AN ABSTRACT OF THE THESIS OF
Jon D. Piganellifor the degree ofDoctor of Philosophyin
Microbiology presented on June 6. 1994 Title:Development of Enteric Protected Vaccines for Aauaculture.
Abstract approved: Stephen L. Kaattari
At present all vaccines for fish are primarily delivered either by injection or immersion which introduces added stress and labor. A more attractive method of vaccine delivery is oral administration using an enteric protection system, Enteric Coated Antigen Microspheres (ECAMs), which can be utilized for a variety of antigenic forms. Relative efficacy of ECAMs was compared by inducing an antibody response in fish via injection (ip), immersion (im) or by ECAMs. Results showed that ECAMs were as effective as ip and im in inducing an antibody response to lipopolysaccharde and protein antigens. ECAMs were able to protect the protein antigen from gastric degradation. The ECAMs were employed to deliver a prototype vaccine for Renibacterium salmoninarum(Rs), the etiological agent of bacterial kidney disease. Upon characterization of the Rs antigens, one predominant cell- associated and extracellular protein (ECP), p57 was identified. The p57 molecule is elaborated in high concentrations in infected fish and exhibits pathogenic activitiesin vitro whichappear to suppress the immune response. Our studies have revealed that a 370C incubation of R.
salmoninarum cells decreased the amount of p57 by the inductionof an autoproteolytic activity. This activity was exploited to producea prototype
vaccine that was delivered by intraperitoneal injection (ip)and demonstrated a significant increase inmean time of death upon challenge by injection. A second experimentwas conducted using a natural exposure (bath challenge) and the heat treated, p57 -Rs cellswere delivered using ECAMs and ip administration. The results proved thatthe route of immunization was critical with respect to naturalexposure, as animals that received ECAMs with heat-treated p57 minus As cells demonstrated statistically significant reduction in the amount of ECPdetected versus control. The decrease in ECP concentration is indicativeof reduced Rs infection. Those animals vaccinated by ip showedno difference. Finally, in order to assess immunologicalmemory induced by antigenic stimulation, an assay was developed thatmeasured the production of cytokines induced by in vivo primingand subsequent in vitro exposure with the specific antigen. Development of Enteric Protected Vaccinesfor Aquaculture
by
Jon D. Piganelli
A THESIS
Submitted to
Oregon State University
in partial fulfillment of the requirements for the degree of
Doctor of Philosophy
Completed June 6, 1994
Commencement June 1995 APPROVED:
4 Professor of Microbiology in charge of major
He d of the Department of Microbiology
Dean of the Grad aSchool
Date thesis presented June 6. 1994
Typed for researcher by Jon D. Piganelli ACKNOWLEDGEMENTS
I would first like to thank my major professor Dr. Steve Kaattari, for giving me the opportunity to work in his laboratory and also for allowing me to finish my work in his laboratory even after he departed O. S. U. for a new appointment.I have benefited greatly from his un-selfish sharing of knowledge and for this I am indebted. Additionally I would like to thank my committee members Dr. Don Mattson, Dr. J. Mark Christensen, and Dr. Jerry Heidel for helpful comments on this thesis.I especially would like to thank Dr. Jo-Ann Leong and Dr. John Rohovec for looking out for the well being of the lab after Steve left. I would like to thank my fellow lab mates that had to put up with me all these years including Leslie "The Gilkster" Gilkey, Janel"Janelo" Bishop, Dan "The Rockhead"Rockey, Jenefer" The Queen" DeKoning-Loo, Mark "King Spamon" Adkison, John"The Hippie" Hansen, Hank "Hankenstien" Ortega, Patty "Woody" Wood, Dave"The Bigman" Shapiro, oh! lets not forget the ageless wonder Don Chen. Greg Wiens, I can't thank you enough for allowing me to work with you on these projects..... THANKS! To the members of the Leong Lab and especially Barb Drolet for her artistic ability. Dan "Dano" Mourich who taught me almost as much Immunology as Steve. Thanks for being a great friend. Greg "Oh what are those Raymond" Rutherford another great friend. The Rohovec lab ODFW, Harriet. The professional women in the office Carlene, Joy, Jenny, Marylin, Linda, and of course Bonnie, I owe most of my Scientific life to you guys. To my colleague and friend in the Pharmacy department Jia Allen Zhang thanks for all of your help and collaboration. I would like to thank my parents Lucille and Rocco for, always believing in me and letting me express myself. Also, for instilling in me the gift of faith in myself and in God and for their many prayers that I truly know have blessed me throughout my entire life.I Love you both very much. To Timmy, Chrissy and Rocco, my older brothers and Roxanne my older sister thank you for all your prayers and support throughout my years in school. Thanks to Dick and Susie my in-laws for their support. To my friends back home Joe Ed, Russ, Mitch, Mike, Dweeze, Kevin, Sam, Daddy "G". Specifically, I would like to thank my best friend, my wife Cheryl for her unconditional support and love without which I would not be complete. Finally, I would like to thank the one and only and the one and only knows who that is. To be given the gift to do Science is only the tip. For whom you choose to do it for is the iceberg...... THANKS LORDI CONTRIBUTION OF AUTHORS
Dr. S. L. Kaattari was the major advisor on all manuscripts. Dr's. J. M. Christensen and J. C. Leong were minor advisors on those manuscripts that their names appeared on. Jia Allen Zhang was responsible for preparation and feeding of the ECAMs to fish described in Chapter 3 and 5 and also participated in collecting samples. Dr. Greg Wiens was responsible for Tables 4.1, 4.2 and some of the SDS-PAGE work from Chapter 4. He also prepared some of antigen for vaccination in Chapter 5 and helped vaccinate some of the animals. Janel Bishop, Dr. R. Stevenson and L. Gilkey participated in collecting samples in Chapter 5. Dan Mourich was responsible for the fractionation and radiolabeling experiments in Chapter 6, as well as for helping with the oxidative burst assays. Dr. Linda Bootland was responsible for the anti-viral assays in Chapter 6. TABLE OF CONTENTS
CHAPTER Page
1- INTRODUCTION 1
2- LITERATURE REVIEW 3
Aquaculture and Associated Disease Problems Vaccines Primary Cells of the Specific Immune Response The Macrophage / Antigen Presenting Cell B-Cells T-Cells Gut Associated Lymphoid Tissue Viral Vaccines Infectious Pancreatic Necrosis Virus Viral Hemorrhagic Septicemia Virus Infectious Hematopoietic Necrosis Virus Bacterial Vaccines/ Vibrio anguillarum Yersina ruckeri Aermonas salmonicida Renibacterium salmoninarum Vaccination Methods / Delivery Systems
3- Enteric Coated Microspheres As An Oral Method for Antigen Delivery to Salmonids 47
ABSTRACT 48 INTRODUCTION 49 MATERIALS AND METHODS 51 RESULTS 55 DISCUSSION 56 ACKNOWLEDGMENTS 60 4- Activation of an endogenous serine protease as a novel method for removal of p57 from the Renibacterium salmoninarum cell surface 66
ABSTRACT 67 INTRODUCTION 68 MATERIALS AND METHODS 71 RESULTS 73 DISCUSSION 75 ACKNOWLEDGMENTS 77 CHAPTER
5- Evaluation of a protease-modified, Renibacterium salmoninarum whole cell vaccine, deliveredorally and intraperitoneally 84
ABSTRACT 85 INTRODUCTION 87 MATERIALS AND METHODS 90 RESULTS 96 DISCUSSION 98 ACKNOWLEDGMENTS 102
6- Assessment of immunologicalmemory to a T-dependent antigen in rainbow trout (Oncorhynchus mykiss) by thein vitro elicitation of a respiratory burst enhancing cytokine 108
ABSTRACT 109 INTRODUCTION 110 MATERIALS AND METHODS 112 RESULTS 118 DISCUSSION 121 ACKNOWLEDGMENTS 125
7- CONCLUSIONS 131
BIBLIOGRAPHY 135 LIST OF FIGURES
CHAPTER 2 Page
2.1. Diagram of the gastrointestinal tract of salmonid species. 16
2.2. Enteric Coated Antigen Microspheres (ECAMs). 45 CHAPTER 3
3.1. Serum antibody titers expressed in units / ml ofserum from coho salmon immunized with TNP-LPS in the form of. 61
3.2. Serum antibody titers expressed in units / ml ofserum from coho salmon immunized with TNP-KLH in the form of. 63 CHAPTER 4
4.1. Total protein stains of R. salmoninarumcells (A)or supernatants (B) after 16 hour treatment at370C(lane2), 17°C(lane3), 40C (lane 4), or -200C (lane 5), molecular weight markers (lane 1). 78
4.2. Total protein stain (A) and western blot using the anti p57 monoclonal antibody 4D3. (B) of R. salmoninarum cells after co-incubation with PMSF and 370C treatment. 79 4.3. Western blot using the anti p57 monoclonal antibody 4D3 of 370C treated R. salmoninarum cells with culture supernatant (ECP). 80 CHAPTER 5
5.1. Total protein stain(A)and western blot (using the monoclonal anti p57 antibody 4D3)(B)of R. salmoninarum after 370C treatment for 48h followed by 0.3% formalin incubation at 170C for 10 h: molecular weight marker (lane1),untreatedR.salmoninarum cells (lane 2), 370C treated R. salmoninarum cells (lane 3). 103
5.2. Percent survival of coho salmon immunized with saline / FIA(A), ECP / FIA (B), CSE / FIA (C) and p57- cells / FIA (D). 104 CHAPTER 6
6.1. Fluorescence measurement of oxidation of DCFH by rainbow trout pronephric macrophages (PM) in the presence of supernatant factors and the triggering agent (1 ug/ml PMA) after 24 hour incubation with supernatants derived from TNP-KLH ( ),ovalbumin (®), and non-antigen conditioned supernatants (®)cells. 126
6.2. Fluorescence measurement of oxidation of DCFH by rainbow trout HK macrophages in the presence of 1 ug / ml PMA after 24 hour incubation with centricon fractionated supernatants. 128
6.3. Fluorescence measurement of oxidation of DCFH by rainbow trout HK macrophages in the presence of 1 ug / ml PMA after 24 hour incubation with TNP-KLH derived supernatants from enriched cultures. 129
6.4. SDS-PAGE fluoregraph of soluble S35 methionine labeled proteins. 130 LIST OF TABLES
CHAPTER 3 Page
3.1. Immunization schedule for TNP-LPS 64
3.2. Immunization schedule for TNP-KLH 65 CHAPTER 4
4.1. Minimum molar concentrationof ammonium sulfate required for demonstratableR. salmoninarumaggregation at various times and temperature. 81
4.2. Minimum ammonium sulfateconcentration required to effect agglutination of R.salmoninarumafter incubation with PMSF at 370C. 82 4.3. Reconstitution of relative cell surface hydrophobicity by concentrated culture supernatant. 83
CHAPTER 5
5.1. Values expressed are themeans of p57 detected (ng/ml) for each particular treatment at eachsampling date 106
5.2. Values are the means of antibodyunits of activity/,ul serum detected for each particular treament at eachsampling date 107 Development of Enteric Protected Vaccines forAquaculture
CHAPTER 1 INTRODUCTION
Immunoprophylaxis has always been the method of choicefor controlling outbreaks of infectious disease in human medicine.However, the use of vaccination has only recently been acceptedas a viable option to control diseases that plague the aquaculture industry. The firstattempt to vaccinate fish dates back to 1930. Later Duff (1942)attempted to orally vaccinate fish against Bacterium salmonicida.It was not until the mid 1970's that vaccines were commercially produced for theaquaculture market (Fryer, et al., 1977). These early vaccines were bacterinscomposed of Gram negative bacteria. The fish were exposed to these bacterinsby immersion or intraperitoneal injection. Although theseroutes of delivery were successful in affording protection, they resulted in increasedstress incurred by the fish and added labor for the aquaculturist. Amore profitable and efficacious method of delivery for vaccinating fish would be oral administration.The least stressful way to deliver antigens to fish is also via oral delivery (Agius et al., 1989)and immunizing by oral administration is not limitedto size or numbers of fish as are immersion and injection regimes (Fryer et al., 1976). Although oral delivery is by far the most feasible method for fish vaccination,early methods resulted in inconsistencies and non-significant protection(Hart et al., 1988). This could be attributed to degradation of pH-sensitiveantigens by the gastric portion of the fish gut. Because of this chemical barrieroral delivery has remained an under utilized approach. 2
The goal of this research was to developan oral delivery system based on enteric-protected vaccines. These Enteric Coated Antigen Microspheres
(ECAMs) are administered by incorporation in the feedand are protected from the gastric portion of the gut through theuse of pH-reactive polymers. The pH- reactive polymers exhibit stability at low pH protecting. theantigen in the fish stomach (Wong et al., 1991), permittingpassage of antigen into the intestine.
Upon entrance into the intestine, the increase in pH results indissolution of the polymer and release of the antigen to gut associated lymphoid tissue(Wong et al., 1991). The first manuscript, Chapter 3compares the relative efficacy of immunization using ECAMs versus intraperitoneal (ip)and immersion delivery at inducing an antibody response to defined hapten-carrier conjugates.
Chapter 4 describes a novel method of activationof an endogenous serine protease that was exploited to remove p57,a cell associated protein with known immunosuppressive activities, from Renibacteriumsalmoninarum (Rs). Renibacterium salmoninarum is thecausative agent of one of the most prevalent diseases of cultured salmonids, bacterial kidneydisease (Fryer and Sanders, 1981). The eventual goalwas to develop a candidate antigen for vaccination. Chapter 5 describes the efficacy of thep57 - Rs cell as a vaccine. The p57 - Rs cell was delivered using ECAMsor ip injection. Chapter 6 describes an assay that assess immunologicalmemory by detection of a secreted cytokine, with suggestion for adaptation of theassay to study the induction of T-cell memory after vaccination. 3
CHAPTER 2
LITERATURE REVIEW
Aauaculture and Associated Disease Problems
In the last twenty years, aquaculture has increased intoa substantially large industry, encompassing many parts of the world (Ellis, 1988).The farming of marine species is an ancient practice with the firstrecorded attempts dating back to China 2000 B.C., since thenman has cultured many different marine species (Leong and Fryer 1993). The Romanswere the first Europeans to conduct organized aquaculture with the growingof oysters (Avault,1987).
In most developed countries, fish are farmed intensively. Undersuch conditions there is the propensity for high density farming. Anytimethis type of animal husbandry is conducted disease problemscan easily become severe (Ellis, 1988). The increase in aquaculture, coupled with thedecrease in water allotments has created intensive farming conditions.It is this factor that contributes to the increase of multiple waterusage which in turn spawns high density culture conditions. Such intensive culture results instress, ultimately leading to immune-dysfunction anda greater incidence of disease
(Wedemeyer, 1970; Pickering and Pottinger, 1985; Pottingeret al., 1992,
Fevolden et al., 1992; Wise et al., 1993). The increase in farmingis growing at a rate of 8% per year and estimates predict that by the year 2000 the market will have grown from 10 million to 20 million metric tons of product(J., Plumb, USDA Fish Health Research Priorties, 1992). 4
To meet this increased market demand,more efficient methods of raising disease-free animals will be needed. Thiscan be accomplished using more efficient animal husbandry techniques and by controlling disease
problems. Diseases that plague the industrycan be controlled by a number of methods such as; antibiotic administration (eitherprophylactically or therapeutically), breeding for disease resistance,the use of vaccines and stress tolerant strains of fish. The application of antibioticshelps to control many bacterial pathogens, but there are many pitfalls associated with their
overuse. Antibiotic usage can lead to bacterial drug resistance, resulting inthe more intense use of new antibiotics. This, in turn, leads toa cyclic problem and inevitably the generation ofmany dangerously resistant pathogens. No chemotherapeutics exist for the control of the viraldiseases that affect the industry.
The increase in aquaculture production and thelimited amount of feasible avenues to control the escalating diseaseproblems associated with it, necessitates the development of vaccines foraquaculture. Fish health is critical to the well-being of the industry and effectivevaccines are desperately needed. The losses due to diseaseare staggering. For example, 77% of cultured catfish were lost betweenegg hatch and market size, primarily due to disease to disease (J., Plumb, USDA Fish HealthResearch Priorties, 1992). The U.S. Trout Farmers Association Boardestimates that 10% of all cultured aquatic animals are lost as a result of infectious disease(J., Plumb, USDA Fish Health Research Priorties, 1992).
The earliest recorded attempt to immunize fish againsta fish pathogen, Vibrio anguillarum, dates back to the 30's. Later, Duffin (1942) reported on the oral vaccination of trout againstBacterium salmonicida.Although Hayashi et 5
al. in (1964) reported on the successfulvaccination of trout against vibriosis by intraperitoneal injection itwas not until the mid to late 1970's that highly effective commercially marketable vaccineswere produced (Fryer et al., 1977). These vaccines were againstVibrio anguillarum(vibriosis) andYersinia ruckeri(enteric red mouth disease). These vaccineswere relatively inexpensive to produce and could be deliveredin a fashion allowing mass administration by simply dippingor bathing fish in the crude, killed broth
culture (Fryer et al., 1977; Amend et al., 1981). Thesuccess of these vaccines induced many investigators to attemptvaccination against all other major diseases of cultured fish. Theirsuccess in achieving the same results was not as consistent. However, the intensification of researchon the subject led to an increase in beneficial results, suchas improved commercial vaccines that offer higher levels of protection and longer durationof immunity.
Vaccines
A vaccine is a preparation of antigenic materialthat is used to artifically induce an aquired active immuneresponse. Because effective vaccination requires induction of long lasting immunity, theability of vaccines to stimulate both B and T lymphocytes is an importantconsideration in vaccine design (Good et al., 1988). The stimulation of immunecells leads to adaptive immunity, which is specific for the particularpathogen and persists for a long period of time (immune memory) (Gray, 1993). Evidencefor this type of long lasting stimulation of immunity has been demonstratedin fish by Evelyn
(1971). He demonstrated that sockeye salmon (Oncorhynchusnerka) responded to injection of whole Renibacterium salmoninarumcells by producing detectable antibody titers for 16 months. Thirteenmonths after the 6
primary injection, the fish were boosted and bled threemonths later. Those animals given a booster injection demonstratedan enhanced secondary antibody titer while animals that had receivedan initial injection of antigen showed no significant titer. Trump and Hildemann (1970)using bovine serum albumin (BSA), were able to demonstrate that gold fish(Carassius auratus) produced a secondary antibody responsein vivo.They saw that the magnitude of the secondary antibody response increasedand the lag period decreased in comparison to the primaryresponse. Rijkers et al., (1980) demonstrated that dose and route of administrationwere important for the generation of a memory response. They found after intramuscular(IM) priming of carp (Cyprinus carpio) with 109 sheep red bloodcells (SRBC) a high primary response was seen, however the secondaryresponse was low. When fish were primed IM with 105 SRBCs the primaryantibody response was low, however when the animalswere boosted 6 months later they produced higher titers than those animals in the 109 injectiongroup. Vaccines take advantage of two essential elements, specificity andmemory, the adaptive immune response is stronger upon a second encounter with the antigen. Themajority of vaccines made today use antigens derived fromthe pathogenic organisms rendered non-pathogenic by inactivation withchemical reagents such as formaldehyde (Rohovec, 1975; Johnsonet al., 1983; Amend et al.,1983; Bloom 1989), and Beta propiolactone(Chanock; 1984; Bloom,1989)
It is important to have an understanding of the mechanismsinvolved in the overall immune response toan antigen in order to design a protective vaccine for the host before natural insult. However,a lack of immunological understanding in the development ofa vaccine may even lead to an exacerbated pathological conditionupon vaccination. For example, 7
stimulation of a humoral immuneresponse may lead to antibody dependent enhancement (ADE) as seen in HIV infection (Homsyl,et al., 1989). To fully understand what vaccines accomplish, it isnecessary to have some background informationon how the fish immune system reacts to "priming" with an antigen (vaccine). The fish immuneresponse to antigenic stimulation has been reviewed by many authors ( Anderson,1974; Corbel,1975; Ingram,1980; Austin and Austin, 1987; Ellis,1988) and to discuss it in full detail is beyond the scope of this review, yet it isessential to give some background information on important cells and their functionsinvolved in the piscine immune response.
Primary Cells of the Specific Immune Response
All immune responses begin with the activationor induction of immune cells. The immune response in fish is generallyattributed to the same cell-cell interactions as seen in mammaliansystems (Clem et al., 1991). The review will focus on the primary cells involved in thepiscine adaptive immune response. This response enlists three distinct cell types;macrophages, B-cells and T-cells, each of which playsa critical role separately and together to produce an intricate network of signals that leadsto successful immune system activation.
The Macrophage / Antigen Presenting Cell
The macrophage is the most essential cell of theimmune system.It assists both B and T cells in generation of thespecific immune responses and 8
its function can be modulated by bothantibody and T- helper, cells (Street and
Mosmann, 1991).It can also function independent of other cells ina non- specific manner by engulfing and destroyinginvading microbes and other pathogenic agents (Secombes, 1992). This functionis not altered by specific vaccination but can be enhanced, non-specifically,by the use of immunomodulators suchas adjuvants (Anderson, 1992).
The importance of antigen processing andpresentation in eliciting an immune response to T-dependent antigens,(i.e. protein antigens), as in higher vertebrates, is also essential in phylogenetically lowervertebrates such as teleosts (Vallejo et al., 1992).Antigens are processed and presented by antigen presenting cells (APC), suchas monocytes and macrophages, to specific lymphocytes in a MHC, restricted fashion(Yang et al. 1989, Vallejo et al., 1990). Processing proteins isan essential prerequisite for recognition by
T-cells, and ultimately, the induction of antigen-specificT-and B effector functions. These eventsare also critical in the fish immune response. Vallejo et al., (1990) demonstrated that antigen pulsedand paraformaldehyde-fixed peripheral blood lymphocytes (PBLs) from in vivoprimed fish could elicit specific proliferative and antibodyresponses in vitro by autologous PBLs. Prior fixing of these PBLs (before pulsing)abolished the proliferative response. Methods that enhance the uptakeor phagocytosis of antigens should result in enhanced immunity. For example, the particularizationof soluble antigens by adsorption of the antigen onto bentonite allowsthis particulate state to be achieved and consequentlya more rapid uptake and processing of the antigen
(Stott et al., 1986). Although macrophages havemany different functions the two significant roles relevant to immune recognitionare providing the mechanisms for antigen processing, and beingequipped with the MHC class II molecules essential for presentation of processedantigens to specific 9
lymphocytes (Unanue, 1984). Thesecells also generate the necessary cytokine (IL-1) crucial for growth andcostimulatory signals for T-cell activation (Weaver and Unanue, 1990; Ortega, 1993).
B-Cells
The fish B-lymphocytes have been distinguished,as in mammalian systems, as those lymphocytes thatexpress immunoglobulin on their surface and secrete specific antibody inresponse to antigenic stimulation (Smith et al., 1967; McKinney et al.,1977; Simaand Vetvicka, 1990 ). These particular cells have been identified ina number of fish species. Although attributes of piscine and mammalian B-cells are similar, physicalcharacterization of the cells, their distribution, the form of theirresponse, and the antibody product itself do not follow the mammalian model (Kaattari,1992). The ontological development of fish B-cells is still not completely understood.Some evidence indictates that the anterior kidney in fishserves the same function as bone marrow, and is possibly the source of lymphocytic stemcells.Intercellular cooperation of these cells with other cells and the lymphokinesrequired for immune function closely parallels the mammalian model(Kaattari, 1992). Also, the antibody molecule is the actual effector molecule,and its function differs from the action of the T-cell and the macrophage whichdirectly exert effector functions. Antibody can protect the host througha variety of mechanisms. The simplest is neutralization. This is achievedwhen the antibody binds to an antigen and sterically hinders the interactionof that antigen with its particular receptor or cell. For example, the bindingof antibody to toxins inhibits the effect that the toxin may have hadon the host cell if allowed to bind (Abbas,
1991). Neutralization alsooccurs when antibody binds to virus, preventing the 10
virus from binding to its specific hostcell receptor. Antibodies also bind to adhesins of bacteria, these anti-adhesinantibodies can prevent attachment of
the bacteria to host mucosal sites (Jacobset al., 1989 ).Studies of anti- adhesin antibody have warranted further investigationinto the possibility of oral vaccination with the goal of elicitingsecretory immunity in piscine species (Rombout et al., 1989, Rombout and VanDen Berg, 1989). Antibody molecules complexed with antigen leadsto a process called opsinization.
Opsinization serves to enhance the efficiencyof phagocytosis of antigens by monocytes and macrophages (Griffin, 1983). Antibodiescan also agglutinate and precipitate pathogens and their antigens,leading to an enhanced removal of the aggregates by the reticuloendothelialsystem of the host. Recent evidence has revealed that theremay be isotypic differences in fish Immunoglobulin as is seen in mammals(Lobb and Clem, 1982;Tomanaga et al.,1984; Kobuku et al., 1987; Ghaffariand Lobb, 1989). However, functional
differences in isotypes have yet to bedemonstrated in teleosts (Wilson and Warr, 1992).
An important function of B-cells is the abilityto form a memory pool of antigenically primed cells. Upon primaryantigenic stimulation a responding population of cells produces activated B-cellsthat differentiate into antibody forming cells (AFC) (Sprent, 1994). Anotherportion of these cells forms a
memory pool of primed but non-antibody secreting cells whichare specific for the same antigen. These cells remainquiescent until secondary stimulation
with this antigen, at which time they producespecific antibody faster and at a higher magnitude than in the primaryresponse (Sprent, 1994; Gray, 1993). This response is enhanced and is contingentupon two conditions: (1) the participation of T-cells and (2) that the antigencarrier is a protein (T- dependent), either soluble or cell associated(Gray and Sprent, 1990; Vitetta et 11
al., 1991; Tittle, 1978).B-cell memory in fish has been demonstrated (Arkoosh
et al., 1991) however, in contrast to mammalian immunologicalmemory, there is no evidence for affinity maturation during theprimary of secondary response (Wilson and Warr, 1992; Arkoosh et al., 1991). Thereis, however, a decrease in lag time before elicitation of the antibodyresponse to the same antigen (Arkoosh et al., 1991; Cossarini-Dunier, 1986).
T-Cells
It generally agreed that the earliest. phylogeneticmanifestations of the adaptive immune system occurs in fish. The ability offish to produce antibody and reject allografts demonstrates that possession ofboth humoral and cell mediated immunity exists ( Goodrich and Nichols, 1933).That two lymphocyte populations exist in fish has been suggested by studiesinvolving the use of mitogens and existence of the hapten carrier effect (Stolenand Makela, 1975; Cuchen and Clem, 1977; Etlinger et al., 1976). Stolenand Makela, (1975) demonstrated that winter flounder thatwere pre-immunized with a carrier protein prior to receiving the haptenated carrier showedan enhanced antibody response to the hapten. However preimmunization witha different carrier protein demonstrated no enhancement. Theirexplanation of these results was that carrier specific lymphocyteswere primed from the initial carrier injection and they provided help to hapten specific lymphocytesas is seen in higher vertebrates. Proof of the existence of T-cellswas also obtained fromin vitro studies using mitogen activation.In the mammalian model, the various distinctive B-cell mitogens lipopolysaccharide (LPS),poke weed mitogen (PWM) and T-cell mitogens conconavilin A (Con A)or phytohemagglutinin
(PHA) are well known (Harwell, et al., 1976). Mitogenicstudies conducted in 12
the teleost have also providedeven more definitive proof for two lymphocyte populations. Etlinger et al., (1976), demonstrated thatthere was organ specific distribution of mitogenic responsiveness analogousto that in mammals. Studying rainbow trout thymocytes, they provedthat these cells only responded to T-cell mitogens while the cells fromthe anterior kidney responded only to the B-cell mitogens. Ellsaesseret al., (1988) also demonstrated that cells found in the thymus of channelcatfish were predominantly surface immunoglobulin negative (slg-),and these cells only proliferated in the presence T-cell mitogens. Clemet al., (1984) demonstrated the differential temperature sensitivity of T-cells by mitogenicstimulation. The level of response to LPS (B-cell mitogen)was unaffected by the sub-optimal temperature (170C) while the level ofresponse to the T-cell mitogen was suppressed at this temperature. These effects could beovercome by first reacting the cells with Con A ata higherin vitrotemperature. Unequivocal evidence that fish possessed T-lymphocyteswas provided by Miller et al., (1987). These investigators produceda monoclonal antibody designated mAb 13C10, that exclusively reacted with channelcatfish sig- cells. The reagent reacted with most thymocytes, neutrophils, thrombocytesand brain cells. The authors concluded that this antibody reacts with relativelyhigh molecular weight antigens on channel catfish T cells and thatit is an anti-T cell reagent that reacts with a marker comparable to mammalianThy-2.
Evidence for accessory cell function and lymphokinesecretion by fish T- cells as is seen in mammals was demonstrated by Caspiand Avtalion (1984). They reported that activity promoting the growthof carp T-like cells (Cyprinus carpio)was found in supernatants of PHA or allogenic stimulatedcarp leukocyte cultures. The levels of the activitywere increased with the addition of phorbol myristate acetate, (PMA)a compound that mimics a step in the 13
pathway of T-cell activation. This stimulating effectwas also seen when mammalian interleukin-2 (IL-2) was added to the cultures. Thegrowth promoting factor only demonstrated activity in antigen-stimulated,but not resting, carp T-like cells. This resultwas consistent with what has been seen in mammals, (i.e. antigen-stimulated cellsare dependent on exogenously produced interleukin-2 for proliferation, Watson, 1979). Grahamand Secombes (1990) studied the cellular requirementsof lymphokine secretion in rainbow trout (Salmo gairdneri) leukocytes. Theyassessed the ability of different subsets of leukocytes to produce macrophgeactivating factor (MAF) following activation with ConA/PMA. The authors demonstratedthat separation of of leukocytes into surface immunoglobulin positive slg+and surface immunoglobulin negative slg- cells by panning showedthat only the slg- cells could produce the MAF activity and macrophageswere necessary as accessory cells. However, macrophages alone were unable to produce MAF.
