DEVELOPMENT OF THE MARINE MYXOZOAN, KUDOA THYRSITES

(GILCHRIST, 1924), IN NETPEN-REARED ATLANTIC SALMON (SU0

SAUR L.) IN BRlTISH COLUMBIA

Jonathan David William Moran

B-Sc., University of New Brunswick (Fredericton), 1992

M-Sc., University of New Brunswick (Fredericton), 1994

A THESIS SUBMIïTED IN PARTIAL FULFILLMENT OF

THE REQUIREMENTS FOR THE

DEGREE OF DOCTOR OF PHILOSOPHY

in the

Department of Biological Sciences

O Jonathan D.W. Moran 1999

SIMON FRASER UNIVERSITY

May 1999

Al1 rights resewed. This work rnay not be reproduced in whole or in part, by photocopy or other means, without permission of the author. National Library Bibliothèque nationale 1*1 of Canada du Canada Acquisitions and Acquisitions et Bibliographie Services services bibliographiques 395 Wellington Street 395. me WeHingtm OttawaON K1A ON4 OcrawaON KiA OF)4 Canada Canada

The author has granted a non- L'auteur a accordé une licence non exclusive Licence aliowing the exclusive permettant à la National Libras, of Canada to Bibliothèque nationale du Canada de reproduce, loan, distribute or seii reproduire, prêter, distribuer ou copies of this thesis in microfom, vendre des copies de cette thèse sous paper or electronic formats. la forme de microfiche/film, de reproduction sur papier ou sur format électronique.

The author retains ownership of the L'auteur conserve la propriété du copyright in this thesis. Neither the droit d'auteur qui protège cette thèse. thesis nor substantial extracts fiom it Ni Is thkse ni des extraits substantiels may be printed or othenvise de celle-ci ne doivent être imprimés reproduced without the author' s ou autrement reproduits sans son permission. autorisation. ABSTRACT

firdoa rhyrsites (phylum Myxozoa) is of concern to the British Columbia (BC) aquaculture industry because of its association with pst-rnortem myoliquefaction in netpen-reared Atlantic salmon (Sdmo salar L.). This condition, commonly referred to as soft flesh syndrome. significantly decreases the market value of afTected fish products.

The seasonality of K. thyrsites was investigated following natural exposure of

Atlantic saImon in seawater netpens at the Pacific Biologicai Station, in Nanaimo. BC.

Atlantic salmon were exposed for several 8-week penods throughout the year, and it was determined that the infections were contracted in only the summer and Ml. The progression of these infections in both netpen and tank popuiations was followed for up to 20 months. The majority of fish contracted K. rhyrsites infections. Sporulation typically occurred within 6 months (approximately 2000 degree-days) afier transfer to seawater, and the fish almost completely recovered fiom the original infections within a year. Of 3 1 nonsalmonid fish species collected fiom the vicinity of seawater netpens and fiom research cruises off Vancouver Island, five species were identified as potentiai reservoir hosts incl uding rock sole (Pleuronectes bilineatus), tube-snout (Aulorhynchusflavidus), arrowtooth flounder (Atheresthes sromias), and two new host records, lingcod (Ophiodon elongatus) and thread fin sculpin (Icelinusfilamentosus).

Attempts to transmit the parasite directly fiom fish to fish by intubation of fiesh myxospores were unsuccessfûi. However, intraperitoneal injection of blood collected from a coho salmon (Oncorhynchus kisutch) infected with K. thyrsites successfully transmitted the infection. Using light rnicroscopy to investigate the sequential development, K. rhyrsites

infections were detected in the somatic musculature at 13 weeks pst-exposure (pe.). In a subsequent experiment, infections were first detected at 6 weeks p-e- using a PCR test and afier 9 weeks p.e. using light microscopy. The earliest developmental stage detected by histology was a smail plasmodium containhg four nuclei at 9 weeks p.e. No host response was observed histologicaily that was directly related to muscle fibers which contained intact pseudocysts. However, the response associated with ruptured pseudocysts was characterized by chronic, multifocal inflammation between the muscle fibers. DEDICATION

This dissertation is dedicated to my farnily, and especially my parents, who have provided

me with unwavering support throughout my university career. This accomplishment

would not have been possible without their guidance and moral support. 1 would also like

to dedicate this, in part, to the residents of St. Martins, New Brunswick, many of whom

have had a significant impact on my values and beliefs. Al1 men dream: but not equally. Those who drearn by night in the dusty recesses of their

minds wake in the day to find that it wris vanity: but the drearners of the day are

dangerous men. for they may act their dream with open eyes. to make it possible.

T.E. Lawrence ACKNOWLEDGEMENTS

1 wouid like to express my sincere thanks to rny academic supervisor. J.M.

Webster? for his encouragement throughout this study. His financial support for travel to

national and international conferences is laudable. My complete gratitude goes to my

research supervisor. M.L. Kent? without whose enthusiastic assistance and financial

support. this project would never have been able to continue for the past five years. His

unconditional moral support, in times when 1 doubted both myself and others, kept me

headed in the right direction. 1 must also acknowledge my bnef, but priviledged, period with L. Margolis. His sudden passing was certainly a somber event during my tenure at the Pacific Biologicai Station (PBS).

I wouId like to acknowledge the role of several employees of the PBS, without whom this research would not have been able to take place on such a scale. In particular, 1 must thank the director of the PBS, D. Noakes, for allowing me access to the PBS facilities for the past 4 years. Also, the assistance of the fish health and histology section employees is gratefully acknowledged, especially S. Dawe, D. Whitaker, T. McDonald, J.

Richard. J. Bagshaw, and V. Rantis. Laboratory space and technicai assistance with the molecular screening were provided by R.H. Devlin and J. Khattra. Fish maintenance at the experimental farm was performed by R. Kennedy and L. Lewington. Of no iess importance was R. Kennedy's organization of the annual hockey pool.

Scholarships and fellowships were provided by the Science Council of British

Columbia and Simon Fraser University. Research funding was provided by Pacific Aqua

Salmon Farming Partnerships (l3. Hicks), Fisheries and Oceans Canada, and the Natural

vii Sciences and Engineering Research Council.

1 would also like to thank M. Bhakthan and S. Foran of Simon Fraser University for diligently forwarding mail to me in Nanaîmo, and keeping me informed of pertinent

University news-

Permission to reproduce photographs was provided by S. Hallett (University of

Queensland), Y. Maeno (National Research Institute of Aquaculture Fisheries Agency), and H. Yokoyama (University of Tokyo). Kudoo rhyrsires fiom South Afiica was provided by M. Kerstan of the Sea Fisheries Research Institute. The English translation of

Egusa (1 986) was provided by S. Feist (Centre for Environment, Fisheries & Aquaculture

Science (CEFAS)). I would also like to acknowledge the hospitality of S. Feist and M.

