the glycoprotein

only in the fish cell lines.

To address the three-phase protection model suggested by Kurath et al. (2006) –

comprising early, specific and long-term (EAVR, SAVR, LAVR) phases - infection trials

were performed with vaccinated fish at 2, 8, and 78 weeks post vaccination (wpv),

respectively. The DNA vaccine with CMV promoter provided protection at all times,

while vaccination with the DNA vaccine with the Mx promoter only protected the fish at

8 wpv. However, following induction with polyI:C in vaccinated fish one week before

challenge, protection was also evident at the early challenge 2wpv.,

The results revealed a superior consumer safety of the trout Mx promoter compared to

the traditional CMV promoter in the context of DNA vaccination of farmed rainbow

trout. However, improvements will be needed in terms of time course efficacy.

Importantly, the data further suggest a lack of direct interdependency of protection in

the different phases of the immune response to the rhabdovirus G gene DNA vaccines.

68

INTRODUCTION

Infectious diseases are one of the major causes of economic losses in the expanding

aquaculture industry. Application of effective vaccines is fundamental to keep the

industry sustainable. Viral haemorrhagic septicaemia virus (VHSV), a member of the

Rhabdoviridae family (Walker, Benmansour et al. 2000), causes a serious disease in

farmed fish, such as rainbow trout and turbot in Europe, and Japanese flounder in East

Asia (Skall, Olesen et al. 2005). Several vaccination strategies have been tested

against VHSV, including inactivated virus vaccines, live-attenuated virus vaccines, and

recombinant vaccines. However, due to low efficacy, high cost per dose, or safety

limitations, there is currently no commercially available vaccine against VHSV (De

Kinkelin, Bearzotti et al. 1995; Lorenzen and Olesen 1996). In 1998, the testing a DNA

vaccine as an alternative immunization strategy against VHSV revealed establishment

of highly protective immunity (Heppell, Lorenzen et al. 1998). Since then, several

studies have evaluated the protection induced by DNA vaccination against VHSV and

the related infectious haematopoietic necrosis virus (IHNV) under different

experimental conditions, with promising results (Lorenzen and LaPatra 2005; Kurath,

Purcell et al. 2007). These DNA vaccines consisted of a eukaryotic expression vector

encoding the viral surface glycoprotein G, under the control of a cytomegalovirus

promoter (CMV). The immunological protection follows a three-phase scenario (Kurath,

Purcell et al. 2007), which involves:

(i) The early antiviral responses (EAVR), which are a cross-reactive protection

associated with innate antiviral mechanisms. This protective phase starts

shortly after the intramuscular injection of the DNA vaccine, and is

characterized by the overexpression of interferon type I (IFN I), and

consequently the expression of multiple interferon-stimulated genes (ISG)

69

such as: Mx, Vig-1, Vig-8 (Boudinot, Blanco et al. 1998; Acosta, Petrie et

al. 2005; Purcell, Nichols et al. 2006).

(ii) The specific antiviral responses (SAVR), which are associated with adaptive

immunity mechanisms and characterized by the presence of neutralizing

antibodies, and cytotoxic T lymphocytes (CTL) and natural killer (NK)-like

cells (Lorenzen, Einer-Jensen et al. 2000; Kurath, Garver et al. 2006; Utke,

Kock et al. 2008).

(iii) The long-term antiviral responses (LAVR), which follow the SAVR, are

characterized by a slightly lower protection than in SAVR and minimal

detection or absence of neutralizing antibodies (Kurath, Garver et al. 2006;

Kurath, Purcell et al. 2007).

The CMV promoter is the most common promoter used in expression vectors for DNA

vaccines, due to its high expression activity in a broad range of eukaryotic cells.

However, the safety concern of the potential recombination of a CMV promoter

sequence in a DNA vaccine for aquaculture has driven the search of an alternative fish-

derived promoter. Some fish-derived promoters analysed in vivo in a DNA vaccine

were the Interferon regulatory factor 1A (IRF1A) promoter, Mx1 promoter (Alonso,

Johnson et al. 2003), and the carp β-actin (AE6) promoter (Chico, Ortega-Villaizan et

al. 2009).

Some of the examined alternative promoters have shown potential in terms of

protection induced by the related DNA vaccines. However, the reports published so far

have not taken the time-course variation of the protective mechanisms into account.

This is important from a practical point of view, where the ability of the vaccine to

induce a fast, efficient, and long-lasting protection, is essential.

The aim of this work was to analyze the capacity of a VHSV glycoprotein-gene DNA

vaccine with an IFN-inducible trout-derived Mx promoter to induce EAVR, SAVR and

70

LAVR in rainbow trout fingerlings. Moreover, in terms of consumer safety, a promoter

showing minimal or no expression in human cells would be preferable. Therefore, we

also compared expression of the vaccine protein in transfected cell lines derived from

fish and humans respectively.

MATERIALS AND METHODS

Cell lines. EPC cells (epithelioma papulosum cyprinid) (Fijan, Sulimanović et al. 1983),

BF2 cells (bluegill fry fibroblast) (Wolf, Gravell et al. 1966) and HeLa cells (ATCC: CCL-

2) were used in this work. The cells were maintained in minimum essential media

(MEM) supplemented with 10% fetal bovine serum (FBS), 100 U/mL of Penicillin and

100 µg/mL of Streptomycin. EPC and BF2 were maintained at 15°C, and HeLa were

maintained at 37°C.

Virus. A low passaged VHSV isolate (DK3592b) was propagated in BF2 cells by

inoculating freshly passaged cells with a low MOI (multiplicity of infection). The

infected cell cultures were maintained at 15°C. When a complete cytopathic effect was

observed, the supernatant was collected and centrifuged at 5000 r.p.m for 15 min at

4°C. The titer of the virus was determined using the method of 50% tissue culture

infective doses (TCID50) per mL in BF2 cells (Reed and Muench 1938).

Plasmid constructs. An expression vector with Mx promoter (pcDNA3-Mx) was

constructed by replacing the CMV promoter of pcDNA3 (pcDNA3-CMV) (Invitrogen)

with the Mx1 promoter from pGL3-Basic-PrMx1 (Collet and Secombes 2001). The Mx1

promoter was excised from pGL3-Basic-PrMx1 as a 600 bp MluI-BglII fragment and

inserted into pcDNA3 digested with MluI and BamHI. The IFN-regulated expression

vector encoding the VHSV glycoprotein (pMx-vhsG) was obtained by excising the VHSV

71

glycoprotein gene from pcDNA3-vhsG (Lorenzen, Lorenzen et al. 1998) as a 1576 bp

EcoRI-EcoRI fragment and ligating it into pcDNA3-Mx digested with EcoRI (Figure 1).

The glycoprotein gene was from originally derived from VHSV isolate DK3592b. The

empty vectors with CMV and Mx promoters were used as a negative control.

Escherichia coli transformed with each plasmid were propagated overnight in 4 L of LB

medium supplemented with ampicillin. Endofree plasmid purification kit (Gigaprep kit

Qiagen) were used for the further purification of the DNA constructs. The vaccination

trial included the constructs described in Table 1.

Figure 1. Construction of the DNA vaccine with the IFN-inducible Mx promoter.

72

Table 1: Description of the plasmids used for DNA vaccination

Plasmids Promoter Transgene

pCMV-Empty CMV without transgene

pCMV-vhsG CMV VHSV glycoprotein

pMx-Empty IFN-inducible Mx promoter without transgene

pMx-vhsG IFN-inducible Mx promoter VHSV glycoprotein

Expression of the VHSV glycoprotein in cell culture

EPC and HeLa cells were seeded in a 24-well plate (2x105 cell/well) one day before

transfection. The transfection was performed with the plasmids, pCMV-Empty, pCMV-

vhsG, and pMx-vhsG using Superfect (Qiagen) as described by the manufacturer. The

amount of plasmids used for the transfection was 4 µg/mL for EPC cells and 9 µg/mL

for HeLa cells. Twenty four hrs post-transfection, the medium was replaced by medium

with and without poly I:C (Sigma), a potent inducer of IFN, in a final concentration of

10 µg/mL.

Detection of the VHSV glycoprotein

The transfected cells were fixed with 72 hrs post transfection in cold acetone 80%. The

staining was carried out according to the protocol used in (Lorenzen, Cupit et al.

2000). Three mouse anti-VHSV glycoprotein monoclonal antibodies were used in this

work as primary antibody: Mab IP1H3 that recognizes a linear epitope, the 3F1A2 that

recognize a conformation-dependent epitope (Lorenzen, Olesen et al. 1990; Lorenzen,

Cupit et al. 2000). The secondary antibody was HRP-conjugated rabbit anti-mouse IgG

(DAKO P0260).

73

First vaccination trial: Evaluation of EAVR and SAVR

Outbreed, all female rainbow trout between 3 - 5 g were used for the vaccination. The

fish were anesthetized in 0,01% benzocaine and then injected intramuscularly with

1µg of each of the 4 purified DNA plasmids (Table 1) in 25 µL of saline solution (0,9%

NaCl). At one week post-vaccination, the fish vaccinated with pMx-Empty and with

pMx-vhsG were split into two subgroups. One subgroup was injected intraperitoneally

(I.P) with 3,5 µg of poly I:C (Sigma) in 50 µL of saline solution, and the second

subgroup was injected with 50 µL of saline solution. After vaccination, the fish were

maintained in 120 L aerated aquaria supplied with recirculated water at 8-10°C in a

pathogen-free laboratory facility.

The challenge with VHSV isolate DK3592b was performed at two times, at 2 weeks and

12 weeks post-vaccination, respectively, in aerated 8 L aquaria with 3 replicates of 24-

26 fish per treatment. The challenge was carried out by immersion in static water with

an infectious dose of 1x105 TCID50 mL-1 of water. After 2 hrs, the water flow was re-

established. During the following 3 weeks, fish with evident clinical signs were

euthanized with an overdose of benzocaine.

Second vaccination trial: Evaluation of the LAVR

In this trial, we focused on the long-term antiviral responses (LAVR). Outbreed all

female rainbow trout (2 - 5g) were divided into 5 groups, 4 of them injected

intramuscularly with the plasmids in 25µL of saline solution, and one group injected

with a saline solution (0.9% NaCl). The plasmid used are described in Table 1. Each

group contained 160 fish, which were maintained in 120 L aerated aquaria supplied

with recirculated water at 8-10°C in a pathogen-free laboratory facility. At 68 weeks

post-vaccination, the groups of fish previously injected with saline solution (0.9%

NaCl), pMx-Empty, or pMx-vhsG were split into two subgroups, one was injected with

74

saline solution (0.9% NaCl) and the another was injected I.P with 100µg of poly I:C

(Sigma) diluted in 100 µl of saline solution per fish. The fish injected with pCMV-

Empty, and pCMV-vhsG were injected only with the saline solution. At this time, the

weight of the fish was 40-70 g each.

The challenge was performed at 78 weeks post vaccination (10 weeks post-induction

with poly I:C). The infection was carried out like in the previous section. The virus used

was VHSV isolate DK3592b with a titer of 3x104TCID50 mL-1in an 8L aquaria. The

challenged was carried out with 3 replicates of 20 fish per treatment.

The mortality of the challenges was recorded daily for 30 days and the relative

percentage survival (RPS) was calculated: RPS=[1-(% mortality of immunized fish / %

mortality of control fish)]x100.

Figure 1. Scheme of the evaluations of the three phases of the protective immunity

induced by the DNA vaccines against VHSV. (A) EAVR, (B) SAVR, (C) LAVR.

75

RESULTS

Expression analysis in vitro. The EPC cell transfected with pMx-vhsG expressed the

glycoprotein following induction with poly I:C, while no expression was detected in

transfected a human HeLa cell. On the contrary, both EPC and HeLa cell transfected

with the pCMV-vhsG expressed the viral glycoprotein in both human and fish cell lines

(Table 2). In EPC cell, the immunostaining of the glycoprotein was positive using both

MAbs, while in HeLa cells the glycoprotein could be detected only with the MAb IP1H3.

This suggested that the glycoprotein was incorrectly folded.