These results support the contention that fish lymphocytescan be divided into slg- T-cells and slg+ B-cells and that lymphokinesecretion can be added to the function of helper activity of these T-cells.
Gut Associated Lymphoid Tissue (GALT)
The mammalian gut associated lymphoid tissue (GALT)especially Peyers'patches (PP) plays an important role in the immunological defense against pathogens entering the the gastrointestinal tract (Rombout,et al., 1989). Specialized cells (M-cells), present in the epitheliumoverlying the PP, transfer antigens to the mucosal lymphoid tissue in which lymphocytes(B and T), macrophages and follicular dendriticor interdigitating cells co-operate in response to antigen (Rombout et al., 1989). In other vertebrate species, less 14
defined but obvious lymphoid aggregates exist in thegut. For example, Zapata and Solas (1979), demonstrated, in reptiles,numerous aggregates of lymphoid tissue occuring along the entire gut.In the amphibian urodele, Pleurodeles waltl (salamander) large lymphoid aggregatesappear throughout the gut,
especially in the small intestine, as true cell infiltrates within the laminapropria,
similar arrangements have been reported for mammalian Peyer'spatches (Ardavin et al., 1982). Work by Tomonaga et al., (1986)on elasmobranchs demonstrated that massive lymphocyte aggregationswere located in the central region of the spiral intestine. These aggregationwere thought to be primitive forms of mammalian Peyer's patches, and consistedof large intraepithelial leukocytes.
In the teleost, Pontinus and Ambrosius (1972) demonstratedantibody secreting cells in the lamina propria of the pyloric region inperch after immunization with sheep red blood cells (SRBC). Fletcherand White (1973) demonstrated increased antibody titers in the intestinalmucus upon oral immunization with heat killed Vibrio anguillarum in plaice(Pleuronectes platessa L.) Oral administration, of Vibrio bacteria, tocarp (Cyprinus carpio) resulted in an increase in the intraepithelial leukocytesalong the entire intestine of the animal ( Davina et al., 1982).
Although fish do not possess specialized aggregatesof lymphoid tissue analogous to Peyer's patches, they dopossess minor subepithelial lymphoid accumulations which have been demonstrated in the roach, Rutilusrutilus, and perch Perca fluviatilis (Rombout and Van den Berg,1989, Zapata and Solas, 1979).It has been shown that carp do possess scattered, lymphoid-like cells in the epithelium or lamina propria (Davina et al., 1980),but no true cell clusters were found. In most telosts, the second gutsegment specializes in the uptake and processing of antigens (Figure 2.1) (Noaillac-Depeyerand Gas, 15
1973; Strobard and Kroon, 1981; Strobard and Vander Veen,1981; Nagai and Fujino, 1983; lida and Yamamoto, 1985; Georgopoulouet al.,1985,1986; Fuijino et al. 1987, Hart et al., 1988; Mclean and Donaldson,1990). After this uptake and processing of the antigens, the antigensappear in intraepithelial
macrophages located in the second gut segment (Romboutet al., 1985).It is likely that cells called enterocytes in the second gutsegment of carp are analogous to the M-cells in mammals andserve the analogous function of
sampling antigens from the gut environment (Romboutet al.,1993).It has been demonstrated that antigen-specific antibody appeared inthe skin mucus of orally vaccinated fish, but detection ofserum antibody was minimal (Fletcher and White, 1973; Kawai et al., 1981; Lobb, 1986; Romboutet al., 1993). The presence of mucosal antibody without concomitant serum antibody suggests that there are B-cells and plasma cells in themucosa, this would explain the mucosal response seen upon oral immunization. Romboutet al. (1993) reported, using monoclonal antibodies and single and doublecell-staining techniques, that intraepithelial lymphoid cellsare composed of slg- putative T and, NK cells, slg+ (B) cells and Ig-binding, antigen-presentingmacrophages. Davidson et al. (1993), have reported that the routeof immunization determines whether antibody secreting cells will be generatedin the gut of rainbow trout (Oncorhynchus mykiss). Specific antibodysecreting cells were detected in both head kidney (HK) and intestinalmucosa after either intraperitoneal (ip) injection or oral intubation ofAeromonas salmonicida antigen. Antibody was detected in both HK and intestinalmucosa but, depending on the route of delivery, important differenceswere seen. 16
Figure 2.1. Diagram of the gastrointestinaltract of salmonid species. The large intestine is responsible for uptakeand transport of antigens. Enlargement shows lumen where antigensare transported to large intraepithelial macrophages in the columnarepithelium (ce). Macrophages process and present antigens to other lymphoid cells.Although clear Peyer's patch-like structuresare not seen many lymphoid like cells are present, scattered throughout the large intestine(McLean and Donaldson, 1990). 17
Pyloric caeca
.columnar epithelium
laminpropria
basement membrane
Figure 2.1 18
Intraperitoneal (ip) injection produced antibody secreting cells in both HKand intestinal mucosa with peak antibody secretion at two weeks in the HK,but not until week 7 in the intestinal mucosa. The peak antibody production in animals in the oral group was apparent three weeks after immunization in both the intestinal mucosa and head kidney (HK). A second peak of antibody secretion was observed in those animals which were ip injected and may be attributed to adjuvant use. No secondary antibody peak was observed in the peroral group but this group did not receive adjuvant with their immunization (Davidsonet al., 1993). Davidson's data suggest that a separate mucosal immunesystem
exists in rainbow trout, as in other teleosts. Further evidence for the theory ofa separate mucosal immune system has been demonstrated by the differential magnitude of the response seen in the gut, paralleling that which isseen in the head kidney of ip immunized animals. The mucosal immunecompartment seems to be an active producer of antibody secreting cells which possesses different kinetics than the systemic immune compartment. In conclusion, the evidence presented suggests that all cellsnecessary for a local or mucosal immune response appear to be present in the gastrointestinal tract of the teleost fish.
Viral vaccines
There are currently three types of antigens used in viral vaccines; killed or attenuated viruses and antigen sub-units. Killed vaccines are prepared by growing large quantities of the virus in cell culture (Fryer et al., 1976)or in live animals. The virus particles are then purified from the cells and inactivated with either formalin (Fryer et al., 1976), beta propriolactone ( Amend, 1976; White and Fenner, 1986), tri-N-butyl phosphate (Bloom, 1989),or urea. Care 19
must be taken to ensure that the virus is rendered inactive, yet stillpossesses full immunogenic activity.Killed vaccines are thought to be safer than
attenuated vaccines as there is no chance of the virus reverting toa virulent form. The disadvantage associated with killed viral vaccines is the need for repeated immunization. These vaccines are particulate antigens and,
therefore are processed by macrophages and presented in the context of class II MHC molecules stimulating a strong humoralresponse but very weak cellular response. As such killed virus does not exploit MHC I presentation,
which requires a live virus and stimulates T cytotoxic cell activation dueto endogenous processing of the antigens.
Attenuated vaccines have the advantage of mimickinga natural infection by the wild type virus. The attenuated vaccinesare constructed to infect the host and stimulate both humoral and cellulararms of the immune system. However, either single or double mutations/deletions,are required so that the virus can no longer replicate satisfactorily and will fail tocause disease (White and Fenner, 1986). This type of vaccination has been successful in mammals (Chanock and Lenner, 1984) and has the advantage of delivering protective antigens to the appropriate site and in theproper context to stimulate the correct immune response. Although the live vaccines stimulatecomplete immunity, there are some major draw backs thatare associated with their use. There is always a possibility that the virus will revert to the virulent wildtype strain (Fryer et al.,1976; Rohovec, et al., 1981). The virusmay be avirulent in the species it was specifically designed to vaccinate but in another bystander species it may be highly virulent causing widspread infection (Fryer et al.,1976;
Rohovec, et al., 1981). Caution must be exercised if live virusesare to be used in aquaculture to ensure that the water used for vaccination is not releasedas normal effluent flow until necessary decontamination proceduresare 20 employed, release of unsterilized water could allow for live viruses to be released into the watershed. Subunit vaccines or recombinant vaccines have been derived through the use of cloned DNA encoding the sequences for antigenic determinants. This DNA is isolated and cloned into bacteria, yeast, or insect cells usinga baculovirus vector (Gilmore et al., 1988; Koener and Leong, 1990). The first subunit vaccine was developed by cloning the gene for the major surface antigen of hepatitis B virus (HBs Ag) in yeast cells. The first successful recombinant vaccine was developed for the major surface antigens (vpl) of foot-and-mouth disease virus (Kleid, 1981). The advent of cloning these antigens using recombinant vectors has also led to the use of synthetically derived peptide vaccines that are constructed from a geneticsequence encoding specific B and T cell epitopes. Fusion of such sequences results ina linear stretch of amino acids. The advantage which peptide vaccines have over live attenuated and killed vaccines is that a specific immune response against the pathogen can be induced without exposing the host to the pathogen (Hilleman, 1985). The disadvantage of recombinant synthetic vaccines is that contaminating products, such as lipopolysaccarhide from the recombinant vector may cause toxic effects when administered and if not
expressed on a live virus vector it won't make use of MHC I presentation.
Infectious Pancreatic Necrosis Virus
Infectious Pancreatic Necrosis Virus (IPNV), was first identified by (Wolf et al., 1960). IPNV is an acute contagious systemic birnavirus disease of fingerling-trout fry (Wolf, 1988). The virus is of worldwideconcern as it has been recognized in every country where salmonids are present (Dorson,1988). 21
The virus was first isolated from brook trout (Salvelinusfontinalis) (Wolf et al., 1960) suffering from a severe disease. The infected fishdemonstrated the following clinical signs: swollen belly, loss ofequilibrium, whirling and corkscrew swimming followed by death (Mi Gonigle,1941). The disease
derives its name from the gross necrosis of theexocrine pancreas as seen upon histopathological examination. Although the diseasewas first seen in the brook trout, IPNV is a serious pathogen ofmany farm-reared salmonids;
rainbow trout (O.mykiss) (Parisot et al.,1963), browntrout (Salmo trutta)
(McKnight and Roberts,1976) and cutthroat (O.clarki)(Parisot et al.,1963).It can also cause mortality in amago salmon (0. rhodurus) (Sanoand Yamasaki, 1973), sockeye salmon (0. nerka) (Sano and Yamasaki,1973)and Arctic char (Salvelinus alpinus) (Ljungberg and Jorgensen,1973). IPNV kills these animals when they are in the fry stage of development.Shifts in water temperature play a role in the development of thedisease, and IPNV can be considered a cold water disease, occurringat temperatures below 160C although this may not be true for all virus strains(Frantsi and Savan, 1971;
Dorson and Torchy, 1981). The survivors of IPNVinfection can become life long carriers continously shedding the virusto progeny.
The virus affects fish as youngas yolk sac fry but highest mortalities are seen at the first feeding stage ( Bootland et al., 1990). Usinga live avirulent IPNV isolate, McAllister (1984) reportedprotection in 10 day post-hatch brook trout, challenged three weeks post vaccination. However,it was questioned as to whether the protection was specificor due to interference of the avirluent strain with the challenge virus. This possibility, howeverwas not investigated. Protection has not always been induced with liveIPNV vaccines (Hill et al.,1980, Dorson, 1988), and furthermore the difficultyassociated with licensing such a vaccine is always a factor to consider. Manyinactivated IPNV vaccines 22
have been tested in 4-8 week post-hatch trout (Dorsonand de Kinkelin, 1977; Hill et al., 1980). Efficacy has been demonstratedin these animals only after vaccination by intraperitoneal injection, but injectionmethod of delivery is not feasible for large numbers of fish in the 4-8 weekpost hatch range. Attempts to vaccinate fish with inactivated virus by immersionhave failed to elicit protective immunity (Dorson and de Kinkelin, 1977; Hillet al., 1980). In some instances the immersion method appeared to exacerbatethe effects of subsequent viral challenge (Bootland et al., 1986). Bootlandet at., (1990), suggested that size and age are critical determinants when vaccinatingby immersion against IPNV. Brook trout ranging from 1-8 weekspost hatch were immunized by immersion using a formalin inactivated IPNVpreparation at a water temperature of 100C. Fish were subsequently challengedfour weeks later with 105 pfu/ml IPNV; relative percent survival(RPS) 60 days post challenge was greatest in those fish in the 2 and 3 week post hatchgroup (Bootland et al,. 1990). There was no significant differenceversus controls in those fish at one week and four weeks post hatch. A detailed analysis of thegrowth rate suggested that protection against IPNV requires immunizationin the eleutheroembryo stage of development,a stage of slow growth (Bootland et al., 1990).
Subunit vaccines produced by recombinant DNAmethods for IPNV have been studied by a number of laboratoriesand from this work attention has focused on the major capsid protein VP2,as it has been associated with protective immunity (Lawerence et al., 1989; Manningand Leong 1990;
Havarstein et al., 1990). Manning and Leong, (1990),reported a successful experiment of inducing protection usinga recombinant DNA-derived subunit vaccine for IPNV. The recombinant protein containedthe 3 major capsid proteins : VP2, NS, and VP 3. The protein is producedas a poly-protein with 23
the NS protein between VP2 and VP3. The NSautocleaves the poly-protein into the respective proteins VP2,VP3 and NS (Linda Bootland,Department of Microbiology, Oregon State University Corvallis, OR).The proteins were delivered in a crude bacterial lysate by immersionand four weeks post vaccination the fish were challenged with 104 pfu/mlIPNV for 18 hours at 200C. Relative percent survivalwas 91 % in the vaccinated fish at 30 days post challenge. These results were encouragingas it proved that biological activity of the recombinantly derived proteinswas retainedin vivoand the proteins did not require purification (Manning and Leong,1990).
Although the results using IPNV subunit vaccines haveproven successful in demonstrating protection against IPNVchallenge (Linda Bootland, personal communication). Studiesmust be conducted to determine efficacy of IPNV subunit derived vaccines againststrains other than the one used to derive the cloned DNA in order to determineif the protective antigens have conserved epitopes in other viral strains.
Viral Hemorrhaoic Septicemia Virus
Viral Hemorrhagic Septicemia Virus (VHS),causative agent of Viral Hemorrhagic Septicemia, is caused bya rhabdovirus that mainly infects young trout. This disease presents the most serious problemin continental Europe and is mainly known for its economic impact in thetrout industry (de Kinkelin, 1988). Infection usually results in death due to extensivehemorrhaging. The virus initially infects the cells of blood capillaries,hematopoietic tissue, nephrons and leukocytes (de Kinkelin,1988). Aconservative estimate of losses in the trout industry due to VHS isapproximately 40 million tons of 24
marketable product / year valued at $ 80 million (Hill, 1991). Thevirus was once thought to cause disease only in rainbow trout but recently it has been
shown to infect a wide variety of other species suchas lake trout (Salvelinus
namaycush), brown trout (Salmo trutta) (de Kinkelin and Le Berre,1977), grayling (Thymallus thymallus) (Wizigman et al., 1980), whitefish (Coregonus
sp.) (Ahne and Thosen, 1985), turbot (Scopthalmus maximus),( Castric and de Kinkelin, 1984) sea bass (Dicentrarchus labrax) (Castricand de Kinkelin,
1984) and sea bream (Esox lucius) (Meier and Vestergaard-Jorgensen,1980). Virulent virus is shed in the urine (Neukrich,1985) andsex fluids. The virus is most abundant in the kidney, spleen, brain,and digestive tract (Wolf, 1988). Transmission is horizontal andmay be vector-mediated (Wolf, 1988).
Asymptomatic carriers can spread the disease to hatcheriesvia water flow (Hill et al. 1991). Eggs can be contaminated by virus in infectedspawners, but this can be controlled by disinfection using iodine (Jorgensen, 1970). This disease
is somewhat like IPNV, theyounger fish are more susceptible to the disease. Therefore the vaccine would ideally be givento young fish before they are most susceptible. The vaccination would have thegreatest impact on fish between 0.5-1.5 grams. However, fish in this size rangecan not be efficiently mass vaccinated by intraperitoneal injection and immersion methodsfor delivery of this virus have thus far,proven not to afford protection. Early vaccine work on VHSV was conducted out by de Kinkelinand LeBerre (1977). Theyprepared beta-propiolactone inactivated virus and eitherip injected or immersed 2 gram fish. They determined thatone ip injection of 2 x 106 pfu g fish -1 afforded protection upon challenge with 5x104pfu/ml water. Immersion of 2 gram fish in an inactivated virus suspension for3 hours proved to be less effective upon challenge. Neutralizing antibody couldnot always be detected in those animals demonstrating resistance, whichmay mean that interference 25
by the vaccinating virus may playa role in protection against challenge. Since a VHS vaccine must be given to very young fish, the killed preparation
delivered ip proved impractical and the researchgroups concentrated on a live vaccine which could be given by immersion,a more practical delivery
approach (Leong and Fryer, 1993). The live attenuated vaccinestrains were produced by successive passage of the virus at progressivelyincreasing temperature to 250C in epithelioma cells (cyprinid EPC) cells.The first live attenuated strain was REVA, which was derived froman F1 serotype produced in RTG (rainbow trout gonad) cells andwas attenuated by 240 successive sub- cultures in these cells at 140C (Jorgensen, 1976; Jorgensen,1982). The vaccine was optimally effective when fishwere immersed at 1 x 104 pfu ml-1 in water for 1 hour, with care to makesure that the temperature of the water remained at 100C. This vaccine was intended for fishup to 100 g and protection lasted between 120-150 days post vaccination(de Kinkelin and Bearzotti-LeBerre, 1981; Bernard et al., 1983). Thesecond attenuated vaccine F 25 was produced in EPC cells by multiplepassages at elevated temperatures.It induced protection against three VHS serotypes, 07.71,23.75, and a wild type strain belonging to serotype I (deKinkelin et al., 1981; Bernard et al., 1983). The final attenuated strainwas a 07.71 variant produced by multiple passage.It afforded protection for 100 days at 100C. Although the attenuated strains allowed immersion delivery with protectionthey were not totally safe as some mortality was recordedas a result of vaccination alone (Bernard et al., 1983).
The use of recombinant subunit vaccine technology hasonly recently made the use of vaccines an economically feasible methodof reducing VHSV infection of trout. A subunit vaccine for VHS has beenproduced using molecular techniques. Researchers had evidence that antibodydirected 26
against the glycoprotein on the surface of the virus couldneutralize the effects of the virus and stop the progression of the infectivecycle (de Kinkelin et al., 1984;Lorenzen et al., 1990). Therefore the glycoproteingene was cloned from VHS and the neutralizing epitopes determined(Thiry et al., 1990). The gene was expressed in a bacterium(Escherichia coil),yeast(Saccharomyces cerevisiae),and insect cell cultures to produce large quantities of this
glycoprotein (Lorenzen et al., 1990; Jorgensen,1992).In a recent trial
conducted by N. V. L. Arhus and the Laboratory of Gene Expression,University 0 of Arhus, the gene for the intermediary protein of the VHSwas expressed in E. coilas a fusion protein and induced virus-neutralizing antibodies. These
antibodies were detectable by western blots and immunofluorescencein those trout injected (Lorenzen, 1990). Early results indicatedthat the recombinant protein also induced protection following injection.It has yet to be demonstrated that these proteins induce protectionwhen delivered by the immersion method. It is likely that immersionwould work as a similar recombinant system for infectious hematopoieticnecrosis virus, another fish rhabdovirus, elicits protection when deliveredby immersion (Gilmore et al., 1988, Jorgensen, 1992).
Infectious Hematoooietic Necrosis Virus
Infectious Hematopoietic Necrosis Virus (IHNV),a fish rhabdovirus, is the causative agent of infectious hematopoieticnecrosis. IHNV causes an acute, systemic and virulent disease in salmonid speciesin the North American Pacific rim from California to Taiwan(Wolf, 1988). The disease occurs naturally in the wild but is mainly seen inan epizootic magnitude among young salmonids reared in the hatchery environment ( Wolf,1988). 27
The losses in hatchery epizooticscan reach devastating proportions. For example, at Fraser River Hatchery,B.C. there was a 96% loss attributed to IHN (Amend, 1969). In some instances losses havebeen reported to require total destruction of the infected populationas the only means of control. The
disease was first reported in 1953 in Washington stateas a cause of death in sockeye salmon (Oncorhynchus nerka) (Rucker,1953). Sockeye salmon eggs from Alaska introduced the disease to Japan inthe early 70's; the disease eventually spread to rainbow trout fryon Honshu Island in Japan (Sano, 1976). Although the incidence of IHN hasgrown at an increasing rate, the only effective means of control has been the destructionof infected fish populations and sterilization of the affected hatchery. Thesemeasures result in potentially severe economic impacts. At present there areno licensed chemotherapeutic drugs available for the control of the disease. Thislack of chemotherapeutics has increased the need for the developmentof vaccines to combat IHN (Leong et al., 1988).
Several vaccines have been developed foruse against IHN. Amend (1976) and Nishimura et al., (1985) reportedhaving success with vaccination of rainbow trout using killed preparations of IHNV.Amend used beta- propiolactone to inactivate virus and immunizedtrout by intraperitoneal injection. He found that fish challenged afterimmunization were protected against a lethal challenge dose of IHNV. Nishimuraexperimented with different methods of formalin inactivation of virusto produce the optimum inactivation regime. Several formalin-killed vaccineswere used to protect juvenile trout against lethal challenge. The vaccinewas most effective when delivered by injection but hyperosmotic immersionwas capable of stimulating limited immunity. Fryer et al., (1976) developedan attenuated strain of INHV by passing the virus multiple times in steelheadtrout cell cultures. LD50 28
studies revealed that the attenuation reduced the virulenceapproximately 100- fold. This vaccine proved to be effective ineliciting protective immunity with
only 5% mortality in the vaccinatedgroup and 90% in the controls.It was also shown that one preparation was capable of givingprotection when delivered by immersion; however some residual virulencewas seen in some trout (Fryer
et al 1976). These results prompted researchers to halt furtherexperiments as the commercial industry expressedconcern about difficulties of licensing an attenuated live vaccine (Fryer et al., 1976).
Recombinant DNA-derived subunit vaccines have beendeveloped exploiting the knowledge that the viral glycoprotein purifiedfrom one isolate of IHNV would induce protective immunity toa wide variety of IHNV isolates (Engelking,1989; Leong et al., 1988; Mourich, 1991 ).This information led Leong's group to express,as a fusion protein with the trpE protein of E. coli, an epitope of the glycoprotein of IHN (Gilmore et al.,1988). Crude lysates of bacteria that expressed this fusion proteinwere used to immunize fish by immersion; protection was observed in fishafter challenge with virulent IHNV (Xu et al., 1991). The mortality values rangedfrom 0-19 % for vaccinates versus 64-92 % for controls and protectionwas seen in a number of different species of salmonids as smallas 0.4 g at the time of vaccination. The cloned product was examined further to reveal what portionsof the fusion protein contained the putative B and T-cell immunodominantepitopes (Xu et al., 1991). A field trial conducted ata site in Idaho using 100,000 1 g rainbow trout vaccinated with the IHNV subunit vaccine and100,000 control fish showed protection after fish were ponded in water froma source containing resident IHNV-infected fish (JoAnn Leong personalcommunication, Department of Microbiology Oregon State University, Corvallis,OR). At 80 days post-ponding the cumulative percent mortality (cpm) in thecontrol ponds was 27% and the 29
cpm for the vaccinates was 4%. These results demonstrated the effectivnessof the subunit vaccine in the most crucial trial of all,a large scale field study at a commercial site.
The development of viral vaccines for theaquaculture industry comes under much scrutiny with regard to safety,cost and effectiveness. Previous to the advent of molecular biology techniques fishfarmers had to rely mainly on good animal husbandry as wellas careful screening of stocks to avoid viral epizootics. However, the advent of recombinantsubunit vaccines, that avoid the need of attenuated live vaccines foreffectiveness, fully meet the criteria for safety and effectiveness. The cost of developingviral vaccines will not be as inexpensive as the bacterial vaccinesas the nature of technology employed to design and implement viral vaccines is farmore difficult than that of the bacterial vaccines.
Bacterial Vaccines / Vibrio anguillarum
Vibriosis is a bacterial disease of salt-waterand migratory fish (Anderson and Conroy 1970).It was first described as "Red Pest" by Bonaveri during the 18th and 19th centuries (Rohovec,1975). The causative agent of "Red Pest" was isolated by Canestrini (1893)who cultured it from eels and named it Bacterium anguillarum(Rohovec, 1975).It was later renamedVibrio anguillarum after Bergman isolated it from infectedeels in Sweden (Bergey Manual of Determinative Bacteriology, 1971). There are at least eight members of thegenus Vibrio that have been associated with disease in fish. This review will focusprimarily onVibrio anguillarum. The severity of the disease correlateswell with the incidence of overcrowding, stress, handling and elevation inwater temperature (Fryer et 30
al.,1972; Anderson, 1970). Vibriosis has global distribution and outbreaksmay occur even in immunized fish with mortalities greater than 70 % in high risk areas (Cisar and Fryer, 1969). Antibiotics such as oxytetracycline (terramycin
TM50), and sulfonamides have been used to control epizooticsbut their use ultimately can lead to the development of drug resistant strains ofbacteria which may pose an even greater threat,as evident in Japan (Aoki et al., 1974). The use and success of the Vibrio vaccines has circumventedthe need for antibiotic use in almost every case where vibriosis control is ofconcern.
The commercial vaccines against V. anguillarum requiredno prior knowledge of virulence factors or protective antigens. Thevaccines produced usually contain inactivated V. anguillarum both whole cells andextracellular products (Smith, 1988).
The extracellular products contain lipopolysacchride (LPS) whichmay either remain cell associated or be released into the culturesupernatant (Chart and Trust, 1984; Evelyn, 1984). The LPSseems to be the protective antigen of
V anguillarum and because the vaccines used today induceadequate stimulation, it is unlikely that these vaccines will requiremore sophisticated production of antigens. However, future workmay be done to incorporate adjuvants or immuno-enhancers to decrease the need for boostingdoses and to increase duration of protection. Vibrio vaccinesare delivered by three main methods, including injection, oral, and variations of the immersionmethod including, hyperosmotic immersion, immersionor dip, bath vaccination, spray or shower and automated immersion (Fryer et al., 1976; Evelyn, 1984; Austin and Austin, 1987). All methods have been succesful with the intraperitoneal injection most effective and oral delivery least effective (Agiuset al., 1983, Antipa et al., 1980). 31
The protective response in fish vaccinated with Vibriovaccines is likely to be antibodymediated,as passive transfer is sufficient for protection. Harrell et al.(1975),showed that passively immunizing rainbow trout with immune
serum protected them from challenge with viablebacteria.Protection was specific as the effect could be abolished when thesera were incubated with Vibrio bacterin.These experiments proved that humoral immunitywas
important in protection(CroyandAmend, 1977; Fryer et al.,1977; Gould et al., 1978).
The Vibrio vaccines are one the most effectiveaquaculture vaccines to date. The data concerning dose, duration and vaccinationtimes are well established and researched (Fryer et al., 1976). The Vibriovaccines have the advantage of being easy to mass produce and deliverymethods are effective in a variety of ways. The only limitations that influencethe effectiveness of the these vaccines are the size and ontologicalstage of development of the fish vaccinated. Fish must be between 1.0 and 2.5grams when vaccinated for protection to be conferred, and duration of immunity isalso dependent on concentration and size of fish (Johnson et al., 1982a;1982b; 1981)
Yersina ruckeri
"Enteric Red Mouth" (ERM) is caused by Y. ruckeri,a Gram negative motile rod ( Ross etal.,1966).The disease has been known in the United States since the 1950's, but the earliest publicationon the pathology of the disease was from Ross and Rucker in1966.It was once thought that ERM was principally a disease of rainbow trout, but it isnow considered a potential pathogen for all salmonidspecies.The bacterium has several different serotypes.Serotype I (Hagermanstrain)is highly virulent and vaccines 32
derived from this serotype haveproven to be protective against challenge by other serotypes (Bullock and Anderson, 1984). In theearly 1970's ERM was responsible for the greatest loss of developing rainbowtrout (Hester 1973). The first ERM vaccine describedwas a phenol-inactivated whole cell preparation that was incorporated into the feedas wet packed Y. ruckeri cells
in thediet (Ross and Klontz, 1965). The results of theERM experiment showed
90 percent relative survival of the vaccinates. This earlywork proved that a marketable vaccine was possible and by the late 1970'scommercially available vaccines brought the ERM disease outbreaksunder control. The success of the immersion Vibrio vaccines provided themeans for the highly effective immersion ERM vaccine which ismore effective than the orally delivered ERM vaccine (Amend et al., 1983). ERM, likethe Vibrio vaccine, has been delivered a number of ways and, analogousto the Vibrio vaccine, the ip injection method is most effective. Although lesseffective than the ip method, oral vaccination against ERM was effective whendoses exceeded those of the ip or immersion method. However, if the rateof feeding was reduced, the efficacy of protection soon dissipated (Ross and Klontz,1965).If the vaccine was administered by anal intubation, higher levels of protectionwere produced, suggesting that partial degradation of theprotective antigens may have taken place when the antigenswere delivered by the oral method (Anderson and Nelson, 1974). Methods of protectingthe antigen from gastric degradation affects the efficacy of orally delivered ERMvaccines.
The success and ease of development of the ERMvaccine have proven invaluable for the commercial fish farmer and, thusfar, the serovar used in the vaccines has afforded cross-protection against otherstrains (Stevenson, R. Department of Microbiology, College of BiologicalScience, University of Guelph, Ontario, Canada). 33
Aeromonas salmonicida
Aeromonas salmonicida, causative agent of furunculosis insalmonids, is a Gram negative non-motile rod.It is a pathogen of both salmon and trout
and is endemic to North America, Europe and Japan. The diseaseis named for the raised liquefactive muscular lesions producedby infection with the
pathogen (McCarthy and Roberts 1980). Although the lesionsappear in those animals that are chronically infected, theyare rarely seen in the acute infection which is characterized by the rapid onset ofa lethal septicemia. The pathogen causes severe epizootics in wild fish populations, as wellas causing massive losses of cultured salmonid species (Mackie et al., 1935).Most, if not all, species of salmonids ranging from alevin to adultare susceptible to furunculosis (Fryer, J. L., Department of Microbiology,Oregon State University, Corvallis, OR). Atlantic salmon and brown troutseem to be the most susceptible and, in a few instances, rainbow trout have beenshown to be resistant (Cipriano, 1982).