Longshaw during my visit to CEFAS.

viii TABLE OF CONTENTS

Page ... ABSTRACT ...... IH

DEDICATION ...... v

QUOTATION ...... vi .. ACKNO WLEDGEMENTS ...... vu

LIST OF TABLES ...... xiv

LIST OF PLATES AND FIGWS ...... xv

CHAPTER 1 - INTRODUCTION ...... 1

Salmonid aquaculture in British Columbia ...... 2

Effect of fish pathogens on the BC aquaculture ïndustry ...... 4

Drug treatments ...... 7

Vaccines ...... 7

The phylum Myxozoa Grassé, 1970 ...... 8

Life cycle and development of myxosporeans ...... 1 1

The genus Kudoa Meglitsch, 1947 ...... 15

Kudoa thyrsires and the BC aquaculture industry ...... 16

Objectives of this study ...... 16

CHAPTER 2 - GENERAL METHODS ...... -1 8

1. LocaIity ...... 19

2. Holding facilities ...... 19 Page

2.1. Seawater netpens ...... ,., ...... 19

2.2. Freshwater tanks ...... 19

2.3. Seawater tanks ...... 19

3 . Hosts ...... 23

3.1 .Exposure to the parasite ...... -23

3 .2.Sample collection ...... 24

3 -3.Tissue processing ...... -24

CHAPTER REVIEW THE GENUS KUDOA

Introduction ...... -27

Taxonomy of Kudoa ...... -29

Host and geographic distribution ...... -33

Development of Kudoa ...... 34

Host-parasite interactions ...*...... -37

Post-mortern myoliquefactive autolysis ...... -41

Detection methods and control strategies ...... 45

7.1 .Gross examination ...... *...... -46

7.Z.Microscopic examination ...... 46

7.3.PCR assays ...... 47

7.4.Immunological techniques ...... 47

7.5.Control strategies ...... -48 Page

8 . Synopsis of described Kudoa species ...... -50

8.1 . Kudou species infecting the somatic musculature ...... 50

8.2. Kudou species fiom sites other than the somatic musculature ...... 64

8.3. Reports of undescribed Kudoa spp ...... 72

9 . Conclusion ...... 73

CHAPTER 4 .PROGRESSION OF NATURAL INFECTIONS ...... 75

Introduction ...... -76

Methods ...... -77

2.1 .Seasonality of the infective stage ...... 78

2.2.Progression of naturai infections ...... -80

2.2.1 . Seawater netpens ...... -80

2.2.2. Seawater tanks ...... 82

2.3 .Survey of potential reservoir hosts ...... -82

Results ...... 82

3.1 .Seasonality of the infective stage ...... -82

3.2.Progression of naturai infections ...... -84

3.3. Survey of potential reservoir hosts ...... -84

Discussion ...... 87 Page

CHAPTER 5 .TRANSMISSION ...... 92

1 . introduction ...... 93

2 . Methods ...... 94

2.1 .Exposure to filtered seawater ...... 95

2.2. Development in fiesh water ...... -96

2 -3.Transmission by injection of blood ...... -97

2.4.Transmissionby intubation of myxospores ...... -97

3 . Resul ts ...... -98

3.1 .Exposure to filtered seawater ...... 98

3.2.Developrnent in fiesh water ...... 98

3.3 .Transmission by injection of blood ...... 98

3.4.Transrnission by intubation of myxospores ...... -99

4 . Discussion ...... -99

CHAPTER 6 .SEQUENTIAL DEVELOPMENT ...... 104

1 . Introduction ...... 105

2 . Methods ...... 106

2.1 .1995-97 Expriment ...... 106

2.2.1997 Experiment ...... 107

3 . Results ...... 108

3.1 Light microscopic observations ...... 108

3.2. 1 995-97 Experiment ...... -109

xii Page

3.3. 1997 Expriment ...... 115

4 . Discussion ...... 115

CHAPTER 7 .DISCUSSION ...... 124

CHAPTER 8 .CONCLUSIONS ...... 139

CHAPTER 9 - LIST OF REFERENCES ...... 142

xiii LIST OF TABLES

Page

Table 4.1. Seasonality of infection by Kudoa thyrsites in Departure Bay, near Nanaimo, British Columbia ...... 83

Table 4.2. Prevaience of Kudoa thyrsites (No. positivehio. exarnined) in wild-caught non-saimonid fishes collected from the coastal waters of

Vancouver Island. British Columbia, offshore or near netpens ...... 86

xiv LIST OF PLATES AND FIGURES

Page

Plate 1 ...... 1 3

Figure 1.1. Diagrammatic representation of the life cycle of fieshwater

myxozoans, as seen in Myxobolus spp. (fiorn Kent and Poppe 1998).

Plate II ...... 20

Figure 2.1. Map of Vancouver Island, British Columbia. showing the

Nanaimo location of the Pacific Biological Station.

Plate III ...... 21

Figure 2.2. Map of Departure Bay. Nanaimo, British Columbia, showing

the location of the experimental seawater netpens and the seawater intakes,

at a depth of 22 m. in relation to the Pacific Biological Station.

Plate IV ...... 22

Figure 2.3. Schematic of the experimental filtration system that was used to

determine the system's efficiency at removing the Kudoa thyrsites fiom the

seawater that is supplied to the Pacific Biological Station. Page

Plate V. Macroscopic observations of various fish species infected with

Kudoa spp...... -28

Figure 3.1 a. Pseudocysts of Kudoa spp. in Pacific hake (Merluccizis

producrus).

Figure 3.1 b. Srnoked Atlantic salmon (Salmo salar) infected with Kudoa

fhvrsites.

Figure 3.1~.Fresh Atlantic salmon held on ice for 5 days showing severe

post-mortem myoliquefaction.

Figure 3.1 d. Indo-Paci fic whiting (Sillago sp. infected wi th Kudoa ciliatae.

Figure 3.1 e. Buri (Seriola qtrinqueradiata) heavily infected with Kudoa

amamiensis.

Plate VI. Wet mount and stained preparations of myxospores of various Kudoa

spp. demonstrating the variation in myxospore morphology...... 3 1

Figure 3.2a. Apical view of quadrate myxospore of Kudoa inlestinalis.

Figure 3.2b. Lateral view of K. inlestinalis myxospore.

Figure 3.2~.Myxospores of Kudoa miniauriculatu showing uplifted tips

of valve termini,

Figure 3.2d. Quadrate myxospores of Kuha puniformis.

Figure 3.2e. Stellate myxospores of Kudoa thyrsires showing unequal size

of polar capsules.

Figure 3.2f. Stained preparation of K rhyrsites myxospores (Giemsa).

xvi Page

Plate VII. SEM preparations of Kudoa spp...... 32

Figure 3.3a. Lateral view of Kudoa iniestinalis myxospore showing apical

projections.

Figure 3.3 b.Apical and posterior views of Kudoa miniauriculata myxospores

showing uplifted tips.

Figure 3.3~.Apical view of Kudoa panformis myxospore.

Figure 3.3d. Posterior view of Kudoa thyrsiies myxospore.

Plate VI11 ...... 36

Figure 3.4. TEM preparation of Pacific hake (Merluccius productus)

Infected with Kudoa paniformis.

Plate K. Inflammatory responses of fish hosts to infections by Kudûa spp...... 38

Figure 3.5a. Final stages of inflammatory response to Kudoa sp. in Pacific

hake (Merluccius productus).

Figure 3Sb. TEM preparation of host response to Kudoa sp. in Pacific hake.

Figure 3 Sc. Inflammatory response of Atlantic salmon to Kudoa thyrsites

infection.

Figure 3.5d. Increased magnification of the inflammatory response in Atlantic

salmon to K. thyrsites infection.

Figure 3 Se. Inflammatory response in lingcod (Ophiudon elongatus) infected

with K. thyrsites.

xvii Page

Plate X. Histoiogical preparations of Kudoa spp. in various fish species (H&E) .....40

Figure 3.6a. Kudoa panr~orrnisin Paci fic hake.

Figure 3.6b. Co-infection of Kudoa thyrsites and K. pan~~orrnis.

Figure 3.6~.Kudoa thyrsites pseudocyst in Atlantic salmon.

Figure 3.6d. Early K. thyrsites infection in Atlantic saimon.

Figure 3.6e. Kzidoa thyrsites pseudoc yst in lingcod (Ophiodon elongaius).

Figure 3.6f. Kudoa thyrsites in cardiac muscle of coho salmon

(Oncorhynchus kisutch).

Figure 3.6g. Kudoa ciliatae in intestinal musculature of Indo-Pacific

whiting (Sillago spp.).

Figure 3.6h. Unidentified Kudoa infection in lingcod (0.elongutus).