Table 2. Expression of the VHSV glycoprotein in fish and human cell culture

EPC cells HeLa cells

IP1H3 3F1A2 IP1H3 3F1A2

Poly I:C

10 µg/mL + - + - + - + -

pCMV-Empty -* - - - - - - -

pCMV-vhsG ++ ++ ++ ++ + + - -

pMx-vhsG + - + - - - - -

* No expression detected: -, intermediated expression: +, high expression:++

First vaccination trial: Evaluation EAVR and SAVR

When the challenge was performed at 2 wpv, the fish immunized with the DNA vaccine

with the IFN-inducible Mx promoter (pMx-vhsG), induced protection only in fish

injected with poly I:C. This protection was not due to an antiviral mechanism induced

poly I:C, but an effect of the expression of the glycoprotein since fish injected with the

control plasmid pMx-Empty and with poly I:C were not protected. When this DNA

vaccine was evaluated at 8 wpv, protection was evident independently of the poly I:C

stimulation. The reference DNA vaccine with CMV promoter induced high protection

without stimulation with poly I:C, at both 2 and 8 wpv, respectively. The plasmids

without the G gene did not induce any protection against VHSV (Table 3).

76

Table 3: Early and specific antiviral responses induced by the DNA vaccines against

VHSV

Groups Induction

1wpv

Challenge

2wpv

(RPS)

Challenge

8wpv

(RPS)

pCMV-vhsG NI* 80,7 94,9

pMx-Empty Poly I:C 0 0

pMx-Empty 0.9% NaCl 0 3,1

pMx-vhsG Poly I:C 79,6 77,6

pMx-vhsG 0.9% NaCl 33,7 79,6

*NI: No induction

Second vaccination trial: Evaluation of the LAVR

Vaccination with the DNA vaccine with the CMV promoter provided good protection

against VHSV challenge at 78 weeks post-vaccination. However, the fish immunized

with the DNA vaccine with Mx promoter did not show any protection. Table 4 shows

the average cumulative mortality in the three replicates of each group. The fish

immunized with the pMx-vhsG stimulated with poly I:C showed a lower cumulative

mortality than the control, it was due to only one of the three replicates, which might

be due to an inter-replicate variability and not to a real protective effect (Table 4).

77

Table 4. Long-term antiviral responses induced by the DNA vaccines against VHSV

Groups Induction

68wpv

Challenge 78wpv*

RPS

Saline 0.9% NaCl 8,3 (6,3; 18,7; 0)

Saline Poly I:C 14 (12,4; 23,3; 6,3)

pCMV-Empty 0.9% NaCl 3,9 (5,5; 6,3; 0)

pCMV-vhsG 0.9% NaCl 81,3 (87,5; 62,5; 93,8)

pMx-Empty Poly I:C 14,6 (25,0; 18,7; 0)

pMx-Empty 0.9% NaCl 17,7 (46,9; 0; 6,3)

pMx-vhsG Poly I:C 30,6 (23,3; 49.9; 18,7)

pMx-vhsG 0.9% NaCl 16,6 (0; 0; 49,9)

* RPS values are the average of the three replicates showed in the table.

DISCUSSION

In this study, we performed a functional characterization of a DNA vaccine with an IFN-

inducible Mx promoter derived from rainbow trout. This characterization involved, first,

an analysis of the expression activity in both a human and a fish cell line, and second,

an evaluation of the protection during the three immune response phases, classified

according to the timing and nature of the presumed protective mechanisms and named

EAVR, SAVR and LAVR (Kurath 2006, Kurath and Purcell 2007).

Our in vitro results showed that the DNA vaccine with the IFN-inducible Mx promoter

from rainbow trout was able to promote expression of the VHSV glycoprotein in fish

cells, but not in a human cell line, while the vaccine with the CMV promoter induced

78

expression of the glycoprotein in both fish and human cells. The Mx promoter driven

expression in the EPC cells was dependent on IFN as induced by poly I:C treatment.

Similar results were obtained with a more sensitive luciferase reporter gene setup (not

shown). Taking into account that the intracellular pathway for activating expression of

IFN-inducible elements is highly conserved between teleost and higher vertebrates

(Robertsen 2008; Collet 2014), it seems unlikely that lack of functionality of the trout

Mx promoter in the HeLa cells should be due to a species-specific transcription factors.

However, since the fish EPC cells were grown at 18°C and HeLa cells at 37°, one

possibility could be that temperature dependent conformation of the Mx promoter

region interfered with activation of transcription in the latter. This needs further

analyses to be determined, but lack of activity of the trout Mx promoter in human cells

makes it attractive for use in farmed fish, since this would eliminate concerns about

potential side effects of transgene expression in consumers eating vaccinated fish.

The efficacy of a DNA vaccine depends on several factors. These include the

immunogenicity of the antigen, the degradation rate of the DNA vaccine in the host,

the expression level of the transgene, and the quality of the host immune responses

(Hølvold, Myhr et al. 2014). Since the initial reports of the high efficacy of the CMV

promoter-based DNA vaccines against fish rhabdoviruses (Anderson, Mourich et al.

1996; Lorenzen, Lorenzen et al. 1998), considerable efforts have been made to

identify alternative promoters, not derived from a human pathogenic virus (Alonso,

Johnson et al. 2003; Chico, Ortega-Villaizan et al. 2009; Martinez-Lopez, Chinchilla et

al. 2013). However, these studies have either been based on quantitative expression

analyses in vitro in transfected cell cultures, or included only a single test point in vivo

in terms challenge post-vaccination. Our results demonstrate, that in order to evaluate

the practical application of a DNA vaccine, it is necessary to examine protection against

disease in vivo in all three phases of the immune response to the vaccine.

79

Back in 2003, Alonso et al. reported rather low protection (RPS=16) against IHN in 0.4

g rainbow trout fry following DNA vaccination with a trout Mx promoter IHNV–G gene

construct resembling that used for VHSV-G in the current study. However, despite the

fact that poly I:C was shown to upregulate expression by a factor 7 in vitro, the effect

was not analyzed in vivo. Since the challenge was performed at one month post-

vaccination with fish kept at 13°C, the authors could not exclude that the observed

protection with the CMV promoter reference construct might be due to innate

mechanisms, i.e. the EARV. Accordingly, we here observed protection at the presumed

time of EARV only when pMx-vhsG vaccinated fish were given poly I:C one week prior

to challenge.

Thus, while the Mx promoter has been observed to have a certain base-line activity,

even without stimulation (Alonso, Johnson et al. 2003; Collet, Boudinot et al. 2004;

Martinez-Lopez, Chinchilla et al. 2013), this appeared insufficient to activate a

protective innate response. In contrast, the high protection against VHSV challenge at

8 wpv found in the current study in fish given the pMx-vhsG, even without poly I:C

stimulation, suggested that the base-line activity of the Mx promoter was sufficient to

trigger a protective SAVR. Since the VHSV glycoprotein has been shown to be able to

induce up-regulation of IFN by itself in transfected cell cultures (McLauchlan, Collet et

al. 2003; Acosta, Petrie et al. 2005) an autocrine stimulation might also have been

involved. At the virus challenge performed 78 wpv the response induced by pMX-vhsG

failed to protect the fish, suggesting an insufficient activation of the LAVR as compared

to the high protection among fish given pCMV-vhsG. While it is general knowledge that

IFN and related innate antiviral immune mechanisms are important not only for

protection at the early stage of infection, but also for paving the way for an efficient

adaptive response (Coffman, Sher et al. 2010), our results suggest a lack of direct

interdependency of establishment of protective immunity during the EAVR, SAVR and

80

LAVR phases. It may thus be anticipated that there are qualitative and/or quantitative

differences between the protective -, and the SAVR/LAVR -promoting mechanism of

the EAVR. Similarly, there may be distinct requirements to reach a protective LAVR on

top of, or beside, those needed for a protective SARV. Further experiments including

examination of the specificity of the protection induced by the pMx-vhsG vaccine at

8wpv along with immune gene expression analyses are needed to further resolve this

aspect. Interestingly, Chang et. al. in a recent report showed that while one type of

IFN I was able to provide protection against infectious salmon anaemia virus (ISAV) in

Atlantic salmon, other types of IFN promoted generation of a protective immune

response to a co-injected antigen (Chang, Sun et al. 2015).

In applied terms, our initial hypothesis was that without poly I:C induction shortly after

vaccination of fish with pMx-vhsG, it would be possible to delay the elimination of cells

harboring the plasmid by the local inflammatory response described in fish vaccinated

with pCMV-vhsG (Lorenzen, Lorenzen et al. 2005). This could then allow a later

dynamic management of the fish immune status by activation of the antigen

expression by stimulation of an IFN response, e.g. by feeding the fish with

immunostimulants before moving them to an endemically infected environment. This

strategy clearly failed, possibly because the induced base-line expression of the

antigen by the Mx promoter was sufficient to activate cytotoxic immune reactions to

eliminate transfected cells. However, taking the very long lasting protection reported

here for the pCMV-vhsG vaccinated fish (78 wpv), and the typical 2-3 year lifespan for

cultured rainbow trout into account, there is limited need for improved duration of

immunity. With poly I:C administration shortly after pMx-vhsG vaccination, protective

EAVR and SAVR comparable to those obtained with pCMV-vhsG were reached. Future

experiments should therefore address whether this also counts for LAVR to evaluate

the applied potential an Mx promoter DNA vaccine construct. Finally, it might be

81

possible to combine the attractive safety aspect of the trout Mx promoter with the high

expression capacity of other promoters by making hybrid promoters as recently

reported by Martinez-Lopez et al (2013) (Martinez-Lopez, Chinchilla et al. 2013).

Acknowledgements

This study was supported by Chilean National Scholarship Program for Graduate

Studies Conicyt for DS, and by the European Commission contract FP7311993

TargetFish.

The authors gratefully acknowledge the excellent assistance from the technical staff of

the Fish Health Section, Department of Animal Science, University of Aarhus.

REFERENCES

Acosta, F., A. Petrie, et al. (2005). "Kinetics of Mx expression in rainbow trout (Oncorhynchus mykiss) and Atlantic salmon (Salmo salar L.) parr in response to VHS–DNA vaccination." Fish & shellfish immunology 18(1): 81-89.

Alonso, M., M. Johnson, et al. (2003). "A fish specific expression vector containing the interferon regulatory factor 1A (IRF1A) promoter for genetic immunization of fish." Vaccine 21(15): 1591-1600.

Anderson, E., D. Mourich, et al. (1996). "Genetic immunization of rainbow trout (Oncorhynchus mykiss) against infectious hematopoietic necrosis virus." Molecular Marine Biology and Biotechnology 5(2): 114-122.

Boudinot, P., M. Blanco, et al. (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(2): 297-306.

Chang, C.-J., B. Sun, et al. (2015). "Adjuvant activity of fish type I interferon shown in a virus DNA vaccination model." Vaccine 33(21): 2442-2448.

82

Chico, V., M. Ortega-Villaizan, et al. (2009). "The immunogenicity of viral haemorragic septicaemia rhabdovirus (VHSV) DNA vaccines can depend on plasmid regulatory sequences." Vaccine 27(13): 1938-1948.

Coffman, R. L., A. Sher, et al. (2010). "Vaccine adjuvants: putting innate immunity to work." Immunity 33(4): 492-503.

Collet, B. (2014). "Innate immune responses of salmonid fish to viral infections." Developmental & Comparative Immunology 43(2): 160-173.

Collet, B., P. Boudinot, et al. (2004). "An Mx1 promoter–reporter system to study interferon pathways in rainbow trout." Developmental & Comparative Immunology 28(7): 793-801.

Collet, B. and C. J. Secombes (2001). "The rainbow trout (Oncorhynchus mykiss) Mx1 promoter." European Journal of Biochemistry 268(6): 1577-1584.

De Kinkelin, P., M. Bearzotti, et al. (1995). "Eighteen years of vaccination against viral haemorrhagic septicaemia in France." Veterinary research 26(5-6): 379-387.

Fijan, N., D. Sulimanović, et al. (1983). Some properties of the epithelioma papulosum cyprini (EPC) cell line from carp Cyprinus carpio. Annales de l'Institut Pasteur/Virologie, Elsevier.

Heppell, J., N. Lorenzen, et al. (1998). "Development of DNA vaccines for fish: vector design, intramuscular injection and antigen expression using viral haemorrhagic septicaemia virus genes as model." Fish & Shellfish Immunology 8(4): 271-286.