The lack of a successful vaccine for furunculosis, incontrast to the vibrio and ERM vaccines, has been attributed to the difficulty inunderstanding the virulence factors associated with the disease and whateffect they have on the host during the infection (Hastings and Ellis, 1985). Thesimple bacterin approach that was successful for the previous vaccines hasnot been successful for furunculosis ( Michel, 1985). Efforts to vaccinate fishagainst A. salmonicida began with Duff, (1942), who demonstrated that longterm oral exposure to killed A. salmonicida protected cutthroat trout. However, the results of Duffs' experiments could not be consistently repeated.Promising results were also achieved when adjuvantswere combined with injectable 34
vaccine preparations. The adjuvants enhanced immunity andmay have allowed for a depot effect of the antigen (Patterson et al.,1974). Udey and Fryer, (1978) reported that fresh A. salmonicida cells, suspended in saline autoagglutinated, but multiple passage of these cells in the laboratory caused this autoagglutination to cease. This loss of autoagglutination correlated well with a lack of virulence. Udey and Fryer determined that this autoagglutination and virulence were dependent on the production ofa cell surface layer external to the outer membrane. They termed this layer the A-layer. The A- layer is composed of a 50 KDa protein or A-protein (Kay et al., 1981). The A- layer was discovered to play an important role in virulence andwas shown to be a protective antigen (Munn et al., 1982; Udely and Fryer, 1978). This discovery led investigators to dissect the wall components of A. salmonicida in an effort to find other key antigens important in protection against the
bacterium. The A-protein has been shown to aid in survival of the bacteriumin the presence of immune and non-immuneserum and provides resistance to complement-mediated lysis (Munn et al., 1982). Vaccines composed of A- layer + (McCarthy et al, 1983) and A-layer- live cells (Cipriano and Starliper, 1982) have been successful in affordingsome protection but the protection was generally considered to be marginal. Thornton et al., (1991), isolated two mutants ofA. salmonicida,a slow growing, amino-glycoside-resistant mutant and a rapidly-growing pseudo-revertant. Both of these mutants continuedto exhibit classical virulence factors associated with A.salmonicida pathogenesis. However, they differed morphologically from the wild-type with respect to the organization of the A-layer. Both mutantswere avirulent and incapable of sustaining infection. The rapidly-growing, antibiotic-sensitive pseudo-revertant, delivered either intraperitoneallyor by immersion, protected fish from challenge with a wild type virulent strain of A. salmonicida. The 35
resistance generated by this live attenuated strain warrants further field trialsto determine it potential use as a candidate vaccine. These live strains wouldbe essential for eliciting both the cell-mediated and humoral immuneresponses, which seem to be critical for full protection against the pathogen (Niklet al., 1991; Rockey, 1989). The development of cellular immuneresponse assays will be needed in order to determine what potential candidateantigens are involved in eliciting protective cellular immunity. Once these antigenshave been identified they can be isolated to producea vaccine.
Renibacterium salmoninarum
Bacterial Kidney Disease (BKD) is caused bya fastidious, slow growing
bacterium, Renibacterium salmoninarum (Elliott et al., 1989; Fryerand Sanders, 1981). BKD is one of the most prevalent diseasesof cultured salmonids (Fryer andSanders, 1981),mortality due to the disease can occur in both fresh and salt water life stages (Earp etal.,1953, Banner et al. 1983). Actual losses attributed to BKD have not beencalculated,however the disease is said to be one of the most important bacterial diseases affectingresident and anadromous salmonid stocks in the Pacific Northwest(Fryerand Sanders, 1981). In spite of its economic importance thereare limited effective methods for controlling BKD (Kaattari et al., 1990).
One reason for this difficulty is the bacteria presents itselfas a facultative intracellular parasite which has the ability to survive and multiplywithin the phagocytic cell. The intracellular nature may allow the bacteriato escape the effects of the humoral arm of the immuneresponse.Also Bandin et al., (1993) has demonstrated that several strains of Renibacteriumsalmoninarum can 36
resist killing by rainbow trout macrophages and possibly multiplywithin the phagocytic cells for up to 3-4 days.
Current approaches at management include stress reduction, quarantine, chemotherapy, culling and total destruction of the population
infected with complete sterilization of the facilities harboring theseanimals. As of yet, no efficacious vaccines exist. Controlling strategiesare primarily limited due to a lack of understanding of the mechanisms of pathogenesis,and how the salmonid responds to infection. Much of the research hasbeen impeded by the technical difficulties related to the culturing of the organism (Fryerand Sanders, 1981; Daly and Stevenson, 1988). These includedifficulty in primary pure culture isolation, slow generation time of 24 hours and the long incubation times (1-4 months) required for experimental challengeof fish (Fryer and Sanders, 1981).
The development and evaluation of potential vaccine candidates has focused on preventive strategies (Bruno, 1988; Elliottet al, 1989 ; Kaattari et al,
1990). However, marginal sucess of many tested BKD treatmentsmay have been due to a lack of thorough examination of possible antigensand the use of immune modulators to enhance their efficacy (Kaattariet al, 1990). The potential of BKD vaccination was first suggested by Evelyn (1971), when he demonstrated that specific agglutininswere produced in juvenile sockeye salmon injected with heat killed Renibacterium salmoninarum.
Although some favorable results were reported forsuccess in vaccination by some investigators, (Patterson et al. 1981; McCarthyet al.; 1984; Shieh 1989; Patterson et al.; 1981), others have not confirmedthese results ( Sakai, et al., 1989; Evelyn, et al., 1988; Kaattari, et al., 1990).
McCarthy et al., 1984, reported a vaccination trial with twopreparations of 37
formalin-inactivated cells ofRenibacterium salmoninarum.The two bacterins used were modified in four different ways; formalin-killed intact cells,double strength concentrate of formalin killed cells and pH-lysed versionsof the previous bacterins. The bacterins were administered withoutadjuvant by ip injection, immersion, or two step hyperosmotic infiltration. Thefish were challenged by ip injection 36 days after vaccination. No significantprotection was afforded by any preparation delivered by hyperosmotic infiltrationor by immersion. However, those fish vaccinated by ip injection,then ip challenged, demonstrated protection against infection. The pH-lysedbacterin was the more beneficial of the modified bacterins. These resultswere promising, however, McCarthy et al. (1984), did notmeasure antibody titers and used Gram staining as a method of measuring the disseminationof the infection. Therefore these results must be interpreted cautiouslyas Gram staining lacks the sensitivity of other more exact methods of diagnosis.Kaattari et al. (1990) utilized a number of products and cell wall fractions includingintact cells, fractured cells, extracellular products and cell wall fractions.The products were delivered either alone or with other bacterial antigens. The immunogens were administered ip, orally, and by immersion, with and without Freund's complete adjuvant None of these early preparations protectedfish.In fact, some of the preparations actually may have exacerbated the disease because some vaccinated groups demonstrated less mean time to death than the controls. This exacerbation was thought to be associated withsome immunosuppressive action of the bacterial productsincluding the p57 component of the extracellular products, or the possible induction of hypersensitivity. Turaga et al. (1987) reported that theproduction of antibodies in vitroby artificially stimulated normal coho salmon lymphocyteswas suppressed when a soluble antigen fraction derived froma culture of R. 38
salmoninarum was incorporated into medium at concentrations of 10µg ml-1 and 100 µg ml -1. Suppression of antibody production was similar in cultures
of lymphocytes from BKD-infected coho salmon and could be correlated with soluble antigen levels between 3 and 80 µg ml -1 in the blood of infected fish.
The components expressed in vivo and in vitromay be of importance to the progression of disease by suppressing the humoral response and
macrophage function (Bandin et al., 1993) in vivo. Therefore, it is imperative that this area of research receives further investigation, in particular the
contribution of antigens that do not cause immune suppressionor dysfunction but, stimulatee cell-mediated immunity (CMI) (Evenden et al. 1993). Further vaccine research would gain considerably from the development of a standardized bath challenge procedure, that mimicsa more natural route of infection than does ip injection (Elliott et al., 1991; Murrayet al.,1992). Also, cohabitation Murray et al., (1992) might also sufficeas a natural challenge method. The ip challenge method bypasses the mucosal immune response totally, leaving only a systemic response (Murray et al.,1992).
Monitoring protection becomes difficult however, under bathor cohabitation challenges, as mortailities may take months toaccrue. Alternatively, protocols such as the soluble antigen ELISA that monitor the production of soluble antigen produced by Renibacterium salmoninarum could be employed to accurately monitor the disease throughout the entire challenge period. This procedure would give earlier assessment of infection and could detect carriers amid the survivors (Rockey et al., 1991; Murray et al., 1992). This type of monitoring may also be used to evaluate the efficacy of candidate antigens used in vaccination.
Bacterial kidney disease of salmonids is a disease which exploitsa variety of pathogenic mechanisms. These mechanismsare correlated with a 39
number of putative virulence factors. Daly and Stevenson (1987),reported on the ability of extracelluar factors to agglutinate rabbiterythrocytes. Catalase, DNase, hemolytic, proteolytic and exotoxin activities havealso been described (Bruno and Munro, 1982; Shieh, 1988). The hemolysinhas been cloned and is 1.6 kb in length (Evenden et al, 1990). This hemolysinhas activity against rainbow trout erythrocytes and the clonedsequence hybridized with four other
isolates of Renibacterium salmoninarum. Rockey et al. (1991)discovered a
proteolytic enzyme produced by the bacterium (Bruno and Munro,1986;
Rockey et al., 1991). Using substrate gel electrophoresisRockey et al.(1991) identified two bands one > 100 kDa and the second< than 20 kDa. The 100 kDa band demonstrated proteolytic activity against p57and against denatured ovalbumin and bovine serum albumin. However,a direct role of the proteases in virulence has not yet been established. The abilityof the bacteria to survive
(Bandin et al., 1993), escape the phagolysosome (Gutenberger,S.Department of Microbiology, Oregon State University, Corvallis, OR)and multiply within the macrophage (Munro and Bruno, 1988)may also aid in establishing infection.
An understanding of these mechanisms is essential fordevelopment of efficacious vaccines.It is well documented that the soluble proteins produced by R.salmoninarum have toxigenic potential (Kaattariet al., 1990; Shieh, 1988;
Bruno and Munro, 1986) The removal of these productsmay produce a less pathogenic strain that may be usedas vaccine. A thorough understanding of the host immune response to the different antigensassociated with R. salmoninarum infection is also vital in producinga successful vaccine. 40
Vaccination Methods / Delivery Systems
The route of exposure has a direct impacton the degree of immunity elcited and regionality of the response. Theseaspects of immunity are
essential in the design of delivery systems and the timing of immunization.In mammalian vaccines, to stimulate life-long immunity, multiplevaccinations may be required (Salk and Salk, 1984; Ada 1988). Fish must also receive
multiple vaccinations to stimulate life-long immunity; howeverthe route of delivering the vaccine and the ambient water temperatureare essential factors for successful vaccination of fish. Fish vaccinatedat less than physiological temperatures display a delayed onset of protective immunityor may not experience immunity at all.Bly et al. (1992) demonstrated that channel catfish (Ictalurus punctatus) could be immunosuppressed bya rapid decrease in environmental water temperature from 22 to 10CC. This immunosuppression was characterized by the lackof leucocyte infiltrates to fungal-associated skin lesions caused by challengewith saprolegnia species. Results also indicated that lowering the temperaturefrom 23 to 110C over a 24 hour period suppressed both B and T cell functions for3-5 weeks as assessed by in vitro responses (Bly and Clem, 1991). Theroute of immunization most often used in mammals (injection), proves less practical forfish and, therefore, various other techniques have been developed tomass vaccinate fish and these include immersion, hyperosmotic immersion,bath, spray and oral deliveries (Ellis, 1988). The intraperitoneal (ip) method hasalso been used and, in many instances is the most effective method of deliveryto achieve specific protection. Intraperitoneal delivery of adjuvants and antibioticsin concert with antigens also improves the efficacy, yet for large scalevaccination 41
and vaccination of very small fish (0.5-2.0 g) the ip method is impractical.
Intraperitoneal vaccination requires intensive labor,even to immunize relatively small numbers of fish. Also, added handling involvedwith ip injections causes increased stress incurred by the fish whichcan lead to
immune dysfunction and possible disease outbreaks. Recentlysemi- automated devices have been developed thatmay ease the handling burden of this method, but the small size of the animal still remainsa problem.
It was discovered in the 70's that osmotic shock improveduptake of antigenic material from an aqueous environment. Thiswas thought to occurred due to a change in gill permeability (Amendet al., 1976). This technique was called hyperosmotic infiltration (HI) and itproved to be of benefit even when using killed bacterin preparations (Antipa and Amend 1977,
Antipita et al., 1980). Although this method proved to beeffective, it was later determined that sufficient levels of antigenexposure and protection could be achieved by simple immersion techniques without subjecting thefish to HI (Gould, 1978). Immersion has become widely used asa primary means of vaccination.It is most widely used on smaller fish (< 10-15 g). Although this method is not a stress-free method of delivery, it is preferredover HI. Studies conducted by Tatner (1987) and Fryer et al. (1977)on exposure time revealed that when antigen is not the limiting factor, lengthening theimmersion time does not result in a greater antigen uptake. The standard methodis to expose the animals for a minimum of 20 seconds to the vaccine inaerated standing water. The disadvantage of immersion vaccination is that it is limitedby the weight of fish that can be immunizedper unit volume of vaccine. Though immersion vaccination usually provides lower levels of immunitythan injection vaccination, the levels of protection are still high enoughto justify the use of this method of vaccination (Lillehaug, 1989). The bathmethod of vaccination 42
is a modification of the immersion method. The fishare exposed to dilute suspensions of the vaccine for times ranging from 30 minutes to several hours
(Rohovec, J. personnal communication, Department of Microbiology,Oregon State University, Corvallis, OR). Sprayvaccination is another modificationof the direct immersion method of vaccination. The first attempts at this technologywere conducted using a sandblaster that sprayed the fish witha mixture of clay particles and the vaccine (Gould, 1978).It was later determined that the fish could be sufficiently immunizedby just sprayingthe vaccine, as long as the exposure time was adequate (Fryer personnal communication, Department of
Microbiology, Oregon State University, Corvallis, OR).
Oral immunization has a long-established tradition; the firstattempts to vaccinate humans and fish was by the oral methodofdelivery. Immunizing fish
by theoral route is as ideal method for mass administrationto fish ofall sizes without associated handling stress (Fryer et al., 1976;1977). The early oral vaccination experiments consisted of exposing the animalsto the antigen in a form of paste or liquid suspension either coatedonto or milled into feed. These early methods resulted in inconsistenciesin protectionas well as failure to elicit a detectable immune response (Evelyn, 1984; Hart et al., 1988; McLean
and Donaldson,1990) This could beattributed to degradation of pH-sensitive antigens exposed to the gastric portion of the gut. Because of theearly failures,oral deliveryhas remained an under utilizedapproach to immunization. However, with advent ofnew technologies, many new oral delivery systems have been developed thatare designed to cope with the limitations associated with oral vaccination. Wong etal.(1992) reported deliveringVibrio bacterins protected with an enteric coatingas a method of protecting the antigen from the gastric pHof the stomachin an effort to allow 43 intact antigen to reach the second gut segment associated with gut associated lymphoid tissue.
Also, only recently has the important protective functions of secretory immunity at mucosal sites become appreciatedas an integral part of the immune response ( Lobb, 1986; O'Hagan, 1992). Traditionally vaccine research has been concerned with the induction of systemic immunity by parenteral immunization. For diseases in which the infectious agent is introduced parenterally, such as those initiated by trauma to intact skin (i.e. tetanus and malaria) this method of delivery is appropriate. However, it is well known that most infections are acquired naturally through mucosal routes; orally, nasally or genitally (Finlay and Falkow, 1989; McGhee et al., 1991). In these instances, parenterally delivered vaccinesmay not be the best choice of delivery for immune stimulation. This is illustrated by the limited efficacy of parental cholera vaccine ( Holmgren et at., 1989). When compared to parenterally delivered vaccines, oral vaccines offer the advantage against mucosally-acquired infections because of their ability to stimulate mucosal immunity (Georgopoulou et al., 1986; Mcghee et al., 1991; Wonget al, 1992 0' Hagan, 1992). Fish rely heavily on their mucosal coatingas a first line of defense to avoid contact with pathogens; therefore stimulating specific immunity at these mucosal sites should increase protection (Davidsonet at., 1993). Oral immunization is also safer, easier to administer, better tolerated and has potential for easier boosting vaccinations. Oral vaccinesare also less expensive and require less purity and quality control than parenteral vaccines (O'Hagan, 1992).
It is currently known that specific antibody can be detected in fishmucus after oral vaccination or anal intubation (Lobb, 1986; Rombout, 1989, Wonget at., 1992; Davidson et al., 1993 ). Many currently available vaccines for the 44
industry would be improved if they could be administeredorally and,
additionally, many new vaccines could be developed againstagents that are at present poorly controlled.
Recently research has resulted in the development ofseveral novel antigen delivery systems which can bevery efficient at inducing both secretory
and systemic immunity following oral administration (Eldridgeet al., 1989).These new improvements offer considerable promise for thefuture and
may result in the development of new oral vaccines. These improvementsare based upon the use of non-replicating antigen delivery systems, which
involves the incorporation of the antigens with medium whichprotects them
against degradation in the gut (Eldridge, 1991; Eldridge, 1991a; Wong et al., 1992).Microparticles are representative of these delivery systems. Their actions are based on the reported uptake of particles into Peyerspatches or analogouscells of the GALT(O'Hagan, 1990). After uptake, microparticlesare phagocytosed by macrophages where the antigensare processed and presented to lymphoid cells of the gut. Several reports have demonstrated evidence that secretory immunity is stimulatedupon oral administration of microparticles (Challacombe et al., 1991,Eldridge et al. 1989, O'Hagan et al., 1989, Wong et al., 1992).It has been demonstrated that these microparticles can also elicit systemic immunity following oral administration (Challacomb et al. 1991 , Eldridge et al., 1989). Enteric Coated Antigen Microspheres
(ECAMs) (Figure 2.2) are one type of microparticle deliverysystem that exploits the use of pH reactive polymers to protect the antigen beingdelivered from the harsh environment of the gastric portion of the gastrointestinaltract.The pH- reactive polymer coating becomes critical when the antigen inquestion is of protein derivation. Protection allows the antigen to be deliveredin its native state which is likely to be essential for immune recognition. 45
Antigen pH-reactive polymer
< 250 nm
Figure 2.2. Enteric Coated Antigen Microspheres (ECAMs). The ECAMs consist of non pareil dextrose beads that antigen is spray coated onto followed by a pH reactive polymer. The polymer serves to protect the antigen from the gastric portion of the gut.
Figure 2.2 46
The fundamental advantages of microparticles for oral immunization
are: 1) they are taken up into the GALT, 2) entrapped antigensare protected from degradation in the gut, 3) several antigensmay be delivered at one time, 4) adjuvants may be used in concert with specific antigens,5) polymers used for protecting the antigens have been approved for humanuse and are inert with respect to immune response andare, therefore, suited for repeat boosting, 6) control or sustained release formulationcan be used for long term exposure and finally 7) microparticles could be designed to target specific sites
(O'Hagan, 1992). The protocol of oral immunization isvery attractive when vaccination of fish is being considered (Ellis, 1988). With the recent evidence
leading to the importance of the secretory immuneresponse and its role in resolving disease, an increasing emphasis has been placedon oral immunization. The oral delivery of vaccines foruse in the aquaculture industry
is by far the most sought after method of choice for delivery,and with the new developments in delivery systems new and improved oral vaccinesfor aquaculture offer the promise of controlling the diseases that plague the aquaculture industry. 47
CHAPTER 3 Enteric Coated MicrospheresAs An Oral Method For Antigen Delivery To Salmonids
Jon D. Piganelli, 1.3 Jia Allen Zhang 2,3,John M. Christensen 2,3 and Stephen L. Kaattari * 1, 3
Department of Microbiology',and Collegeof Pharmacy2and theLaboratory for Fish Disease Research3, OregonState University, Corvallis, OR 97331.
* To whom correspondence shouldbe addressed
Published in Fish and Shellfish Immunology4,179-188, 1994 48
ABSTRACT
A comparative study was conductedto determine the immunization efficacy of enteric coated antigen microspheres(ECAMs) versus intraperitoneal (ip) and immersionantigen delivery. Anal intubation of antigen
served as a positive control by directly deliveringthe antigen to the hindgut. Naive juvenile coho salmon Oncorhynchuskisutch were exposed to three
different concentrations of trinitrophenlyatedlipopolysaccharide (TNP-LPS)or trinitrophenylated keyhole limpet hemocyanin(TNP-KLH) by the three routes, oral (ECAM), intraperitoneal, immersionor anal intubation. TNP-LPS or TNP- KLH were prepared for oral immunizationby coating onto 0.45 mm dextrose sugar beads followed by an application of EurdragitLD-30 co-polymer. Serum was taken from immunized and non-immunizedcontrol fish fed diet containing non-antigen coated beads. Theserum was assayed by ELISA at 4,6, and 8
weeks post immunization toassess antibody titers.Statistical analysis of the serum showed that oral vaccinationsignificantly increased anti-TNP titers at 4 and 6 weeks post vaccinationas compared with the controls (p< 0.05). The anti-TNP-LPS ip titers after TNP-LPS immunizationwere significantly higher than the controls (p<0.05) at 4,6and 8 weeks post immunization, while the group that was immersed showedno significant difference from the control. There was no significant differencebetween TNP-LPS ip and ECAM immunization methods.
The TNP-KLH immunized ECAMgroup showed significantly elevated levels of anti-TNP antibodiescompared with controls at 6 and 8 weeks, while the ip immunized groupwas significantly higher than the control throughout the 8 week period. The group immunizedby anal intubation had significantly higher titers than controls at 6 and8 weeks but not at 4 weeks. 49
INTRODUCTION
Although there are at least twenty recognized bacterialdiseases of salmonids there are only six commercial vaccinescurrently available. These vaccines, which are simply whole cellbacterins,are composed of gram negative organisms delivered by immersion and intraperitonealinjections (Wong et al.,1992).Although immersion and intraperitoneal vaccination
methods have proved beneficial in the aquacultureindustry,an alternative method of delivery is oral immunization (Hartet al., 1988). Oral vaccination reduces the amount of labor andexpense of administering a vaccine (Hart et al., 1988, Lillehaug, 1989); more importantly, it reducesthe stress incurred by the fish during immunization
Recent studies have demonstrated that thegut associated lymphoid
tissue(GALT)is an important part of the total secretory immunesystem
(Rombout et al,1989).Previous oral immunization studies have focusedon such parameters as the duration of antigen deliveryand the total antigen received by each animal (Ross &Klontz,1965), not on the method of antigen
delivery to GALT. A number of investigators ( Mughal& Manning 1986,
McLean&Ash,1987,McLean&Donaldson,1990,Dogget&Harris, 1991) reported that fish possessGALTthat responds to antigenic stimulation.
Mughal & Manning(1986),demonstrated that oral priming of juvenile thick- lipped grey mullet (Chelon labrosus Risso) withhuman gamma globulin (molecular weight 150,000) resulted inan increase in serum antibody.
An essential factor to consider in the design oforal vaccines is the harsh denaturing acidic environment of the stomachas well as the resident
proteases involved in degradation of complexproteins.Vaccines of this nature 50
should be designed to be protected through theacidic stomach region of the intestinal tract and successfully reach the portionof the hindgut associated with lymphoid tissue. This pointwas confirmed by a recent study showing
enhanced immune response against Vibrio anguillarumafter anal intubation,
which was directly deposited to the second gutsegment (Rombout et al, 1989). Our enteric coated antigen microspheres (ECAMs)serve to protect the antigen so that it can arrive intactto the GALT(Porter, 1985). Incorporation of pH
reactive polymers as coating materials protects theantigen from the low pH of the stomach, permitting its passage into the intestineintact. Upon entrance into the intestine, the increase in pH results indissolution of the polymer and release of the antigen to the GALT (Wong et al. 1992).
The goals of the present study were: 1) tocompare the relative efficacy of ECAMs in inducing an antibody response withrespect to delivery by immersion, intraperitoneal injections and analintubation and; 2) to determine whether ECAMs were effective in delivering haptenatedforms of lipopolysaccharide and protein antigens, the latterbeing more sensitive to the denaturing conditions of the gastrointestinal tract. 51
MATERIALS AND METHODS
Animals. Coho salmon (Oncorhynchus kisutch), weighing 45grams, were obtained from Sandy Fish Hatchery, (Sandy, Oregon). Fish were maintained at the Salmon Disease Research Laboratory at Oregon State
University in 120 C pathogen free water. Fishwere fed a diet of Oregon Moist Pellet (OMP,Bioproducts, Astoria Oregon).
Antigens. Trinitrophenylated-lipopolysaccharide (TNP-LPS)was prepared as previously described, by (Jacobs & Morrison 1975). TNP-KLH was prepared as previously described (Rittenberg & Pratt 1969).
Preparation of Enteric Coated Antigen Microspheres (ECAMs). ECAMs were prepared by spray-coating the antigen,either TNP- LPS or TNP-KLH onto 40-50 mesh dextrose beads then spray-coatingthe antigen coated beads with Eudragit L-30D,a commercially available aqueous latex dispersion of methacrylic-arylic acid co-polymer. The conditionsof the spray coating were similar to that of Hossain & Ayres (1990).
Incorporation of prepared ECAMs into thefood.Daily maximal intake of food by the fish was determined by taking theaverage food consumption of the fish for five days. The average in take of foodwas 10 52
grams of OMP tank -1. Therefore, the ECAMs were incorporated with the food
in a ratio that would insure a full days dose of antigen /tank (see TableI. for total dose /treatment group) The mixturewas allowed to disperse uniformly throughout the mash and distilled waterwas added to form a mull. The mixture was then extruded through a defined pore size extruder (Vitano, EastLake, OH) and was cut into pellets.
Immunization. Twenty five salmon per treatmentwere immunized
with TNP-LPS by three different routes: ECAM, ip, and immersion(Table 3.1 .). Animals immunized ip were injectedone time with 10 µg, 10 ng or 1 pg TNP- LPS using an Eppendorff repeater pipette fitted witha 26 gauge needle. Once a day for 30 days, TNP-LPS ECAM immunized fish were fed ECAMs incorporated into OMP at three different concentrations10 µg pellet -1, 10 ng pellet -1, or 1 pg pellet -1.Immersion immunization was done according to Velji et al. (1990) with the following modifications. Fishwere exposed to three different concentrations of TNP-LPS (Table 3.1) by reducingthe water level in each tank from 125 liters to 25 liters. The flow of waterwas stopped, aeration was started and antigen was added to the respective tanks. The fishwere exposed to the antigen for 15 minutes in standing aeratedwater. The water flow was resumed and the tanks were allowed to fillat a rate of 2.8 liters minute -1, thus removing antigen through normal effluent flow. Controlfish were fed non-antigen coated beads for 30 days.
TNP-KLH immunized fish were divided intogroups of 13 animals /treatment. Treatments were ECAM, ip and anal intubationadministration of antigen. TNP-KLH ECAM-immunized fishwere fed as described with pellets containing antigen at concentrations of 100µg pellet -1, 5 µg pellet -1 and 0.5
µg pellet -1 (Table 3.2). One group of fish was immunized ip with 100µg of 53
TNP-KLH solubilized in 100 µl sterile phosphate bufferedsaline (PBS). Fish immunized by the anal intubation methodwere given 100 µg of TNP-KLH solubilized in 100 µl of sterile PBS delivered withan apex clear pipette tip West CoastScientific, Hayward CA ). The controlfish were fed non-antigen coated beads for 30 days.
Collection of Sera.Individual serum samples were collected from five TNP-LPS immunized fish from eachgroup at 4, 6, and 8 weeks post immunization. TNP-LPS immunized fishwere euthanized by anesthetic overdose in benzocaine (Sigma). The caudal pendunclewas severed and blood collected from the caudal vein in 1.5 ml microfugetubes (West Coast Scientific, Hayward, CA).TNP-KLH immunized fish were individually marked and bled repeatedly from the caudal penduncleat 4, 6, and8,weeks in order to monitor the individual titer of each animal throughoutthe experiment. After collection,blood samples were stored overnight at 40 C to facilitate clotting and separation of serum from the erythrocytes. Tubeswere centrifuged for 2 min. in a Beckman model Emicrofuge,(BeckmanInstruments, Palo Alto, CA) to pellet thecells.Sera were drawn off the pellet and stored in microfuge tubesat -70 0C.
Determination of Antibody Activity. Theconcentration of antibody activity was ascertained by the use ofan Enzyme-linked immunosorbent assay (ELISA) as previously described by Arkoosh and Kaattari(1991), Arkoosh and Kaattari (1990). Briefly, each antisera is titratedon an ELISA plate (Costar E.I.A/R.I.A. certified surface chemistry,Cambridge Ma.) using trinitrophenylated Bovine Serum Albumin (BSA)as the coating antigen. Each plate contained a standard anti-TNP hyper immuneserum titration used for normalization of the 54
data and agreement of units of activity per µlserum. One unit of antibody activity is equivalent to the volume of standard serum required to produce 50 % of the maximum optical density obtainable with the standard antiserum after incubation of the serum dilutions with the antigen-coated plates, the signal in the wells were subsequently developed by theuse of a biotinylated anti-trout- Ig, followed by Streptavidin Horseradish peroxidase (Sigma). The plateswere washed and substrate (ABTS) was added. Optical densitieswere read kinetically for 10 minutes.