Plate XI ...... 79

Figure 4.1. Map of Vancouver Island, British Columbia., showing sarnpling

sites where wild-caught non-salmonid fishes were collected fiom the coastal

waters offshore or near netpens.

Plate XII ...... 8 1

Figure 4.2. Biweekly mean water temperatures at the Pacific Biological

Station experimental seawater netpens (at a depth of 4 m) and in the

seawater tanks between January 1995 and May 1997.

xviii Page

Plate XII1 ...... 85

Figure 4.3. Progression of Kudoa thyrsites infections in Atlantic salrnon

within two seawater netpen populations (1995 and f 996) and one seawater

tank population (1 995).

Plate XIV. Developing plasmodia of Kudoa thyrsites within the musculature of

Atlantic salmon ...... 1 1O

Figure 6.1 a. Presporogonic plasmodium within the somatic musculature.

Figure 6.1 b. Presporogonic plasmodium containing several discemible

generative cells or vegetative nuclei.

Figure 6.1~.Early plasmodial stage developing within the cardiac

musculature.

Figure 6.1 d. increased magni fication of Figure 6.1 c.

Plate XV. Sporogonic plasmodia of Kudoa thyrsites within the somatic

musculature of Atlantic salmon ...... 1 1 1

Figures 6.2a and 6.2b. Early sporogonic plasmodia within the somatic

musculature with developing myxospores.

Figure 6.2~.Apparent double infection of a muscle fiber by plasmodia.

Figure 6.2d. Plasmodium with fùlly-formed myxospores within the somatic

musculature.

xix Page

Plate XVI. Inflammation associated with Kudoa thyrsites infections ...... 1 12

Figure 6.3a. Intact polysporic plasmodium within a muscle fiber.

Figure 6.3b. Granuloma with myxospores between the muscle fikrs.

Figure 6.3~.Chronic, extensive inflammation between the muscle fibers

as seen in cross-section.

Figure 6.M. My'tospores within phagocytes.

Figure 6.3e. Regenerating muscle fibers during resolution of infection.

Plate XVII ...... 113

Figure 6.4. Prevalence of Kudoa thyrsites stages and associated

inflammation in the musculature of Atlantic saimon using a combination

of histology and wet mount preparations in fish first exposed 25 May 1995.

Plate XVIII ...... 114

Figure 6.5. Prevalence of Kudoa thyrsites stages and associated

inflammation in the musculature of Atlantic saimon using both histological

examination and a K. thyrsires-specific PCR test of fish first exposed

09 June 1997.

PIate XIX ...... 1 16

Figure 6.6. PCR of DNA from Atlantic salmon tissues screened using the

Kudoa thyrsites-specific PCR test. CHAPTER 1

INTRODUCTION ' 1. Salmonid aquaculfure in British Columbia

Salmonid aquaculture in British CoIumbia (BC) has expanded significantly over the past decade from a production of 6 600 tonnes in 1988 to 23 800 tonnes in 1995. and has become BCTslargest agicultural expon (British Columbia Environmental

Assessment Office 1997). On a global scale, the BC industry ranks fourth in salmonid production with a market share of 4.3%, behind Norway (45.5%), Chile (22.9%), and the

United Kingdom (1 1.8%) (British Columbia Environmental Assessment Office 1997). in

BC. Atlantic salmon (Salmo salar L.) account for 69% of the annual salmonid production in the provincial aquaculture industry, with chinook salmon Oncorhynchus tshawytscha and coho salmon 0. kisutch, accounting for 29% and 2%, respectively.

Salmon production in 1996 had ân estimated wholesale value of more than $172 million and, therefore, the industxy bas become an integral component of the economies of several coastal comrnunities of BC. This is the result of direct employment within the aquaculture industry and the indirect economic benefits of having the employees reside in these coastal communities. Specifically, the industry provides fiill-time employment to almost 3000 people, with wages and benefits totaling over $78 million annually.

in 1995. a provincial salmon aquaculture strategy was announced jointly by the

BC Ministry of Agriculture. Fish, and Food, and the BC Ministry of Environment, Lands, and Parks. This aquaculture strategy included a moratorium to limit expansion of the aquaculture industry in BC and a recommendation to have an environmental assessrnent of current aquaculture practices. Continual concerns over the commercial rearing of salmon in BC have resulted in the establishment of an expert advisory panel on the aquaculture industry. The panel was comprised of representatives from the federal, provincial and municipal govemments, industry, environrnental organizations, First

Nations. commercial and sports fisheries, tourism, and labor. It was established to prepare

an environrnental assessment of cwrent aquaculture practices in the province and has

addressed concerns over the escapement of Atlantic salrnon from aquaculture sites and

the potential establishment of this species in BC river systems. Other concems included

the risk of disease transmission between farmed and wild fish species, the impact of

waste discharge fiom the aquaculture sites, the e ffect of extensive aquaculture on marine

mammals, and the location of salmon fms(British Columbia Environmental

Assessment Office 1997).

The report prepared by the environmental assessment pane1 made

recommendations that, once addressed, would permit the industry to i) expand with

minimal impact on the environment irnrnediately adjacent to the culture sites. and ii) develop in locations without disease risk to the wild fishenes. In dl, 49 recommendations

were made. nine of which involved addressing the significant number of concerns regarding the health of farmed fish and its potential impact upon the heaith of wild fish populations. including both commercially and non-cornrnerciaily important species.

These recommendations include the following: 1) establishment of a cornmittee to develop a comprehensive fish health policy; 2) strengthened monitoring practices for pathogens; 3) development and enforcement of fish health standards; 4) improved accessibility to, and quality of, fish heaith information; 5) strengthened egg and fish importation policies and programs; 6) strengthened sarnpling and disease reporting requirements for fish to be transferred within the province; 7) enhanced inspection practices at fish processing plants; 8) strengthened control of dmg (e.g., antibiotics) usage on aquaculture sites; and 9) to review issues and public concerns related to the use of these drugs.

As a result of the decreasing price for sahon products, combined with the continued moratorium against acquiRng new netpen sites, the BC salmon fming industry has modified its approach so as to increase efficiency on the sites that are currently in use. This increased eficiency is the result of the change in technology, and in particular, the use of larger netpen structures. This shift fiom 15 m cages to 30 m diarneter cages and to even larger, 90- 120 m, circular cages has resulted in a more efficient use of the current sites resulting in the continued increase in production. Another option of the industry is the use of closed bag containment systems, which permit more environmental control. in particular, control over the depth at which the seawater intake is situated allows adjustment of the water temperature to provide optimal rearing conditions.

As well, the closed bag system affords protection against predators and hannful environmental occurrences such as aigai blooms.

2. Effect of fts/r puthogens on the BC uquaculrure industry

One of the greatest concerns of any agricultwal based industry is loss resulting from pathogenic organisms that either cause mortality, reduce yield, or lower product quality. The risk of disease within netpen populations has been an important concem to the salmon farming industry since its formation. This potential risk of disease exposure in seawater netpens is high due to the unrestricted seawater flow (Kent 1992). Kent (1992) and Kent and Poppe (1 998) have prepared reviews of the diseases that are of concem to the salmon netpen industry which include diseases induced by viral and bacterial pathogens, füngal organisms, protozoan and metazoan parasites, as well as algal blooms and idiopathic diseases.

Bacterial kidney disease (BKD), caused by Renibacterium salmoninarum, has been a chronic problem when rearing Pacific salmon (Oncorhynchus spp.) in BC. With the transition to Atlantic saimon in the mid-1980's, BKD became less of a concern.