Hølvold, L. B., A. I. Myhr, et al. (2014). "Strategies and hurdles using DNA vaccines to fish." Vet Res 45: 21.

Kurath, G., K. A. Garver, et al. (2006). "Protective immunity and lack of histopathological damage two years after DNA vaccination against infectious hematopoietic necrosis virus in trout." Vaccine 24(3): 345-354.

Kurath, G., M. Purcell, et al. (2007). "Fish rhabdovirus models for understanding host response to DNA vaccines." CAB reviews 2(1): 1-12.

Lorenzen, E., K. Einer-Jensen, et al. (2000). "DNA vaccination of rainbow trout against viral hemorrhagic septicemia virus: a dose–response and time–course study." Journal of Aquatic Animal Health 12(3): 167-180.

83

Lorenzen, E., N. Lorenzen, et al. (2005). "Time course study of in situ expression of antigens following DNA-vaccination against VHS in rainbow trout (< i> Oncorhynchus mykiss</i> Walbaum) fry." Fish & Shellfish Immunology 19(1): 27-41.

Lorenzen, N., P. Cupit, et al. (2000). "Three monoclonal antibodies to the VHS virus glycoprotein: comparison of reactivity in relation to differences in immunoglobulin variable domain gene sequences." Fish & Shellfish Immunology 10(2): 129-142.

Lorenzen, N., P. M. Cupit, et al. (2000). "Immunoprophylaxis in fish by injection of mouse antibody genes." Nature biotechnology 18(11): 1177-1180.

Lorenzen, N. and S. LaPatra (2005). "DNA vaccines for aquacultured fish." Revue Scientifique Et Technique-Office International Des Epizooties 24(1): 201.

Lorenzen, N., E. Lorenzen, et al. (1998). "Protective immunity to VHS in rainbow trout (Oncorhynchus mykiss, Walbaum) following DNA vaccination." Fish & Shellfish Immunology 8(4): 261-270.

Lorenzen, N. and N. Olesen (1996). "Immunization with viral antigens: viral haemorrhagic septicaemia." Developments in biological standardization 90: 201-209.

Lorenzen, N., N. J. Olesen, et al. (1990). "Neutralization of Egtved virus pathogenicity to cell cultures and fish by monoclonal antibodies to the viral G protein." Journal of General Virology 71(3): 561-567.

Martinez-Lopez, A., B. Chinchilla, et al. (2012). "Replacement of the human cytomegalovirus promoter with fish enhancer and core elements to control the expression of the G gene of viral haemorrhagic septicemia virus (VHSV)." Journal of biotechnology 164(2): 171-178.

McLauchlan, P., B. Collet, et al. (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 Immunology 15(1): 39-50.

Purcell, M. K., K. M. Nichols, et al. (2006). "Comprehensive gene expression profiling following DNA vaccination of rainbow trout against infectious hematopoietic necrosis virus." Molecular immunology 43(13): 2089-2106.

Reed, L. J. and H. Muench (1938). "A simple method of estimating fifty per cent endpoints." American journal of epidemiology 27(3): 493-497.

84

Robertsen, B. (2008). "Expression of interferon and interferon-induced genes in salmonids in response to virus infection, interferon-inducing compounds and vaccination." Fish & shellfish immunology 25(4): 351-357.

Skall, H. F., N. J. Olesen, et al. (2005). "Viral haemorrhagic septicaemia virus in marine fish and its implications for fish farming–a review." Journal of fish diseases 28(9): 509-529.

Utke, K., H. Kock, et al. (2008). "Cell-mediated immune responses in rainbow trout after DNA immunization against the viral hemorrhagic septicemia virus." Developmental & Comparative Immunology 32(3): 239-252.

Walker, P., A. Benmansour, et al. (2000). "Family rhabdoviridae." Virus taxonomy: 563-583.

Wolf, K., M. Gravell, et al. (1966). "Lymphocystis virus: isolation and propagation in centrarchid fish cell lines." Science 151(3713): 1004-1005.

85

4.3 Manuscript III

Attempt to mimicking antibody-antigen complexes by DNA

vaccination in a fish virus model

Dagoberto Sepúlveda, Jesper Skou Rasmussen, David Parra, Niels

Lorenzen

Intended for submission to Fish and Shell Fish Immunology

86

Attempt to mimicking antibody-antigen complexes by DNA vaccination in a

fish virus model

Dagoberto Sepúlvedaa, Jesper Skou Rasmussena, David Parrab, Niels Lorenzena

a Department of Animal Science, Aarhus University, Denmark.

b Department of Cellular Biology, Physiology and Immunology, Autonomous University

of Barcelona, Spain.

ABSTRACT

DNA vaccines have shown contradictory results inducing protection against fish viral

diseases. While the DNA vaccines against fish rhabdovirus have proved to be highly

protective under different experimental conditions, DNA vaccines against other fish

viral diseases have not shown the same promising results. The low immunogenicity of

the antigen encoded in theses vaccines might be a factor in the induction of low or no

protective immune responses in the host.

Studies in mammals have shown that antigens genetically linked to the

immunoglobulin Fc domain of IgG elicited a superior immune response compared to

the antigen without the Fc domain presumably by promoting an efficient uptake

mediated by Fc-receptors on immune cells. Here, we aimed at applying this approach

to DNA vaccination of fish in order to develop a generic strategy for enhancing the

immunogenicity of the expressed antigen. A low protective DNA vaccine encoding the

secreted form of the VHSV glycoprotein (Gs) was used as a model.

87

While cell cultures transfected with the DNA vaccine plasmids encoding Gs secreted the

recombinant protein into the supernatant, the Fc-fusion proteins could only be

detected in the cell fraction of such cultures.

The in vivo evaluation showed that no protection was induced by the DNA vaccines

encoding either Gs-Fc or Gs. However, due to the absence of secretion in vitro, we

could not conclude that Fc-fusion protein approach did not work as an enhancer of the

immunogenicity of the antigen. Structural differences between fish and mammalian Fc

domains might explain the difficulties with the implementation of this approach in fish.

Therefore, further optimization of the secretion of the fish Fc-fusion protein design is

probably needed.

INTRODUCTION

DNA vaccines against fish rhabdovirus, such as viral haemorrhagic septicaemia virus

(VHSV), infectious haematopoietic necrosis virus (IHNV), and hirame rhabdovirus

(HIRRV) have been able to induce high level of protection under different experimental

conditions (Anderson, Mourich et al. 1996; Heppell, Lorenzen et al. 1998; Takano,

Iwahori et al. 2004). The effectiveness of these vaccines relies on: (i) the high

immunogenicity of the viral glycoprotein encoded by these DNA vaccines (Bearzotti,

Monnier et al. 1995; Winton 1996), and (ii) the fact that DNA vaccination allows the

activation of different protective immune mechanisms in the host (Liu 2011). Shortly

after intramuscular injection of the DNA vaccine against VHSV in rainbow trout, the

expression of the glycoprotein by the host induces an early and non-specific protection,

which is orchestrated by the up-regulation of interferon type I (IFN I), a central

antiviral component of the innate immunity (Boudinot, Blanco et al. 1998; Lorenzen,

Lorenzen et al. 2002; Acosta, Petrie et al. 2005). The non-specific protection is

followed by a specific protection associated with both mechanisms of the adaptive

88

immunity, the cellular and humoral immune responses (LaPatra, Corbeil et al. 2000;

Utke, Kock et al. 2008).

However, the promising results of the DNA vaccine against fish rhabdovirus have not

been replicated for other viruses that affect aquaculture industry. Experimental DNA

vaccines against infectious salmon anemia virus (ISAV) and infectious pancreatic

necrosis virus (IPNV) have induced relatively low protection (Mikalsen, Torgersen et al.

2004; Mikalsen, Sindre et al. 2005; Munang’andu, Fredriksen et al. 2012), while the

DNA vaccines against salmon alphavirus (SAV) and Atlantic halibut nodavirus (AHNV)

have shown not protection at all (Sommerset, Skern et al. 2005; Xu, Mutoloki et al.

2012).

An efficacious vaccine should involve a high immunogenic antigen, which is recognized

as a dangerous molecule by the host immune system, and consequently, activate

multiple protective mechanisms involving both innate and adaptive immune

mechanism. A low protective DNA vaccine against ISAV, encoding the hemagglutinin

had significantly increased efficacy when co-injected with a plasmid encoding type I

interferon (IFN). This indicates that the hemagglutinin by itself was incapable of

inducing the innate antiviral responses necessary to trigger a protective adaptive

response (Chang, Sun et al. 2015). Therefore, enhancing the immunogenicity of an

antigen could be a strategy to improve the protection induced by the related DNA

vaccine against a viral fish disease.

The Fc-fusion proteins could be an alternative to increasing immunogenicity of

antigens. The Fc-fusion proteins consist of an active protein (e.g. antigen) genetically

linked to an Fc domain of an immunoglobulin (Levin, Golding et al. 2015). Initially, the

Fc-fusion proteins were used to extend the half-life of the active protein, prolonging

the activity of therapeutic proteins in the blood stream, because an immunoglobulin Fc

domain avoids the endosomal degradation of the active protein (Roopenian and Akilesh

89

2007). Additionally, the presence of an Fc domain has been used to improve stability

and solubility of some proteins, as well as to allow protein purification by using protein

G/A affinity matrices (Rath, Baker et al. 2013).

Nevertheless, the capacity of the IgFc domain to engage specific cellular receptors

(FcR) and thereby modulate the immune response of the host, is one of the most

important properties of the Fc-fusion proteins, that have driven their application within

different therapeutic purposes. Some of the response mechanisms modulated by Fc-

fusion proteins are; phagocytosis, activation of the antibody-dependent cell mediated

cytotoxicity (ADCC) by natural killer cells, or the complement-dependent cytotoxicity

(CDC) (Shinkawa, Nakamura et al. 2003; Levin, Golding et al. 2015). Specifically for

vaccine application, the Fc domain can improve the efficacy by which the antigen is

recognized by FcRs on antigen presenting cells (APC), thereby enhancing the take up,

process, and antigen presentation (Soleimanpour, Farsiani et al. 2015). Vaccination of

mice with antigens from either HIV or influenza virus linked to the Fc domain of IgG2a,

thus induced a more effective immune response than the antigen alone (Loureiro, Ren

et al. 2011; Zaharatos, Yu et al. 2011).

In this study, we made the first attempts to design and evaluate a DNA vaccine

encoding an Fc-fusion protein for vaccination against VHSV, as a model of a generic

DNA vaccine design for fish. In order to evaluate any improvement in the protection

induced by this approach, a low protective DNA vaccine that encoded the secreted

form of the VHSV glycoprotein (Gs) was compared to DNA vaccines encoding the Gs

genetically linked to the Fc domain (Gs-Fc) of each of the three rainbow trout

immunoglobulin isotypes (IgM, IgT, and IgD).

90

MATERIALS AND METHODS

Cell culture. Epithelioma papulosum cyprinid (EPC) cells (Fijan, Sulimanović et al.

1983) and bluegill fry (BF2) cells (Wolf, Gravell et al. 1966) were used in this work.

Both cell lines were maintained at 15°C in minimal essential medium (MEM)

supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, and 100 U/mL

of Penicillin and 100 µg/mL of Streptomycin antibiotics.

Virus. DK-3292b VHSV isolate was propagated in BF2 cells. The cell culture was

infected with a low multiplicity of infection (MOI). When a complete cytopathic effect

was observed, the supernatant was collected and centrifuged at 4500 x g for 15 min at

4°C. The virus was stored at -80°C. Virus titer was determined using tissue culture

infectious dose per mL (TCID50 /mL) (Reed and Muench 1938).

Amplification of Fc domains of rainbow trout immunoglobulins. Total RNA was

isolated from spleen from rainbow trout stored in RNAlater at -20°C, using the RNeasy

mini kit (Qiagen). The cDNA was synthesized from 500 ng of the isolated RNA with

iScript (Bio-Rad) according to the procedure recommended by the manufacturer. The

PCR amplification was performed with the primers listed in Table 1. The primer design

was based on the sequence of the secreted form of each immunoglobulin isotype of

rainbow trout; IgM (Genebank No: AY870259.1), IgT (Genebank No: AY870268.1),

IgD (Genebank No:JQ003979.1).