Statisticalanalysis.All ELISA datawas analyzedusing One-way analysis of variance (ANOVA) program on the Statgraphics software package, to determine difference based on variation between groups. ANOVA analysis generates an F-statistic that can be analyzed by another test to determine significance (t-test) using the pooled standard deviation(SP2). Ninety-five percent confidence limits were derived using standard error of the mean. 55
RESULTS
TNP-LPS immunized fish yielded significantly higher titers (p<0.05) at 4 weeks for all ECAM doses when compared to controls (Fig 3.1A). Peak
antibody titers in the high dose orally vaccinatedgroupoccurred at 6 weeks and decreased to control values by 8 weeks. The high and medium ip titers were significantly higher than controls at 4 weeks (p<0.05) and also peaked at
6 weeks (Fig 3.1 B). Although the immersion fish hadaverage titers greater than that of thecontrols,at no time were they statistically different from the controls (Fig. 3.1C).Statistical analysis among treatments revealed that there was no significant difference between the high ECAM and high ip group titers at 4 and 6weeks,only the 8 week ip titer values were significantly higher (p< 0.05). The medium and low group titers at 4,6 and 8 weeks for ECAMsversus ip also exhibited no significant difference betweengroups. The TNP-KLH ECAMimmunization revealed that the high and medium antigen concentrations demonstrated no significant increase in titerversus the control throughout the 8 week period; however the low ECAMgroup had a statistically significant increase in titer (p<0.05) at both 6 and 8 weeks above that of the controls (Fig 3.2). All values reportedwere subtracted from prebleed titers of each individual animal in order to control for background titerssome of which were abnormally high. The ip treated fish had statistically significantly higher antibody titers (p< 0.05) than the controls throughout the8 week period. The titers of anal intubation group was significantly higher than thecontrol (p< 0.05) at every sample interval except 4 weeks. 56
DISCUSSION
Extensive efforts have been devoted to the development of vaccinesto combat pathogens that plague the aquaculture industry; howevermore facile
and efficacious methods of delivering these vaccines remainsan obstacle. The least stressful method to deliver antigens would be via theoral route (Agius et al. 1983). Experiments conducted by Sakai et al (1984) demonstrated that oral vaccination has been effective with at leastone pathogen, Vibrio anguillarum. However with the advanceof enteric coated antigen microsphere technology the efficiency of oral vaccinationmay be enhanced.
ECAM administration was comparable to the other traditional,yet invasive, methods of eliciting serum antibody titers (i.e.. ip injection).The results of the TNP-LPS ECAM studies demonstrateda significant difference versus the control at all doses at 4 weeks and with the high dose at 6 weeks.
Comparison of the 4,6 and 8 week results between groupsdemonstrated that there was no significant difference between the high oraland the high ip titers at 4 and 6 weeks, only the 8 week ip titer resultswere significantly higher.
These results confirm that enteric coated antigen microspheresare as efficient in eliciting a detectable serum antibody titeras the ip method. Of some interest was the lack of significant serum antibodies in the TNP-LPS immersion treated fish.Velji et al. (1990), reported that no agglutinating antibody titerwas detectable in any of their Vibrio ordalii LPS treated coho, yetat levels of 1 µg ml-1 immersion concentration the fishwere adequately protected against
Vordalii challenge. Other investigators ( Velji et al. ,1990, Aokiet al. 1984, Kawano et al., 1984, and Sakai et al., 1984) also reported that immersion usually fails to elicit detectable serum agglutinins. Although their resultsare 57
based on data from an agglutinationassay, a subjective and less sensitive assay than the ELISA, our results confirm these observations.
Our results utilizing ECAMs for antigen deliveryare contrary to those of Lillehaug (1989), who observed thata formulation of acid-resistant film or a slow release pellet used to deliver the antigenprotected not only from degradation, but also from absorption by thegut. Lillehaug's results also may be attributed to limited exposure to antigen whichis not considered appropriate for oral immunization (McGhee et al.,1991). We fed antigen for 30 days where as Lillehaug (1989) exposed fishto antigen once. Higher and more frequently administered doses are necessary to immunizeby the oral method (McGhee et al. 1991). Also the slow releaseof antigen from the pellet or a large pellet size may not allow for the movement of the pelletto the second gut segment. Such limited releasemay have inhibited the ability of the fish to respond immunologically. Wong (1991) showedthat ECAMs must be of a minimum size before theyare able to pass through the pyloric sphincter and into the intestinal tract, otherwise the ECAMswill aggregate in the stomach. Using ECAMs labeled with radio-opaque dye,Wong was able to view the passage of ECAMs through the digestive tract of the salmon. ECAMsare protected until the second gut segment is encountered;whereupon there is a rapid dissolution of the antigen into themore alkaline environment of the intestine.
Enteric coated antigen microspheresare a very attractive choice for oral delivery of a protein antigen. Wenneras et al. (1992)report that oral immunization of humans with purified fimbriae fromenterotoxigenic E. coli has been attempted with littlesuccess, probably due to destruction and denaturation by proteolytic enzymes,as well as low gastric pH. Although LPS may be stable at low stomach pH and is unaffected by proteolyticenzymes, 58
proteins must be protected until they reach the lymphoid tissue associatedwith the intestinal tract. The experiments employing TNP-KLHwere conducted to determine if ECAMs were also sufficient to deliver haptenatedprotein antigens to the gut. The fish immunized with high and medium concentrationsof TNP- KLH showed increased serum antibody, butno statistically significant differences from controls. The group of fish receiving the lowest ECAM concentration, however, had statistically higher titers than the controlsat both 6 and 8 weeks. This may be due to dose dependent factors regulating immune response, in that the higher doses may actually be sub-optimal in elicitingan immune response, while the lower dose achieves sucha response. Jenkins et al. 1992, reports that lower concentrations of anally delivered HGG(1mg ml and 2mg ml-1) resulted in low levels of antigen uptake into the plasma.
Maximal absorption occurred with 20mg ml-1 although these levels were not significantly higher than those detected after 10mg ml-1 HGG intubation. Less
HGG was absorbed when a dose was administered (50mg ml-1). The 10 mg ml-1 dose orally delivered was absorbed in the greatest quantities(p< 0.05). This data suggests that the lower dose of antigen is optimaldue to the increase in absorption of proteins from the gastrointestinal tractof fish. Jenkins 1992 et al., reports that levels of protein absorption in humans is approximately0.02% of the ingested dose and 0.01-0.1 % in other mammals, whereas fish show an ability to absorb a much greater portion of the total protein dose delivered,up to 4.0 %.
Alternatively, the high and medium dosesmay have allowed for greater antigen absorption thus causing a state of tolerance. Determinationof tolerance induction would require a secondary challenge forassessment of responsiveness. The fish intraperitoneally immunized responded with significant serum titers throughout the 8 week period, and those in theanal 59
intubation group had significantly elevated titersat all weeks except week 4. The results from this group confirmed that by-passingthe gastric portion of the first gut segment, yielded a significant antibodyresponse to a protein. The fish which received the lowest dose of TNP-KLH ECAMhad higher antibody responses after 8 weeks than did those ip vaccinated at 8 weeks.
Previous studies using ECAMs for vaccinationagainst V. anguillarum
demonstrated that enteric coating provided for enhanchedprotection of serum and mucus antibody over that induced by uncoated oralvaccination Wong et. al. (1992). The increased titerswere likely in excess of what was required for
protection on both coated and uncoated vaccinatesachieved a high level of protection.
Many aspects must be considered when developingoral vaccines including, antigen dose, duration of antigendelivery and site of delivery. The use of ECAMs to deliver defined antigens has demonstrated the possibleutility of ECAMs and provided some information about thestability of protein and LPS antigens delivered by this method. Thesestudies have also provided some preliminary evidence as to efficacious doses of different typesof antigens delivered by this method. Theuse of ECAM technology may also be of benefit for development of vaccines touse in controlling fish diseases that have failed to elicit protective response by other deliverymethods. This technology may also be applied to the delivery of intactproteins in humans and other animals. Eldridge et al. (1991), reported thatthe biodegradable microspheres as a vaccine delivery system, haspotential widespread application for the induction of both systemic andmucosal immunity in mice and that these microspheres have the potentialto be effective in other species. 60
ACKNOWLEDGMENTS
The authors would like to thank Dr. Willem Van Muiswinkel, ValBrooks and Tracie Norris for their excellent technical assistance. The authorswould also like to thank Dr. Sandra Ristow, Dr. John Rohovec, Dr. GregoryWiens, Dr. Mark Adkison, Lesile Gilkey, Janell Bishop, John Hansen, HenryOrtega, David Shapiro and Patty Wood for their critical review of this manuscript.This work was supported by U.S.D.A CSRS 90-37116-534. Oregon Experimental
Station Technical Paperno. 61 Figure 3.1.
Serum antibody titers expressed in units of activity /ml ofserum from coho salmon immunized with TNP-LPS in the form of.(A) Enteric coated antigen microspheres at three different doses; high (U),medium (®) and low (®) (B)
Intraperitoneal injection at three doses; high (p,), medium (© ),and low (1211) and (C) Immersion at three different doses high (O), medium (E7) and low (®.). Values are averages based on n=5 for each dose at each individual time point.
Asterisks denote a statistically significant difference (p< 0.05) between controls Units of antibody activity/ml Units of antibody activity/mi Units of antibody activity/ml E r r r r r r r r'r r r r r r r r , -i , DillIIIIIIIIIIIIi -i -- ' '`' OAti 63
65000 E 60000 55000 50000 45000 40000 35000 30000 25000 20000 15000 10000 5000 0 4wk 6wk 8wk
Figure 3.2
Serum antibody titers expressed in units of antibody / ml serum from coho salmon immunized with TNP-KLH in the form of.ECAMs at high(), medium(®). and low(®) concentrations. Also plottedare valuesfor ECAM controls.(), ip injectedTNP-KLH (®).and anally intubatedTNP-KLH (b).All values were derived from subtracting background titers from pre bleeds of each animal assayed resulting values were averaged. All groups started with n=1 3 initially but due to death during bleeds numberschanged, throughout.High oral final n=12, Medium oral finaln=10,Low oral final n=12, Control finaln=5.Asterisks denote a statistically significant difference(p<0.05)between controls. 64
Table 3.1. Immunization schedule for TNP-LPS:
Method High Medium Low Administered
Oral 10 gga 10 nga 1 pga 30 daysb Ip injection 10 µg 10 ng 1 pg oncec Immersion 10 µg 10 ng 1 pg oncec Control Non-antigen 30 days coated beads
a. Represents dose/ fish/day b Administered once/day for entire 30 day period. c. Administered one time when oral feeding began. 65
Table 3.2. Immunization schedule for TNP-KLH:
Oral 100 µga 5 µga .5 µga 30 daysb Anal 100 ------once intubation µg ip injection 100 gg ------once Control Non-antigen------30 days coated beads a. Represents dose/ fish/day b Administered once/day for entire 30 day period. c. Administered one time when oral feeding began. 66
CHAPTER 4
Activation of an endogenous serine protease as a novel method for removal of p57 from the Renibacterium salmoninarum cell surface
Jon D. Piganelli I., Gregory D. Wiens 1,2 and Stephen L. Kaattari 1,3
'Department of Microbiology and Center for Salmon Disease Research Oregon State University, Corvallis, OR. 97331.
2Present address Oregon Health Sciences Centers Department of Molecular Microbiology and Immunology. Portland OR 97219
3Present addressSchool ofMarineScience,Virginia Institute of Marine Science The College of William & Mary Gloucester Point, VA 23062
To be submitted to: Journal of General Microbiology. 67
ABSTRACT
Renibacterium salmoninarum,a salmonid pathogen, has a hydrophobic
cell surface which correlates with isolate virulency.Incubation of Renibacterium salmoninarum cells at 370 C for> 4 hours decreases relative cell surface
hydrophobicity as measured by the saltaggregation test and with the proteolysis of the 57-58 kDa protein. No significant reductionin hydrophobicity was observed in cells maintained up to 16 hours 200C, 40C or -200C. Complete inhibition of the hydrophobicity,was prevented by preincubation with 5mM
phenylmethylsulfonyl fluoride (PMSF)suggesting that a serine protease is
responsible for the digestion. Relative cell surfacehydrophobicity of 370C
treated bacteria could be specifically reconstitutedby incubation of R. salmoninarum cells with partially purifiedextracellular protein (ECP) from R. salmoninarumcells. This reconstitution of relativecell surface hydrophobicity
correlated with the reappearance of the 57-58kDa band on the cell surface as determine by total protein stain and westernblot. 68
INTRODUCTION
Bacterial kidney disease is a chronic, systemic disease that isoften fatal for salmonids (Fryer and Sanders, 1981). The causativeagent,Renibacterium salmoninarum is a Gram positve, non-motile, slow growing, fastidious,
facultative intracellular parasite ( Fryer and Sanders,1981; Bruno, 1986; Evenden, 1993). Bacterial kidney diseasecan be transmitted directly by the feeding of raw viscera from infected fish (Wood and Wallis,1955) or horizontally from infected to non-infected fish sharing thesame water supply (Mitchum and Sherman, 1981;Bell etal., 1984). In spite of a significant effort and manyyears of research many aspects of the disease and causativeagent are still poorly understood. Many researchers in the field of fish diseaseconsider BKD the most difficult disease to control (Elliott, et al., 1989). Onereason for this difficulty is that the bacterium is a facultative intracellular parasitewhich has the ability to survive and possibly multiply within the phagocyticcell ( Bruno, 1986; Austin and Austin, 1987; Gutenberger, 1992; Bandin, 1993).Another reason for this lack of understanding is the difficulty of primary isolationin pure culture and the long incubation times (1-4 months) required forexperimental challenge of fish (Fryer and Sanders, 1981;Toranzo and Barjo, 1993). Thisprolonged incubation leads to inconsistencies in the production ofdiagnostic antigens and repeatable mortality kinetics (mean times to death), whichconfounds analysis of these components in the laboratory.
Renibacterium,unlike common Gram-negative, named for, fish
pathogens (i. e. Vibrio angiullarum and Aermonas salmonincida)does not immediately cause rapid mortality by disabling the host withthe production of high levels of extoxins.It does, however, exploit a variety of pathogenic mechanisms that contribute to death over the long durationof the disease. The 69
bacterium produces copious amounts of extracellularproteins (ECP) during infection (Turaga, 1987) which consist ofa 57-58 kDa protein (p57) or "antigen
F", as described by Getchell et al. (1985) andits proteolytic breakdown products
(Turga et al., 1987; Wiens and Kaattari,1989). The p57 protein is also
expressed on the bacterial cell surface and is the majorcell surface component
(Wiens and Kaattari, 1989; Daly and Stevenson1990). Turaga et al. (1 987a) demonstrated that incubation of ECP with salmonidlymphocytes caused immunosuppression of the antibodyresponse to a non-related antigen. These
data support the fact that ECP / p57 has biologicalactivity, and that activity could be linked to virulence. A number of otherin vitro functions have been
attributed to p57. These include hemagglutinationof rabbit and other
mammalian erythrocytes (Daly and Stevenson, 1987),agglutination of salmonid spermatoza (Daly and Stevenson, 1989) and the abilityto agglutinate salmonid leukocytes (Wiens and Kaattari, 1991). Additionally,a vascular permeability factor has recently been detected in the ECPwhich may increase vascular permeability in diseased fish (Bandin et al., 1992).
Possibly one of the bacterium's most importantvirulence factors is its hydrophobic cell surface (Daly and Stevenson1987; Bruno, 1988). In 1988, Bruno identified isolates of R. salmoninarumthat did not auto-agglutinate. This correlated with a relatively reduced cell surface hydrophobicityas determined by salt aggregation test. Strains which didnot autoaggluntinate were also less virulent than autoagglutinating strains.It was later determined that this reduction in virulence and hydrophobicitycorrelated with the lack of a saline extractable 57kDa protein (Bruno,1990). Daly andStevenson (1990) found that hydrophobicity could be restored to water-washedR. salmoninarum cells when they were incubated with ECP containing thep57 component. 70
In an attempt to characterize other pathogenic functions of the ECP, Rockey et al., (1991) reported that ECP exhibited proteolytic activity under elevated temperature. The responsible protease digested the majority of the proteins in ECP. The digestion of the ECP also abrogated the immunosuppresive function of ECP reported by Turaga. This proteolyticactivity was sensitive to phenylmethylsulfonly fluoride, ethanol and methanol, and thus found to be a serine protease. Using substrate gel elctrophoresis Rockey defined the molecular weight of the protease to be approximately 100 kDa.
In this manuscript we demonstrate that subjecting Renibacterium salmoninarum whole cells to an elevated temperature activates the endogenous serine protease which cleaves the 57kDa protein from the cell surface greatly reducuing the hyfrophobicity. A practical goal of sucha process would be to produce non-toxic antigens for use in vaccination. 71
MATERIALS AND METHODS
Preparation of Bacterial cells and ECP. Renibacteriumsalmoninarum
ATCC 33209or isolate D6 (originally obtainedfrom C. Banner,Oregon
Department of Fish andWildlife,Oregon StateUniversity, Corvallis,OR) was grown in a one liter volume in 2.5 liter low form, culture flasks(VWR) with intermittent shaking at17CC.KDM-II medium was prepared according to Evelyn
(1977) except without serum supplementation.Bacteria were grown for 7-8
days until an O.D.between 0.4- 0.8 (525nm)wasgenerated.Cells were pelleted at 6000 x g centrifugation for 30 minand resuspended in 100 ml cold PBS(PBS;0.85%NaCl, 10mMNaP04, pH. 7.2).After a second centrifugation
the cells were placed in eppendorf tubes forsalt aggregation test or stored at -20PC.
Extracellular protein (ECP)was extracted from culture supernatants as described by Wiens and Kaattari (1989).
Salt aggregation assay. Phosphate bufferedsaline washed R. salmoninarum cells were pelleted for 2 min ina microfuge (Beckman, model E). The wet weight of the cellswas determined after complete removal of supernatant. Cells were resuspended toa final concentration of 50 mg/mI in
10mM phosphate buffer without saline (pH 7.2).Cells were diluted to yield an O.D. of 0.95 (460 nm). These cellswere then used in the salt aggregation assay according to the method of Lindahl et al. (1981)as described by Daly and Stevenson (1987). Briefly, 25 ul of the cellsuspension was mixed with 25 ul of doubling dilutions of 2.0 M ammonium sulfate inphosphate buffer (adjusted to pH 6.8 with 1 N ammonium hydroxide) using 12well depression micro slides (Clay Adams, NJ). Slides were then agitated for5 min at 100 rpm (Junior Orbit 72
shaker, Labline Inc.) and agglutinationwas determined by observation with a dissecting scope (10x magnification) using oblique lighting.All assays were performed at room temperature.
Effects of time and temperature treatments of R. salmoninarum
cells. R. salmoninarum cells (450 pI of 50mg ml-1 in cell suspension 10 mM Phosphate buffer without salt) were incubated ineppendorf tubes (Intermountain Scientific, Bountiful, Utah) for 2, 4, 6,or 8 h at either -20, 4, 17, 22 or 370C. After incubation cells were microfuged for 2 min andan aliquot of the cells and supernatant were retained for electrophoretic analysis.
Addition of protease inhibitors and antibiotics to R. salmoninarum cells. The protease inhibitor phenylmethlysulfonyl fluoride(Sigma), was added at concentrations of 2 mM to 15 mMor 1 % ethanol was added as a diluent control. Antibiotics gentamycin, chloramphenicol,ampicillin, or tetracycline were incubated with cells at a final concentration of 2.5mg/ml at either 40C or 370C. After 8 h the cells were microfuged for 2 min andresuspended in PBS. Salt aggregation was performed as described.
Polyacrylamide gel electrophoresis and westernblotting. Polyacrylamide gel electrophoresis and westernblotting was performed as previously described (Wiens and Kaattari 1989). Preparationof R. salmoninarumcells for electrophoresis was as described by Wiens and Kaattari (1991). 73
RESULTS
Effects of elevated temperature on R. salmoninarumcell surface hydrophobicity and p57. Exposure of R. salmoninarumto 370C reduced the relative cell surface hydrophobicity after 4 h (Table 4.1). Cellsincubated at -20, 4, 17 or 220C up to 16 h demonstratedno change in relative cell surface
hydrophobicity. Loss of cell surface hydrophobicity alsocorrelated with the decrease of p57 on the bacterial cell surfaceas determined by total protein staining of 10% SDS-polyacrylamide gels (figure 4.1A). Analysisof supernatant revealed staining profiles comparable to p57 and its breakdownproducts (Rockey et al., 1991) (figure 4.1 B). Addition ofantibiotics gentamycin, choloramphenicol, ampicillin, tetracycline,or sodium azide had no effect on the loss of hydrophobicity or protein degradation profilessuggesting that protein systhesis or metabolism is not required (data not shown).
Effects of elevated temperature and co-incubation withprotease inhibitor PMSF on R. salmoninarum cell surfacehydrophobicity and p57. Incubation of PMSF at concentrations from 2 mMto 15 mM prevented the temperature-induced loss of hydrophobicity(Table 4.2). Heattreated (37°C) R. salmoninarum cells were completely protected from the lossof hydrophobicity if co-incubated with high concentrations (10-15 mM) of PMSF.Lesser amounts of PMSF offered lesser degrees of protection. This increasein relative cell surface hydrophobicity with increasing concentration of PMSFcorrelated with the partial inhibition of the loss of the p57 protein at 2 mMand complete protection of p57 from proteolysis at 15 mM asseen by total protein (figure 4.2A) and western blot
(figure 4.2B). 74
Reconstitution of relative cell surface hydrophobicitywith ECP. Relative cell surface hydrophobicity could be restoredby the addition of 1.0 mg/ml ECP to 370C treated cells (Table 4 3). Co-incubationof ECP with 370C treated cells restored the cell surface associatedp57 as seen by western blot (Fig. 4.3 A) and the cellular hydrophobicity. Treatmentof ECP at 370C was able to abrogate this restorative function. Restoration of hydrophobicitywas specific because heterologous proteins bovineserum albumin, hen egg albumin, or rainbow trout serum were unable to completelyrestore cell surface hydrophobicity. 75
DISCUSSION
Cell surface hydrophobicity isan important bacterial property which
facilitates anti-phagocytic activity (Van Oss1971), as well as attachment to substrates (Finlay and Falkow, 1989). A numberof investigators have
documented the hydrophobic nature of R. salmoninarumby the salt aggregation test, adherence to hydrocarbons, and bindingto nitrocellulose filters (Daly and Stevenson, 1987; Bruno, 1988; and Badin1989). This hydrophobic nature has
been correlated with isolate virulence,as those Renibacterium cells which lack
hydrophobicity are less virulent (Bruno, 1988).This lack of hydrophobicity also has been shown to be correlated with the lossof the 57-58 kDa protein (Bruno, 1990). Daly and Stevenson, (1987) have foundthat cell surface hydrophobicity
was decreased after cells were trypsin or protease K treated. Theloss of hydrophobicity due to protease treatmentcorrelated with the loss of the 57-58 kDa protein (Daly and Stevenson, 1990).
We demonstrated that .subjecting R. salmoninarumcells to heat treatment (370C) results in the proteolysis of the57kDa protein from the cell surface. This degradation is caused by the activationof an endogenous serine protease, as this proteolysis could be partially blockedwith the addition of PMSF at 2 mM and totally inhibited at 15 mM. These resultsare consistent with those of Rockey et al., (1991), characterizing theprotease associated with purified R. salmoninarum ECP, in that incubation of R. salmoninarumECP at 370C resulted in the degradation of the p57 protein fromthe purified ECP. This degradation of the p57 protein also reduced the invitro immunosuppressive activity of the ECP. 76
In these experiments itwas found that 370C 10 h incubation eliminated
cell surface hydrophobicity and enhanceddegradation of cell-associated p57.
The relative cell surface hydrophobicitycould be reconstituted with ECP butnot
heterologous proteins. Thissuggests a specific non-covalent attachmentof p57 to R. salmoninarum cell surface.Thiscorrelates the findings of Daly and Stevenson (1990).
The traits of a accomplished pathogenare the ability to perform
necessary functions including: gaining entry intoa host, finding an unique niche within the host, escaping, deceiving,or exploiting innate host defense
mechanisms, proliferating; and finally,exiting the host in a manner designedto maximize transmission toa new susceptible host (Falkow, 1991). To achieve
these goals the microorganismmust posses an armament of factors. These
factors are considered virulencefactors and are essential for thesuccess of the pathogen.It is obvious that Renibacteriumsalmoninarum possess putative virulence factors necessaryto achieve infection.
In order to elicit a protectiveresponse by vaccination againsta pathogen it is necessary to design thevaccine to be devoid ofany virulent activity, yet still
possess the immunogenic properties. Ashas been demonstrated in thisreport and other reports from researchers,the biological effects of ECP andmore
importantly the p57 moiety havebeen implicated in number of pathogenicand immunosuppressive activities. Therefore, we have exploited theactivation of a cell associated serine proteasethat cleaves the p57 moiety from thecell surface of Renibacterium salmoninarum, thus removing this putative virulencefactor. 77
ACKNOWLEDGMENTS
This research was supported inpart by Bonneville Power Administration award
#DE-PG79-89BP95906, USDA CSRSgrant #92-34123-7665, and from a Sigma Xi grant-in-aid of research awardedto Greg Wiens. The authors wish to thank
Drs. Jo-Ann Leong, J. Mark Christensen,Jerry Heidel and Don Mattson for their critical review of this manuscript. OregonAgricultural Experimental Station TechnicalReport #. Figure 4.1
A. B. 345 97 66 43 -57 31
22 15
Figure 4.1. Total protein stainsofR. salmoninarumcells (A) or supernatants (B) after 16 hour treatment at 370C (lane2), 17°C (lane 3), 40C (lane 4),or -200C (lane 5), molecular weight markers(lane 1). Figure 4.2
A. B. 2
-
57 18Ci 122 89 66 58 36 27
Figure 4.2. Total protein stain (A)or western blot using the anti p57 monoclonal antibody 4D3. (B) of R. salmoninarum cells after co-incubation with PMSFand 370C treatment. R.salmoninarumno heat treatment (Lane 1), R. salmoninarum cells 370C treated (lane 2), R. salmoninarum, and 1 % ethanolcontrol diluent (lane 3),R. salmoninarum370C treated and 2 (Lane 4), 5 (Lane 5), 10 (Lane 6), and 15 mM PMSF in ethanol lane (Lane 7). (B) Molecular weightmarkers, (Lane 1) R. salmoninarum no heat treatment (lane 2), R. salmoninarumcells 370C treated (lane 3), R. salmoninarum, and 1 % ethanol control diluent (lane4), R. salmoninarum370C treated and 2 (lane 4), 5 (lane 5),10 (lane 6), 15 mM PMSF in ethanol (lane 7). Figure 4.3
A.
1 2 3 4 5 6
180 123: 89 67' 50 38 34
Figure 4.3. Western blot using the anti p57 monoclonalantibody 4D3 of 37 oC treated R. salmoninarum cells with culture supernatant (ECP). Molecular weight markers (lane 1), whole cellsno heat treatment (40C control) (lane 2), whole cells no heat treatment(4oC controlformalin fixed) (lane 3), 370C treated cells (lane 4), 370C treated cells and ECP added back (lane5), ECP alone (lane 6). 81
Table 4.1. Minimum molar concentration of ammonium sulfate required for demonstrable R. salmoninarum aggregation at various time and temperatures. Aggregation of cells was observed at (10x magnification).
Time(hr) 37°C 20°C 17°C 4°C -20°C
0.5 0.004 0.004 0.004 0.004 0.004 1.0 0.032 0.004 0.004 0.004 0.004 1.5 0.25 0.004 0.004 0.004 0.004 2.0 0.50 0.004 0.004 0.004 0.004 4.0 2.00 0.004 0.004 0.004 0.004 8.0 2.00 0.004 0.004 0.004 0.004 16.0 >2.00 0.004 0.004 0.004 0.004 82
Table 4.2 Minimum ammonium sulfate concentration requiredto effect agglutination of R. salmoninarum after incubationwith PMSF at 370C. Agglutination (+)or lack thereof (-) was recorded after an eight hour incubation. Cells were observed under10X magnification. PMSF (mM) NH4SO4 0 2 5 10 15 0.004 - - - + + 0.032 - - - + + 0.25 - - + + + 0.50 - - + + + 2.0 + + + + + 83
Table 4.3. Reconstitution of relative cell surface hydrophobicity by concentrated culture supernatant. Cells were incubated at 37°C for 12 h, washed and incubated with the following concentration of protein.
Protein Added Temperature [M salt] for to RS cells treatment of RS cells aggregation of RS cells
Buffer alone 4°C 0.062 Buffer alone 37°C 2.0
1.0 mg/ml ECP 4°C 37°C 0.004 0.1 mg/ml ECP 4°C 37°C 1.0 0.01 mg/ml ECP 4°C 37°C 1.0
1.0 mg/ml ECP 37°C 37°C 1.0 0.1 mg/ml ECP 37°C 37°C 1.0 0.01 mg/ml ECP 37°C 37°C 1.0
1.0 mg/ml OVA 37°C 0.5 0.1 mg/ml OVA 37°C 1.0 0.01 mg/ml OVA 37°C 1.0
1.0 mg/ml BSA 37°C 2.0
1 /10 R BT Sera 56°C 37°C 2.0 1 /10 FBS Sera 56°C 37°C 0.5 84
CHAPTER 5 Evaluation of a protease-modified,Renibacterium salmoninarum whole cell vaccine, delivered orally and intraperitoneally
Jon D. Piganelli 1,2, Gregory D. Wiens 1,3Jia Allen Zhang 4,, John M. Christensen 4. and Stephen L. Kaattari 1,2,5
'Department of Microbiology and 2Centerfor Salmon Disease Research
4College of Pharmacy
Oregon State University, Corvallis, OR.97331.
2Present address Oregon HealthSciences Centers Department of Molecular Microbiology and Immunology.
Portland OR 97219
5Present address School Of MarineScience, Virginia Institute of Marine Science
The College of William & Mary
Gloucester Point, VA 23062 85
ABSTRACT
A whole cell Renibacterium salmoninarumvaccine was developed using heat-induced elimination of p57(a suspected virulence factor) from the cell surface. Coho salmon (Oncorhynchuskisutch) were immunized with this
p57- vaccine by either intraperitoneal (ip)injection or per os.In the first experiment the vaccine was administeredip and the fish were challenged by ip injection.In a separate experiment the p57-vaccine was administered either orally (in the food) or by intraperitonealinjection. The animals were challenged by bath immersion. Vaccinationof coho salmon (0. kisutch) in the first experiment by ip administrationof p57-, formalin fixed, R. salmoninarum conferred a statistically significant increasein mean time to death when ip challenged with R. salmoninarumat 4.1 x 106 colony forming units (cfu).