However, other diseases and parasites that are native to the Pacific Ocean but were not considered a significant detriment to the cufture of Pacific salmon have become significant in the rearing of Atlantic salmon in seawater netpens. These diseases and pathogens include infectious hematopoietic necrosis 0,fiirr?nculosis (Aeromonas salmonicida), myxobacteriosis (Cytophaga/FZavobacteriumspp.)' sea lice (e-g.,

Lepeophtheirus salmonis and Caligus elongatus), and "sofi flesh syndrome", caused by

Ktrdoa rhyrsires (Gilchrist, 192J), the focus of this dissertation.

IHN is an extremely problematic viral disease that causes significant losses in susceptible fish, particularly the young salmon, and was first reported in pen-reared

Atlantic salmon in BC in 1992 (Traxler et al. 1993). This pathogen causes mortality in young fish by causing severe necrosis in the hematopoietic tissue of the kidney and spleen

(Yasutake 19701, and has become a significant disease threat to Atlantic salmon fmsin

BC.

Another pathogen that bas affected Atlantic salmon culture in BC is the bacteriwn

A. salmonicida, which causes the disease commonly referred to as funinculosis. Both acute and chronic foms of the disease may occur (Evelyn et al. 1998). Acute cases may show symptoms such as anorexic or lethargic fish, darkening of the skin, and hemorrhages of the skin and fins. Chronic tuninculosis may be manîfest as "furuncles" underlying the skin, or as large, bloody ulcers (Evelyn et al. 1998).

Myxobacteriosis (CytophagdFlavobacieriurn spp.) has ken associated with high rnortalities in pen-reared Atlantic salmon on the BC Coast. Two types of myxobacterial infections are recognized; one of which causes mouth lesions and the other causes skin lesions (Evelyn et al. 1998). Myxobacteriai stomatitis is manifest as macroscopic yellow patches in the mouth and lethargic. emaciated. and anorexic fish. Severe infections may cause complete erosion of the mandible. Skin lesions resulting fiom myxobacterial infections are seen as large, white patches on the posterior of the body. Complete areas of the skin may be eroded exposing the musculature and either permitting establishment of secondary infections or aecting the fish's ability to osmoregulate (Evelyn et al. 1998).

The sea lice, Lepeophtheirus and Caligus spp., gaze ectoparasiticalty upon the surface of the fish, feeding on mucus, skin, and blood. Heavy infections may result in erosion of the epidermis and sub-epidemis, leading to secondary infections and osmoregulatory problems (Johnson and AIbright 1992). It has been demonstrated that sea lice may act as vectors for both viral and bacterial pathogens, such as infectious salrnon anemia (ISA) (see Johnson 1998).

The recent review (Kent et al. 1994b) of myxosporeans of concern to both the commercial salmonid fisheries and the aquaculture industry in BC included the proliferative kidney disease organism (PKX)and Ceratomyxa shasta as pathogenic species found in BC waters. Of no less importance to the industry are those myxosporean species that affect product quality, either as a result of degradation of the flesh (e-g., K. ihyrsites) or due to large unsightly cysts (e-g., Henneguya salminicola). However, myxosporean infections have ken used also as biological tags in BC fisheries management (e.g ., H. salminicola; Myxobotus arcticus) and in regulatory enforcement

(M. arcticus) (cf. Margolis 1982, 1993).

Hea~yinfections of the myxozoan K. thyrsites may cause pst-mortem myoliquefaction, commonly referred to as sofi flesh syndrome, in several marine fish species. These infections do not cause fish rnortality, but instead, significantly decrease product quality afier the saimon are harvested.

3. Drug treatments

Several dmgs are in use to combat bacterial, fungai, and parasitic diseases in cultured fishes. The antibiotics used include oxytetracycline, potentiated sulfonamides

(e-g..Romet 30), florfenicol, and erythrornycin (Brackett and Karreman 1998). Those to control sea lice include Azimethiphos and Ivermectin, but their use is permitted only under limited registration or with veterinary approvai. Formalin has been used to control

Freshwater ectoparasites (e.g., monogeneans) and fimgal growth on eggs (E3rackett and

Karreman 1998). Experimental drug treatments using TNP-470, an analogue of the drug fumagillin-DCH, are highly effective against microsporeans such as Loma satmonae and

Nzrcleospora salmonis, and against some myxosporean infections (e-g., PKX organism)

(Higgins and Kent 1998; Higgins et al. 1998). However, TNP470 is limited to erperimental use and cannot be used to protect any animai produced for human consurnption. To date, there is no dmg available for treating K. thyrsites infections.

3. Vaccines

Fortunately, significant advances in vaccinology have permitted the development of multivalent vaccines to protect against a suite of organisms pathogenic to fish (e.g.,

funrnculosis caused by A. sahonicida and vibriosis caused by Vibrio spp.). Vaccines designed against several other viral (e-g., infectious pancreatic necrosis (PN),MN) and bactenal (e-g-. BKD, bacterial mouthrot, rickettsia) infections are either in development or in experimental use (Brackett and Karreman 1998). The possibility of developing vaccines against myxosporeans (e-g., PKX organism. K. hyrsires) is king explored.

5. The phylum Myxozoa Grassé, 19 70

Myxosporeans (phylum Myxozoa) are an assemblage of multicellular organisms recognized primarily as parasites of teleost fishes (Lom 1987). However, rnyxosporean infections have been reported in other vertebrates (e.g., elasmobranchs (Stofiegen and

Anderson 1 990; Heupel and Bennett 1996), aquatic reptiles (Kudo 191 9; Johnson 1969). and arnphibians (McAllister and Trauth 1995)). Myxosporean infections have ken described also from invertebrate hosts (e-g., bryozoans (Canning et al. 1W6), annelids

(Kudo 19 19), insects (Thélohan 1895), and digeneans (Overstreet 1976)).

Levine et al. (1 980) published a revised classification for al1 protozoans that included the phylum Myxozoa as a matter of convenience, even though their multicellular origin was recognized and accepted. Within the phylum were the class Myxosporea

Bütschli, 188 1, which included the histozoic and coelozoic myxozoan parasites of poikiiotherrnic vertebrates and some invertebrates, and the class Actinosporea Noble in

Levine et al., 1980, which included those myxozoans parasitizing annelids. Stolc (1899) first suggested the metazoan affinity of the myxozoans. With the advent of molecular systematics Smothers et al. (1 994) investigated the phylogenetic position of three bivalvulid myxozoan genera arnongst several eukaryotic species using the 18s ribosomal

gene sequence, and confmed that myxozoans constitute a metazoan lineage. However,

they claimed that there was no evidence for a cornmon evolutionary history between the

myxozoans and midarians (phylum Cnidaria), and instead that myxosporeans were

closely related to the bilateral animais. Comparable results have been achieved in the

phylogenetic analysis of Myxidium lieberkuehni and Tetracupsula bryozoides (cf.

Schlegel et al. 1 996; Anderson et al. 1 998). Ultrastructural analysis of Thelohanellus

nickolskii by Siddall et al. (1995) confinned the metazoan lineage. Howver, their

phylogenetic analysis of Henneguya doori suggests a taxonornic affinity with the

cnidarian Polypodiurn hydriforme. Additional phylogenetic analyses of myxozoans fiom

other representative groups are required in order to settle this contentious systematics

issue.

The construction of phylogenetic trees using the most complete smail-subunit

(SSU) rDNA sequences produces results that generaliy agree with the phylogenetic

hypotheses of Shulman (1 966). According to Shulman (1966), the first myxozoans were

coelozoic in marine teleosts, and later evolved to fom histozoic species. As well, he

suggested that the ancestral bipolarid genera gave rise to the platysporinid genera in fiesh

water and that the multivalvulids, including Kudoa, evolved fiom ancestors comparable to Ceraromyxa. Presently, SSU rDNA sequence is available fiom 20 myxozoan species

from seven genera (Kent and Palemela 1999). The continued use of SSU rDNA in phylogenetic analyses will provide more definitive answers as the SSU rDNA sequences of more myxozoan species fiom representative genera become available.