For each fish immunoglobulin, two versions of the Fc domain were amplified. One

version consisted of the Fc region of the 3 C-terminal CH domains of the secreted

immunoglobulin molecule, here called “long-Fc”(LFc). The second version was the

short-Fc (ShFc), which included only the most C-terminal CH domain of the Fc region

(Figure 2). For the amplification of each immunoglobulin Fc isotype, two forward

primers (short-Fc and long-Fc) and one reverse primer with the stop codon were used.

91

The forward primers contained the Sal I restriction site while the reverse primer

contained the EcoRI restriction site to allow insertion into the expression vector.

The PCR amplification was carried out with the high fidelity DNA polymerase, Herculase

II Fusion (Agilent Technology) according to the recommendations of the manufacturer.

The estimated annealing temperature of the primers was 57°C. The size of the

amplicons was checked by agarose gel electrophoresis.

Table 1. Primers used to amplify the different Fc domains of fish immunoglobulins

Primer name Sequence 5’-3’

IgM-short-Fc- SalI-For ATATAT GTCGAC CAGCGTCCATCTGTCTTTC

IgM-long-Fc- SalI-For TTATAT GTCGAC CAGCCGTCTCTTTACGTAATG

IgM-Stop-EcorRI-R CTGCAT GAATTC CCTCTACTGGGCCATGCATC

IgT-short-Fc- SalI-For ATATAT GTCGAC TCTGTGTCCGTCCACATTC

IgT-long-Fc- SalI-For TTATAT GTCGAC CTGATGTGTATGATTGAAGATTTC

IgT-Stop-EcorRI-R ATATAT GAATTC TTACTTGTCTTCACATGAGTTAC

IgD-short-Fc- SalI-For ATATAT GTCGAC TCAGTGACCCCCTCCGC

IgD-long-Fc- SalI-For ATATAT GTCGAC GAACTCCTTCTAGTCCCCAG

IgD-Stop-EcorRI-R ATATAT GAATTC TTATATCAGAATTGAGTGAACGGAC

Bold sequence: endonuclease restriction sites

Underline sequence: stop codon

Plasmids. The pVax1 vector (Thermo Fisher Scientific) was used as the expression

vector in this work. According to Figure 1, the Fc-fusion proteins had three sections

inserted between HindIII and EcoRI of pVax1. The first section, named Gene1,

consisted of the secreted form of the VHSV glycoprotein (Gs), which consisted in the

sequence from the position 1 to 1356 of the full-length glycoprotein gene. The second

section consists of a flexible linker to avoid interferences between active protein (Gs)

92

and the Fc domain. The linker sequence involved 3 repetitions of the amino acidic

sequence Gly-Ser-Ser-Ser. The third section, named Gene2, which consist of the

different Fc domains amplified from different fish immunoglobulins. One construct only

contained Gs and the linker, to evaluate if the linker interfered with the Gs secretion.

Additionally, two DNA vaccines were used as a reference, first a highly protective DNA

vaccine that encodes the full-length VHSV glycoprotein (1521bp) and second, the DNA

vaccine that encode only the Gs. All constructs used in this work are summarized in

Table 2. Conventional cloning procedures were used to make all constructs. The

purification of the constructs were performed with Endofree plasmid Megaprep kit

(Qiagen). All constructs were sequenced using generic external primers for the CMV

promoter and polyadenylation signal, and with the primers listed in Table 1.

Figure 1: Scheme of the different sections of the Fc-fusion constructs inserted into the

expression vector. The antigen part is where both the full length and the secreted form

of the VHSV glycoprotein were inserted. The Fc domain part is where both versions of

the Fc domain of each fish immunoglobulin isotypes were inserted.

93

Table 2. Constructs for vaccination

Name Antigen part

(5’-end) Linker

Fc domain part

(3’-end)

pVax-vhsG Gfulla ---- ----

pVax-Gs Gsb ---- ----

pVax-Gs-Lc Gs (GSSS)3 ----

pVax-Gs-L-ShFc-IgM d Gs (GSSS)3 ShFc-IgM (short)

pVax-Gs-L-LFc-IgMe Gs (GSSS)3 LFc-IgM (long)

pVax-Gs-L- ShFc-IgT Gs (GSSS)3 ShFc-IgT (short)

pVax-Gs-L- LFc -IgT Gs (GSSS)3 LFc-IgT (long)

pVax-Gs-L- ShFc-IgD Gs (GSSS)3 ShFc-IgD (short)

pVax-Gs-L-LFc-IgD Gs (GSSS)3 LFc-IgD (long)

a Gfull: it is full-length of the VHSV glycoprotein gene (1521 bp) including the

transmembrane domain.

b Gs: secreted form of the VHSV glycoprotein gene (1356 bp) without the

transmembrane domain.

c ”L” refers to the linker peptide.

d “ShFc” refers to the short variant of the Fc domain part (Fig 2)

e “LFc” refers to the long variant fo the Fc domain part (Fig 2)

94

Figure 2. Short and long versions of the Fc domain in the Fc-fusion proteins. (A)

Structure of the heavy chain of a fish immunoglobulin. The secreted forms of IgM and

IgT have the same structure showed in this scheme, while the secreted form of IgD

possess 7 CH domains instead of 4. (B) Short and long versions of the Fc-fusion protein

encoded in the DNA vaccines. The Gs in the figure correspond to the extracellular

domain of VSV glycoprotein in order represent the Gs part of the Fc-fusion protein. The

glycoprotein was taken from (Garry and Garry 2008).

Transfection. In a 24-well plate, 2x105 cells/well were seeded 24 hrs. before the

transfection. The day of the transfection, the medium of each well was replaced for

250 µL of fresh MEM with 10% serum. The transfection solution was performed mixing

0.75 µg of each construct and 2µg of PEI, in a total volume of 250 µL of MEM without

serum. The transfection solution was incubated for 20 min at room temperature and

then added to one well, to have a final volume of 500 µL. This procedure was carried

out with each construct described in Table 2. The transfected cell cultures were

incubated at 28°C for 5 hrs. Then the medium of each well was replaced for 400 µL of

95

fresh MEM with 10% serum and maintained at 15oC until the evaluation of the

supernatant and the cell monolayer were performed.

ELISA. At 7 days post-transfection, the supernatant and cell lysate were collected to

evaluate the amount of Fc-fusion proteins in each fraction. The cells were lysed using

in each well 150 µL of lysis buffer (150 mM NaCl, 10 mM Tris, 1% Triton x-100, and

protease inhibitors) after removing the supernatant. In a 96-well plate coated with

rabbit anti-VHSV glycoprotein, 50 µL of either supernatant or cell lysate was added to

each well and incubated for 1hr at room temperature. For detection of bound VHSV G

protein, the primary antibody used was a mouse monoclonal anti-VHSV glycoprotein

IP1H3 (Lorenzen, Olesen et al. 1988). An HRP-conjugated rabbit anti-mouse IgG

(DAKO P0260) was used as secondary antibody.

Vaccination. Outbreed rainbow trout (5-10 gr.) were anesthetized in 0.01%

benzocaine. The immunization was performed with 3 µg of plasmids dissolved in 50µL

of saline solution (0,9% NaCl) per fish. The fish were divided into two groups; one

group was vaccinated with an intraperitoneal injection (I.P), while another group with

an intramuscular injection (I.M). The fish were kept in aerated 120L aquaria with

recirculated water at 8-10°C. The immunization was performed with a mixture

containing 1 µg of each of the three constructs with the short-Fc, as one treatment.

The same was performed with the long-Fc constructs.

Challenge. At 7 weeks post vaccination, the fish were transferred to aerated 8 L

aquaria in a contained facility. Each treatment was tested in duplicate with 18-20 fish

per aquarium. The infection was performed by immersion in static water with the VHSV

isolated DK3592b, used with an infectious dose of 1x105 TCID50 mL-1. The water flow

was re-established after 3 hrs. The fish were monitored 3 times per day for 30 days.

Fish with evident clinical sign of the disease or with compromised welfare were

euthanized by an overdose of benzocaine.

96

Statistical analysis

The statistical analysis was performed using t-test with R. Statistical significance was

considered when p<0.05.

RESULTS

Evaluation of the antigen localization in transfected cell cultures

The comparative analysis showed that while the Gs expressed by transfected cell was

mostly secreted into the supernatant, the Gfull, was mostly detected in the cellular

fraction, as expected and presumably due to the presence of the transmembrane

domain, (Figure 3A).

At 7 days post-transfection, the cell cultures transfected with the DNA vaccine

encoding the Gs (pVax-Gs) and Gs with the linker (pVax-Gs-L) had recombinant G

protein mainly in the supernatant. However, the cell cultures transfected with the DNA

vaccines encoding the various Fc-fusion proteins (Gs-Fc) we all had the fusion proteins

retained in the cellular fraction (Figure 3B).

97

Figure 3. Secretion profile in vitro of the DNA vaccines encoding: (A) Gs and Gfull,

and (B) the Fc-fusion proteins. The difference between the secreted and retained

fraction was statistically significant (p<0,05).

Evaluation of the immune protection induced by a DNA vaccine encoding the

Fc-fusion proteins

The DNA vaccines that encode any of the Fc-fusion proteins were not able to induce

protection against VHSV in rainbow trout fingerlings. Moreover, no difference was

observed when the immunization was performed either by I.P or I.M. The DNA vaccine

that encode the full-length glycoprotein (pVax-vhsG) induced protection when was

injected I.M (Figure 6), while no protection was observed by I.P injection. In one of the

replicate aquaria with fish injected with pVax-vhsG a tail-biting activity of some of the

fish led to termination of more than 50% of the population due to welfare concerns.

This aquaria was excluded from the data.

98

Figure 4. Protection against VHSV challenge induced by the DNA vaccine encoding the

Fc-fusion proteins. Each DNA vaccine was injected intraperitoneally (I.P) and

intramuscularly (I.M).

DISCUSSION

The immunogenicity of antigen is fundamental to trigger the innate and adaptive

immune responses establishment of protection against disease (Loureiro, Ren et al.

2011; Zaharatos, Yu et al. 2011; Thim, Villoing et al. 2014; Chang, Sun et al. 2015).

In this work, we evaluated whether it was possible to improve the protection induced

by a DNA vaccine that encodes a secreted antigen (Gs) by linking the antigen to a fish

immunoglobulin Fc domain.

Because, this was the first attempt to develop the Fc-fusion protein approach applied

to fish viral diseases, we took into account:

(i) It is known that an Fc domain of a specific immunoglobulin isotype has specific

effector functions. For instance, an antigen linked to an Fc domain from IgG2a elicited

superior humoral responses compared to the antigen linked to the Fc domain of IgG1

in mice (Zaharatos, Yu et al. 2011). The Fc-fusion proteins developed here comprised

the Fc domain of all three fish immunoglobulin isotypes (IgM, IgT, and IgD).

(ii) In mammals, the first constant domain of the immunoglobulin heavy chains (CH1)

binds a chaperone protein (BIP) in the endoplasmic reticulum (ER) (Elkabetz, Argon et

al. 2005). BIP releases the immunoglobulin only when heavy chain binds light chain.

Heavy chains without the CH1 domain can be secreted as a monomer without the

presence of the light chain. Whether fish immunoglobulin heavy chains follow the same

mechanisms is unknown. Therefore, two versions of the fish immunoglobulin Fc

domain were linked to the antigen and tested in this study. The long-Fc version

99

corresponded to the full-length Fc domain of IgM and IgT, respectively (Figure 2B).

The short-Fc version was designed to evaluate whether a potential retention of the

heavy chain could be due to other regions beside CH1 (CH2, CH3). Because, it has to be

taken into account the absence of hinge region in fish immunoglobulins, a natural

separation between CH1 and the Fc domain.

The in vivo trial showed no protection induced by the vaccination with any of the DNA

vaccines encoding the Fc-fusion proteins. Additionally, no differences were observed

related to the route of delivery (I.P vs I.M). Similarly, the DNA vaccine encoding the

non-linked Gs was not able to induce protection against VHSV. The DNA vaccine that

encodes the full-length glycoprotein was only able to induce protection in fish when it

was injected I.M. Taking into account that the effect of certain vaccines is different

according to the delivery route, here we evaluated both I.P and I.M immunizations

(McLauchlan, Collet et al. 2003; Martinez-Lopez, García-Valtanen et al. 2013).