There was no significant differencedemonstrated by immunization with preparations containing water washedextracellular protein (ECP) or non-water washed ECP (ECP is composedof p57 and its proteolytic breakdown products) (Rockey et al., 1991).
The second experiment determinedvaccine efficacy by monitoring for the elaboration of p57 in the kidneysof vaccinated and control fish. As the infection progresses, p57 producedaccumulates in the infected salmon. Fish were sampled at time points of 0 (pre-challenge),50, 90, or 150 days post bath immersion challenge. The fishwere exposed to 4.2 x 106 cfu / ml on day 0. Fish orally vaccinated with p57-Renibacterium cells, using pH protected, enteric-coated antigen microspheres(ECAMs) as the delivery vehicle, demonstrated a significant difference(p<0.01) at day 150 post challenge in the p57 levels versus controls. Fish receivinga non-pH protected p57- 86
Renibacterium cells also showed a significant difference (p<0.03)versus control. Fish ip injected with the p57-cells or fish fed non-370C heat treated whole Renibacterium cells using ECAMs demonstratedno significant difference with respect to soluble antigen levels (p>0.05)versus controls. Intraperitoneally vaccinated fish demonstrated high circulatingantibody titers throughout the entire experiment. 87
INTRODUCTION
The first reports of bacterial kidney disease (BKD) of salmon were made almost 60 years ago ( Belding et al., 1935). To date the disease remains one of the most formidable foes of the aquaculture industry. The responsible organism, Renibacterium salmoninarum (Rs) is a fastidious, slow growing, non-motile, Gram-positive, facultative intracellular parasite (Fryer and Sanders, 1981) which causes a chronic and often fatal disease in a wide range of wild and cultured salmonid species (Austin, 1987). The disease can cause severe losses among intensively cultured salmonids (Bullock and Herman, 1988) and can be transmitted directly by the feeding of raw viscera (Wood and Wallis, 1955) or horizontally from infected fish sharing the same water supply (Mitchum and Sherman, 1981; Bell et al., 1984). Reliance on diet modification, chemotherapy, and selective segregation of infected brood stock have been the only effective, albeit limited means of controlling disease outbreaks (Elliott et al., 1989).
Prophylactic treatment through vaccination offers an ideal alternative for the control of BKD. However, investigators have had limited success at achieving a protective immune response against (Rs) using standard methods for vaccine preparation (Patterson et al., 1981, McCarthy et al., 1984; Sakai et al., 1989, Bruno, 1988, and Evenden et al., 1993). Reasons for this limited success may be do to our general lack of knowledge as to the mechanisms of pathogenesis and salmonid defense mechanisms (Kaattari et al., 1988). Also, failure to establish a reliable and natural challenge procedure has limited the success in this arena (Evenden et al., 1993).
Elimination of virulence factors from prototype vaccine material may benefit vaccine design. Renibacterium salmoninarum posses a number of 88
putative virulence factors that may contribute to its pathogenesis. Daly and Stevenson (1987) have reported on the ability of extracellular factors to agglutinate rabbit erythrocytes. Catalase, DNase, hemolytic, proteolytic and exotoxin activities have also been described (Bruno and Murno, 1982, Shieh, 1988; Evenden et al., 1993). Bruno 1988, found that isolates which do not auto-agglutinate have a reduced cell surface hydrophobicity andare less virulent. These isolates also lack a saline extractable 57 kDa protein (Bruno, 1990). Previous work in our lab (Rockey et al., 1991), characterizing the role of p57, has revealed that a novel endogenous serine proteasecan be activated by heat treating (Rs) cells at 37 oC for > 10hrs. This active protease then digests p57. Removal of this putative virulence factormay stimulate a protective immune response as postulated by Kaattari et al. (1988). Until recently the lack of a natural challenge procedure has also impeded vaccine research (Elliott et al., 1991; Murray et al., 1992). Previous challenge procedures have utilized ip injection of live cells (Bell et al., 1984). However, this method bypasses the skin andmucus, the first line of defense of the fish, and therefore, is not relevant to naturalexposure (Murray et al., 1992). Alternatively new challenge procedures consisting of bath challenges and prolonged incubation periods, which may more closely simulate natural exposure, have been performed (Elliott et al., 1991; Murray et al., 1992). Also, protocols such as the soluble antigen ELISAs, that accurately monitor the production of expressed antigen can be used to detect the progression of the disease throughout the entire challenge period (Pascho et al., 1987; Rockeyet al., 1991). This latter procedure eliminates the necessity of solely noting the mean day to death for evaluation of vaccine efficacy.
In this manuscript we tested the efficacy of the 370C treated (p57- ) cell in affording protection upon intraperitoneal (ip)as well bath challenge. 89
The prototype vaccine was administered intraperitoneally as well as orally, using enteric coated antigen microspheres (ECAMs). 90
MATERIALS AND METHODS
Animals. Coho salmon (Oncorhynchus kisutch)were obtained from the Oregon Department of Fish and Wildlife hatchery at Sandy, Oregon.Fish were kept at the Salmon Disease Laboratory, Oregon State University, in 120C
water. Fish were fed Oregon Moist Pellet (OMP) (Bioproducts, Astoria,OR) daily.
Preparation of bacterial cells and ECP.Isolate D6 (originally obtained from C. Banner, Oregeon Department of Fish and Wildlife, Oregon
State University, Corvallis, OR) was grown in 7, 1 liter volumes in 2.5 lowform, culture flask ( VWR) with continuous shaking at 170C. Briefly, the harvestof cultures was extended to growing sevenone liter volumes of bacteria in 2.5 liter low form, brand culture flasks with constant shaking in KDM-IImedium according to Evelyn (1977) except withoutserum. Bacteria were grown for 7-8 days until O.D. between 0.4-0.8 (525) was generated. Cellswere pelleted by 6000 x g centrifugation for 30 min and resuspended in 100 mlcold phosphate buffered saline, (PBS; 0.85% NaCl, 10 mM NaPO4, pH. 7.2), (PBS).After a second centrifugation the cells were placed in microfuge tubes and frozenat - 700C for further use. Extracellular proteinwas extracted from culture supernatants as described by Wiens and Kaattari(1989).
Preparation of Bacterial cell surface extract (CSE). Acell surface extract (CSE) was prepared according to the method of Dalyand Stevenson (1989). Briefly, 2-4g of wet weight bacterial cells were washed with 100 ml of sterile phosphate buffered saline (PBS) and pelletedat 6000 x g for 30 min. Cells were resuspended in 100 ml of ddH2O for 1 hon ice and the 91
cells were repelleted, the supernatant removed and precipitated with the addition of powdered ammonium sulfate to,yield a final 50% solution. The CSE was dialyzed 3x against (PBS) overnight and filter sterilized (0.45/1,
Gelman Scientific). Protein concentrationwas determined by the method of Lowry et al. (1951).
Vaccine preparation. Previously harvested cellswere thawed from -70CC, microfuged for 2 min (6000xg), weighed and resuspendedto 200 mg/ ml in sterile, cold PBS. The cells were placed at 370C for 48 h.Upon completion of incubation cells were microfuged and resuspended into3% formalin-PBS at 170C for 10 h. Cellswere washed 3x with 1 ml PBS and re weighed.
Antigen Preparation for the ip vaccination / challenge experiment. Antigens for the intraperitoneal vaccination (ip) and challenge experiment were prepared as follows: Antigenswere emulsified in Freund's incomplete adjuvant (FIA) for 4 min at 100 unitson a Virtis "23" mixer (Virtis Co., Gardiner, NY). PBS and PBS/FIA alone servedas controls. The emulsified antigens were as follows: 370C heat treated p57-Renibacterium cells (500 /1g),cell surface extract (CSE) (50 /1g) from water washed whole cells, and soluble protein (50/1g) from extracellular protein extractedfrom culture supernatants. Injections were initially given, ina total a volume of 0.1 ml, ip and intramuscularly, 0.05 ml in each site usinga 22g needle. Boosts were followed 45 days after primary injection and consisted ofone half the quantities used in the primary injection, contained in thesame total volume. A secondary boost was given 10 days after the first boost; the concentrations 92
remained the same as in the first boost. Each of the five treatmentswere performed in triplicate with a total of 45 fish / tank.
Intraperitoneal challenge. R. salmoninarum used in the challenge was grown for 7 days in KDM-II and washed 1 x in PBS. Cellswere resuspended in sterile PBS to a final O.D.(525) of 0.2 and delivered ipto the
fish. Plate counts confirmed the challenge dose to be 4.1 X 10 6cfu / ml. Mortalities were recorded daily. Gram staining and indirect immunofluorescent assay (IFA) (Bullock et al., 1980) was used to confirm the presence of R. salmoninarum in a portion of the mortalities.
Oral vaccine preparation. Previously harvested cellswere thawed for use in preparing the respective antigens for vaccination. Theantigens for the oral ECAM experiment were as follows: ECAMswere prepared by spray- coating the antigen, either 370C p57- cells (oral p57-) or non heat treated cells ( oral p57+) onto 40-50 mesh dextrose beads to administera dose of 100 Ng / fish / day. The beads were then spray-coated with Eudragit L-30D,a commercially available aqueous, pH-reactive latex dispersion of methacrylic- arylic acid co-polymer. The 370C p57- cellswere also coated, to administer 100 jig / fish/ day without the pH-reactive polymer toserve as an a non pH -protected control (oral NC p57-). The final control consisted ofnon-antigen coated, pH protected ECAMs (control). The fishwere fed the respective ECAMs incorporated in their diet. The ECAMswere mix uniformly throughout the mash of Oregon Moist Pellet (BioProducts, Astoria, OR) anddistilled water was added to form a mull. The mixture was then extruded througha defined pore size extruder (Vitano, EastLake, OH) and was cut into pellets. The fish received ECAM incorporated feed onan every other day basis for a total of 30 93
days. An intraperitoneal injection treatment was incorporated and consistedof 370C heat treated p57- emulsified in FIA (ip57-)as described, with modifications to injection. Fish received 500 jig injected ina volume of 0.1 ml anterior to the pelvic fin using a 26 gauge needle. The boost schedule forthe ip group followed the same scheduleas the previously described ip injection experiment. After ECAM and ip injection schedules were complete the fish were rested 20 days and then challenged. Each of the five treatments were performed in triplicate with a total of 25 fish / tank. To monitor the humoral responses and pre-challenge soluble antigen titers, five fish per tank were sacrificed, sera and kidney samples were collected prior to challenge.
Bath challenge for oral vaccine experiment. R. salmoninarum
(D-6 strain) was grown as previously described. The growth from3 one liter flask was combined. Fish were exposed to R. salmoninarum bybath challenge as described by Elliot and Pascho (1991) with modifications. The water level in each tank was first reduced from 125 liters to 25 liters. The flow ofwater was then stopped, supplemental aeration initiated. ViableRenibacterium salmoninarumwas added to the tanks to give a final concentration of 4.2 X 106 cfu/ ml as determined by plate count. The fishwere exposed to the bacteria for 22 hr in the standing aerated water. Following the 22 hrexposure the water flow was resumed and the tanks were allowed to fill ata rate of 2.8 liters / minute, thus removing bacteria through normal effluent flow. One replicate tank of non-heat treated (OC4°C) vaccinated fishwas lost due to interruption of air to the tank.
Determination of Antibody Activity. Antibody activity titerswere ascertained by the use of an enzyme-linked immunosorbentassay (ELISA) as 94
previously described by (Arkoosh and Kaattari 1990; Arkoosh andKaattari, 1991; Piganelli, et al., 1994) with modifications. Briefly, eachantiserum was titrated on an ELISA plate (Costar E.I.A/R. I. A. certifiedsurface chemistry, Cambridge, MA) using formalin fixed Renibacteriumsalmoninarum (Rs) as the coating antigen (150 pg /ml). Each plate containeda titration of an anti-Rs hyperimmune serum. This latter titration permits normalizationof the data and a standardized estimation of the units of activity per µl serum.
ELISA-based monitoring of diseaseprogress.Monitoring the progress of infection after challenge, was accomplished using the monoclonal antibody-based ELISA protocol described by Rockeyet al. (1991), with modifications. The ELISA precisely quantifies theamount of p57, a major component of the extracellular proteins produced by R. salmoninarum.Briefly, five fish from each triplicate treatmentwere sacrificed in order to monitor levels of soluble antigen. Samples were takenwere at 0 (pre-challenge), 50, 90 and 150 days post challenge. Kidney sampleswere collected from fish using fresh collection tools for each animal sampled to minimizecross-contamination. The kidney samples were placed into microfugetubes held on ice and then mixed 1:1 (weight : volume) with cold 1% bovineserum albumin in Tween 20- tris buffered saline (BSA/TTBS) (50 mM Tris, 1 mM EDTA, 8.7%NaCl and 0.1% Tween 20 [pH 8.0]). The sampleswere then homogenized via repeated passage through a 1 ml syringe. Supernatants were collectedas described by Rockey et al., (1991). The ELISAswere performed on all samples according to the protocol of Rockey et al., 1991, with modifications. Allreagents were loaded onto the plate manually instead of using the Autodropplate loader (Flow Laboratories, McLean, VA). Standard concentrationsof p57 were run on every plate in order to generate a standard curve. Incubation timeswere 95
followed as described and optical densities (absorbance)at 405 nm (A405) were measured on a Titertek Multiscan Plus plate reader (Flow Laboratories).
Derivation and analysis of the linear equation from theaverage of triplicate wells for each sample as well as the standardcurve were conducted using the
DeltaSoft ELISA program software package (BioMetallics). Theconcentration of p57 in each sample was calculated as described by Rockey et al., (1991),
using optical densities values generated from the standard curve.Theassay has a baseline detection limit of 3 ng /ml.
Polyacrylamide gel electrophoresis and western blotting. Polyacrylamide gel electrophoresis and western blottingwas performed as previously described (Wiens andKaattari, 1989).Preparations of R. salmoninarum cells used as antigens in the vaccine experimentswere electrophoretically analyzed as described in Wiens and Kaattari (1991).
Statistical analysis.All p57 montoring data were analyzed using one-way analysis of variance (ANOVA) program on the Statgraphics software package. ANOVA analysis generatesa F-statistic which is subsequently analyzedby thet-test to determine significance using the pooled standard
deviation (SP2). Ninetynine and ninety five percent confidence limitswere derived using standard error of the mean. 96
RESULTS
Total protein and western blot analysis of whole Renibacterium salmoninarum cell vaccine preparations.Total
protein stain (Figure 5.1 A) and western blot (5.1 B) of equal amounts (50pg)of treated and untreated Renibacterium salmoninarum cells. The removal of p57 after 48 h incubation at 370C followed by formalin treatment is demonstrated.
Effect of ip vaccination and challenge. The first vaccine experiment compared the effect of R. salmoninarum cell surface extract (CSE),
ECP and 370C p57- whole cells. Fish were ip challenged with 4x 106 CFU / fish of live R. salmoninarum cells. The challenge dose resulted in complete mortality in all tanks by 43 days. No difference in mortalitywas observed in mean time to death between CSE and ECP vaccine treatments and the saline control, (figures 5.1, B, C). However a significant extension of themean time to death was observed (p<0.05) in the p57- whole cell vaccine treatment (figure 5.1,D).
Effect of oral (ECAM) or ip vaccination and bath challenge. Fish which were vaccinated either by ECAM (oral)or ip injection delivery were bath challenged. The progress of the infection was monitoredover time by the determination of p57 production. Due to the variance in p57 levelsamong individual fish, all data was log transformed. Statistical analysis of the data collected from the sampling dates showed that the therewere no significant differences between treatments versus non-antigen coated control beadsat (0) pre, 50 and 90 days, post challenge (Table 5.1). Althoughmore fish in the 97 control elaborated p57 as opposed to the oral p57- group, the variance was too high. However, statistical analysis of the 150 day sampling demonstrated that oral p57- cell treatment demonstrated significantly less p57 compared to the controls (P<0.01), the oral NC p57- also demonstrated significantly less p57 versus controls (P<0.03) (Table 5.1). The ip p57- and the oral C p57+ groups showed no significant difference versus the controls (P>0.05). Serum antibody titers were also monitored at each sample day and showed that the ip p57- treatment had produced specific serum antibody titers throughout the sample period (Table 5.2). The antibody titers in this groups were much greater than found in the ECAM vaccinated groups (Table 5.2). 98
DISCUSSION
We have examined the efficacy of a vaccines based on the removal ofa virulence factor (p57) from Renibacterium salmoninarum cells in affording protection upon intraperitoneal (ip) or bath challenge. The candidate vaccine was administered intraperitoneally by injection as well as orally using enteric coated antigen microspheres (ECAMs). The first experiment strictly employed ip injection with an ip challenge, and demonstrated that a significant increase in mean day to time to death (p<0.05) was observed with the 370C treated whole cells. However, there was no significant difference between fish immunized with cell surface extract (CSE) or extracellular protein (ECP) alone versus the control. ECP did not increase the mortality rate as observed by other investigators (Tuaga et al., 1986) compared with the salineor saline /FIA controls.
Although a significant delay in the mean day to death was observed using the p57- cells, the ip exposure chosen for challenge resulted in complete mortality. This route of challenge bypasses the outer integument of the fish and introduces the pathogen in an unnatural way. In fish, unlike mammals, the entire interface with the environment consists of a mucosal epithelium, therefore it would seem that these mucosal surfaces would playa critical role in fish's first line defense (Davidson et al., 1993). Therefore the second experiment was conducted in order to assess vaccine efficacy utilizinga more natural route of pathogen exposure, bath challenge. We also attempted to assess efficacy of a feasible and stress free method of vaccine delivery feeding ECAMs. Challenge methods based on bath immersion have been established and proven to be effective model for studying Renibacterium infection ina laboratory setting (Murray et al., 1992). The disadvantage of these challenge 99
methods is that considerable time is needed to reachsignificant mortality. However, this disadvantage has beenovercome with the advent of diagnostic ELISAs that can effectively monitor the production ofantigens elaborated by Renibacterium salmoninarum in situ. Suchassays circumvent the need to solely rely on mean day to deathas the means of assessing the infection
status, and provide a more natural, informative and convenientmeans of determing vaccine efficacy.
Fish in the second experiment were vaccinated by ECAMfeeding or by conventional ip injection. All fish were then bath challenged. Theresults of this experiment demonstrated route of administrationwas important when
natural challenge conditions were employed. Fish fed ECAMscoated with p57- whole cells (oral p57-) demonstrateda significant difference (p<0.01), at 150 day post challenge, with respect to the amount ofsoluble antigen that was detected versus controls. The increase in p57 concentrationsin the kidney of control fish is indicative of Renibacterium infection. Those fishfed unprotected p57- cell coated beads (without the pH polymer protections)also showed a significant decrease (p<0.03) in p57 concentrationversus control, however the significance was not as great as with the pH protected ECAMs.The fish fed ECAMs coated with the non-heat treated cells (oral p57+)showed no statistical difference versus controls with respect to p57 levels throughoutthe entire experiment. Those animals receivingan ip injection p57- cells (ip p57-) demonstrated measurable and, on theaverage, higher units of anti- Renibacterium serum antibody throughout the entireexperiment. However, this serum antibody response did notappear to have protective value when the pathogen was administered by bath immersion. At day 90 theip p57- vaccinated animals demonstrated higher concentrationsof p57 than all other groups including control. These resultsare consistent with other reports which 100
relate that fish exposed to Renibacteriumcan mount an immune response as indicated by antibody titers, however thepresence of this has not correlated
with protection against the disease(Evelyn, 1971,Paterson et al.,1981,Bruno, 1987).Recently it has been reported by Davidson et al., (1993)that route of immunization has an impacton the generation of antibody secreting cells in the gut of rainbowtrout (0.mykiss). They demonstrated that oral immunization
of fish produced a faster peak antibody titer ( threeweeks)in both the intestinal mucosa and head kidney as compared to those fish whichwere ip injected (sevenweeks).This rapid induction of a peak antibody titer in the intestinal
mucosa may explain the efficacy of the oralp57- vaccine.The mucosal response may have allowed for protection at the mucosal epithelium which is crucial to diseaseresistance,when the route of exposure is through mucosal sites (Finlay andFalkow, 1989,McGhee etal., 1991).Lobb etal.,(1986) demonstrated that bath immunization of channel catfish(Ictalurus punctatus) produced cutaneous mucus antibody in five of six animalsfollowing immersion in dinitrophenylated-horse serum albumin (DNP-HSA).However only one out of five fish demonstrated a marked increase inserum anti-DNP antibody following the bathimmersion.These studies demonstrated that the channel catfish immune system could respond differentially dependingon the route of exposure.Lobb also showed that the channel catfishpossess numerous lymphocytes associated with theepidermis.There have also been reports that have serum antibody titers of bathor orally immunized fish do not correlate with protection to against certainpathogens.Forexample, Croy and Amend (1977) observed significant protectionwas achieved with bath immunization against vibriosis in the absence of serumtiters.Because Renibacterium is a facultative intracellular parasite the role of protection bycellular immunity should not not be over looked, howevermore work and development of cell 101
mediated assays are needed to explore this possibilty.Clearly, if oral delivery of the ECAMs coated with p57- (oral p57-) elicitedmucosal immunity then
protection may have been achieved at the initial siteof entry. In conclusion, the results presented here demonstratethat the removal of a putative virulence factor p57 and route of deliveryare critical factors to consider if protection against Renibacterium infection isto be achieved. The orally delivered p57- cells (oral p57-)afforded significant protection when fish were exposed to viable Renibacterium by bath immersion, while the ipp57- and oral p57+ cells did not. Although in the first experimentthe ip p57- vaccine demonstrated a significant delay inmean time to death after ip challenge, the results of the second experiment proved that whenanimals are challenged by a natural route this method immunization was significantly less effective.
These results show promise for elucidating theideal candidate antigen and method of delivery of that antigen to stimulatea protective immune response in salmonids against BKD. 102
ACKNOWLEDGMENTS
This research was support in part byUSDA-CSRS #92-34123-7665 and USDA (WRAC) # 91-38500-6078.The authors thank Drs. J. Mark Christensen, J. Heidel, Don Mattsonand J. C. Leong for their critical review of this manuscript. Oregon AgriculturalExperimental Station TechnicalReport #. Figure 5.1
A. B.
1 2 3 1 2 3
97 66 f57 43
31
22 14
Figure 5.1. Total protein stain (A) and western blot (using the monoclonalanti p57 antibody 4D3) (B) of R. salmoninarum after 370C treatment for 48h followed by 0.3% formalin incubation at 170C for 10 h: molecular weight marker (lane 1), untreated R. salmoninarum cells (lane 2), 370C treated R. salmoninarum cells (lane 3). 104
Figure 5.2. Percent survival of coho salmon immunized withsaline, FIA (A),
ECP /FIA (B), CSE / FIA (C) and p57- cells / FIA (D). Forsimplicity each treatment was graphed againstthe FIA control. Three tanks of 40 fish were used for each treatment. Mean percent survival and standard deviationwere calculated for each day after challengewith 4.1 X 106 cfu. Asteriskdenotes significant diference at the 0.05 level (*= p<0.05). A. 10 Days After Challenge 20 30 40 50 0 10 Days After Challenge 20 30 40
10 Days After Challenge 20 30 40 Figure 5.2 10 Days After Challenge 20 30 40 106
Table 5.1.
Values expressed are the means of p57 detected (ng/ml) for each particular treatment at each sampling date. Standard errors are in parentheses. Asterisk= denotes significant difference from control p<0.01, = p< 0.03 versus control. Treatment number of fishmean pre- mean 50 daysmean 90 daysmean 150 days challenge post challenge post challenge post challenge aControl 15/sample day<3 ng /ml # 2.4 (.34) 351 (352) 2070 (1600) bOral p57- 15/sample day<3 ng /ml 1.2 (.2) 20 (18) 1.9(.419)- cNPP p57- 15/sample day<3 ng /ml 1.9 (.51) 21 (17) 2.9(.59)-- dOral p57+ 1 0/sample day<3 ng /ml 1.3 (.3) 8701 (8600) 8403 (5603) 00_p57- 15/sample day<3 ng /ml 2.14 (.94) 12900 (6400) 220 (173) a= control non-antigen coated beads. b= ECAM delivered p57- whole cells c= Non-pH protected p57- whole cells. d= p57+ whole cells. e= intraperitoneal injected p57- whole cells. #= detection limit of assay. 107
Table 5.2.
Values expressed are the means ofserum antibody units of activity / µl serum detected for each particular treatment at each sampling date. Standarderrors are inparentheses. Treatment number of fishmean pre- mean 50 daysmean 90 daysmean 150days- challenge post challengepost challenge post challenge 8Control 15/sample day-ND- -ND- 379 (272) 2060 (2276) bOral p57- 15/sample day-ND- -ND- -ND- 126(106) cNPP p57- 15/sample day-ND- -ND- 938 (570) 1827 (738) dOral p57+ 10/sample day-ND- -ND- 500 (320) 3423 (603) eip p57- 15/sample day5800, (748) 42400 (13432)82000 (56312)14776 (6119)
a= control non-antigen coated beads. b= ECAM delivered p57- whole cells c= Non-pH protected p57- whole cells. d= p57+ whole cells. e= intraperitoneal injected p57- whole cells. -ND-= not detectable. 108
CHAPTER 6 Assessment of immunological memory to a T-dependent antigen in rainbow trout (Oncorhynchus mykiss) by the in vitro elicitation ofa respiratory burst enhancing cytokine
Jon D.Piganellil, D.V. Mourich1, J.C. Leong1 and S.L. Kaattaril.2
'Department of Microbiology and The Center for Salmon Disease Research Oregon State University, Corvallis, OR. 97331
2Present address School of Marine Science, Virginia Institue of Marine Science College of William & Mary
Gloucester Point, VA 23062
To be submitted to: Developmental and Comparative Immunology. 109
ABSTRACT
A model assay is described forassessment of cell-mediated immunological memory in fish. Peripheralblood lymphocytes from rainbow trout (Oncorhynchus mykiss) primedwith trinitrophenylated keyhole limpet
hemocyanin(TNP-KLH) were stimulated in vitro withthe same antigen to produce a secreted cytokinewith respiratoryburst enhancementactivity. Respiratoryburst enhancementactivitywas measured by incubating naive pronephriccells with the supernatant factor(SF). Factoractivatedpronephric cell respiratory burst levelswere assayedby the oxidation of a fluorochromic reagent 2', 7' -dichlorodihydrofluoresceindiacetate (DCFH-DA). The induction
of cytokine productionwas specific since in vitro exposure witha non-related antigen failedto produceelevated enhancementof respiratory burst activity. Dose dependency was exhibitedas decreasing concentrationsof SF gave decreasingrespiratory burstactivity.In order to estimate the molecular weight of theSF, in vitro radiolabeling and subsequentfractionationby filtration exclusion was performed. The results suggested a protein species near 30 kDa was responsible for the SFactivity. Electrophoretic distribution of the radiolabeled, de novo synthesized'proteins produced by the stimulated cells indicated thatthere were protein speciesat 34, 36and 89kDa. These were only seen in those cultures producingthe activity. To determine if the SFwas a gamma interferon-like factor, anti-viralassays were conducted. Supernatant derived factorconferred no anti-viral activity againstIHNV. Enrichment for surface immunoglobulin (slg+)or negative (slg-) cells and the comparison SF production in these two cultures indicatedthat slg- were likely responsible for the SF production. 110
INTRODUCTION
It has been well established, in both mammals and fish that T-cell dependent antigens (i.e. those antigens for which B-cells require T-lymphocyte
help) are proteins. These antigens require processingand presentation in
order to induce the antigen specific T lymphocyte (Unanue,1984; Schwartz, 1985; Vallejo et al., 1990 ). The processing and presentationevents occur within antigen presenting cells (APC), whose interaction with T-cellsis crucial to the establishment of immunological memory (Unanue, 1984; Schwartz,
1985). Evidence for the induction ofa memory state may be obtained from the examination of both humoral and cellular immuneresponses. In fish, such examination has been confined to the humoralresponse (Miller and Clem, 1984; Tatner, 1986; as reviewed by Kaattari, 1992; Wilsonand Warr, 1992). Fish B-cell memory is usually characterized byan increase in specific antibody titer after secondary exposure to antigen (Trump and Hildeman,1970; Desvaux and Charlemagne, 1981; Arkoosh and Kaattari,1991). However, unlike the mammalian response, antibody isotype switching(Lobb and Olson, 1988; Killie, et al., 1991) and affinity maturation (Russelet al., 1970; Voss, et al., 1978;0'leary, 1980, Makela and Litman, 1980) havenot routinely been observed in fish.In this report, an assay that assesses immunologicalmemory in fish by measuring the production ofa cytokine(s) upon in vitro antigen specific stimulation is described.
There have beenmany reports of the production of cytokines in several species of fish after stimulation with mitogens like concanavalinA (ConA)
(Secombes, 1992) or with exposure to other antigens (Caspiand Avtailon, 1984; Blazer et al, 1984; Graham and Secombes, 1988; 1990).Furthermore it 111
has been shown in fish, that cytokinesare necessary for in vitro immune responses. Some investigators have shown thatan exogenous factor similar to interleukin-1 is able to replaceaccessory cell function (Sizemore et al., 1984, Ortega, 1993). These studies indicate thatfish, produce cellular factors that induce a response in fish macrophages ina manner similar to mammalian cytokines. Based on such findings,we have assessed memory to a specific antigen by measuring the production of cytokinesinduced by subsequent in vitro exposure to that specific antigen. Restingpopulations of peripheral blood cells obtained from adult fish thatwere previously exposed to antigen were found to liberate a cytokine-like factorupon in vitro addition of the original priming antigen. When this factorwas then used to activate a naive population of pronephric lymphoid cells, an enhanced respiratoryburst was observed. 112
MATERIALS AND METHODS
Animals. Shasta strain rainbow trout(Oncorhynchus mykiss)were obtained from Dr. Jerry Hendricks, Departmentof Food Science and Technology, Oregon State University and maintainedat the Salmon Disease Laboratory in Corvallis, OR. This facility receivesfish pathogen-free well water at a constant temperature of 120C. The fishwere kept in aquaria receiving water at 0.5 gallons per minute and fed Oregon Moist Pellet(OMP) commercial fish diet.