Recent evidence has shown that members of the class Actinosporea are in fact alternate stages in the life cycle of fieshwater myxosporeans, which led Kent et al.

(1 994a) to suppress the class Actinosporea in favor of the more senior class Myxosporea.

As a result. al1 but one of the families previously contained within the class Actinosporea

were suppressed, the exception king the family Tetractinomyxidae, which was

subsequently transferred to the class Myxosporea. According to Kent et al. (1994a), the

majority of the members of the class Actinosporea should remain as species inquirendue

until their associated myxosporean phase has ken identified- Nevertheless, species

descriptions of actinosporeans continue to be published (e.g., description of

Sphaeractinomyxon ersei by Hallett et al. 1998).

At present, the class Myxosporea is comprised of more than 1330 described

species (Lorn and Dykova 1992). Species identification of myxosporeans is dependant

upon the observation of the myxospore stage. Myxospores are composed of two to seven

myxospore shell valves, one to two infective sporoplasms, and one to seven polar

capsules (Lom and Dykova 1992). The class is divided into the two orders Bivalvulida

(spores with two valves) and Multivalvulida (spores with three or more valves). Moser

and Kent (1 994) provided a comprehensive summary of the taxonomy of the class

Myxosporea identifying 12 families within the order Bivalvulida and five families within

the order Multivalvulida.

The order Bivdvulida is compnsed of those species of myxosporeans whose

spores possess two valves. According to Moser and Kent (1994): the order Bivalvulida is comprised of 40 genera, and are represented by such well-known genera as Myxidium,

Mqisobolus, Henneguyo, and Ceratomyxa. Other characteristics of the order include the valves meeting in a single circumsporal suture and the nurnber of polar capsules king usually two (Lom and Dykova 1992). However, some bivalvulid genera have spores with

only a single polar capsule (e-g., Coccomyxa) and others have as many as four polar

capsules (e-g., Chloromyxum) (cf. Lom and Dykova 1992). The rnajority of myxosporean

species that have been described beiong to this order.

The order Multivaivulida is comprised of those species of myxosporeans whose

spores possess three or more valves. To date. the maximum nurnber of valves described

from any one species of myxosporean is seven, as is observed within the genus

Septemcapsuh. Lorn and Dykova (1 992) describe the charactenstics of the order

Multivalvulida which include: spores of radial syrnmetry composed of three to seven

valves meeting in three to seven sutures; each valve containing a polar capsule; and polar

capsule situated at the spore's apex. The lone exception is the genus Unicapsulu, which

has myxospores possessing three valves and a single, fùlly-formed, polar capsule.

6. Life cycle and developmenf of rnyxosporeans

The life cycles of myxosporeans at first were believed to be either direct, or

required a penod of ageing of the spores in the environment, before the spores were

infective to other fish. Within the past two decades, the life cycles of several fieshwater

myxosporean species have been elucidated. Markiw and Wolf (1983) were the first to

identiQ that Myxobot~cerebralis, the myxosporean species that causes salrnonid whirling disease, exhibited a heteroxenous (Le., multiple host) life cycle, and specifically, that transmission of the infection fiom fish to fish does not occur.

In the life cycle of M. cerebralis, the spores that are produced in salmonids (e-g., rainbow trout Oncorhynchus mykiss) are released into the environment after the death of the fish and cannot infect other fish. These "myxospores" must be ingested by a suitable

al temate host which, in the life cycle of M. cerebralis, is Tublyex tubtyex, a freshwater

oligochaete (phylum Annelida). Mer a 3-4 month development period (at 12°C) within

the oligochaete. "actinospores" are released into the water and infect salrnonid fry after

the actinospores either come into contact with the skin or are ingested. Afier the infective

sporoplasm penetrates the epithelium, the parasite then migrates dong the peripherai

nemes to the site of infection (Le., cartilage of the salmonid fiy) to continue its

development.

To date, a heteroxenous life cycle (see Fig. 1.1, life cycle of Myxobolus spp.) has

been demonstrated in species of the genera Myxobolus, Hoferellus, Cerafomyxa,

ZschokeZZa, Myxidiurn, and Henneguya (cf. Lom et al. 1997). Recently, polychaetes (e-g.,

:Munayunkia speciosa) have been implicated as alternate hosts in the myxozoan life cycle

(Bartholomew et al. 1997)-

It is unknown whether marine myxozoans exhibit a similar type of life cycle to those in fiesh water. Both oligochaetes and polychaetes are comrnon in the marine environment and actinosporean infections have been reported in marine oligochaetes

(Marques 1984; Hallett et al. 1998). Moreover, the marine myxosporean Myxidiurn leei has been directly transmitted to gilthead brearn (Sparus aurarus) and red drum (Sciaenops ocellatus) (cf. Diamant 1997, 1998). Although it is possible that some marine myxozoans have the ability to infect fish directly, claims of direct transmission must be investigated further because it is not clear whether the infection in these cases was transmitted by myxospores or trophozoites.

Early development in myxozoans has been investigated in several bivalvulid PLATE 1

Figure 1.1. Diagrammatic representation of the life cycle of fieshwater myxozoans, as

seen in iMyxobolus spp. (fiom Kent and Poppe 1998). a and b - vegetative

development producing daughter cells in a multinucleated plasmodium; c -

sporogenesis forming multicellular spores; d - myxospore (Le., stage infective

to annelid) released fiom the fish; e - alternate annelid host; f - actinospore

(i.e., stage uifective to fish) released from the alternate host. genera such as Myxobofw, Sphaerospora, Sphaeromyxa, Myxidiurn, Hoferellus, and

Cerarornyxa (see reviews by Lom and Dykova 1992; Moser and Kent 1994; Kent and

Paienzuela 1 999). In some genera (e.g., Myxobolus, Sphaerospora), a proliferative phase of development, referred to as an extrasporogonic phase, may occur in a site other than that of sporulation (Lom and Dykova 1992). Similarly, the PKX myxozoan represents the extrasporogonic stage of an undescribed species (Kent and Hedrick 1986). El-Matbouli et al. (1995) provided the most detailed description to date on the early development of myxosporeans within the fish host in their investigations of M. cerebralis infections in rainbow trout. After penetrating the fish's epitheliurn (e.g., skin or gut), the sporoplasm divides within the epithelial cells, and these extrasporogonic forms continue to divide during their migration to the site of spodation (Le., the cartilage with M. cerebralis infections) via the peripheral nerves. Other known sites of extrasporogonic development include the blood, swimbladder, and rete mirabile of the eye for some Sphaerospora infections, the kidney interstitiuni for PKX. and the kidney glomerulus for M. lieberklrehni (cf. Lom and Dykova 1992). The early development of members of the order

Multivalvulida, including Kudoa, has not been described. Also, an extrasporogonic phase of development has not been observed in any member of this group.

Lom and Dykova (1 992) recognize three types of sporogonic development: 1) small, mono- or disporic trophozoites that produce one or two spores; 2) mictosporic species which have mono-, di-, or polysporic plasmodia that produce one, two, or several spores (usually observed in coelozoic myxosporeans); and 3) polysporic plasmodia that contain many vegetative nuclei and generative cells (observed in both coelozoic and histozoic myxosporeans). Sporogenesis in myxozoans may occw either by development in the pansporoblasts or by direct spore morphogenesis within plasmodia (Lom and

Dykova 1992). Pansporoblasts originate by the union of nvo generative cells and usuaily

develop to produce two spores (Lom and Dykova 1992). This process is seen in those

myxosporeans that produce large plasmodial trophozoites (e.g., Sphaeromyxa,

~My.robolus.Myxidium). in direct spore morphogenesis. sporogonic cells produce

sporoblasts that give rise to valvogenic, capsulogenic, and sporoplasmic cells, each of

which has a predetermuied role in spore formation (Lom and Dykova 1992). Direct spore

morphogenesis has been observed in a pseudoplasmodia for Sphaerospora and

Cerarornyxa,and in some histozoic multivalvulid species (Lom and Dykova 1988).