Our results confirm the assumption that the immune mechanism by which the

immunogenicity of an antigen is enhanced using the Fc-fusion protein approach

requires secretion of the antigen, thereby allowing the interaction of Fc domain with Fc

receptors on specific immune cells. Therefore, due to the absence of secretion

observed in cell culture and likely in vivo, we could not conclude that the Fc-fusion

protein approach in fish was not able to induce protection in rainbow trout fingerlings.

Optimizing the secretion of the Fc-fusion proteins will be necessary to assess their real

capacity to induce protection in fish.

It has been described that the amino acidic sequence in the C-terminal of the Fc

domain, determine the secreted or membrane bound form of the immunoglobulins.

However, for IgM and IgA of mammals, the C-terminal has also been associated with a

control mechanism to secrete immunoglobulin with the correct polymerization (Guenzi,

Fra et al. 1994). Beside the CH1 domain, a cysteine close to the C- terminal of the

100

secreted form of the heavy chain cause cellular retention. Similarly, the secreted form

of fish immunoglobulins also has a cysteine close to the C-terminal that could be

causing the cellular retention of the Fc-fusion proteins designed here (Table 3).

Therefore, to perform an evaluation of the protection induced by a DNA vaccine

encoding the Gs-Fc is fundamental to optimize the secretion of the different Fc-fusion

protein. In order to do so, the first step would be to modify the cysteine in the C-

terminal of the Fc domains of the fish immunoglobulins.

Table 3. C-terminal tail of murine and fish immunoglobulins.

Isotype Sequence

Murine IgG Murine IgE Murine IgD Murine IgM Murine IgA RT IgM RT IgT RT IgD

G K G K N T S L R P S G C Y H L L P E S D G F S R R P D G P A L A G K P T L Y N V S L I M S D T G G T C Y G K P T N V S V S V I M S E G D G I C Y I D R T S N Q P N L V N L S L N V P Q R C M A Q G S D N S T S P K E M S V S K S T G N S C E D K L A L N I S K P G V C L S V H S I L I

RT: Rainbow Trout

Cysteines (C) in the C-terminal tail of the murine and fish immunoglobulins are

underlined.

However, it is important to consider that fish immunoglobulins are not structurally

identical to mammals’ immunoglobulins. For instance, fish antibodies lack the hinge

region, and the J peptide, and while the most prevalent form of fish IgM in serum is

tetrameric, it is pentameric in mammals. Therefore, it is possible that, compared to

mammalian Ig, other elements in the fish immunoglobulin Fc domains could affect the

secretion of Fc-fusion proteins.

This work represents the first attempt to apply the Fc-fusion protein approach in a

vaccine against a fish viral disease. Taking into account that in mammals, the Fc-fusion

proteins have not only showed potential as a therapeutic strategy, but also as a

molecular tool to determine how the Fc domain works modulating the immune system,

101

a functional Fc fusion protein would have a huge potential to elucidate important

aspects of the effector mechanisms driven by the humoral immune responses in

teleost.

Acknowledgements

This study was supported by the Chilean National Scholarship Program for Graduate

Studies (Conicyt) for DS, and by the European Commission contract FP7311993

TargetFish.

The authors gratefully acknowledge the excellent assistance from the technical staff of

the Fish Health Section, Department of Animal Science, University of Aarhus.

REFERENCES

Acosta, F., A. Petrie, et al. (2005). "Kinetics of Mx expression in rainbow trout

(Oncorhynchus mykiss) and Atlantic salmon (Salmo salar L.) parr in response to VHS–

DNA vaccination." Fish & Shellfish Immunology 18(1): 81-89.

Anderson, E., D. Mourich, et al. (1996). "Genetic immunization of rainbow trout

(Oncorhynchus mykiss) against infectious hematopoietic necrosis virus." Molecular

Marine Biology and Biotechnology 5(2): 114-122.

Bearzotti, M., A. Monnier, et al. (1995). "The glycoprotein of viral hemorrhagic

septicemia virus (VHSV): antigenicity and role in virulence." Veterinary research 26(5-

6): 413-422.

Boudinot, P., M. Blanco, et al. (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(2): 297-306.

Chang, C.-J., B. Sun, et al. (2015). "Adjuvant activity of fish type I interferon shown in

a virus DNA vaccination model." Vaccine 33(21): 2442-2448.

Elkabetz, Y., Y. Argon, et al. (2005). "Cysteines in CH1 underlie retention of

unassembled Ig heavy chains." Journal of Biological Chemistry 280(15): 14402-

14412.

102

Fijan, N., D. Sulimanović, et al. (1983). Some properties of the epithelioma papulosum

cyprini (EPC) cell line from carp Cyprinus carpio. Annales de l'Institut Pasteur/Virologie,

Elsevier.

Garry, C. E. and R. F. Garry (2008). "Proteomics computational analyses suggest that

baculovirus GP64 superfamily proteins are class III penetrenes." Virology journal 5(1):

28.

Guenzi, S., A. Fra, et al. (1994). "The efficiency of cysteine‐mediated intracellular

retention determines the differential fate of secretory IgA and IgM in B and plasma

cells." European journal of immunology 24(10): 2477-2482.

Heppell, J., N. Lorenzen, et al. (1998). "Development of DNA vaccines for fish: vector

design, intramuscular injection and antigen expression using viral haemorrhagic

septicaemia virus genes as model." Fish & Shellfish Immunology 8(4): 271-286.

Hoffmann, B., M. Beer, et al. (2005). Fish rhabdoviruses: molecular epidemiology and

evolution. The World of Rhabdoviruses, Springer: 81-117.

LaPatra, S. E., S. Corbeil, et al. (2000). "The dose-dependent effect on protection and

humoral response to a DNA vaccine against infectious hematopoietic necrosis (IHN)

virus in subyearling rainbow trout." Journal of Aquatic Animal Health 12(3): 181-188.

Levin, D., B. Golding, et al. (2015). "Fc fusion as a platform technology: potential for

modulating immunogenicity." Trends in biotechnology 33(1): 27-34.

Liu, M. A. (2011). "DNA vaccines: an historical perspective and view to the future."

Immunological reviews 239(1): 62-84.

Lorenzen, N., E. Lorenzen, et al. (2002). "Immunity induced shortly after DNA

vaccination of rainbow trout against rhabdoviruses protects against heterologous virus

but not against bacterial pathogens." Developmental & Comparative Immunology

26(2): 173-179.

Lorenzen, N., N. J. Olesen, et al. (1988). "Production and characterization of

monoclonal antibodies to four Egtved virus structural proteins." Dis Aquat Org 4: 35-

42.

Loureiro, S., J. Ren, et al. (2011). "Adjuvant-free immunization with hemagglutinin-Fc

fusion proteins as an approach to influenza vaccines." Journal of virology 85(6): 3010-

3014.

103

Martinez-Lopez, A., P. García-Valtanen, et al. (2013). "Increasing Versatility of the

DNA Vaccines through Modification of the Subcellular Location of Plasmid-Encoded

Antigen Expression in the." PloS one 8(10).

McLauchlan, P., B. Collet, et al. (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

Immunology 15(1): 39-50.

Mikalsen, A. B., H. Sindre, et al. (2005). "Protective effects of a DNA vaccine

expressing the infectious salmon anemia virus hemagglutinin-esterase in Atlantic

salmon." Vaccine 23(41): 4895-4905.

Mikalsen, A. B., J. Torgersen, et al. (2004). "Protection of Atlantic salmon Salmo salar

against infectious pancreatic necrosis after DNA vaccination." Diseases of Aquatic

organisms 60: 11-20.

Munang’andu, H. M., B. N. Fredriksen, et al. (2012). "Comparison of vaccine efficacy

for different antigen delivery systems for infectious pancreatic necrosis virus vaccines

in Atlantic salmon (Salmo salar L.) in a cohabitation challenge model." Vaccine 30(27):

4007-4016.

Rath, T., K. Baker, et al. (2013). "Fc-fusion proteins and FcRn: structural insights for

longer-lasting and more effective therapeutics." Critical reviews in biotechnology(0): 1-

20.

Reed, L. J. and H. Muench (1938). "A simple method of estimating fifty per cent

endpoints." American journal of epidemiology 27(3): 493-497.

Roopenian, D. C. and S. Akilesh (2007). "FcRn: the neonatal Fc receptor comes of

age." Nature Reviews Immunology 7(9): 715-725.

Shinkawa, T., K. Nakamura, et al. (2003). "The absence of fucose but not the presence

of galactose or bisecting N-acetylglucosamine of human IgG1 complex-type

oligosaccharides shows the critical role of enhancing antibody-dependent cellular

cytotoxicity." Journal of Biological Chemistry 278(5): 3466-3473.

Soleimanpour, S., H. Farsiani, et al. (2015). "APC targeting enhances immunogenicity

of a novel multistage Fc-fusion tuberculosis vaccine in mice." Applied microbiology and

biotechnology: 1-14.

Sommerset, I., R. Skern, et al. (2005). "Protection against Atlantic halibut nodavirus in

turbot is induced by recombinant capsid protein vaccination but not following DNA

vaccination." Fish & shellfish immunology 18(1): 13-29.

104

Takano, T., A. Iwahori, et al. (2004). "Development of a DNA vaccine against hirame

rhabdovirus and analysis of the expression of immune-related genes after vaccination."

Fish & Shellfish Immunology 17(4): 367-374.

Thim, H. L., S. Villoing, et al. (2014). "Vaccine Adjuvants in fish vaccines make a

difference: comparing three adjuvants (Montanide ISA763A Oil, CpG/Poly I: C Combo

and VHSV Glycoprotein) alone or in combination formulated with an inactivated whole

salmonid alphavirus antigen." Vaccines 2(2): 228-251.

Utke, K., H. Kock, et al. (2008). "Cell-mediated immune responses in rainbow trout

after DNA immunization against the viral hemorrhagic septicemia virus."

Developmental & Comparative Immunology 32(3): 239-252.

Wolf, K., M. Gravell, et al. (1966). "Lymphocystis virus: isolation and propagation in

centrarchid fish cell lines." Science 151(3713): 1004-1005.

Xu, C., S. Mutoloki, et al. (2012). "Superior protection conferred by inactivated whole

virus vaccine over subunit and DNA vaccines against salmonid alphavirus infection in

Atlantic salmon (Salmo salar L.)." Vaccine 30(26): 3918-3928.

Zaharatos, G. J., J. Yu, et al. (2011). "HIV-1 and influenza antigens synthetically linked

to IgG2a Fc elicit superior humoral responses compared to unmodified antigens in

mice." Vaccine 30(1): 42-50.

105

5. Part IV: General discussion and perspectives

Since the first time that the DNA vaccines against VHSV and IHNV were successfully

evaluated in fish in the late 1990’s (Anderson, Mourich et al. 1996; Heppell, Lorenzen

et al. 1998), several studies have contributed to understand many aspects of

protection induced by DNA vaccines, among them (i) what mechanisms were triggered

by DNA vaccines, (ii) what environmental or physiological factors could affect the

protection induced by these vaccines, and (iii) the duration of each type of protective

mechanism. At the beginning, the focus of these studies was driven by the commercial

purpose of developing an immunization strategy to fight serious viral diseases that

affect aquaculture industry (Corbeil, LaPatra et al. 1999; Corbeil, Kurath et al. 2000;

LaPatra, Corbeil et al. 2000; Lorenzen, Einer-Jensen et al. 2000). All these knowledge

led to license and use of a DNA vaccine against IHNV in 2005 (Salonius, Simard et al.

2007). Besides the commercial aspect of developing effective prophylactic measures to

control viral diseases, DNA vaccines have also been a useful tool to obtain a better

understanding of how the fish antiviral immunity works (Lorenzen, Lorenzen et al.

2002; Kurath, Purcell et al. 2007).