Antigens and immunization. Trinitophenyl keyholelimpet hemocyanin (TNP-KLH) was preparedas previously described (Rittenberg & Pratt 1969). Ovalbumin was obtained from (Sigma).Antigens were adsorbed
to bentonite according to the protocol of Buttke,1987. Rainbow trout (600 g) were injected intraperitoneallywith 100 µg TNP- KLH emulsified in Freund's complete adjuvantand boosted at 9 weeks with 50 µg TNP-KLH Freund's incomplete adjuvant. The animalswere rested for >40 weeks after the last injection beforeuse in the in vitro analysis.
Cell preparation, tissue culture and factorproduction. Peripheral blood leukocytes and plasmawere obtained from trout >40 weeks post final injection. The fish were anesthestized in benzocaine(ethyl p- aminobenzoate, Sigma) (Kaattari and Irwin, 1985)and blood was taken from the caudal vein as described by DeKoningand Kaattari (1991). The blood was centrifuged at 500 x g to separate plasma fromcell pack. Plasma was removed and diluted to 10% with RPMI 1640 withoutphenol indicator (Sigma, MO) and placed in a water bath for 45 minutesat 42°C, to inactivate 113
complement. The cell pack was diluted 1:5 v/vwith RPMI and layered over an equal volume of Histopaque-Ficoll 1077 (Sigma).The tubes were centrifuged and the layer of cells at the interface (bullycoat) was removed and washed as described by DeKoning and Kaattari (1991). Cellswere then resuspended in 10 ml RPMI and cell viabilitywas determined by trypan blue dye exclusion. After enumeration, cells were resuspendedto a final concentration of 2 x 107 cells / ml in RPMI 1640 minus phenolindicator containing 4% heat inactivated autologous plasma and 0.05µg/ml gentamycin sulfate (Whittaker Bioproducts Inc.). Aliquots of 50 µlwere cultured in individual wells of a 96
well flat bottom tissue culture plate (Nunc,IL). Peripheral blood leukocytes were pulsed with 50 µl either TNP-KLHor ovalbumin adsorbed to bentonite at a concentration of 30 ug / ml and incubated inan incubator culture chamber (Model 624, C.B.S. Scientific Co., Del MarCA) at 17°C, in a blood-gas environment containing 10% CO2, 10% 02,80% N2. After a 24 h period the cells were washed twice to remove antigenand fresh medium was then added. Forty-eight hours later supernatant fluidswere harvested and centrifuged (300 x g) to remove cell debris. Thesupernatant fluids were filter-sterilized, (0.45pm filter, Gelman Scientific) divided intoaliquots and frozen at -200C for later use.
Factor indicator assay, induction ofoxidative burst activity in pronephric macrophages incubated withcultured supernatant. A naive rainbow trout was sacrificed and thefirst third of the anterior kidney was removed for preparation ofa single cell suspension. The cell suspension was layered over Histopaque-Ficoll 1077 (Sigma)and the buffy coat was washed twice. Then, the cells were enumeratedand checked for viability by trypan blue exclusion. Cell numbers were adjustedand plated at 1 X 106 cells / well 114
in a 96 well plate.Fifty microliters of the supernatant fluid generated by antigen stimulation or control (unstimulated supernatantfluid) was added to the cells for 24 hrs. At that time thesupernatant fluid was removed and the cells were prepared for the oxidative burst fluorometricassay. The assay employs a cytoplasmic fluorescent indicator,2', 7' -dichlorodihydrofluorescein diacetate (DCFH-DA) (MolecularProbes, Eugene,OR), that is added to thepronephric-derived macrophages at a final concentration of 5 µM / well for 30 minutes. The DCFH-DA isa non- fluorescent, membrane permeable agent thatdiffuses readily into the cytoplasm. Once inside the cell DCFH-DA is cleavedby non-specific cellular esterases to remove the diacetate residue trapping DCFHwithin the cell in an non-permeable, but still non-fluorescent reducedstate (Ryan et al., 1990; Bass et al, 1983). The cells were triggered by the additionof phorbol myristate acetate (PMA), to induce respiratory burst, (Sigma)at a final concentration of 100 µg / ml. Upon triggering with phorbol myristateacetate, oxidation products in macrophages are generated. Theseproducts in turn oxidize DCFH to its fluorescent form 2, 7' -dichlorodihydrofluorescin.The reaction can be monitored with a fluormeter. A Cytofluor 2300 FluorescentMeasurement System was used which providesan excitation wavelength at 485 nm and measures specific fluoresence of thereactive fluorchrome at a wavelength of 530 nm (Wan et al., 1993; Ryanet al., 1990).
Size fractionation of active supernatants.Antigen stimulated and control cell supernatants were fractionated ina stepwise fashion using centricon microconcentrators (Amicon, Inc. Beverly, MA)following the manufacturer's protocol. The first separation excludedmolecules larger than 100,000 M.W. This was followed by successive filtrationthrough filters that 115
excluded 30,000 M.W. and then 10,000 M.W species. Bothfiltrates and retentates were saved after each fractionation and stored frozenuntil testing in the oxidative burst fluorometricassay. Retentates and filtrates were concentrated 5x and greater as a result of the fractionationprocedure. Sterile conditions were maintained throughout the procedures.
35S-methionine in vitro labeling and SDS-PAGEde novo synthesized soluble proteins. Control PBLsor PBLs pulsed in vitro with 30 µg / ml TNP-KLH were radiolabeled with 0.4 pCui 35S-methionine/ well (Amersham). The cells were incubated for24 hours at 170C, washed twice and then radiolabel was added foran additional 48 hours. Supernatant fluids from the labeled cells were then harvested, separated intodifferent molecular weight fractions and concentrated with the Centriconmicroconcentrators. Fifty microliters of the concentrated sampleswere mixed with a loading dye containing B-mercaptoethanol. Each samplewas boiled for 5 min and loaded into separate wells of a 10% SDS-PAGE gel(12 cm X 15 cm). Current was applied at 25 constant volts overnight at 40C. Thegels were removed and enhanced using the fluorography enhance kit (Dupont).Gels were subsequently dried and placedon hyperfilm (Amersham) in the dark for 10 days.
Enrichment of (surface Immunoglobulin) slg-cells and slg+ cells by Dyna bead Biomagnetic separation.Harvested peripheral blood leukocytes from an antigen primed rainbowtrout > 40 weeks prior were counted and then added to a 75cm2 flask, incubatedat 170 C for 24 hours to allow for adherent cells to adhere. At this timenon-adherent cells were gently removed by washing with medium and transferredto a 50 ml conical tube, 116
washed, enumerated and resuspended ata final concentrationof 2 x107 cells / ml. The remaining adherent cellswere removed with a cell scraper (Falcon,
VWR), transferred to a 50 ml conical tube, washedand resuspended to a final concentration of 5 x 105 cells / ml. Wells ina 6 well plate (Nunc, IL) previously coated with bovine fibronectin (10 Ng/ml) (Sigma) received5 x 105 cells each and the cells were allowed to incubate for4 hours at 170C(Ortega, 1993).
After this incubation the adherent cells Were scraped off theplastic well,
enumerated and checked for viability then resuspended to 6 x105cells / ml and 500 µl /well was plated in a 24 well flat bottom plate.
Dynal tosylactivated magnetic beads (M-280,Dynal, Lake Success NY) were washed and coated with 1 / 14 (Deluca et al., 1983),a monoclonal antibody against trout immunoglobulin. Non-adherent lymphocyteswere added to 1 / 14 coated magnetic beads and gentlyrotated for 45 minutes at 17oC. After incubation, cellswere removed to 1.5 ml microfuge tubes and
placed into a magnetic particle concentratorfor 10 min at 40C. After concentration unbound cells were removed, centrifugedenumerated and checked for viability then resuspended toa final concentrationof 2 x 107 cells / ml. Magnetic particles were washed twice toensure that all non-specfically bound cells were removed.
SurfaceIg+cells bound to magnetic beads were harvested using the
Detacha Bead Reagent (Dynal). The slg+ cellswere washed and resuspended to a concentration of 2 X 107 cells / ml.
Non-adherent slg+ or slg- cellswere individually pipeted onto the pre- plated adherent cells or combined ata final number of 1 X107 cells / well. These cells were exposed to antigenas described to determine whether the production of the trout cytokine was associated with slg+,slg- or adherent cells. Supernatant fluids from individual wellswerecentrifuged,removed and frozen 117
at -20 0C. Samples were thawed and used in the oxidative burstfluorometric assay as described.
Test for antiviral activity of the antigen-induced factor.A direct neutralization of infectious hematopoietic necrosisvirus (IHNV, Rangen isolate) by the factor was ascertained. Stock factorat a 40x concentrate was diluted 1:10 or 1:100 and incubated overnight with 1000or 100 PFU of IHNV in a total volume 300 pi HBSS medium. One hundredmicroliters of each incubation solution (300 or 30 PFU)were placed on a confluent monlayer of RTG cells in triplicate wells and allowed to incubate for1 hr at 17°C. The wells were then overlayed with methylcellulose containing medium and incubated for an additional 7 days after which viral plaqueswere enumerated. An indirect or interferon-like activityassay was also performed as described (deKinkelin and Dorson, 1973 and Tengelsenet al. 1989). Briefly, factor containing supernatant fluidwas diluted 1: 10 and applied to preformed monolayers of RTG cells ina 96 well microtiter plate (MTP). Poly l:C as a positive control was added to monolayers in two fold serialdilutions starting with 100 to 12.5 ng / well. Cells were incubated for 48 hrat 170C after which IHNV was added at 4.5 x 102or 4.5 x 101 TCID50 / well. Plates were incubated for an additional 7 days. Cellswere fixed, stained and CPE was scored as positive or negative by comparison to control cells receivingno treatment. 118
RESULTS
Detection of antigen-induced macrophage acitivity. The macrophage activation assay was used to determine thespecificity of recognition by in vivo primed lymphocytes. To achieve this,pronephric kidney leukocytes were incubated with the priming antigen (TNP-KLH)or a control antigen (ovalbumin). Oxidation of DCFH to its reactive fluorescentform, 2', 7' -dichlorodihydrofluorescein, by factor-activated naive pronephricmacrophages (PM) was measured with the Cytofluor 2300 Fluorescent Measurement
System. The (PM) that received TNP-KLH derivedsupernatant from TNP-KLH stimulated cultures demonstrated a statistically significantenhancement in respiratory burst activity with respect to control antigenand non-antigen stimulated culture supernatants (figure 6.1A). The activitywas also titratable (figure 6.1 B). The control supernatant fluid from in vitrounstimulated, TNP- KLHin vivoprimed, cells demonstrated a negative regulatory affect with respect to respiratory burst activityas the PM with PMA triggering agent alone demonstrated a greater activation (figure 6.1 A).
Approximation of factor molecular weight.In order to determine the approximate molecular weight of the factor,supernatant fluids were subjected to centricon filter fractionation. The filtrate andretentate from the fractionation steps were assayed activity. Theassay confirmed that a significant portion of the activity was found in the retentateof the 30 k fraction (i.e. > 30,000d). (figure 6.2). 119
Determination of lymphocytesource of the factor source. Cell panning techniques were employed to determine the typeof lymphocyte that was responsible for production of the factor. Using 1 /14 Dynal coated magnetic beads, lymphocytes from antigen-primedtrout were separated based upon adherence, slg+ or sig- markers. Separated lymphocyte populationwere co-cultured as described with TNP-KLH for factor production.The supernatant derived from slg- enriched cultures produceda significant increase in factor activity versus all groups (figure6.3). While the slg+enriched cultures produced significant factor above control itwas not significant compared to unfractionated cell culture derived supernatant. Althoughboth slg- and slg+ enriched cultures produced measurable levels of factoractivity, the slg- cultures were significantly higher, suggesting that the sig-cells are the source of the factor.
Analysis of anti-viral activity in factor. To determineif the factor contains inteferon-like activity, anti-viralassays were performed. First, the factor did not have any direct effecton IHNV. Virus incubated at a final concentration 300 or 30 PFU / well with the 4xor 0.4x concentrated supernatant gave the same number of virus plaquesas did untreated virus. Cell monolayers were also treated with the factorcontaining supernatant fluid to determine if an anitviral state could be induced. Cytopathiceffect was observed in the the factor treated cells withno apparent reduction compared to the controls. Poly l:C was usedas a positive control for the induction of an antiviral state. A 50 % reduction in CPEwas seen in both viral infective doses tested, down to a concentration 25ng / ml Poly l:C. 120
Analysis of de novo protein synthesis after in vitro antigen pulsing. To analyze thede novosoluble proteins liberated into the supernatant fluid after incubation with specific antigen, 35S methioninein vitro radiolabeling experiments were conducted under factor productionconditions. The radiolabeled protein profile for the TNP-KLH stimulatedcells was distinctively different from the unstimulated cells (figure 6.4). Denovo synthesis of proteins demonstrated exclusive bands at 90,36.5 and 34 in those fliter fractionated supernatant fluids from cells thatwerein vitrostimulated with antigen, produced different banding patterns exclusive to theirtreatment. 121
DISCUSSION
The immune response in fish is generallyattributed to the same cell-cell interaction as observed in mammaliansystems (Clem et al., 1991). The specific adaptive response minimally enlists threedistinct cell types; macrophages, B-cells and T-cells. Each of theseplays a critical role to form
an intricate network of signals that leads to a successful immunesystem response.
The macrophage is an essentialauxiliary cell forthe generation of a specific immune response.It assists both B and T cells and its activitycan inturn be modulated by both antibody fromB cells and cytokines from T cells (Street and Mosmann, 1991). As in mammals,the role of antigen processing and presentation by macrophages is essential forteleosts in eliciting a
response to T-dependent antigens(Vallejo et al., 1992). Antigensare processed and presented by macrophages andother antigen presenting cells (APC) such as monocytes, to T cells ina MHC restricted fashion (Yang et al. 1989, Vallejo et al., 1990). The prerequisiteof processing proteins is required for recognition by T-cells, and ultimately,the induction of antigen-specific T and B effector functions.These cells also generate thenecessary cytokines crucial for growth and costimulatory signals for T-cellactivation (Weaver et al., 1990; Ortega, 1993).
A memory response to an antigen is dependentupon all the components described above such as antigen processing,cell to cell interaction for presentation, production ofand cognate activation by various cytokines. Memory responses in mammals havebeen found to be contingent upon two conditions: (1) the participation of T-cellsand (2) the antigen or 122
carrier is a protein (T-dependent), either solubleor cell associated (Gray and Sprent, 1990; Vitetta et al., 1991; Tittle, 1978).
It has been demonstrated in this study that in vivo primed and rested
rainbow trout, when pulsed in vitro with thesame protein antigen, demonstrate specific lymphocyte induction. The induction evidenced by the production
macrophage activaing function, onlyseen in those leukocytes that received a
secondary in vitro exposure to the priming antigen. A control antigen didnot induce this activivty (figure 6.1). The SF behaved ina dose dependent fashion as activity decreased with decreasing concentration of factor (figure 6.1 B).
A molecular weight of approximately 30 KDa was deduced by filter fractionation of the SF activity (figtre6.2) and in vitro radiolabeling
experiments. The in vitro labeling experiments also demonstrated thatprotein
bands at 34, 36 and 89 kDa were exclusively found in TNP-KLHpulsed cultures (figure 6.4). Cytokine molecules typically exhibit molecularweights between 17 to 39 kDa (Balkwill 1991). The proteins producedde novo when exposed to antigen fall with in this range. Also, the activity of the factorfrom size fractionation experiments suggesta factor in this range. Both the activity and molecular weight suggest a cytokine that has not yet been describedin the current fish literature.
In order to determine the lymphocytic origin of the SF,cell enrichment experiments were conducted. Peripheral blood leukocyteswere separated based on adherence via slg+.It has been shown by Vallejo et al., (1990) that in order for channel catfish lymphocytes to respond in vitroto TD antigens, accessory cell functions such as processing and presentationare required. Therefore in these experiments, adherent leukocyteswere included with all cell fractions. Alone however, adherent leukocyteswere incapable of producing the activating macrophage SF. It is likely that the slg+ cultures were 123
less efficient at producing the SF,since they were depleted of the majority of T- lymphocytes. The activity ofthesig-cultures was significant versus the
unfractionated cultures suggesting thatthe sIg- cultures were the origin of the SF.
T cell activation is accomplished throughthe recognition of MHC II and peptide fragments derived from the nativeprotein antigen by the T-cell receptor (Germain 1994) and by soluble cytokinefactors (Janeway and Bottomly 1994). The T cell directs the form of the immuneresponse differentially by producing distinct cytokine signals. For example,resting T cells can producegamma interferon or IL-4. Gamma interferonhas a positive influence on NK cells anda negative influence on the productionof antibody (Street and Mosmann 1991), thus preferentially inducing cell-mediatedimmunity. In contrast, resting T cells also produce IL-4, which influences otherT cells to produce a number of cytokines which predominantly affect Bcell growth and differentiation (Trinchierli 1994). T-cell productionof IL-4 and gamma interferon activitycan also differentially affect antiviral activity.IL-4 does not share gamma interferon's anti-viral properties. However,these two cytokines do sharea respiratory burst activity (Balkwill 1991).
Macrophage activating activity hasbeen elicited from fish lymphocytes using non-specific inducing agentssuch as T cell mitogens (Graham and Secombes 1990). These authorscontend that the factor produced from concanavalin A (Con A) stimulated PBLs hasgamma interferon-like activity, based upon induction ofan anti-viral state in rainbow trout gonad cells incubated with the factor. Theantigen-induced SF, however,was unable to produce either a direct effecton virus or induce an anti-viral state in viral susceptible cells. The cells were receptiveto interferon-like activity as Poly I:C 124 sucessfully induced this activity. Thereforethe SF, appears not to possess antiviral activity at any concentration tested.
Assessment of immunological memory in fishhas been limited to measurement of the humoral response (Arkooshet at, 1991). In these studies it has been demonstrated that T-dependentmemory function can be independently assessed in trout by the employmentof an in vitro system. Thus memory to a specific antigen can be assessed by themeasurement of cytokine activity from memory T-cells.
This form of immunologicalmemory assessment should be benefical for vaccine design. Effective vaccination requiresinduction of long term immunity and the ability to stimulate both B and T lymphocytes.We feel that this assay system can be modified and utilized to studycandidate antigens for vaccines, and their potential to producea sufficient immunological response that leads to the induction of immunologicalmemory. 125
ACKNOWLEDGMENTS
This research was support in part by USDA-CSRS#92-34123-7665 and USDA(WRAC) #91-38500-6078. The authors thank DR. J. Mark Christensen, J. Heidel, Don Mattson and J. C. Leongfor their critical review of this manuscript. Oregon Agricultural ExperimentalStation TechnicaiReport #. 126
Figure (6.1). Fluorescence measurement of oxidation of DCFH by rainbow
trout pronephric macrophages (PM) in thepresence of supernatant factors and the triggering agent (1 ug/ml PMA) after 24 hour incubation withsupernatants derived from.TNP-KLH (0),ovalbumin (®), and non-antigen conditioned supernatants (® )cells. Basal respiratory burst activity is considered indicator
cells (PM) with PMA only, no supernatant (U ).Figure (6.1 B) Titration of TNP-
KLH derived supernatants (M) undiluted, (®) 1:4 dilution and (®)1:16 dilution.Basal respiratory burst activity is indicated by PMA onlystimulation (HK) (®) results are the means of triplicate readings expressed/106cells after 40 minutes. There was a significant difference (p<0.05) between TNP-KLH versus the ovalbumin and PMA only control,but no difference between the ovalbumin and the PMA only control. Figure 6.1 A
7000 6500 6000 5500 5000 4500 4000 3500 3000 2500 0 2000 U- 1500 1000 500 0 I
Figure 6.1 B.
6500
6000
5500 5000 4500
4000 3500 3000 2500
2000 1500 1000
500
0 128
-oX 3000 2500 2000 1500 1000 5500 5000 4500 4000 3500 o0.Ln. - a U- Figure 6. 2 Figure 6. Figure (6.2). Fluorescence measurement of DCFH oxidation by rainbow trout Figure (6.2). Fluorescence measurement Results of fractionated and concentrated pronephric macrophages (PM). (M), > 30 KDa (®), supernatants. Unfractionated PMA only (0). There burst activity is indicated by macrophages incubated with (p<0.05). <30 KDa (®), < 10 KDa and 0 supernatant (®). Results are the means of <30 KDa (®), < 10 KDa and 0 cells after 40 minutes. Basal respiratory triplicate readings expressed/106 to all groups was a significant difference in > 30KDa supernatant compared i Figure 6.3
4000
3500
3000
2500
2000
1500
1000
500
0
Figure (6.3). Fluorescence measurement of DCFH oxidation by rainbow trout pronephric macrophages from slg- or slg+ enriched cell cultures. Surface Ig- enriched cell culture (0), surface lg+ enriched cell culture (®), adherent cell culture (U), unfractionated cell culture (®) and control indicator cells (PM) with PMAonly, no supernatant ([I). Results are the means of triplicate readings expressed/106 cells after 40 minutes. There was a significant difference in the slg- derived supernatant compared to all other derived supernatants (p<0.05). Also, there was no significant difference between the slg+ derived supernatant versus unfractionated derived supernatant(p>0.05). ISO
123 at 41
so
37.5
34
Figure 6.4. SDS-PAGE fluoroegraphof soluble S35 methionine labeled proteins. Lanes1:Pre-stained molecular weight markers(Sigma). 2: 1Ox concentrated TNP-KLH stimulated cell supernatant.3:1 Ox concentrated non- stimulated cell supernatant. 4: 5x concentratedTNP-KLHstimulated cell supernatant. 5: 5x concentrated non-stimulated cellsupernatant. 6: 5x concentratedTNP-KLHstimulated cell supernatant 100 to 30 KDamaterial. 7: 5x concentrated non-stimulated cell supernatant-100 to +30KDa material.
Arrows indicate soluble proteins at 89, 36,and 34common to active fractions. 131
CHAPTER 7 CONCLUSIONS
The commonality of the studies in this thesis is basedupon the development of a more attractive delivery system foraquaculture vaccines. The conventional methods of immersion and intraperitoneal(ip) injection delivery employed thus far have been sufficient, however,these methods become limited when large numbers and size of theanimals are considered. Intraperitoneal vaccination requires intensive laborto immunize relatively
small numbers of fish, and for the vaccination of small fish (0.5-2.0g) this method is impractical. Immersion vaccination methods,although not as labor intensive as ip delivery, still require that the animalsbe crowded into a holding pond where they are exposed to vaccine in standing aeratedwater. The disadvantage of this method is that it is limited to the weightof the fish that can be immunized per unit volume and the addedstress that is incurred by the animals upon crowded.
Immunizing fish by the oral route of delivery isan ideal method for mass vaccination to fish of all sizes without an increase in laboror stress incurred by the fish. The normal feeding schedulescan be used to also vaccinate the fish. The early use of oral deliveryas a means of vaccination however, met with limited successas initial vaccine preparations were
produced by adding the vaccine directly in the milled feed.These early methods resulted in inconsistencies in protectionas well as lacking a
detectable immune response (Evelyn, 1984; Hart et al., 1988). Theearly inconsistencies seen with oral deliverymay have been attributed to degradation of pH-sensitive antigens exposed to the gastricportion of the gut. 132
Therefore in these studies we developed an oral delivery system basedon the use of enteric protection, utilizing enteric coated antigen microspheres
(ECAMs). In chapter 3 we demonstrated that our ECAMswere as effective as ip and immersion in inducing an antibody response to lipopolysaccharide and protein antigens.It was also demonstrated that the ECAMs were able to
protect the protein antigen from gastric degradation. The ECAMswere employed to deliver a prototype vaccine for Renibacterium salmoninarum (Rs), the etiological agent of bacterial kidney disease (BKD). However, in order for us to develop an orally delivered vaccine for BKD it was necessary make the candidate antigen devoid of any virulent activity. Therefore, in chapter 4 characterization of the Rs antigens revealed one predominant cell-associated
and extracellular protein (ECP), p57. The biological effects of ECP andmore
importantly the p57 moiety have been implicated in number of pathogenicand immunosuppressive activities. Therefore in order to developa candidate antigen, removal of this immunosuppressive moietywas necessary. Our studies revealed that subjecting Rs cells to a 370C incubation decreased the
amount of this cell associated p57 from the cell surface by the induction ofan autoproteolytic activity. We therefore exploited the activation of this
autoproteolytic activity that cleaves p57 moiety in the production ofa vaccine. This p57- cell was used as a prototype vaccine that was delivered using
ECAMs and ip injection.In the first experiment the vaccine was administered
ip and the fish were challenged by ip injection.In a separate experiment the p57- reduced vaccine was administered either orally by incorporation in the food or intraperitoneal injection. The animalswere challenged by bath
immersion. Vaccination of coho salmon (0. kisutch) in the first experimentby
ip administration of p57-, formalin fixed, R. salmoninarum conferreda 133
statistically significant increase inmean time to death when ip challenged with R. salmoninarum at 4.1 x 106 colony forming units (cfu).
The second experiment determined vaccine efficacyby monitoring for the presence of p57 elaboration in the kidneysof vaccinated and control fish. Soluble p57 is elaborated by live R. salmoninarumas the infection progresses. Fish were sampled at time points of 0 (pre-challenge),50, 90, or 150 days post bath immersion challenge. The fishwere exposed to 4.2 x 106 cfu / ml on day
0. Fish orally vaccinated with p57- Renibacterium cells,using pH protected, enteric-coated antigen microspheres (ECAMs)as the delivery vehicle, demonstrated a significant difference (p<0.01)at day 150 post challenge with
respect to p57 levels versus controls. Fish receivinga non-pH protected version of the p57- Renibacterium cells also showeda significant difference (p<0.03) versus control. Fish ip injected with thep57-cells or fish fed non- 370C heat treated whole Renibacteriumcells using ECAMs demonstrated no significant difference with respect to soluble antigenlevels (p>0.05) versus controls.In conclusion, the results presented here demonstratethat the removal of a putative virulence factor p57 and routeof delivery are critical factors to consider if protection against Renibacteriuminfection is to be achieved. The orally delivered p57- cells (oral p57-)afforded significant protection when fish were exposed to viable Renibacteriumby bath immersion, while the ip p57- and oral p57+ cells did not. Althoughin the first experiment the ip p57- vaccine demonstrateda significant delay in mean time to death after ip challenge, the results of the second experimentproved that when animals are challenged by a natural route this methodimmunization was significantly less effective. These studies lend promisefor elucidation of the ideal candidate antigen and method of deliveryof that antigen to stimulate a protective immune response in salmonids against BKD. 134
Finally in chapter 6 a model assay was developed toassess immunological memory in fish to T-dependent type antigensby measuring the production of a cytokine upon in vitro antigenic specific stimulation. These studies demonstrated that T-dependent memory functioncan be independently assessed in trout by the employment of an in vitro culture system. Thus memory to a specific antigen can be assessed by the measurement of cytokine activity from memory T cells. These studies of immunologicalmemory assessment can potentially be beneficial for vaccine design and testing to determine if candidate antigens will elicita sufficient immunological response that leads to the induction of immunologicalmemory. 135
BIBLIOGRAPHY
Abbas, A.K., Lichtman, A.H., Pober, J.S. (1992). Cellularand Molecular Immunology. 417 pp.
Ada,G.L.(1988).What to expect of a good vaccine and how to achieveit. Vaccine. 6, 77-79.
Agius C.,Horne, M.T. and Ward, P.D. (1983).Immunizationof rainbow trout, Salmo gairdneriRichardson,againstvibriosis: comparison'of an exact antigen with whole cell bacterin by oraland intraperitonealroutes. Journal of Fish Diseases 6, 129-134.
Ahne, W. and Thosen, I. (1985). Occurrenceof VHS in wild white fish (coregonus sp.). Zentrabl. Veterinary Med. 32,73-75.
Amend,D.F., Yasutake, W.T. andMead,A.(1969).A hematopoietic viral disease of rainbow trout (Salmo gairdnern). Trans AmericanFisheries Society. 98, 796-804.
Amend, D.F. (1976). Prevention and controlof viral disease of salmonids. Journal of Fisheries Research Board, Canada.33, 1059-1066.
Amend, D.F., Finder, D.C. (1976). Uptakeof bovine serum albumin by rainbow trout from hyperosmotic solutions:a model for vaccinating fish. Science. 192, 793-794.
Amend, D.F. and Johnson, K.A. (1981). Currentstatus and future needs ofVibrio anguillarum. bacterins. Developmentof Biological Standards. 49, 403- 417.
Amend, D.F., Johnson, K.A., Croy, T.R. and McCarthy,D.H. (1983). Some factors affecting the potency ofYersinia ruckeribacterins. Journal of Fish Diseases. 1983, 6, 337-344.
Anderson, J.W. and Conroy, D.A. (1970). Vibriosisdiseases in fishes. In: A Symposium on Diseases of Fishes and Shellfishes,(ed.) S.F. Snieszko, Washington, DC: American Fisheries Society,Special Publication No. 5, 226- 272.
Anderson, D.P. (1974). Immunology: Diseases of fishesbook 4: Fish Immunology. TFH Publications, Neptune City.
Anderson, D.P. and Nelson, J.R. (1974). Comparisonof protection in rainbow trout(salmo gairdneri)inocculated with and fed Hagerman's red mouth bacterins. Journal Fisheries Research Board of Canada.31, 214-216. 136
Anderson, D.P. (1992). Immunostimulants, adjuvants, and vaccinecarriers in fish: applications to aquaculture. In: Annual Reviewof Fish Diseases. Faisal, M. and Hetrick, F.M. [eds.]. 2, 281-309.
Antipa, R., Amend, D.F. (1977). Immunizationof Pacific salmon: Comparison of intraperitoneal injection and hyperosmoticinfiltration ofVibrio anguillarum and Aeromonas salmonicida bacterins. Journal of FisheriesResearch Board of Canada. 34, 203-208.