Pseudoplasmodia are defined as a sporogonic stage which has a single vegetative nucleus

and produces one or two spores (Lorn and Dykova 1995). In both Kudoa lunata and

Kudoa paniformis, sporogenesis occurs without the formation of pansporoblasts within a

large plasmodium (Lom and Dykova 1992). Therefore, it has been suggested that most, if

not all. of the members of the genus Kudoa undergo direct spore morphogenesis.

7. The genus Kudoa Meglitsch, 1947

Meglitsch (1 947) established the genus Kudoa to include histozoic myxosporeans that are typically parasitic within the skeletal muscle of fish. Kudoa spp. are characterized by producing quadrate or stellate spores with delicate spore membranes and indistinct sutures dividing the spores into four shell valves, and each valve has a polar capsule. The spore contains two uninucleate sporopiasms, in which one sporoplasm envelops the other.

The type species of the genus is Kudoa dupeidire.

The genus Kudoa is comprised of 44 described species and its species are distributed throughout marine and estuarine waters worldwide (Moran et d. 1999a).

These myxosporeans are primarily histozoic, intracelluiar parasites of the musculature.

However. since the establishment of the genus, coelozoic species have been described

(see reviews by Egusa 1986; Moran et al. 1999a).

8. Kudoa thyrsites and the BC aquaculrure industry

Kudoa thyrsifes has been reported as a parasite of the musculature of 18 genera of

marine fishes around the world (see Moran et al. 1999a). Severe infections by this parasite result in post-mortem myoliquefaction in fmedAtlantic and coho salmon.

Infections by K. rhyrsires in salmonid fishes are not known to cause mortalities or physiological or behavioral abnormalities, even though infections of the cardiac muscle have been reported (Kabata and Whitaker 1989; Moran et al. 1999a). Instead, the detrimental effect of this parasite results in economic losses to the industry resulting from daims of poor product quality by retailers and the subsequent pour market perception of

BC reared salmon products. Marsal (1 994) concluded that direct losses in 1993 to the BC aquaculture industry were approximately US$682,000.

9. Objectives of thls study

The primary objective of this study was to investigate aspects of the biology of the marine myxosporean parasite K. fhyrsites affecting seawater, netpen-reared, Atlantic salmon in BC.

This research project was designed to provide new insights into the biology of K. rhyrsires with a view to helping to control the disease. The thirc! chapter (published in Aquaculture) is a review of the available literature on the genus and its 44 species. The molecuIar techniques that are currentiy king developed, should help ciariw some of the taxonomic issues within the group that are identified in this chapter. The fourth chapter

(accepted for publication in the Journal of Aquaric Animal Nealth) describes the nature and seasonality of K. thyrsites infections within the Atlantic salmon netpen populationso and describes a swey for reservoir hosts. The fifth chapter (accepted for publication in the Journa2 of Aquafic Animal Heakh) reports on an investigation into the modes of transmission of the parasite to Atlantic salmon, identifies possible routes of infection, and the fate of infections in salmon transferred back to fresh water afier a brief seawater exposure. The sixth chapter (accepted for publication in Diseuses of riquatic Organisms) reports on investigations, using traditional and polymerase chain reaction (PCR) techniques. into the development of the parasite within Atlantic salmon and subsequent host response.

This dissertation is a compilation of my research, much of which is king pu blished as collaborative papers in different peer-reviewed journals. The collaboration reflects a limited amount of assistance and contribution by the other authors. 1 have formatted al1 chapters to conform with the guidelines established by the Canadian

Jorrrnal of Fisheries and Aquatic Sciences. However, the overall thesis conforms with the requirements of the Graduate Studies Programme of Simon Fraser University. CHAPTER 2

GENERAL METHODS 1. Localiîy

The Pacific Biological Station (PBS) is situated in Nanaimo, BC (Fig. 2.1). The

experimental seawater netpens are located dong the western shore of Brandon Island

(Fig. 2.2) within easy access of the PBS- On-site tank facilities are located within the J.R.

Brett Fish Culture Building at the PBS.

2. Holdingficiîifies

2.1. Seawater netpens

Facilities at the experimental station include netpens of either 3 x 3 x 6 m (length

s width x depth) or 5 x 5 x 6 m in size. Seawater temperatures at the netpen site,

measured at 4 m depth, typically range fiom 6 - 18°C throughout the year (Fig. 4.2).

2.2. Freshwater tanks

The fi-eshwater tank facilities of the PBS (ei ther flow-through or recirculating)

include a range of tank sizes (e.g., 400,4000. and 4500 L) which are supplied with either heated. chilled, or ambient, dechiorinated fiesh water at 12 Wmin. Arnbient fieshwater tank temperatures typically range fkom 8 - 18°C throughout the year.

2.3. Seawater tanks

The seawater tank facilities include flow-through tanks in a range of sizes (e.g.,

400,4000, and 4500 L), which can also be supplied with heated, chilled, or ambient seawater. Seawater is obtained directly from Departure Bay at a depth of 22 m (Fig. 2.2).

Incoming seawater passes through sand filters (A) (Fig. 2.3) prior to king pumped into PLATE n

Figure 2.1. Map of Vancouver Island, British Columbia, showing the Nanaimo location

of the Pacific Biological Station.

PLATE LI1

Figure 2.2. Map of Departure Bay, Nanaimo, British Columbia, showing the location of

the experimental seawater netpens and the seawater intakes, at a depth of 22

m. in relation to the Pacific Biological Station. 1 Pacific Biological

1 Departure Bay PLATE IV

Figure 2.3. Schematic of the experimentd seawater filtration system that was used to

determine the system's eficacy at removing the Kudoa thyrsites infective

stage fiom the seawater that is supplied to the Pacific Biological Station.

.4rrows indicate the direction of flow of the seawater. Al1 incoming seawater

passes through sand filters (A). Additional filtration is provided by 25, 5, and

1 pn bag filters (B) and a I pm cartridge filter (C). SAND FILTER the PMfish culture building. Sand-filtered seawater that was used for several of these

experiments was treated with additionai filtration. This additional treatment included

passing through either a series of three stainless steel filter housings (B) (Mode1 FI -50,

Tate Andale Canada inc., Concord, Ontario, Canada) fitted with bag filters of pore sizes

25. 10. and 1 Fm, or a combination of these sequential bag filters plus a carûidge filter

(C) with a pore size of 1 pm (Fig. 2.3). Ambient seawater tank temperatures typically

range from 8- 12°C throughout the year (Fig. 4.2).

3. Wosk3

Al1 Atlantic salmon hosts used in this research were reared at, and obtained fiom,

either federal (Fisheries and Oceans Canada Rosewall Creek, BC) or private (Freshwater

Farrns. Duncan, BC) fieshwater hatcheries located on the east Coast of Vancouver Island.

Fish were maintained in fieshwater recirculating tanks prior to their experimental use. Al1

fish were fed an artificial, commerciaily-prepared feed (Ewos Canada Ltd., Surrey, BC)

daily. to satiation.

3.1. Exposure to the parasite

Atlantic salmon were exposed to K. rhyrsites by various means according to

experimental protocol and as specified in each chapter.

Salmon were exposed naturally to K. rhyrsites at the PBS while they were held either in the experimental seawater netpens or in flow-through, seawater tanks. Natural

exposures were initiated on 26 May 1995.