This thesis was divided into two sections according to the addressed topics. The first

part, include manuscript I, aimed at evaluating of whether repeated passaging of VHSV

in vaccinated fish would promote the virus to break through the protection and cause

disease. The second part used the DNA vaccine against VHSV as a model to evaluate

two new DNA vaccine designs, which could potentially improve the safety and general

applicability of DNA vaccine in farmed fish.

106

5.1 Section I

5.1.1 Manuscript I

The results suggested that there is a low probability that VHSV might generate and

select a variant that escapes from any of the protective mechanisms induced by the

DNA vaccine in a short time: the early antiviral responses, the specific antiviral

responses, and the neutralizing antibodies from immunized rainbow trout sera.

Although this is an advantage in term of applicability of DNA vaccine, one factor to

consider is that the experimental setup used here was designed to evaluate the

selection of escape mutants under highly protective immunization conditions, which

included:

(i) The vaccination was carried out with a highly protective dose of the DNA

vaccine.

(ii) The DNA vaccine used encoded the homologous VHSV glycoprotein to the one

encoded by the virus used in the challenge.

However, it is important to take into account that this work did not cover the

possibility of VHSV could generate and select escape mutants upon less protective

vaccination conditions or after more passages of VHSV infection in vaccinated fish.

5.2 Section II

This part focused on the characterization of the immune protection induced by two

innovative DNA vaccine approaches. These approaches took advantage of our

knowledge of the fish immune system and applied immune elements from the innate

and adaptive arms, respectively, to modulate the vaccine effect in order to improve

safety and applicability of DNA vaccines in general.

107

5.2.1 Manuscript II

The analysis of the first vaccine approach was described in the manuscript II. In

agreement with safety guidelines, the DNA vaccine tested in this manuscript consisted

of an expression vector with a trout-derived interferon-inducible Mx promoter, which

control the expression of VHSV glycoprotein in the host, instead of the human

cytomegalovirus (CMV) promoter, which has been used in the extensively analyzed

DNA vaccine against VHSV (Heppell, Lorenzen et al. 1998; Lorenzen, Einer-Jensen et

al. 2000).

The manuscript II showed a time-course study to evaluate whether the fish immunized

with DNA vaccine with Mx promoter was able to mount the three phases of antiviral

responses: early antiviral responses (EAVR), specific antiviral responses (SAVR), and

long-term antiviral responses (LAVR). Furthermore, because of the inducible feature of

this DNA vaccine, it was also possible to understand how different expression levels of

the glycoprotein affect the induction of the different protective phases.

The results in manuscript II agreed with the model that DNA vaccination induces three

phases of antiviral responses (Kurath, Purcell et al. 2007), also indicating that each

phase needs to be activated independently. In the specific case of the activation of

SAVR and LAVR, both antiviral responses are associated with the adaptive immune

responses. However, this does not mean that the activation of an efficacious SAVR

involves the activation of an efficacious LAVR. Indeed, the DNA vaccine with Mx

promoter without stimulation with poly I:C showed no induction of LAVR even when an

efficient SAVR was induced. Many vaccine evaluations published are performed

between 1-3 month post vaccination or 300-900 degree days. However, in order to

evaluate different protective mechanisms is suggested that the evaluation of any

vaccine for aquaculture should include a challenge to test specifically the LAVR.

108

It is well known, that to mount an efficacious adaptive immunity by vaccines is

necessary to trigger mechanisms associated with innate responses (Coffman, Sher et

al. 2010). Although it requires further experiments, the results of the protection

induced by the DNA vaccine with CMV and Mx promoter, suggested that the innate

antiviral responses (EAVR) could be either protective or the non-protective, and both

could be able to trigger a protective SAVR. However, the activation of LAVR might

require triggering additional and more complex mechanisms that are present only in

the protective innate antiviral responses (Figure 7).

Figure 7. Model of induction of SAVR and LAVR by innate responses. (A) without

innate responses is not possible to activate SAVR or LAVR. (B) non-protective innate

antiviral responses are able to trigger the activation of SAVR. (C) Protective innate

antiviral responses are able to trigger the activation of SAVR and LAVR.

109

Future perspectives

The fish-derived promoter and the selective expression of the antigen in fish cell and

not in human cell line are two preferable characteristics in a DNA vaccine developed for

aquacultured fish. However, it seems that the low expression activity of Mx promoter

under stimulation of poly I:C is not sufficient to mount an effective long-term antiviral

response. An alternative strategy, based on the work presented in Martinez-Lopez et al

(2012), could be design a hybrid promoter combining the enhancers or the promoter

cores from other fish-derived promoters in order to keep the preferable features of the

Mx promoter, but with a high expression activity to effectively trigger all three phases

of the host antiviral responses, similar to the protection induced by the DNA vaccine

with the CMV promoter (Martinez-Lopez, Chinchilla et al. 2012).

Additionally, further experiments should involve a gene expression analysis comparing

the innate responses induced by the DNA vaccine with the Mx promoter and the CMV

promoter. These results would help to understand the specific innate responses

required to induce SAVR and/or LAVR. This knowledge might be applied to adjuvants

development, which would be able to induce specific pathways of the innate immunity

according to the characteristic of the vaccine, reducing the side effects of the

traditional adjuvants used in fish vaccination.

5.2.2 Manuscript III

The second DNA vaccine approach evaluated in this section was described in the

manuscript III. This manuscript showed the first attempt to apply the Fc-fusion protein

approach to increase the immunogenicity of the fish-derived antigen. In order to apply

this approach, the DNA vaccine against VHSV was used as a model for further

development of a generic vaccination strategy that can be used against other viruses

that affect finfish aquaculture.

110

The secretion of Fc-fusion proteins is fundamental to allow the Fc domains to engage

FcR on immune cells. Taking into account that cell cultures transfected with the DNA

vaccines encoding the Fc-fusion proteins showed no secretion of these proteins, it is

logical to think that an in vivo infection trial should fail in inducing protection.

Nevertheless, the preliminary in vivo trial was carried out for two reasons: First, it

might be that undetectable levels of the Fc-fusion proteins were indeed secreted in cell

culture, which could be sufficient to induce protection against VHS. Second, although

the Fc fusion protein were retained inside the cells, these proteins could be released

upon cellular lysis associated with the production of the Fc-fusion protein, which would

allow the Fc domains bind the FcRs on immune cells and trigger the immune response.

The in vivo and in vitro results suggested that the retention of the Fc-proteins was

likely caused by a region in the C-terminal of the Fc domains. This agreed with the fact

that IgM and IgA from mammals are retained inside the cell even after removing the

CH1 domain, similar to what occurs with the constructs studied in this study (Sitia,

Neuberger et al. 1987; Guenzi, Fra et al. 1994). The analysis of mammals´ IgM and

IgA showed that this retention was due to a cysteine closer to C-terminal. Fish

immunoglobulins also have a cysteine close to the C-terminal, which might be

interfering with the secretion.

Future perspectives

The next steps to develop the Fc-fusion proteins approach against fish viruses should

include modifying the C-terminal of the Fc domain of each fish immunoglobulin isotype

in order to optimize the secretion of these proteins in cell culture (Sitia, Neuberger et

al. 1987; Sitia, Neuberger et al. 1990). After the optimization of the Fc-fusion protein

secretion, it should be necessary to carry out a new in vivo infection trial, but in this

experiment, each vaccine will be evaluated individually, instead of mixing them like in

the preliminary trial described in manuscript III. This analysis could also provide some

111

knowledge about the specific function of Fc-domain of each fish immunoglobulin

isotype.

Additionally, it has been reported that certain mutations in the Fc-domain of IgG have

allowed modulating specific immune responses according to the therapeutic

applications. Therefore, it should be interesting to explore this principle with the fish Fc

domains to understand effector mechanisms triggered by fish.

112

6. Concluding remarks

The three manuscripts, on which this Ph.D. thesis is based, provide significant

new knowledge on functional aspects of the immune protection induced by

DNA vaccination of rainbow trout against VHS virus under experimental

conditions. The main results of the studies are listed in the following.

Section I

Manuscript I: “Can VHS virus bypass the protective immunity induced by

DNA vaccination in rainbow trout?”

- After repetitive passages of VHSV in rainbow trout immunized with the

DNA vaccine, the virus did not generate variants capable of bypassing

the innate and adaptive protection induced by the DNA vaccine.

- No neutralization resistant VHSV escape variants were found after

repetitive passaging of VHSV in cell culture in the presence of serum

from rainbow trout immunized with the DNA vaccine.

- When examined shortly after inoculation with VHSV, some vaccinated

healthy fish appeared to be carriers of infection. Such carriers could

transmit the infection to cohabitant naïve fish.

113

Section II

Manuscript II: “Time-course study of the immune protection induced by an

interferon-inducible DNA vaccine against viral haemorrhagic septicaemia

virus in rainbow trout”

- The DNA vaccine with Mx promoter allowed differential antigen

expression in cell lines of fish and human origin, respectively. Only the

transfected fish cell line expressed the glycoprotein upon stimulation

with IFN inducer poly I:C.

Both fish and human cell lines transfected with the DNA vaccine with the

CMV promoter constitutively expressed the glycoprotein.

- Immunization with the DNA vaccine with the Mx promoter induced

protective EAVR and SAVR upon stimulation with poly I:C, while only

protective SAVR was induced without stimulation.

- Despite stimulation of a protective SAVR, the DNA vaccine with Mx

promoter was not able to induce a similar long-term protection, as

observed in the DNA vaccine with the CMV promoter.

- The results suggest a lack of direct relationship between protection in

the three phases of the immune response to the DNA vaccination. This

stresses the importance of testing new vaccines throughout the full

response profile.

114

Manuscript III: “Attempt to mimicking antibody-antigen complexes by DNA

vaccination in a fish virus model”

- None of the Fc-fusion proteins designed in this work were secreted from

transfected cell cultures.

- No protection against VHS was induced in rainbow trout fingerlings by

the DNA vaccines encoding the Fc-fusion proteins or the Gs alone.

- Optimization of the fusion gene design to obtain antigen that can be

secreted is probably necessary to perform a proper evaluation of the

prophylactic potential of Fc-fusion proteins as vaccines against fish

diseases.

115

7. References

Acosta, F., A. Petrie, et al. (2005). "Kinetics of Mx expression in rainbow trout (Oncorhynchus mykiss) and Atlantic salmon (Salmo salar L.) parr in response to VHS–DNA vaccination." Fish & shellfish immunology 18(1): 81-89.

Adelmann, M., B. Köllner, et al. (2008). "Development of an oral vaccine for immunisation of rainbow trout (Oncorhynchus mykiss) against viral haemorrhagic septicaemia." Vaccine 26(6): 837-844.

Adetokunboh, O., O. Atibioke, et al. (2015). "Antiretroviral Treatment and Resistance Patterns in HIV-Infected Children." Current infectious disease reports 17(10): 1-9.

Alonso, M., M. Johnson, et al. (2003). "A fish specific expression vector containing the interferon regulatory factor 1A (IRF1A) promoter for genetic immunization of fish." Vaccine 21(15): 1591-1600.

Ammayappan, A. and V. N. Vakharia (2011). "Nonvirion protein of novirhabdovirus suppresses apoptosis at the early stage of virus infection." Journal of virology 85(16): 8393-8402.

Anderson, E., D. Mourich, et al. (1996). "Genetic immunization of rainbow trout (Oncorhynchus mykiss) against infectious hematopoietic necrosis virus." Molecular Marine Biology and Biotechnology 5(2): 114-122.

Anderson, E., D. Mourich, et al. (1996). "Gene expression in rainbow trout (Oncorhynchus mykiss) following intramuscular injection of DNA." Molecular Marine Biology and Biotechnology 5(2): 105-113.

Ángeles Esteban, M. (2012). "An overview of the immunological defenses in fish skin." ISRN Immunology 2012.

Barouch, D. H., J. Kunstman, et al. (2003). "Viral escape from dominant simian immunodeficiency virus epitope-specific cytotoxic T lymphocytes in DNA-vaccinated rhesus monkeys." Journal of virology 77(13): 7367-7375.

Barouch, D. H., J. Kunstman, et al. (2002). "Eventual AIDS vaccine failure in a rhesus monkey by viral escape from cytotoxic T lymphocytes." Nature 415(6869): 335-339.

Bearzotti, M., B. Delmas, et al. (1999). "Fish rhabdovirus cell entry is mediated by fibronectin." Journal of virology 73(9): 7703-7709.