Antipa, R., Gould, R., Amend, D.F. (1980). Vibrio anguillarumvaccination of sockeye salmon, Oncorhynchus nerka (Walbaum) by directand hyperosmotic immersion. Journal of Fish Diseases. 3, 161-165.
Aoki, T., Egusa, S. and Arai, T. (1974). Detection R factorin naturally occurring vibrio anguillarum strains. Antimicrobial Agents and Chemotherapy.6, 534- 538.
Aoki, T.,Sakai, M.and Takahashi, S. (1984). Protective immunity inayu, Plecoglossus altivelis, vaccinated with Vibrio anguillarum.Fish Pathology 19,181-185.
Ardavin, C. F., Zapata, A., Vellena, A. andSolas, M.T.(1982). Ultrastructure of gut-associated lymphoid tissue (GALT) in the amphibian urodelepleurodeles waltii. Cell Tissue Research. 224, 663-671.
Arkoosh, M. R. and Kaattari,S.L.(1990). Quanitation offish antibody to a specific antigen by an Enzyme linked immunosorbentassay (ELISA). In: Stolen, J.; Fletcher, T. C. ; Robertson, B.S.; van Muiswinkel, W.B.,editors. Techniques in Fish Immunology.New Jersy: S.O.S. Publications;15-24.
Arkoosh, M. R. and Kaattari, S.L. (1991). Developmentof immunological memory in rainbow trout (Oncorhynchus mykiss) I. An imuunochemicaland cellular analysis of the B-cell response. Developmentaland Comparative Immunology 15, 279-293.
Austin, B. and Austin, D.A. (1987). Bacterial Fish Pathogens:Diseases in Farmed and Wild Fish. Ellis Norwood Limited, Chichester,U.K. 346 pp. Avault, J. (1987). Aquaculture. In: Lackey, R.T., Nielsen,L. (eds.) Fisheries management. Blackwell Scientific Publishers, Boston,pp. 379-411.
Balkwill, F. R. (1991). Cytokines: A practical approach.(ED) Rickwood, D. and Harries, B. D. Oxford University Press New York, NY.pp. 349.
Bandin I.,Santos, Y., Barja, J. L., Magarinos, B. and A.E. Toranzo (1992).The detection of two antigenic groupsamong Renibacterium salmoninarum isolates. Federation of European Microbiological Societies94, 105-110. 137
Bandin, I., Ellis,A.E.,Barja, J.L., Secombs, C.J. (1993).Interaction between rainbow trout macrophages andRenibacterium salmoninarumin vitro. Fish and Shellfish Immunology. 3, 25-33.
Banner, C.R., Rohovec, J.S. and Fryer, J.L. (1983). Renibacterium salmoninarum as a causeof mortality among Chinook Salmonin salt water. Journal of World Mariculture Society. 14, 236-239.
Bass, D. A., Parce, W., Dechatelet, L. R., Szejda, P., Seeds,M. C. and Thomas, M. (1983). Flow cytometric studies of oxidative product formationby neutralphils:a graded responseto membrane stimulation. Journal of Immunology 130, 1910-1917
Belding, D.L. and Merrill, B. (1935). A preliminaryreport upon a hatchery disease of the Salmonidae. Trans American Fisheries Society.65, 76-84
Bell, G. R., Higgs D. A. and Traxler, G. S. (1984). Theeffect of dietary ascorbate, zinc andmanganeseon the development of experimentally induced bacterial kidney disease in sockeye salmon(Onchorhynchus nerka). Aquaculture 36, 291-311.
Bernard, J., de Kinkelin, P. and Bearzotti, M. (1983).Viral hemorrhagic septicemia of trout: relation between the G Polypeptide,antibody production and protection of the fish following infection with F25attenuated variant strain. Infection and Immunity. 39, 7-14.
Blazer, V.S., Bennett, R.O. and Wolke, R.E. (1984).The cellular immune response of rainbow trout (Salmo gairdneri Richardson) to sheep red blood cells. Developmental and Comparative Immunology.8, 81-88.
Bloom, B. (1989). Vaccines for the Third World. Nature342, 115-120.
Bly, J.E. and Clem, L.W. (1991). Temperature mediatedprocesses in teleost immunity: In vitro immunosupression induced by invivo low temperature in channel catfish. Veterinary Immunology and Immunopathology.3-4, 365-377. Bly, J.E., Lawson, L.A. and Clem, L.W. (1992). Temperatureeffects on channel catfish to a fungal pathogen. Diseases of Aquatic Organisms.13, 155-164.
Bootland, L.M., Stevenson, R.M., and Dobos, P. (1986).Experimental induction of the carrier state in yearling brook trout:a model challenge protocol for IPNV immunization. Veterinary Immunology and Immunopathology.12, 365-372.
Bootland, L.M., Dobos, P., Stevenson, R.M.W. (1990). Fryage and size effects on immersion immunization of brook trout, Salvelinus fontinalis Mitchell, against infectious pancreatic necrosis virus. Journal of Fish Diseases.13, 113- 125. 138
Bruno, D.W. and Munro, A.L.S. (1982). Detection of thecausative agent of bacterialkidneydisease. Bulletin of EuropeanAssociation of Fish Pathology. 2, 10-12.
Bruno D.W. (1986). Histopathology of bacterial kidney disease inlaboratory infected rainbow trout, Salmo gairdneri (Richardson),and Atlantic salmon, Salmo salar L., with reference to naturally infected fish. Journalof Fish Diseases. 9:523-537.
Bruno, D.W. and Munro, A.L.S. (1986). Haematologicalassessment of rainbow trout, Salmo gairdneri Richardson and Atlantic salmon, Salar L., infectedwith Renibacterium salmoninarum. Journal of Fish Diseases. 9, 195-204.
Bruno D.W. and A.L.S. Munro. (1986). Uniformity in the biochemicalproperties of Renibacterium salmoninarum isolates obtained from severalsources. FEMS Micro. Lett. 33:247-250.
Bruno, D.W. (1987). Serum agglutinating titers against Renibacterium salmoninarum the causative agent of bacterial kidney disease, in rainbow trout, Salmo gairdneri Richardson, and Atlantic salmon, Salmo salarL. Journal of Fish Biology. 30, 327-334
Bruno, D.W. (1988). The relationship between auto-agglutination,cell surface hydrophobicity and virulence of the fish pathogen Renibacterium salmoninarum. Federation of European Microbiological Societies.51, 135- 140.
Bruno D.W. (1990). Prescence ofa saline extractable protein associated with virulent strains of the fish pathogen Renibacterium salmoninarum.Bulletin of European Association of FishPathology. 10,8-10.
Bullock, G.L. and Stuckey,H.M. (1975). Fluorescentantibody identification and detection of the Corneybacterium causing bacterial kidneydisease. Journal of Fisheries and Research board of Canada 32, 2224-2227.
Bullock, G.L. and Anderson, D.P. (1984). Immunizationagainst Yersinia ruckeri, cause of Enteric Red Mouth Disease. In: Symposiumon Fish Vaccination ed. De Kinkelin, P. OIE, Paris, 151-166:
Bullock, G.L. and Herman, R.L., (1988). Bacterial kidneydisease of salmonid fishes caused by Renibacterium salmoninarum. Fish diseaseleaflet 78, U.S. Fish and Wildlife Service, Washing ton D.C. 10p.
Canestrani, G. (1893). La malattia dominate delle anguille.Atti Institute Vento Service, 7 809-814.
Caspi, R.R. and Avtailon, R.R. (1984). Evidence for the existenceof an IL-2 like lymphocyte growth promoting factor ina bony fish, Cyprinus carpio. Developmental and Comparative Immunology. 8, 51-60. 139
Castric, J. and de Kinkelin, P. (1984). Experimental study of the susceptibilityof two marine fish species, sea bass (Dicentrachus labrax) and turbot (Scophthalums maximus) to viral hemorrhagic septicemia. Aquaculture.41, 203-212.
Challacombe, S.J., Rahman, D., Jeffery, H., Davis, S.S. and O'Hagan, D. T., (1991). Biodegradable microparticles for oral immunization. Immunolgy2, 239-242.
Chanock, R.M. and Lerner, R.A., eds. (1984). Modern approachesto vaccines. Cold Spring Harbor, Cold Spring Harbor, New York.
Chart, H. and Trust,T.J.(1984). Characterization of the surface antigen of marine fish pathogens, Vibrio anguillarum. and Vibrio ordalli. Canadian Journal Microbiology. 30, 703-710.
Cipriano, R.C. (1982). Resistance of salmonids to Aeromonassalmonicida: relation between agglutinins and neutralizing activities. Trans. Amer. FishSoc.
Cipriano, R.C. and Starliper, C.E. (1982). Immersion and injectionvaccination of salmonids against furunculosis with an avirulent strain of Aeromonas salmonicida. Progress Fish-Culture. 44, 167-169.
Cisar, J.O. and Fryer, J.L. (1969). An epizootic of vibriosis in chinooksalmon. Bulletin of Wildlife Diseases. 5, 73-76.
Clem,L.W., Faulmann, E., Miller, N.W., Ellsaesser,C.,Lobb, E.J. and Cichens, M.A. (1984). Temperature-mediated processes in teleost immunity: differential effects of in vitro and in vivo temperatureson mitogenic responses of channel catfish lymphocytes. Developmental and Comparative Immunology. 8,313- 320.
Clem, L.W., Miller, N.W., Bly, J.E. (1991). Evolution of lymphocyte subpopulations, their interactions and temperature sensitivities. In Phylogenesis of immune functions. Edited by Warr, G.W., Cohen, N.Boca Raton, FL: Press Inc. 191-213.
Corbel, M.J. (1975). The immune response of fish:a review. Journal Fish Biology. 7, 539-560.
Cossarini-Dunier, M., Desavaux, F.X., Dorson, M. (1986). Variability in humoral responses to DNP-KLH or rainbow trout (Salmo gairdneri). Comparison of antibody kinetics and immunoglobulin spectrotypes between normal trouts and trouts obtained by gynogenesis or self-fertilization. Developmental and Comparative Immunology 10, 207-217. 140
Croy, T.R. and Amend, D.F. (1977). Immunization of sockeye salmon (Oncorhynchus nerka) against vibriosis using the hyperosmotic infiltraiton technique. Aquaculture. 12, 317-325.
Cuchens, M. A., Clem, L. W. (1977). Phylogeny of lymphocyte heterogeneity II. Differential effects on fish T-like and B-like cells. Cellular Immunology. 34, 219- 230.
Daly, J.G. and Stevenson, R.M. (1987). Hydrophobicity and Haemagglutinating properties of Renibacterium salmoninarum. Journal of General Microbiology. 133, 3575-3580.
Daly J.G. and R.M. Stevenson. (1987). Hydrophobic and haemagglutinating properties of Renibacterium salmoninarum. Journal General Microbiology 133:3575-3580.
Daly, J.G. and Stevenson, R.M. (1988). Inhibitory effects of salmonid tissueon the growth of Renibacterium salmoninarum. Diseases of Aquatic Organisms. 4, 169-171.
Daly J.G. and R.M. Stevenson (1989). Agglutination of salmonid spermatozoa by Renibacterium salmoninarum. Journal Aquatic Animal Health.1, 163-164.
Daly J.G. and R.M. Stevenson (1990). Characterization of the Renibacterium salmoninarum haemagglutinin. Journal General Microbiology 136, 949-953.
Davidson, G.A., Ellis, A.E., Secombes, C.J. (1993). Route of immunization influences the generation of antibody secreting cells in the gut of rainbow trout (Oncorhynchus mykis). Developmental and ComparativeImmunology. 17, 373-376.
Davina, J.H.M., Rijkers, G.T., Rombout, J.H.W.M., Timmermans, L.P.M. and Van Muiswinkel, W.B. (1980). Lymphoid and non-lymphoid cells in the intestine of cyprinid fish. In Development and Differentiation of the Vertebrate Lymphocytes. (J.D. Horton, ed.) 129-140. Amsterdam: Elsevier/North-Holland Biomedical Press.
Davina, J.H.M., Parmentier, H.K. and Timmermans, L.P.M. (1982). Effects of oral administration of Vibrio bacteria on the intestine of Cyprinid fish. Developmental and Comparative Immunology. Suppl. 2, 157-166. deKinkelin, P. and Dorson, M. (1973). Interferon production in rainbow trout (Salmon gairdnen) experimentally infected with egtved virus. Journal of General Virology 19, 125-127. de Kinkelin, P. and Le Berre, M. (1977). Demonstration de la protection de la truite Arc-en-ciel contre la SHV, par I'administration d'un virus inactive. Bulletin de ('Office Internation des Epizooties. 83 (5-6), 401-402. 141
de Kinkelin, P. and Bearzotti-Le Berre, M. (1981). Immunizationof rainbow trout against viral hemorrhagic septicemia (VHS) witha thermoresistant variant of the virus. Development in Biological standards. 49, 431-439.
de Kinkelin, P., Bernard, P., Hattenberger-Baudouy, A.M. (1984).Immunization against viral diseases occurring in cold water. Symposiumon Fish Vaccination. Office International des Epizootics (OIE), Paris.
de Kinkelin, P. (1988). Vaccination against viral hemorrhagicsepticemia virus (VHS). In: Ellis, A.E. (ed.) Fish Vaccination. Academic Press,London. 172-192.
DeKoning-Loo, J. and Kaattari, S. L. (1991). Mitogenesisof rainbow trout peripheral blood lymphocytes requires homologous plasmafor optimal responsiveness.In vitro Cellular and Developmental Biology 27A, 381-386.
Deluca, D. Wilson, M. and Warr, G. W. (1983). Lymphocyteheterogeneity in the trout (Salmo gairdneri), defined with monoclonalantibodies to IgM. European Journal of Immunolgy 13, 546-551.
Desvaux, F. X. and Charlemagne J. (1981). The goldfishresponse.I. Chareacterization of the humoralresponse to particulate antigens. Immunology 43, 755-762.
Dogget, T. A. and Harris, J. E. (1991). Morphology of thegut associated tissue ofOreochromis mossambicusand its role in antigen absorption Fish and Shellfish Immunology 1,213-227.
Dorson, M. (1988). Vaccination against infectiouspancreatic necrosis. In Ellis, A.E. (ed.): Fish Vaccination. Academic Press, London.162-171.
Dorson, M. and de Kinkelin (1977). Overview of viraldiseases of salmonids. In proceedings from the International Symposiumon Diseases of Cultured Salmonids. April 4-6, 1977, Seattle, Washington.91-145. Dorson, M. and Torchy, C. (1981). The influence of fishage and water temperature on mortalities of rainbow trout (Salmo gairdneriRichardson) caused by an European strain of infectious pancreatic necrosisvirus. Journal of Fish Diseases. 4, 213-221.
Duff, D.C.B. (1942). The oral immunization of trout againstBacterium Salmonicida. Journal of Immunology. 44, 87-94.
Earp, B.J., Ellis, C.H. and Ordal, E.J. (1953). Kidney diseasein young salmon. Special Reports, Ser. No. 1. State of Washington, Departmentof Fisheries. 73.
Eldridge, J.H., Meulbroek, J.A., Staas, J.K., Tice, T.R.and Gilley, R.M. (1989). Vaccine-containing biodegradable microspheresspecifically enter gut- 142
associated lymphoid tissue following oral administrationand induce disseminated mucosal immune response. In: Immunobiologyof Proteins and Reptides (ed.) Atass, V.M.Z. Plenum Press, NY. P.191-202.
Eldridge, J. H.,Staas, J. K.,Meulbroek, J. A.,McGhee, J. R.,Tice, T. R. and Gilley, R. M. (1991). Biodegradable microspheresas a vaccine delivery system. Molecular Immunology 28, 287-294.
Eldridge, J.H., Staas, J.K., Meulbroek, J.A., Tice, T.R.-andGilley, R.M. (1991). Biodegradable and biocompatible poly (DL-Lactide-CO-Glycolide) microspheresas anadjuvant for staphylococcal enterotoxin B toxoid which enhances the level of toxin-neutralizing antibodies. Infectionand Immunity. 9, 2978-2986.
Eldridge,J.H.,Staas,J.K.,Meulbroek, J.A., Tice,T.R. and Gilley, R.M. (1991a). Biodegradable microspheres asa vaccine deliverysystem.Molecular Immunology. 3, 287-294.
Elliott, D.G., Pascho, R.J., Bullock, G.L. (1989). Developmentsin the control of bacterial kidney disease of Salmonidfishes.Disease of Aquatic Organisms. 6, 201-215.
Elliott, D.G., Pascho, R.J., McKibben, C.L. and Thomassen,M.C. (1991). Development of a waterborne challenge procedure forinfecting salmonids with Renibacterium salmoninarum. Abstract, 14th AnnualAFS/FHS Meetings, 32nd Western Fish Disease Conference, Newport, Oregon,July 31-Aug. 3. P. 22. Ellis, A.E., Hastings, T.S. and Munro, A.L.S. (1981).The role of Aeromonas salmonicida:extracellular products in the pathology of furunculosis. Journalof Fish Diseases. 4, 41-51.
Ellis, A.E. (1988). Current aspects of fish vaccination.Diseases of Aquatic Organisms. 4, 159-164.
Ellis, A.E. (1988). Fish Vaccination. Academic PressSan Diego, CA,: 255 pp.
Ellsaesser, C.F., Bly, J.E., Clem, L.W. (1988). Phylogenyof lymphocyte heterogeneity: the thymus of the channel catfish. Developmentaland Comparative Immunology. 12, 787-799.
Engelking, H.M. and Leong, J.C. (1989). The glycoproteinof infectious hematopoietic necrosis elicits neutralizing antibodyand protective responses. Virus Research. 13, 213-230.
Etlinger, H.M., Hodgins, H.O., Chiller, J.M. (1976). Evolutionof the lymphoid system I. Evidence for lymphocyte heterogeneity in rainbowtrout revealed by the organ distribution of mitogenicresponses. Immunology. 116, 1547-1553.
Evelyn, T. P. T. (1977). An improved growth medium forthe kidney disease bacterium and some notes on using the medium. Bull. Int.Epiz. 78, 511-513. 143
Evelyn, T.P.T. (1971). The agglutinins response in sockeye salmon vaccinated intraperitoneally with a heat killed preparation of the bacterium responsiblefor Salmon kidney disease. Journal of Wildlife Diseases. 7, 328-335.
Evelyn, T.P.T. (1984). Immunization against pathogenic vibriosis.In Symposium of Fish Vaccination, ed. P. deKinkelin. Paris: Off. Int.Epiz. 121- 150.
Evelyn, T.P.T., Ketcheson, J.E., Prosperi-Porta, L. (1988). Trials withanti- bacterial kidney disease vaccines in two species of Pacific salmon. In: Conference handbook. International Fish HealthConference,Vancouver, B.C., Fish Health Section, American Fisheries Society, Washington, D.C.38.
Evenden, A.J., Gilpin, M.L. and Munn, C.B. (1990). Cloning andexpressing of a gene encoding haemolytic activity from the fish pathogen Renibacterium salmoninarum. Federation of European Microbiological Societies. 71,31-34.
Evenden, A.J., Grayson, T.H., Gilpin, M.L., Munn, C.B. (1993). Renibacterium Salmoninarum and bacterial kidney disease- the unfinished jigsaw. Annual Review of Fish Diseases. 88-104.
Fevolden, S. E., Refstie, T. and Roed, K.H. (1992). Disease resistancein rainbow trout (Oncorhynchus mykiss) selected for stressresponses. Aquaculture 104, 19-29.
Finlay, B.B. and Faldow, S. (1989). Common themes in microbial pathogenicity. Microbiological Reviews. 53, 210-230.
Fletcher, T.C. and White, A. (1973). Antibody production in thePlaice, Pieuronectes platessa L. after oral and parenteral immunization withVibrio anguillarum antigens. Aquaculture. 1, 417-428.
Frantsi,C. andSaven, M. (1971).Infectious pancreatic necrosis virus. Temperature and age factors in mortality. Journal of WildlifeDiseases. 7, 249- 255.
Fryer, J.L., Nelson, J.S. and Garrison, R.L. (1972). Vibriosis infish. P. 129-133 in R.W. Moore (3d.) Progress in fishery and food science, Vol.5. University of Washington Publication in Fisheries, New Series, Seattle.
Fryer, J.L., Rohovec, J.S., Tebbit, G.L., McMichael, J.S.and Pilcher, K.S. (1976). Vaccination for control of infectious diseases in Pacificsalmon. Fish Pathology. 10, 155-164.
Fryer, J.L., Amend, D.F., Harrel, L.W., Novotony, A.S., Plumb, J.A.,Rohovec, J.S. and Tebbit, G.L.(1977). Development of Bacterins and vaccinesfor control of Infectious DiseasesofFish. Oregon State University. Sea Grant College Program. Published No. Oresu T 77-012. 144
Fryer, J.L. and Sanders, J.E. (1981). Bacterial kidney disease of salmonid fish. Annual Review of Microbiology. 35, 273-298.
Fuijino, Y., Ono, S., Nagai, A. (1987). Studieson the uptake of rabbit's immunmoglobulin into the columnar epithelial cells in the gut of the trout, Salmo gairdneri. Bulletin of the Japanese Society of Scientific Fisheries 53, 367-370.
Georgopoulou, U., Sire, M.F. and Vernier, J.M. (1986). Immunological demonstration of intestinal absorption of proteins by epithelial cells of the posteriorintestinal segmentand their intracellular digestion in rainbow trout. Ultrastructural and Biochemical Study. Biology of the Cell. 5, 269-282.
Georgopoulu, U. and Vernier, J.M. (1986). Local immunologicalresponse in the posterior intestinal segment of the rainbow trout after oral administration of macromolecules. Developmental and Comparative Immunology. 10, 529-537.
Germain, R. N. (1994). MHC-dependent antigen processing and peptide presentation:Providing ligandfor T-lymphocyte activation. Cell 76,287-299.
Getchell, R.G., J.S. Rohovec, and J.L. Fryer. (1985). Comparison of Renibacterium salmoninarum isolates by antigenicanalysis.Fish Pathology. 20(2/3):149-159.
Ghaffari, S.H. and Lobb, C.J. (1989). Nucleotidesequences of channel catfish heavy chain cDNA and genomic blot analyses. Implications for the Phylogeny of Ig heavy chain. Journal of Immunology. 143, 2730-2736.
Gilmore, R.D., Engelking, H.M., Manning, D.S., Leong, J.C. (1988). Expression in Escherichia coli of an epitope of the glycoprotein of infectious hematopoietic necrosis virus protects against viral challenge. Biotechnology. 6, 295-300.
Good, M.F., Berzofsky, J.A. and Miller, L.H. (1988). The T-cellresponse to the malaria circumsporozoite protein. An immunological approach to vaccine development. Annual Review of Immunology. 6, 663-688.
Goodrich, H.B. and Nichols, R. (1933). Scale transplantation in the goldfish, Carassius auratus, I. Effects of chromatophores; II. Tissue rejection. Biological Bulletin. 56, 253-259.
Gould, R.W. (1978). Development of a new vaccine delivery system for immunizing fish and investigation of the positive antigens in Vibrio anguillarum. Ph.D. Thesis, Oregon State University, Corvallis, OR.
Gould, R.W., O'Leary, P.J., Garrison, R.L., Rohovec, J.S. and Fryer, J.L. (1978). Spray vaccination: a method for the immunization of fish. Fish Pathology. 13, 63-68. 145
Graham, S. and Secombes C. J. (1988). The production of a macrophage- activating factor from rainbow trout(Salmo gairdneri)leucocytes. Immunology 65, 293-297.
Graham, S. and Secombes C. J. (1990). Cellular requirements for lymphokine secretion by rainbow trout (Salmo gairdneri) leucocytes. Developmental and comparative and Comparative Immunology 14, 59-68. Graham, S. and Secombes, C.J. (1990). Cellular requirements for lymphokine secretion by rainbow trout Salmo gairdneri leucocytes. Developmental and Comparative Immunology. 14, 59-68.
Grandel, J.L. and Harmsen, E.G.M. (1984). Phylogeny of interleukins: growth factor produced by leukocytes of the cyprinid fish, Cyprinus carpio. Immunology. 52, 477-483.
Gray, D. (1993). Immunological memory. Annual Review of Immunology. 11, 49-77.
Gray, D. and Sprent, J. (1990). Immunological memory. Curr. Top. Microbiology and Immunology 159, 1-141.
Griffin, B.R. (1983). Opsonic effect of rainbow trout (Salmo gairdneri) antibody on phagocytosis of Yersinia ruckeri by trout leukocytes. Developmental and Comparative Immunology. 7, 253-259.
Griffiths, S.G., and W.H. Lynch. (1991). Instability of the major soluble antigen produced by Renibacterium salmoninarum. Journal of Fish Disease. 14:55- 66.
Gutenberger, S. K., Dimstra, J. S., Rohovec, J. L., Fryer, J. L. (1991). Intracellular survival ofRenibacterium salmoninarumin trout macrophages. Abstract 14th Annual AFA/FHS meeting and the 32nd Western Fish Disease Conference, New Port, Oregon. July 31st-Aug. 3rd.
Harrell, L.W., Etlinger, H.M. and Hodgins, H.O. (1975). Humoral factors important in resistance of Salmonid fish to bacterial diseases I. Serum antibody protection of rainbow trout against vibriosis. Aquaculture. 6, 211-220.
Hart, S., Wrathmell, A.B., Harris, J.E. and Grayson, T.H. (1988). Gut immunology in fish: a review. Developmental and Comparative Immunology. 12, 453-480.
Harwell, L., Kappler, J.W. and Marrack, P. (1976).Antigen-specific and non- specific mediators of T-cell/B-cellcooperation. III. Characterization of the nonspecific mediators from different sources. Journal of Immunology. 116, 1379-1385. 146
Hastings, T.S. and Ellis, A.E. (1985). Differences in the production of haemolytic and proteolytic activities by various isolates of Aeromonas salmonicida. Pp. 69-77 In A.E. Ellis (ed.) Fish and Shellfish Pathology. Academic Press, London.
Havarstein, L.S., Kalland, K.H., Christie, K.E., Endresen, C. (1990). Sequence of the large double-stranded RNA segment of the N. strain of infectious pancreatic necrosis virus: A comparison with other Birnaviradae. Journal of General Virology. 71, 299-308.
Hayashi, K., Kobayashi, S., Kamata, T. and Ozaki, H. (1964). Studieson the Vibrio-diseaseof rainbow trout(Salmo gairdnen)II. Prophylactic vaccination against the Vibrio-disease. Journal of Faculty of Fisheries, Perfectural University of Mie. 6, 181-191.
Hester, F.E. (1973). Fish Health: a nationwide survey of problems and needs. Progress in Fish-Culture. 35, 11-18.
Hill, B.J., Dorson, M., Dixon, P.F. (1980). Studieson immunization of trout against IPN. In Fish Diseases [ed.] W. Ahne. Springer-Verlag Berlin,pp 29-36.
Hill, B. (1991). Impact of viral diseases on salmonids in Europe. In Kimura, T. (ed.) Proceedings of the OJI International Symposiumon Salmonid Diseases, 22-25, October 1991, Hokkaido University Press, Sapporo, Japan.pp 48-59.
Hilleman, M.R. (1985). New directions in vaccine development and utilization. Journal of Infectious Diseases. 151, 407-414.
Holmgren, J., Clemens, J., Sack, D.A., Svennerholm, A.M.(1989). New cholera vaccines. Vaccine. 7: 94-96, 1989.
Homsyl, J., Meyer, M., Tuteno, S., Clarkson, and Levy J.A. (1989). The Fc and not the CD4 receptor mediates antibody enhancement of HIV infection in human cells. Science 244, 1357-1360.
Hossain, M. and Ayers, J. (1990). Variables that influence coat integrity ina Laboratory Spraycoater. Pharmaceutical Technology 10, 72-82.
Howell, C.J. (1984). A chemotactic factor in the carp, Cyprinus carpio. Developmental and Comparative Immunology. 11, 139-145. lida, H., Yamamoto, T. (1985). Intracellular transport of horseradish peroxidase in the absorptive cells of goldfish hindgut in-vitro, with special referenceto the cytoplasmic tubeles. Cell Tissue Research. 240, 553-560.
Ingram,G.A.(1980). Substances involved in the natural resistanceof fish to infection - a review. Journal Fish Biology. 16, 23-60. 147
Jacobs,D. M. andMorrsion, D. C. (1975).Stimulation of a T-independent primary anti-hapten responsein vitro byTNP-lipopolysaccharide(TNP-LPS). Journal of Immunology 114, 360-364.
Jacobs, E., Gerstenecker, B., Macher, B., Huang, C.H., Hu, P.C., Halter, R. and Brecht, W. (1989). Binding sites of attachment inhibiting monoclonal antibodies and antibodies from patients on peptide fragments of the mycoplasma pneunoniae adhesin. Infection and Immunity. 57, 685-688
Janeway, C. A. and Bottomly, K. (1994). Signals and signs for lymphocyte response. Cell 76, 275-285.
Jenkins, P. G., Harris, J. E. and Pulsford, A. L. (1992). Quantitative serological aspects of the enhanced enteric uptake of human gamma globulin by Quil-A saponin in Oreochromis mossambicus. Fish and Shellfish Immunology 2, 193- 211.
Johnson, D.F. and Johnson, K.A. (1981). Duration of immunity in Salmonids vaccinated by direct immersion with Yersnia ruckeri and Vibrio anguillarum. bacterins. Journal of Fish Disease. 5, 207-213.
Johnson, K.A. and Amend, D.F. (1983). Comparison of eff icacy of several delivery methods using Y. ruckeri bacterin on rainbow trout, Salmo gairdneri Richardson. Journal of Fish Diseases. 6, 337-396.
Johnson, K.A., Flynn, J.K. and Amend, D.F. (1982). Onset of immunity in salmonid fry vaccination by direct immersion inVibrio anguillarum.and Yersinia ruckeri bacterins. Journal of Fish Diseases, 5, 197-205.