Fish were exposed expenmentally to the parasite while they were held in 400 L flow-through, fkeshwater tanks. This experirnental exposure to K. thysites was by means

of direct intubation of fkeshly harvested myxospores and by the injection into the

intraperitoneal cavity of blood collected directly fiom the caudal vessels of infected coho

salmon in 10 ml heparinized VACUTAINER@tubes (Becton Dickinson and Company,

Franklin Lakes. New Jersey, USA).

3.2, Sample collection

Samples of the treated fish were collected fiom either the netpens or tanks using dipnets, and they were then transferred to the PBS Parasitology wetlab facilities for processing. Screening for K. rhyrsites infections (e.g., seasonaiity and progression of natural infection experiments) was done by preparing wet rnounts or histological sections of tissue samples fiom fish that were killed by a sharp blow to the salmon head at the seawater netpen site. Muscle samples were removed from the left abdominal wall using dissecting instruments, placed in individual whirl-paks, and either held at 4°C temporarily or stored fiozen at -20°C until the tisssue codd be screened. Atlantic salmon that required immediate pst-mortem processing of tissues (Le., fixation for histology and

PCR) were transported live to the wetlab facilities, killed, and sampled in the laboratory irnrnediately prior to fixation of the tissues. The tissues were then processed as described in the appropriate section.

3.3. Tissue processing

Wet mount preparations of salmon muscle tissue were examined using phase contrast microscopy (Leitz Wetzlar Dialux 20) at 320x, as described in Moran and Kent ( 1999). For histological examinations, tissues were fixed in Davidson's solution

(Humason 19791, embedded in paraff~n.sectioned at 5 Pm, and stained with hematoxylin and eosin (H&E). Histological sections were examined at both lOOx and 320x using bright field microscopy.

A comprehensive methodology of each experiment performed as part of this research project is included in the appropriate section of tbis dissertation. REVIEW OF THE GENUS KUDOA

"Reprinted from Aquaculture, Volume 172, Moran, J.D. W., Whitaker, D.J., and Kent, M.L., A review of the myxosporean genus Kudoa Meglitsch, 1947, and its impact on the international aquaculture industry and commercial fisheries, Pages No. 163- 196, Copyright 1999, with permission fiom Elsevier Science". 1. Introduction

The genus Kudoa (Myxozoa: Myxosporea) is detrimental to fish products as a result of their site of infection. primarily the musculature. Thus these parasites are of concem to aquaculture and the wild fish fisheries because of their impact on product quality (Fig. 3.1). There cm be a negative visual impact on product quality due to large, unsightly cysts or regions of lysis within the musculature (e-g., Kudoa amamienris in buri

(Seriola quinqueradiata) (Fig. 3.1e); K. thyrsifes in smoked Atlantic salmon (Fig. 3.1 b)).

Probably more important is the association of Kudoa with accelerated muscle degeneration (post-mortem myoliquefaction) (e-g., K. thyrsites in pen-reared Atlantic salmon) (Fig. 3. lc). This muscle degeneration. also referred to as '%oft flesh, is probably the result of proteolytic enzymes released by the parasite (Tsuyuki et al. 1982). These enzymes are used by the parasite to break down host tissues for its growth and development. Many Kudoa species have been associated with soft flesh syndrome in cornmercially important marine fishes. Some of the most noted species are K. thyrsites in fmed Atlantic salmon, Kudoa musculoliquefaciens in swordfish (Xiphias gludius), K. paniformis in Pacific hake (Merluccius productus), and K. clupeidae in Atlantic hemng

(Clidpea harengus). For many other Kudoa species, the effect of the infections have yet to be investigated (e.g., heart and centrai nervous system infections). Several species of wild

Pacific salmonids (Oncorhynchus spp.) were identified as harboring K. thysifes infections within the cardiac muscle (Kabata et al. 1986; Kabata and Whitaker 1989).

Cardiac output in heavily uifected fish may be compromised; therefore, making them more susceptible to predation. PLATE V

Figure 3. I . Macroscopic obsewations of various fish species infected with Kudoa spp.

a) Pseudocysts of Kudoa spp. in Pacific hake (Merluccius productus), White

arrow indicates living parasite. Black arrow shows effect of host response on

parasite.

b) Smoked Atlantic salmon (Salmo salai-) infected with Kudoa thyrsites.

Arrow indicates area of parasite-induced lysis.

C) Fresh Atlantic salmon held on ice for 5 days showing severe pst-mortem

myoliquefaction.

d) Indo-Paci fic whiting (Sillago sp.) infected with Kudoa ciliatae. Arrow

indicates cyst in intestinal wal1 (fiom Hallett et al., 1997).

e) Buri (Seriola quinqueradiata) heavily infected with Kudoa amamiensis.

Arrow indicates cyst within the somatic musculature (courtesy of H.

Yokoyama).

This review deals primarily with those species of Kudoa that affect the product quality of fishes reared in aquaculture and in comrnercially important fisheries. Where possible. the common and scientific names of host fishes used are as recommended by

Robins et al. ( 1991 a, b). The objective of this chapter is not to provide a taxonomie revision of the genus, but rathet to prepare a comprehensive review of the literature currently available on species of the genus Kudoa. In addition to this review of Kudoa spp. and their relationship to soft flesh, a synopsis is included of species described to date. Based upon the information accumulated in the preparation of this chapter, an investigation into the validity of some species descriptions and host records is warranted.

2. Taxonomy of Kudoa

The class Myxosporea is divided into the orders Bivalvulida Shulman. 1959

(myxospores with two valves) and Multivalvulida Shulman, 1959 (rnyxospores with three or more valves). The latter order contains the family Kudoidae, and its single genus,

Kudoa. Lom and Dykova (1992) describe the order Multivalvulida as containing those species having myxospores of radial symmetry composed of three to seven valves meeting in three to seven sutures. Another important characteristic of the order includes a polar capsule in each valve, situated at the spore's apex. The only exception is the genus

Unicapsula Davis, 1924, which has a single polar capsule and three spore valves.

However. the degenerative remnants of two additional polar capsules are found in the region of the spore apex.

Egusa (1 986) prepared a comprehensive review of the order Multivalvulida, which included brief descriptions of several of the known multivalvulid species.

According to Lom and Dykova ( i 992), six genera are contained within the order

Multivalvulida and include: Trilospora Noble, 1939: Unicupsula Davis, 1924; Kudoa

Megl itsc 11. 1 947: Penfacupsuiu Naidenova and Zai ka. 1 970; kacapsula Arai and

Matsumoto, 1953; and Septerncapsuiu Hsieh and Chen, 1984. Recent evidence has

determined that members of the class Actinosporea are in fact alternate stages in the life

cycle of fieshwater myxosporeans, which led Kent et ai. (1 994a) to suppress the class

Actinosporea in favor of the more senior class Myxosporea. With the suppression of the

cIass Actinosporea, the genus Terractinomyxon Uceda, 1 9 12 was transferred to the

myxosporean order Multivalvulida. With the establishment of the genus Tefracapsula by

Canning et al. (1 996), there is a total of eight genera within the order.

The original members of the genus Kudoa were first included in the genus

Chloromyxum, based on the comparable number of polar capsules. Meglitsch (1947)

recognized that some species of Chloromyxum had myxospores with four valves, which

was contrary to the genus description. As a result, Meglitsch (1947) transferred those

species of Chloromymrrn with four myxospore valves to the newly established genus

Kudoa (type species: K. clupeidae (Hahn, 19 17)). The onginai generic diagnosis, as provided by Meglitsch (1 947), characterizes members of Kudoa as histozoic, and typically parasites of the skeletal musculature of fish. Kudoa species are also characterized by having quadrate or stellate myxospores with delicate membranes and indistinct sutures between the four valves; each valve possessing one polar capsule.