116

Bearzotti, M., A. Monnier, et al. (1995). "The glycoprotein of viral hemorrhagic septicemia virus (VHSV): antigenicity and role in virulence." Veterinary research 26(5-6): 413-422.

Boudinot, P., M. Blanco, et al. (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(2): 297-306.

Brudeseth, B. E., H. F. Skall, et al. (2008). "Differences in virulence of marine and freshwater isolates of viral hemorrhagic septicemia virus in vivo correlate with in vitro ability to infect gill epithelial cells and macrophages of rainbow trout (Oncorhynchus mykiss)." Journal of virology 82(21): 10359-10365.

Brudeseth, B. E., R. Wiulsrød, et al. (2013). "Status and future perspectives of vaccines for industrialised fin-fish farming." Fish & shellfish immunology 35(6): 1759-1768.

Coffman, R. L., A. Sher, et al. (2010). "Vaccine adjuvants: putting innate immunity to work." Immunity 33(4): 492-503.

Collet, B. (2014). "Innate immune responses of salmonid fish to viral infections." Developmental & Comparative Immunology 43(2): 160-173.

Collet, B., P. Boudinot, et al. (2004). "An Mx1 promoter–reporter system to study interferon pathways in rainbow trout." Developmental & Comparative Immunology 28(7): 793-801.

Collet, B. and C. J. Secombes (2001). "The rainbow trout (Oncorhynchus mykiss) Mx1 promoter." European Journal of Biochemistry 268(6): 1577-1584.

Corbeil, S., G. Kurath, et al. (2000). "Fish DNA vaccine against infectious hematopoietic necrosis virus: efficacy of various routes of immunisation." Fish & shellfish immunology 10(8): 711-723.

Corbeil, S., S. LaPatra, et al. (1999). "Evaluation of the protective immunogenicity of the N, P, M, NV and G proteins of infectious hematopoietic necrosis virus in rainbow trout oncorhynchus mykiss using DNA vaccines." Diseases of Aquatic organisms 39(1): 29-36.

Chang, C.-J., B. Sun, et al. (2015). "Adjuvant activity of fish type I interferon shown in a virus DNA vaccination model." Vaccine 33(21): 2442-2448.

117

Chico, V., M. Ortega-Villaizan, et al. (2009). "The immunogenicity of viral haemorragic septicaemia rhabdovirus (VHSV) DNA vaccines can depend on plasmid regulatory sequences." Vaccine 27(13): 1938-1948.

Dale, O. B., I. Orpetveit, et al. (2009). "Outbreak of viral haemorrhagic septicaemia (VHS) in seawater-farmed rainbow trout in Norway caused by VHS virus Genotype III." Diseases of Aquatic organisms 85(2): 93.

De Kinkelin, P., M. Bearzotti, et al. (1995). "Eighteen years of vaccination against viral haemorrhagic septicaemia in France." Veterinary research 26(5-6): 379-387.

Desvignes, L., C. Quentel, et al. (2002). "Pathogenesis and immune response in Atlantic salmon (Salmo salar L.) parr experimentally infected with salmon pancreas disease virus (SPDV)." Fish & shellfish immunology 12(1): 77-95.

Dixon, B. (2012). "Vaccines for finfish aquaculture: What do we need to know to make them work?" Electronic Journal of Biotechnology 15(5): 14-14.

Domingo, E., C. K. Biebricher, et al. (2001). Quasispecies and RNA virus evolution: principles and consequences, Landes Bioscience Austin.

Domingo, E., J. Sheldon, et al. (2012). "Viral quasispecies evolution." Microbiology and Molecular Biology Reviews 76(2): 159-216.

Elkabetz, Y., Y. Argon, et al. (2005). "Cysteines in CH1 underlie retention of unassembled Ig heavy chains." Journal of Biological Chemistry 280(15): 14402-14412.

EURL-FISH (2015). "19th Annual Meeting of the National Reference Laboratories." European Union Reference Laboratory for Fish Diseases and National Veterinary Institute, Technical University of Denmark.

Evelyn, T. (1996). "A historical review of fish vaccinology." Developments in biological standardization 90: 3-12.

FAO (2014). "The State of World Fisheries and Aquaculture " Opportunities and challenges"." Food and Agriculture Organization of the United Nations.

Fijan, N., D. Sulimanović, et al. (1983). Some properties of the epithelioma papulosum cyprini (EPC) cell line from carp Cyprinus carpio. Annales de l'Institut Pasteur/Virologie, Elsevier.

118

Fischer, U., E. O. Koppang, et al. (2013). "Teleost T and NK cell immunity." Fish & shellfish immunology 35(2): 197-206.

Garry, C. E. and R. F. Garry (2008). "Proteomics computational analyses suggest that baculovirus GP64 superfamily proteins are class III penetrenes." Virology journal 5(1): 28.

Garver, K. A., S. E. LaPatra, et al. (2005). "Efficacy of an infectious hematopoietic necrosis (IHN) virus DNA vaccine in Chinook Oncorhynchus tshawytscha and sockeye O. nerka salmon." Diseases of Aquatic organisms 64(1): 13-22.

Gomez-Chiarri, M., S. K. Livingston, et al. (1996). "Introduction of foreign genes into the tissue of live fish by direct injection and particle bombardment."

Grischkowsky, R. S. and D. F. Amend (1976). "Infectious hematopoietic necrosis virus: prevalence in certain Alaskan sockeye salmon, Oncorhynchus nerka." Journal of the Fisheries Board of Canada 33(1): 186-188.

Guenzi, S., A. Fra, et al. (1994). "The efficiency of cysteine‐mediated intracellular retention determines the differential fate of secretory IgA and IgM in B and plasma cells." European journal of immunology 24(10): 2477-2482.

Hall, S. J. (2011). Blue frontiers: managing the environmental costs of aquaculture, WorldFish.

Harmache, A., M. LeBerre, et al. (2006). "Bioluminescence imaging of live infected salmonids reveals that the fin bases are the major portal of entry for Novirhabdovirus." Journal of virology 80(7): 3655-3659.

Haugarvoll, E., I. Bjerkås, et al. (2008). "Identification and characterization of a novel intraepithelial lymphoid tissue in the gills of Atlantic salmon." Journal of anatomy 213(2): 202-209.

Heppell, J., N. Lorenzen, et al. (1998). "Development of DNA vaccines for fish: vector design, intramuscular injection and antigen expression using viral haemorrhagic septicaemia virus genes as model." Fish & Shellfish Immunology 8(4): 271-286.

Holmes, E. C. (2009). Evolution and emergence of RNA viruses, Oxford University Press.

Hølvold, L. B., A. I. Myhr, et al. (2014). "Strategies and hurdles using DNA vaccines to fish." Vet Res 45: 21.

119

Huang, C., M.-S. Chien, et al. (1996). "Mapping the neutralizing epitopes on the glycoprotein of infectious haematopoietic necrosis virus, a fish rhabdovirus." Journal of General Virology 77(12): 3033-3040.

Isshik, T., T. Nishizawa, et al. (2001). "An outbreak of VHSV (viral hemorrhagic septicemia virus) infection in farmed Japanese flounder Paralichthys olivaceus in Japan." Diseases of Aquatic organisms 47(2): 87-99.

Kibenge, F. S., M. G. Godoy, et al. (2012). "Countermeasures against viral diseases of farmed fish." Antiviral research 95(3): 257-281.

Koppang, E. O., U. Fischer, et al. (2010). "Salmonid T cells assemble in the thymus, spleen and in novel interbranchial lymphoid tissue." Journal of anatomy 217(6): 728-739.

Kurath, G., K. A. Garver, et al. (2006). "Protective immunity and lack of histopathological damage two years after DNA vaccination against infectious hematopoietic necrosis virus in trout." Vaccine 24(3): 345-354.

Kurath, G., M. Purcell, et al. (2007). "Fish rhabdovirus models for understanding host response to DNA vaccines." CAB reviews 2(1): 1-12.

Kurath, G. and J. Winton (2011). "Complex dynamics at the interface between wild and domestic viruses of finfish." Current Opinion in Virology 1(1): 73-80.

Lambkin, R., L. McLain, et al. (1994). "Neutralization escape mutants of type A influenza virus are readily selected by antisera from mice immunized with whole virus: a possible mechanism for antigenic drift." Journal of General Virology 75(12): 3493-3502.

LaPatra, S. E., S. Corbeil, et al. (2000). "The dose-dependent effect on protection and humoral response to a DNA vaccine against infectious hematopoietic necrosis (IHN) virus in subyearling rainbow trout." Journal of Aquatic Animal Health 12(3): 181-188.

Lecocq-Xhonneux, F., M. Thiry, et al. (1994). "A recombinant viral haemorrhagic septicaemia virus glycoprotein expressed in insect cells induces protective immunity in rainbow trout." Journal of General Virology 75(7): 1579-1587.

Leong, J., E. Anderson, et al. (1996). "Fish vaccine antigens produced or delivered by recombinant DNA technologies." Developments in biological standardization 90: 267-277.

120

Leung, T. L. and A. E. Bates (2013). "More rapid and severe disease outbreaks for aquaculture at the tropics: implications for food security." Journal of applied ecology 50(1): 215-222.

Levin, D., B. Golding, et al. (2015). "Fc fusion as a platform technology: potential for modulating immunogenicity." Trends in biotechnology 33(1): 27-34.

Liu, H., Y. Liu, et al. (2011). "Clathrin-mediated endocytosis in living host cells visualized through quantum dot labeling of infectious hematopoietic necrosis virus." Journal of virology 85(13): 6252-6262.

Liu, M. A. (2011). "DNA vaccines: an historical perspective and view to the future." Immunological reviews 239(1): 62-84.

Lorenzen, E., K. Einer-Jensen, et al. (2000). "DNA vaccination of rainbow trout against viral hemorrhagic septicemia virus: a dose–response and time–course study." Journal of Aquatic Animal Health 12(3): 167-180.

Lorenzen, E., N. Lorenzen, et al. (2005). "Time course study of in situ expression of antigens following DNA-vaccination against VHS in rainbow trout (< i> Oncorhynchus mykiss</i> Walbaum) fry." Fish & Shellfish Immunology 19(1): 27-41.

Lorenzen, N., P. Cupit, et al. (2000). "Three monoclonal antibodies to the VHS virus glycoprotein: comparison of reactivity in relation to differences in immunoglobulin variable domain gene sequences." Fish & Shellfish Immunology 10(2): 129-142.

Lorenzen, N., P. M. Cupit, et al. (2000). "Immunoprophylaxis in fish by injection of mouse antibody genes." Nature biotechnology 18(11): 1177-1180.

Lorenzen, N. and S. LaPatra (2005). "DNA vaccines for aquacultured fish." Revue Scientifique Et Technique-Office International Des Epizooties 24(1): 201.

Lorenzen, N., E. Lorenzen, et al. (1999). "Genetic vaccination of rainbow trout against viral haemorrhagic septicaemia virus: small amounts of plasmid DNA protect against a heterologous serotype." Virus research 63(1): 19-25.

Lorenzen, N., E. Lorenzen, et al. (1998). "Protective immunity to VHS in rainbow trout (Oncorhynchus mykiss, Walbaum) following DNA vaccination." Fish & Shellfish Immunology 8(4): 261-270.

121

Lorenzen, N., E. Lorenzen, et al. (2002). "DNA vaccines as a tool for analysing the protective immune response against rhabdoviruses in rainbow trout." Fish & shellfish immunology 12(5): 439-453.

Lorenzen, N., E. Lorenzen, et al. (2002). "Immunity induced shortly after DNA vaccination of rainbow trout against rhabdoviruses protects against heterologous virus but not against bacterial pathogens." Developmental & Comparative Immunology 26(2): 173-179.

Lorenzen, N. and N. Olesen (1996). "Immunization with viral antigens: viral haemorrhagic septicaemia." Developments in biological standardization 90: 201-209.

Lorenzen, N., N. J. Olesen, et al. (1993). "Molecular cloning and expression in Escherichia coli of the glycoprotein gene of VHS virus, and immunization of rainbow trout with the recombinant protein." Journal of General Virology 74: 623-623.