Jorgensen, P.E.V. (1992). Recent advances in surveillance and control of viral hemorrhagic septicemia (VHS) in trout. In: Kimura, T. (ed.) Proceedings of the OJI International Symposium on Salmonids Diseases, 22-25. October 1991, Hokkaido University Press. Supora, Japan. 60-71.
Kaattari, S. L. and Irwin, M. J. (1985). Salmonid spleen and anterior kidney harbor populations of lymphocytes with different B-cell repertoires. Developmental and Comparative Immunology 9, 433-444.
Kaattari, S., Chen, D., Turaga, P., Weins, G. (1988). Development ofa vaccine for bacterial kidney disease. Bonneville Power Administration, Portland, OR. 55. Project 84-46, Annual report 1987, Portland, OR.
Kaattari, S.L., Rockey, D.D., aliens, G., Turga, P.S.D. and Rohovec, J. (1988). Bacterial kidney disease and furnculosis vaccines. Proceedings of the Aquaculture International Congress and Exposition. Vancouver, British Columbia. Sept. 6-9. 148
Kaattari, S., Turaga, P., Weins, G. (1990). Development ofa vaccine for bacterial kidney disease in salmon. Final Report, August 1989.Bonneville Power Administration, Portland, OR. 55.
Kaattari, S.L. (1992). Fish B Lymphocytes: defining their formand function. Annual Review of Fish Diseases. 2, 161-180.
Kawai, K.K., Kusuda, R., Itam, T. (1981). Mechanism of protectionin ayu orally vaccinated for vibriosis. Fish Pathology. 15, 257-262.
Kawano, K.,Aoki, T. and Kitao, T. (1984). Duration of protection against vibriosis in Ayu Plecglossus altivelis vaccinated by immersionand oral administration withVibrio anguillarum.Bulletin of the Japanese Society of Scientific Fisheries 50, 771-774.
Kay, W.W., Buckley, J.T., Ishiguro, E.E., Phipps, B.M., Monette, J.P.L.and Trust, T.J. (1981). Purification and disposition ofa surface protein associated with virulence of Aeromonas salmonicida:. Journal of Bacteriology.147, 1077- 1084.
Kleid, D.G. (1981). Cloned viral protein vaccine for foot-and-mountdisease: responses in cattle and swine. Science. 214, 1125-1129.
Kobuku, G., Hinds, K., Litman, R., Shamblott, M.J. and Litman,C.W. (1987). Extensive families of constant regiongenes in a phylogenetically primitive vertebrate indicates an additional level of immunoglobulincomplexity. Proceedings National Academy of Science. USA. 85,5868-5872
Koener, J.F. and Leong, J. (1990). Expression of the glycoproteingene from a fish rhadovirus by using baculovirus vectors. Journal of Virology.1, 428-430.
Lawrence, W.R., Nagy, E., Duncan, R., Krell, P. Dobos, P. (1989).Expression in E. coli of the major outer capsid protein of infectious pancreaticnecrosis. Gene. 79, 369-374.
Leong, J.C. (1993). Molecular and biotechnologicalapproaches for fish vaccines. Current Opinion in Biotechnology, 4, 286-293
Leong, J.C. and Fryer, J.L. (1993). Viral vaccines for aquaculture.Annual Review of Fish Diseases. 3, 225-240.
Leong, J.C., Fryer, J.L., Winton, J.R. (1988). Vaccinationagainst infectious hematopoietic necrosis virus. In Ellis, A.E. (ed.) Fish Vaccination.Academic Press, London. B 193-202.
Lillehaug, A. (1989). A cost-effectiveness study of three differentmethods of vaccination against Vibriosis in Salmonids. Aquaculture. 83, 227-236. 149
Lillehaug, A. (1989). Oral immunization of rainbow trout,Salmo gairdneri Richardson, against vibriosis with vaccines protected against digestive degradation. Journal of Fish Diseases 12, 579-584.
Lindahl M., A. Faris, T. Wadstrom,andS.Hjerten(1981).A new test based on "salting out" to measure relative surface hydrophobicity of bacterial cells. Biochemica et Biophysica Acta. 677:471-476.
Ljungberg,O. andVestergaard Jorgensen,P.E.(1973). Infectious pancreatic necrosis (IPN) of salmonids in Swedish fish farms. In symposiumon the major Communicable Fish Diseases in Europe and Their Control. (ed.) W.A. dill. European Inland Fisheries Advisory Committee. 17, 67-70.
Lobb, C.J. and Clem, L.W. (1982). Fish lymphocytes differ in the expression of surface immunoglobulin. Developmental and Comparative Immunology. 6, 473-479.
Lobb, C.J. (1986). Structural diversity of channel catfish immunoglobulins. In: Fish Immunology Elsevier. 443.
Lobb,C.J. and Olsen, M.O.J. (1988). Immunoglobulin heavy H chain isotypes in a teleost fish. Journal of Immunology 141, 1236-1245.
Lorenzen, N., Olsen, J.F., Vestergaard-Jorgensen, P.E. (1990). Neutralizing of egtved virus pathogenicity to cell cultures and in fish by monoclonal antibodies to the viral G-Protein. Journal of General Virology.
Lowry, O.H., N.J. Rosebrough, D.L. Farr, and R.J. Randall (1951). Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry. 193:255-276.
Killie, J.K., Espelid, S., and Jorgensen, T. O. (1991). The humoral immune response in Atlantic salmon (Salmo salar L.) against the hapten carrier antigen NIP-LPH; the effect of determinant (NIP) density and the isotype profile of anti- NIP antibodies. Fish and Shellfish Immunology 1, 33-46.
M'Gonigle, R.H. (1941). Acute catarrhal enteritis of salmonid fingerlings. Trans American Fisheries Society. 70, 297-303.
Mackie, T.J., Arkwright, J.A., Pryce-Tannatt, Mottram, J.C., Johnston, W.D. and Menzies, W.J.M. (1935). Final reports of the Furunculosis Committee.
Manning, D.S., Leong, J. (1990). Expression in Escherichia coli of the large genomic segment of infectious pancreatic necrosis virus. Polyprotein Virology. 179, 9-15.
McAllister, P.E. (1984). Infectious pancreatic necrosis immunization. American Fisheries Society Fish Health Newsletter. 12, 236. 150
McCarthy, D.H. and Roberts, R.J. (1980). Furunculosis of fish and thepresent state of ourknowledge. In: Advances in Aquatic Microbiology, ed. M.R. Droop and H.W. Jannasch. London: Academic Press, 293-341.
McCarthy, D.H., Amend, D.F., Johnson, K.A. and Bloom, J.V. (1983). Aeromonas salmonicida: determination ofan antigenassociated with protective immunity and evaluation of experimental bacterin. Journal of Fish Diseases. 6, 155-174.
McCarthy, D.H., Croy, T.R., Amend, D.F. (1984). Immunization of rainbowtrout Salmo gairdneri Richardsonagainst bacterial kidney disease: preliminary efficacy evaluation. Journal of Fish Diseases. 7, 65-71.
McGhee, J.R., Mestcky, J., Dertzbaugh, M.T., Eldridge, J.H., Hirasawa, M.and Kiyono, H. (1991). The mucosal immune system: from fundamentalconcepts to vaccine development. Vaccine. 2, 75-89.
McKinney, E.C., Ortiz, G., Lee, J.C., Siegel, M.M., Lopes, D.M., Epstein, R.S. and McLeod, T.F. (1977). Lymphocytes of fish: multipotenialor specialized? In Weigut, R.K. and Cooper, E.L. (eds.). Phylogeny of thymus and bonemarrow- bursa cells. Elsevier Nor-Holland Biomedical Press. Amsterdam, The Netherlands. 73-82.
McLean, E. and Ash, R. 1987. Intact protein (antigen) absorptionin fishes: mechanisms and physiological significance. Journal of Fish Biology.29, 177- 186.
McLean, E. and Donaldson, E. (1990). Absorption of bioactive peptidesby the gastrointestinal tract offish:areview.Journal of Aquatic AnimalHealth. 2, 1-11.
Meier, W. andVestergaard-Jorgensen,P.E.(1980).Isolation of VHS virus from pike fry (Esox lucius) with haemorrhagic symptoms. In: Fish Diseases,ed. W. Ahne, Proceedings in Life Sciences, Berlin: Springer Verlag.pp.8-17.
Michel,C.(1985).Failure of anti-furunculosis vaccination of rainbow trout (Salmo gairdnen), using extra-cellular products of Aeromonassalmonicida: as an immunogen. Fish Pathology. 20, 445-451.
Miller, N.W. and Clem, L.W. (1984). Microsystem for invitro primaryand secondary immunization of channel catfish Ictalurus punctatus leukocyteswith hapten-carrierconjugates.Journal ImmunologicalMethods.72, 367-378.
Miller, N.W., Bly, J.E., VanGinkel, F., Ellsaesser, C.F., Clem, L.W. (1987). Phylogeny of lymphocyteheterogeneity:identification and separation of functionally distinct subpopulation of channel catfish lymphocyteswith monoclonalantibodies.Developmental and Comparative Immunology. 11, 739-747. 151
Mitchum, D. L. Sherman, L. E. (1981). Transmission of bacterial kidney disease from wild to stocked hatchery trout. Can. J. Fish. Aquat. Sci. 38, 547- 551.
Mourich, D.V., Leong, J.C. (1991). Mapping of the immunogenic regions of the IHNV glycoprotein in rainbow trout and mice. Proceedings of the Second International Symposium on Viruses of Lower Vertebrates, 29-31 July 1991, Corvallis, OR, 93-100.
Mughal, M. S. and Manning, M. J. (1986). The immune sytem of juvenile thick- lipped grey mullet, Chelon labrous Risso: antibody response to soluble protein antigens. Journal of Fish Biology 29, 177-186.
Munn, C.B., Ishiguro, E.E., Kay, W.W. and Trust, T.J. (1982). Role of surface components in serum resistance of virulent Aeromonas salmonicida:. Infection and Immunity. 36, 1069-1075.
Munro, A.L.S., and Bruno, D.W. (1988). Vaccination against bacterial kidney disease. 124-134. In: A.E. Ellis (ed.) Fish Vaccination. Academic Press Inc. San Diego, CA.
Murray, C.B., Evelyn, T.P.T., Beacham, T.D., Barner, L.W., Ketcheson, J.E., Prosperi-Porta, L. (1992). Experimental induction of bacterial kidney disease in Chinook Salmon by immersion and cohabitation challenges. Disease of Aquatic Organisms. 12, 91-96.
Nagai, A. and Fujino, Y. (1983). On the experiment of antigen-uptake through the gut in rainbow trout and ayu-fish. Proceeding 2nd N. Pacific Aquaculture Symposium, 191-205.
Neukirch, M. (1985). Uptake, multiplication and excretion of viral hemorrhagic septicemia in rainbow trout (Salmo gairdnen). pp.295-300 in A.E. Ellis, (ed.) Fish and Shellfish Pathology. Academic Press, London.
Newman, S.G. (1993). Bacterial vaccines for fish. Annual Review of Fish Diseases. 3, 145-185.
Nikl,L., Albright, L.J. and Evelyn,T.P.T. (1991). Influence ofseven immunostimulants on the immune response of coho salmon to Aeromonas salmonicida. Diseases of Aquatic Oraganisms 12, 7-12.
Nishimura, T., Sasaki, H., Ushiyama,M.,Inoue, K., Suzuki, Y., Ikeya, F., Tanaka,M.,Suzuki, H., Kohara, M., Arai,M.,Shima, N., Sano, T. (1985). A trial of vaccination against rainbowtrout fry with formalin IHN virus. Fish Pathology. 20, 435-443.
Noaillac-DePeure, J. and Gas, N. (1973). Absorption of protein macromolecules by the enterocytes of the carp (cyprinus carpio, L.). Zeitschrifit. Zellforschung and Mikroskopische Anatomie 146, 525-541. 152
O'Hagan, D.T., Palin, K.J., Davis,S.S. and Artursson, P. (1989). Microparticles as potentially orally active immunological adjuvants.Vaccine. 7, 421-424.
O'Hagan, D.T. (1990). Intestinaltranslocation of particulates: implicationsfor drug and antigen delivery. AdvancedDrug Delivery Reviews. 5, 265-295
O'Hagan, D.T. (1992). Oral Deliveryof vaccines: Formulation and clinical pharmacokinetic considerations. ClinicalPharmacokinetics. 22, 1, 1-10.
O'leary, D.J. (1980). Partial characterizationof high and low molecular weight immunoglobulin in rainbow trout (Salmogairnderi). Ph. D. Thesis Oregon State University 97331. pp72.
Ortega, H. (1993). Mechanismsof accessory cell function in rainbow trout (Oncorhynchus mykiss). Masters Thesis.Oregon State University.pp 109.
Parisot, T.J., Yasutake, W.T. andBressler, V. (1963). A new geographicand host record for infectious pancreaticnecrosis. Trans American Fisheries Society. 91, 63-66.
Pascho, R.J., Mulcahy, D. (1987).Enzyme-linked immunosorbentassay for soluble antigen of Renibacteriumsalmoninarum, the causative agent of bacterial kidney disease. CanadianJournal of Fisheries and Aquatic Sciences. 44, 183-191.
Patterson, W.D. and Fryer, J.L. (1974).Immune response of juvenile coho salmon (Oncorhynchus kisutch)to Aeromonas salmonicida: cells administered intraperitoneally in Freunds CompleteAdjuvant. Journal of Fisheries Research Board of Canada. 31, 1751-1755.
Patterson, W.D., Desautels, D.,Weber, J.M. (1981). The immuneresponse of Atlantic salmon, Salmo salar L.,to the causative agent of bacterial kidney disease, Renibacterium salmoninarum.Journal of Fish Diseases. 4, 99-111.
Pickering, A.D. and Pottinger, T.G.(1985). Cortisol can increase the susceptibility of brown trout, Salmon TruttaL., to disease without reduceing the white blood cell count. Journalof Fish Biology. 27, 611-620.
Piganelli, J. D., Zhang, J.A., Chirstensen,J.M. and Kaattari, S.L. (1994). Enteric coated micropheresas an oral method for antigen delivery to salmonids. Fish and Shellfish Immunology.4, 179-188.
Plumb, J. (1992). USDA Fish HealthResearch Priorities minutes from panal session.
Pontinus, H. and Ambrosius, H. (1972).Contribution to the immune biologyof poikiothermic vertebrates. IX. Studieson the cellular mechanisms of humbral immune reactions in perch, Percafluviatilis L. Acta Biol Med Ger. 29, 319-339. 153
Porter, S. C. (1985). Coating of Pharmaceutical Dosage Forms. In Remington's Pharmaceutical Sciences. The Philadelphia College of Pharmacy and Science, 1633.
Pottinger, T. G., Pickering, A.D. and Hurley, M. A.(1992). Consistency in the stress response of individuals of two strains of rainbow trout, (Oncorhynchus mykiss). Aquaculture 103, 275-289.
Rijikers, G.T., Frederix-wolters,L.M.H., van Muiswinkel, W.B. (1980). The immune system of cyprinid fish. The effect of antigen dose and route of administration on the development of immunolgical memory of the carp (Cyprinus carpio). (ed) Marget Manning. Elsevier/North-Holland Biomedical Press. pp. 83-92.
Rittenberg, M. B. and Pratt, K. (1969). Anti-trinitro-phenyl (TNP) plaqueassay. Primary response of BALB/c mice to soluble and particulate antigens Proceedings of the Socitey of Experimental Biological Medicine. 132, 575-581.
Rockey, D.D. (1989). Virulence factors of Aeromonas salmonicida: and their interaction with the salmonid host. Ph.D. Thesis, Oregon State University.
Rockey, D.D., Gilkey, L.L. Wiens, G.D. and Kaattari, S.L. (1991). Monoclonal antibody analysis of the Renibacterium salmoninarum P57 protein in spawning chinook and coho salmon. Journal of Aquatic Animal Health. 3, 23-30.
Rockey, D.D., Turaga, P.S.D., aliens, G.D., Cook, B.A., Kaattari, S.L. (1991). Serine proteinase of Renibacterium salmoninarum digestsa major autologous extracellular and cell-surface protein. Canadian Journal of Microbiology. 37, 758-763.
Rohovec,J.S.(1975). Oral and parenteral immunization for the controlofVibrio anguillarum, the etiological agent of Vibriosis in Salmonid fish. Ph.D Thesis Oregon State University. 3-5.
Rohovec, J.S., Winton, J.R., Fryer, J.L. (1981). Bacterins and vaccines for the control of infectious disease in fish. Proceeding of Republic of China-United States Cooperative Science Seminar on Fish Diseases, Msc Symposium Series No. 3, National Science Council, Republic of China.
Rombout, J.H.W.M., Lamers, C.H.J., Hellfrich, M.H., Dekkee, A., and Taverne- Thiele,J.J.(1985). Uptake and transport of intact macromolecules in the intestinal epithelium of the carp (Cyprinus carpio) and the possible immunological implications. Cell Tissue Research. 239, 519-530.
Rombout, J.H.W.M., Van Den Berg, A.A., Van Den Berg, C.T.G.A., Witte, P. and Egberts, E. (1989). Immunological importance of the second gut segment of carp, III. Systemic and/or mucosal immune responses after immunization with soluble or particulate antigen. Journal Fish Biology.35, 179-186. 154
Rombout, J.H.W.M. and Van Den Berg,A.A.(1989). Immunological importance of the second gut segment of carp, I. Uptake and processing of antigens by epithelial cells and macrophages. Journal of Fish Biology. 35, 13-18.
Rombout, J.H.W.M., Taverne-Thicle, A.J. and Villena, M.I. (1993). The gut- associated lymphoid tissue (GALT) of carp (Cyprinus carpio L.): An immunohistochemical analysis. Developmental and Comparative Immunology. 17, 55-66.
Ross, A.J. and Klontz, G.W. (1965). Oral immunization of rainbow trout (Salmo gairdneri) against an etiological agent of 'Red Mouth Disease.' Journal Fisheries Research Board of Canada. 22, 713-719.
Ross, A.J., Rucker, R.R., Ewing, W.H. (1966). Description ofa bacterium associated with redmouth disease of rainbow trout. Canadian Journal of Microbiology. 12, 763-770.
Rucker, R.R., Whipple, W.J., Parvin, J.D., Evans, C.A. (1953). A contagious disease of salmon possible by viral origin. U.S. Department of Interior Fish and Wildlife Service, Fisheries Bulletin. 76, 54. 35-46.
Russel, N.J., Voss, E.W. and Sigel, M.M. (1970). Some characteristics ofanti- dinitrophenyl antibody in the gray snapper. Journal of Immunology 105,262- 264.
Ryan, T. C., Weil, G. J., Newburger, P. E., Haugland, R. and Simons, E. R. (1990). Measurements of superoxide release in the phagovacuoles of immune complex-stimualted human neutrophils. Journal of Immunological Methods 130, 223-233.
Sakai, M., Atsuta, S., Kobayashi, M. (1989). Attempted vaccinationof rainbow trout, (Oncorhynchus mykiss) against bacterial kidney disease. Bulletin of the Japanese Society of Scientific Fisheries. 55, 2105-2109.
Saki, M. I. ,Aoki, T.,Rohovec, J. S. and Fryer, J. L. (1984). Comparison of the immune cellular response of fish vaccinated by immersion and injection of Vibrio anguillarum. Bulletin of the Japanese Society of Scientific Fisheries50, 1187-1192.
Salk, D. and Salk, J. (1984). Vaccinology of poliomyelitis. Vaccine.2, 59-74. Sano, T. 1976. Viral diseases of culture fish in Japan. Japanese Fish Pathology. 10, 221-226.
Sano, T. and Yamazaki, T. (1973). Studieson viral diseases of Japanese fishes infectious pancreatic necrosis virus ofamago trout. Bulletin of the Japanese Society of Scientific Fisheries. 39, 477-480. 155
Schwarts, R. H. (1985). T-lymphocyte recognition of antigen in association with gene productsof the major histocompatibility complex. Annual Review of Immunology. 3,237-261.
Secombes, C.J. (1992). The role of phagocytes in protective mechanisms of fish. In: Annual Review of Fish Diseases. Faisal, M. and Hetrick, F.M. [eds.]. 2, 53-73.
Shieh, H.S. (1988). An extracellular toxin produced by fish kidney disease bacterium, Renibacterium salmoninarum. Microbiology Letters. 37, 141-145.
Shieh, H.S. (1989). Protection of Atlantic salmon against bacterial kidney disease with Renibacterium Salmoninarum extracellular toxin. Microbiology Letters. 41, 69-71.
Sima, P. and Vetvicka, V. (1990). Evolution of immune reactions. CRC Press, Boca Raton, Florida. 171-227
Sizemore, R. C., Miller, N. W., Cuchens, M. A. Lobb, C. J., Clem, L. W. (1984). Phylogeny of lymphocyte heterogeneity: The cellular requirements for in vitro mitogenic responses of channel catfish leucocytes. Journal of Immunology 133, 2920-2924.
Smith, A.M., Potter, M. and Merchant, E.B. (1967). Antibody-forming cells in the pronephros of the teleost Lepomis macrochirus. Journal of Immunology 99, 876-882.
Smith, P.D. (1988). Vaccination agaisnt Vibriosis.In Fish Vaccinataion.(ed) Ellis.A.E. Academic Press San Diego, CA,: 255 pp
Sprent, J. (1994). T and B Memory Cells. Cell, 76, 315-322
Stolen, J.S., Anderson, D.P., Van Maiswinkel, W. B. (1986). Fish Immunology. Elsevier, New York, 443 pp.
Stolen, J.S., Makela, O. (1975). Carrier preimmunization in the anti-hapten response of a marine fish. Nature. 234, 718-719.
Stott, E.J., Tyrrell, D.A. (1986). Applications of immunological methods in virology. In Weir, D.M. (ed.) Applications of immunological methods in biomedical sciences. Blackwell Scientific Publications, Oxford. 4, 120-1204.
Street, N.E., Mosmann, T.R. (1991). Functional diversity of T-lymphocytes due to secretion of different cytokine patterns. FASEB. 5, 171-177.
Stroband, H.W.J. and Kroon, A.G. (1981). The development of the stomach in clarias lazera and the intestinal absorption of protein macromolecules. Cell Tissue Research. 215, 397-425. 156
Stroband, H.W.J. and Vander Veen, F.H. (1981).The localization of protein adsorption during the transport of food along theintestine of the grass carp, Ctenopharyngodon idella (val.) Histochemistry.64, 235-249.
Tatner, M. R. (1986). The ontogeny of hummoralimmuity in rainbow trout. Salmo gairdneri. Veterinary Immunologyand Immunopathology 12, 93-105.
Tatner, M.F. (1987). The quantitative relationshipbetween vaccine dilution, length of immersion time and antigenuptake, using radiolabelled Aeromonas samonicida bath in direct immersion experimentswith rainbow trout, Salmo gairdneri. Aquaculture. 62, 173-185.
Tengelsen, L. A., Anderson, E. Leong, J. (1989).Variation in fish interferon-like actviity: Cell line production an IHN virus isolatesensitivity. Fish Health Science / American Fisheries Society NewsLetter 17, 4-5.
Thiry, M., Lecocq-Xhonneux, F., Dheur, I., Renard,A., de Kinkelin, P. (1990). Molecular cloning of the mRNA coding for theG protein of the viral hemorrhagic septicemia (VHS) of salmonids.Veterinary Microbiology. 23, 221- 226.
Thornton, J.C., Garduno, R.A., Newman, S.G.,Kay, W.W. (1991). Surface- disorganized attenuated mutants of Aeromonassalmonicida as furunculosis live vaccines. Microbial Pathogenisis.11, 85-99.
Tittle, T (1978). Functional Differentiationof Memory B lymphocytes. Ph.D. Thesis University of Oregon Health SciencesCenter. 139 pp.
Tittle, T. V., and Rittenberg, M. B. (1980).IgG memory B-cell population subdifferences in susceptibility to stimulationby TI-1 TI-2 antigens. Journal of Immunology 124, 202-206.
Tomanaga, S., Kobayashi, K., Kajii, T.,Awaya, K. (1984). Live populations of immunoglobulin forming cells in theskate, Raja kenojii: their distribution and characterization. Developmental and ComparativeImmunology. 8, 803-808.
Tomonaga, S., Kobayashi, K., Hagiwara,K., Yamaguchi, K. and Awaya, K. (1986). Gut-associated lymphoid tissue (GALT)in elasmobranchs. Zoological Science. 3, 453-458.
Toranzo, A. E. and Barja,J. L.(1993). Virulece factors of bacteria pathogenic for coldwater fish. Annul. Rev. Fish Dis.5-3
Trinchieri, G. (1994). Interleukin-12 and itsrole in the generation of TH1 cells. Immunology Today. 157
Trump,G.N. and Hildemann, W.H. (1970).Antibody responses of goldfish to bovine serum albumin primary and secondaryresponses. Immunology 19, 621-627.
Turaga, P., Weins, G., and Kaattari, S. (1987).Bacterial kidney disease: the potential role of soluble protein antigen(s). Journalof Fish Biology. 31 (Suppl. A): 191-194.
Udey, L.R. and Fryer, J.L. (1978). Immunizationof fish with bacterins of Aeromonas salmonicida:. Marine FisheriesReview. 40, 12-17.
Unanue, E.R. (1984). Antigen presentingcell function of the macrophage. Annual Review of Immunology. 2, 395-428.
Vallejo, AN, Miller, N.W., Jorgenson, T.,Clem, N.W. (1990). Phylogeny of immune recognition: antigen processing/presentationin channel catfish immune response to hemocyanins.Cellular Immunology. 130, 362-369.
Vallejo, AN, Miller, N.W. and Clem, W.L.(1992). Antigen processing and presentation in teleost immuneresponses. Annual Review of Fish Diseases. 2, 73-89.
Veliji, M. I.,Albright, L. J. and Evelyn, T.P.T. (1990). Protectiveimmunity in juvenile coho salmon ( Oncorhynchuskisutch) following immunization with Vibrio ordalii lipopolysaccharideor from exposure to live V. ordalli cells. Diseases of Aquatic Organisms 9,25-29.
Vestergaard-Jorgensen, P.E. (1970).The survival of viral hemorrhagic septicemia (VHS) virusassociated with trout eggs. Riv. Ital. Piscicol. Ittopatol.5, 13-14.
Vitetta, E.S., Berton, M.T., Burger, C.,Kepron, M., Lee, W.T. and Yin, X.M. (1991). Memory B and T Cells. AnnualReview of Immunology 9, 193-217.
Voss, E. W., Groberg, W.J. and Fryer,J.L. (1978). Binding affinity ofa tetrameric coho salmon Ig anti-haptenantibodies. Immunochemistry 15, 459- 464.
Wan, C. P. Mayung, E. and Lau, B. H.S. (1993). An automated micro- fluorometric assay for monitoringoxidative burstactivityof phagoycytes. Journal of Immunological Method159, 131-138.
Watson, J. (1979). Continous proliferationof murine antigenic-stpecific helper T-lymphocytes in culture. Journalof Experimental Medicine. 150, 15/0-/515.
Weaver,C.T.,Unanue, E.R. (1990).The costimulatoryfunction of antigen presenting cells. Immunology Today. 11,49-54. 158
Wedemeyer, G. A. (1970). The role of stress in the disease resistance of fishes. American Fisheries Society Special Publication. 5, 30-34
Wenneras, C.,Svennerholm, A.,Ahren, C. and Czerkinsky, C. (1992). Antibody-secreting cells in human peripheralbloodafter oral immunization with inactivated enterotoxigenic Escherichia coli vaccine. Infection and Immunity 60, 2605-2611.
White, D.O., and Fenner, F.J. (1986). Medical Virology, Third Edition. Academic Press, Orlando, Florida. 635 pp.
Wiens, G.D., and Kaattari, S.L. (1989). Monoclonal Antibody Analysis of Common Surface Protein(s) of Renibacterium salmoninarum. Fish Pathology. 24(1) 1-7.
Wiens, G.D., and Kaattari, S.L. (1991). Monoclonal Antibody Characterization of a Leukoagglutinin produced by Renibacterium salmoninarum. Infection and Immunity. 59:631-637.
Wilson, M.R. and Warr, G.W. (1992). Fish immunoglobulins and thegenes that encode them: In Annual Review of Fish Diseases. Faisal, M. and Hetrick, F.M. (eds.) 2, 201-223.
Wise, D.J., Schwedler, T.E. and Otis, D. O. (1993). Effects of dtresson susceptibility of naive channel catfish in immersion challenge with (Edwardsiella ictalun). Journal of Aquatic Animal Health 5, 92-97.
Wizigmann, G., Baath, G. and Hoffman, R. (1980). Isolierung des frusder viralen hamorrhagischen septikamie (VHS)aus regenbogen-forellen, Hecht- und Aschenbrut. Zentralbl. Veterinarmed. 27, 79-81.
Wolf, K. (1988). Fish Viruses and Fish Viral Disease. Cornstock Publishing Associates, Cornell University Press Ithaca, NY. 476pp
Wolf, K., Sniesko, S.F., Dunbar, C.E. and Pyle, E. (1960). Virusnature of infectious pancreatic necrosis in trout.
Wong, G., Kaattari, S.L. and Christensen J.M. (1992). Effectiveness ofan oral enteric coated vibrio vaccine for use in salmonid fish. Immunological Investigations 21, 353-364.
Wood, J.W., and Wallis, J. (1955). Kidney disease in adult chinook salmonand its transmission by feeding to young chinook salmon. Research Briefs, Fisheries Commission of Oregon. 2 32-40.
Xu, L., Mourich, D.V., Engleking, H.M., Ristow, S., Arnzen, J., Leong, J.C. (1991). Epitope mapping and characterization of the infectious hematopoietic necrosis virus glycoprotein, using fusion protein synthesized in E co/i. Journal of Virology. 65, 3, 16/1-/615. 159
Yang, M.C.W., Clem, N.W., Buttke, T.M. (1989). Unsaturated fattyacids inhibit IL-2 production in thymus-dependent antibodyresponses in vitro. Immunology. 68, 181-185.
Zapta, A. and Solas, M.T.(1979). Gut associated lymphoid tissue(GALT) in reptilia:structure of mucosalaccumulations.Developmental and Comparative Immunology. 39, 477-487.