Lorenzen, N., N. J. Olesen, et al. (1988). "Production and characterization of monoclonal antibodies to four Egtved virus structural proteins." Dis Aquat Org 4: 35-42.

Lorenzen, N., N. J. Olesen, et al. (1990). "Neutralization of Egtved virus pathogenicity to cell cultures and fish by monoclonal antibodies to the viral G protein." Journal of General Virology 71(3): 561-567.

Loureiro, S., J. Ren, et al. (2011). "Adjuvant-free immunization with hemagglutinin-Fc fusion proteins as an approach to influenza vaccines." Journal of virology 85(6): 3010-3014.

Lumsden, J., B. Morrison, et al. (2007). "Mortality event in freshwater drum Aplodinotus grunniens from Lake Ontario, Canada, associated with viral haemorrhagic septicemia virus, Type IV." Diseases of Aquatic organisms 76(2): 99.

Ma, C., J. Ye, et al. (2013). "Differential compartmentalization of memory B cells versus plasma cells in salmonid fish." European journal of immunology 43(2): 360-370.

Maclachlan, N. J. and E. J. Dubovi (2010). Fenner's veterinary virology, Academic press.

Magnadottir, B. (2010). "Immunological control of fish diseases." Marine biotechnology 12(4): 361-379.

Manrubia, S. C. and E. Lázaro (2006). "Viral evolution." Physics of Life Reviews 3(2): 65-92.

122

Martinez-Lopez, A., B. Chinchilla, et al. (2012). "Replacement of the human cytomegalovirus promoter with fish enhancer and core elements to control the expression of the G gene of viral haemorrhagic septicemia virus (VHSV)." Journal of biotechnology 164(2): 171-178.

Martinez-Lopez, A., B. Chinchilla, et al. (2013). "Replacement of the human cytomegalovirus promoter with fish enhancer and core elements to control the expression of the G gene of viral haemorrhagic septicemia virus (VHSV)." Journal of biotechnology 164(2): 171-178.

Martinez-Lopez, A., P. García-Valtanen, et al. (2013). "Increasing Versatility of the DNA Vaccines through Modification of the Subcellular Location of Plasmid-Encoded Antigen Expression in the." PloS one 8(10).

McCallum, H., D. Harvell, et al. (2003). "Rates of spread of marine pathogens." Ecology Letters 6(12): 1062-1067.

McLauchlan, P., B. Collet, et al. (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 Immunology 15(1): 39-50.

Meyer, F. P. (1991). "Aquaculture disease and health management." Journal of animal science 69(10): 4201-4208.

Midtlyng, P. J., K. Grave, et al. (2011). "What has been done to minimize the use of antibacterial and antiparasitic drugs in Norwegian aquaculture?" Aquaculture Research 42(s1): 28-34.

Mikalsen, A. B., H. Sindre, et al. (2005). "Protective effects of a DNA vaccine expressing the infectious salmon anemia virus hemagglutinin-esterase in Atlantic salmon." Vaccine 23(41): 4895-4905.

Mikalsen, A. B., J. Torgersen, et al. (2004). "Protection of Atlantic salmon Salmo salar against infectious pancreatic necrosis after DNA vaccination." Diseases of Aquatic organisms 60: 11-20.

Munang’andu, H. M., B. N. Fredriksen, et al. (2012). "Comparison of vaccine efficacy for different antigen delivery systems for infectious pancreatic necrosis virus vaccines in Atlantic salmon (Salmo salar L.) in a cohabitation challenge model." Vaccine 30(27): 4007-4016.

Mutoloki, S. and Ø. Evensen (2011). "Sequence similarities of the capsid gene of Chilean and European isolates of infectious pancreatic necrosis virus point towards a common origin." Journal of General Virology 92(7): 1721-1726.

123

OIE (2012). "Viral Haemorrhagic Septicaemia." Manual of Diagnostic Test for Aquatic Animals- World Organization of Animal Health.

Olesen, N. J. (1998). "Sanitation of viral haemorrhagic septicaemia (VHS)." Journal of applied ichthyology 14(3‐4): 173-177.

Olesen, N. J. and P. Jørgensen (1986). "Detection of neutralizing antibody to Egtved virus in rainbow trout (Salmo gairdneri) by plaque neutralization test with complement addition." Journal of applied ichthyology 2(1): 33-41.

Olesen, N. J., H. F. Skall, et al. Eradication of viral haemorrhagic septicaemia in Danish aquaculture (O-129). 16th International Conference on Diseases of Fish and Shellfish.

Pierce, L. R. and C. A. Stepien (2012). "Evolution and biogeography of an emerging quasispecies: diversity patterns of the fish Viral Hemorrhagic Septicemia virus (VHSv)." Molecular phylogenetics and evolution 63(2): 327-341.

Purcell, M. K., K. M. Nichols, et al. (2006). "Comprehensive gene expression profiling following DNA vaccination of rainbow trout against infectious hematopoietic necrosis virus." Molecular immunology 43(13): 2089-2106.

Rath, T., K. Baker, et al. (2013). "Fc-fusion proteins and FcRn: structural insights for longer-lasting and more effective therapeutics." Critical reviews in biotechnology(0): 1-20.

Read, A. F., S. J. Baigent, et al. (2015). "Imperfect Vaccination Can Enhance the Transmission of Highly Virulent Pathogens." PLoS Biol 13(7): e1002198.

Reading, S. and N. Dimmock (2007). "Neutralization of animal virus infectivity by antibody." Archives of virology 152(6): 1047-1059.

Reed, L. J. and H. Muench (1938). "A simple method of estimating fifty per cent endpoints." American journal of epidemiology 27(3): 493-497.

Rice, J., M. L. Dossett, et al. (2008). "DNA fusion gene vaccination mobilizes effective anti‐leukemic cytotoxic T lymphocytes from a tolerized repertoire." European journal of immunology 38(8): 2118-2130.

Robertsen, B. (2008). "Expression of interferon and interferon-induced genes in salmonids in response to virus infection, interferon-inducing compounds and vaccination." Fish & shellfish immunology 25(4): 351-357.

124

Roopenian, D. C. and S. Akilesh (2007). "FcRn: the neonatal Fc receptor comes of age." Nature Reviews Immunology 7(9): 715-725.

Ross, K., U. McCarthy, et al. (1995). "A outbreak of viral haemorrhagic septicaemia (VHS) in turbot (Scophthalmus maximus) in Scotland." Bulletin of the European Association of Fish Pathologists 14(6): 213-214.

Salinas, I., Y.-A. Zhang, et al. (2011). "Mucosal immunoglobulins and B cells of teleost fish." Developmental & Comparative Immunology 35(12): 1346-1365.

Salonius, K., N. Simard, et al. (2007). "The road to licensure of a DNA vaccine." Current Opinion in Investigational Drugs 8(8): 635.

Schlotfeldt, H., W. Ahne, et al. (1991). "Occurrence of viral haemorrhagic septicaemia in turbot (Scophthalmus maximus)-a natural outbreak." Bulletin of the European Association of Fish Pathologists (United Kingdom).

Schönherz, A. A., N. Lorenzen, et al. (2015). "Ultra-deep sequencing of VHSV isolates contributes to understanding the role of viral quasispecies." Veterinary research.

Shinkawa, T., K. Nakamura, et al. (2003). "The absence of fucose but not the presence of galactose or bisecting N-acetylglucosamine of human IgG1 complex-type oligosaccharides shows the critical role of enhancing antibody-dependent cellular cytotoxicity." Journal of Biological Chemistry 278(5): 3466-3473.

Sitia, R., M. Neuberger, et al. (1990). "Developmental regulation of IgM secretion: the role of the carboxy-terminal cysteine." Cell 60(5): 781-790.

Sitia, R., M. S. Neuberger, et al. (1987). "Regulation of membrane IgM expression in secretory B cells: translational and post-translational events." The EMBO journal 6(13): 3969.

Skall, H. F., N. J. Olesen, et al. (2005). "Viral haemorrhagic septicaemia virus in marine fish and its implications for fish farming–a review." Journal of fish diseases 28(9): 509-529.

Smail, D. A. and M. Snow (2011). "3 Viral Haemorrhagic Septicaemia." Fish Diseases and Disorders 3: 110.

Soleimanpour, S., H. Farsiani, et al. (2015). "APC targeting enhances immunogenicity of a novel multistage Fc-fusion tuberculosis vaccine in mice." Applied microbiology and biotechnology: 1-14.

125

Sommerset, I., E. Lorenzen, et al. (2003). "A DNA vaccine directed against a rainbow trout rhabdovirus induces early protection against a nodavirus challenge in turbot." Vaccine 21(32): 4661-4667.

Sommerset, I., R. Skern, et al. (2005). "Protection against Atlantic halibut nodavirus in turbot is induced by recombinant capsid protein vaccination but not following DNA vaccination." Fish & shellfish immunology 18(1): 13-29.

Takano, T., A. Iwahori, et al. (2004). "Development of a DNA vaccine against hirame rhabdovirus and analysis of the expression of immune-related genes after vaccination." Fish & shellfish immunology 17(4): 367-374.

Thim, H. L., S. Villoing, et al. (2014). "Vaccine Adjuvants in fish vaccines make a difference: comparing three adjuvants (Montanide ISA763A Oil, CpG/Poly I: C Combo and VHSV Glycoprotein) alone or in combination formulated with an inactivated whole salmonid alphavirus antigen." Vaccines 2(2): 228-251.

Tomé, L., S. Frabasile, et al. (2012). "Selection and characterization of human respiratory syncytial virus escape mutants resistant to a polyclonal antiserum raised against the F protein." Archives of virology 157(6): 1071-1080.

Tort, L., J. Balasch, et al. (2003). "Fish immune system. A crossroads between innate and adaptive responses." Inmunología 22(3): 277-286.

Ulmer, J. B., J. J. Donnelly, et al. (1993). "Heterologous protection against influenza by injection of DNA encoding a viral protein." Science 259(5102): 1745-1749.

Utke, K., H. Kock, et al. (2008). "Cell-mediated immune responses in rainbow trout after DNA immunization against the viral hemorrhagic septicemia virus." Developmental & Comparative Immunology 32(3): 239-252.

Vallejo, A. N., N. W. Miller, et al. (1991). "Phylogeny of immune recognition: processing and presentation of structurally defined proteins in channel catfish immune responses." Clinical and Developmental Immunology 1(3): 137-148.

Walker, P., A. Benmansour, et al. (2000). "Family rhabdoviridae." Virus taxonomy: 563-583.

Walker, P., A. Benmansour, et al. (2000). "Virus taxonomy: classification and nomenclature of viruses. Seventh report of the International Committee on Taxonomy of Viruses."

126

Winton, J. (1996). "Immunization with viral antigens: infectious haematopoietic necrosis." Developments in biological standardization 90: 211-220.

Wolf, K., M. Gravell, et al. (1966). "Lymphocystis virus: isolation and propagation in centrarchid fish cell lines." Science 151(3713): 1004-1005.

Wolff, J. A., R. W. Malone, et al. (1990). "Direct gene transfer into mouse muscle in vivo." Science 247(4949): 1465-1468.

Workenhe, S. T., M. L. Rise, et al. (2010). "The fight between the teleost fish immune response and aquatic viruses." Molecular immunology 47(16): 2525-2536.

Xu, C., S. Mutoloki, et al. (2012). "Superior protection conferred by inactivated whole virus vaccine over subunit and DNA vaccines against salmonid alphavirus infection in Atlantic salmon (Salmo salar L.)." Vaccine 30(26): 3918-3928.

Ye, J., I. M. Kaattari, et al. (2013). "The teleost humoral immune response." Fish & shellfish immunology 35(6): 1719-1728.

Zaharatos, G. J., J. Yu, et al. (2011). "HIV-1 and influenza antigens synthetically linked to IgG2a Fc elicit superior humoral responses compared to unmodified antigens in mice." Vaccine 30(1): 42-50.

Zapata, A., B. Diez, et al. (2006). "Ontogeny of the immune system of fish." Fish & shellfish immunology 20(2): 126-136.

Zhang, Y.-A., I. Salinas, et al. (2010). "IgT, a primitive immunoglobulin class specialized in mucosal immunity." Nature immunology 11(9): 827-835.