Figures 3.2 and 3.3 show examples of the variations of myxospore morphology. Another important feature of the genus is that the myxospore has two uninucleate sporoplasms, in PLATE VI

Figure 3.2. Wet mount (a-e) and stained (f) preparations of myxospores of various Kudoa

spp. demonstrating the variation in myxospore rnorphology.

a) Apical view of quadrate myxospore of Kudoa infestinalis. Bar = 1 pm

(fiom Maeno et al., 1993).

b) Lateral view of K. intesfinalis myxospore. Bar = 1 pm (fiom Maeno et al.,

1993).

c) Myxospores of Kudoa miniauricuiata showing upIifted tips of valve

tennini. Bar = 5 Pm.

d) Quadrate myxospores of Kudoapanifarmis. Bar = 10 Pm.

e) Stellate myxospores of Kudoa fhyrsires showing unequal size of polar

capsules. Bar = 10 Pm.

f) Stained preparation of K. thyrsites myxospores (Giemsa). Bar = 10 pm

(courtesy of S. Hallett).

PLATE VI1

Figure 3.3. SEM preparations of Kudoa spp.

a) Lateral view of Kudoa intestinalis myxospore showing apical projections.

Bar = 1 pm (fkom Maeno et al., 1993).

b) Apical and posterior views of Kudoa rniniauriculata myxospores showing

uplified tips. Note suture lines dividing myxospore into four valves and the

presence of apical projections. Bar = 10 Pm.

c) Apical view of Kudoapaniformis myxospore, Bar = 10 Pm.

d) Postenor view of Kudoa thyrsifes myxospore. Bar = 10 Pm. which one sporoplasm envelops the other (Stehr 1986; Lom and Dykova 1988).

The genus Kudoa is currently comprised of 44 described species distributed throughout marine and estuarine fishes worldwide. Whereas these myxosporeans are primarit y histozoic parasites of the musculature (i.e.. the parasite sporulates within muscle). a few coelozoic species have been described. These coelozoic species should be examined in greater detail to determine if indeed they should belong to the genus Kudoa.

3. Host and geographic distribution

Al1 known species of Kudoa infect marine and estuarine fishes. Egusa (1986) discussed the distribution of mu1 tivalvulid myxosporeans, and commented on the report by Lom et al. (1983) of an unidentified species of Kudoa in rainbow smelt (Osmerus mordar) collected from a freshwater lake in Canada. Scott and Scott (i988), however. describe rainbow smelt as an anadromous fish species, so this report of a true fieshwater species of Kudoa needs corroboration.

Host specificity varies greatly within the class Myxosporea, but most are genus or species specific (Shulman 1966). This range of host specificity is exemplified within the genus Kudoa. For example, several species of Kudoa exhibit a hi& degree of host specificity and have ody been recovered fiom the type host (e-g., K atropi, K. hanchiata). in contrast, K. thyrsites has a worldwide distribution and has been reported from more than 20 fish species. However, there is a possibility that K. ~hyrsiresis actually an assemblage of several morphologically indistinguishabie species, rather than one species infecting many fishes around the world. Within Canadian waters alone,

McDonald and Margolis (1995) report K. thyrsires in 1 1 fish species. At present, the phylogenetic data available for multivaivulid myxosporeans are incomplete. Hervio et ai. (1997) were the first to obtain sequences fiom the SSU rDNA of rnultivalvulid myxosporeans. Sequences were detemined for three eastem Pacific species of Krtdoa (K. miniauriculata, K. panilormis, and K. rhyrsites) and one western Pacific species (K. amamiensis). Construction of a phylogenetic tree resulted in a monophyletic grouping of the Kudoa spp. The sequences fkom Kudoa spp. fiom the eastern Pacific were found to be approximately 97% identicai, while the lone western Pacific species was only

9 1 % identical to the eastern Pacific species.

Using SSU rDNA sequence, Hervio et al. (1997)reported that K. thyrsites collected fkom three fish hosts (Pacific hake, Atlantic salmon, and tube-snout

(.4rrlorhynchusjlavidus))off the coast of British Columbia appear to be the same species.

Shaw et al. (1 997) investigated K. thyrsites infections in tube-snout; a fish that is commonly reported near commercial seawater netpens of British Columbia salmon fms.

Using examinations of myxospore morphotogy and DNA sequencing, they also conchded that the Masp. infecting tube-snouts was the same species of Kudoa that adversely affects the local aquaculture industry. Furthemore, the SSU rDNA sequence of

K. thyrsites fiom snoek (Thyrsites atun) collected off the coast of South Afnca is essentially identicai (>99%) to the original Atlantic salmon K. rhyrsites isolate collected in British Columbia (R.H. Devlin and J. Khattra, Department of Fisheries and Oceans,

West Vancouver Labratory, pers. comrn.).

4. Developmenf of Kudoa

Lom and Dykova (1992) described the developrnent of the trophozoite stage of myxosporeans within the fish host. During development, there may be a proliferative

phase in tissues or organs different fiom the final site (extrasporogonic phase), which is

independent of the sporogonic phase. This proliferative phase has ken shown to exist in

several genera of myxosporeans, but has yet to be described for Kudoa species. Moran et

al. (1 999b) deterrnined that K. rhyrsites does produce extrasporogonic stages that

circulate throughout the blood and that these stages are transmissible by injecting the

blood of an infected saimon into the intraperitoneal cavity of a naive salmon. in these

experiments, K. rhyrsifes infections were successfully transmitted to 2 of 23 Atlantic

salmon.

Once the parasite becomes established within the muscle fiber, the plasmodium

apparently does not undergo division, but instead grows to an enormous size, replete with

developing myxospores. Within histozoic species, nutrition is achieved by means of

pinocytotic activity across the host-parasite interface (Lom and Dykova 1992) (Fig. 3.4).

Within the genus Kudoa, the process of sporogenesis has been investigated using

transmission electron microscopy for K. lunata (cf. Lom and Dykova 1988) and for K. paniformis (cf. Stehr 1986). Lom and Dykova ( 1988) discussed the formation of large

polysponc plasmodia without the production of pansporoblasts in K. lunafa, K- paniformis, K. thyrsites and an unidentified Kudoa species. Even with Kudoo species that

form srnall trophozoites producing up to eight myxospores, there is no formation of a

pansporoblast. The lack of pansporoblast formation is likely consistent for ail Kudoa

species.

Moran et al. (1998) followed the development of K. thyrsires in Atlantic saimon

that had been naturally exposed to the unknown infective stage of the parasite in seawater PLATE VI11

Figure 3.3. TEM preparation of Paci fic hake (Merhccius producfus) infec ted with Kudoa

paniformis (m - host musculature; s - developing myxospore). Arrow indicates

host-parasite interface. Bar = 5 Pm. tanks. Vegetative stages were detected in the cardiac musculature at 4 months post- exposure (p-e.) and in the somatic musculature by 5.5 months p-e. Fully developed myxospores were observed in 64% of the Atlantic salmon examined at 6 months p.e. A subsequent experiment used both histology and the PCR test developed by Hervio et al.

(1 997) to detect early infections in various tissues and organs (Moran et al. 1998). Tissue samples of al1 major organs were examined using histological sections and samples of blood, gill, muscle, skin, and intestine were screened using the PCR test. Muscle infections were not detected by histology until 9 weeks p.e. at which time 1 of 25 (4%) of the Atlantic salmon was positive for infection. However. using PCR, 8 of 10 (80%) of the

Atlantic salmon were identified as positive for infection as early as 6 weeks p.e. and 7 of

10 (70%) were positive at 9 weeks p.e. Moran et al. (1 999b) also investigated the possibility of transmission of K thyrsites infections between fish. Mature myxospores were intubated into the stomach of Atlantic salmon held in fieshwater tanks. When no infections were detected after 6 rnonths p.e., it was concluded that direct transmission by fresh myxospores does not occur in K. thyrsites.

S. Host-parasite interactions

Inflammation is recognized as the principal defence mechanism utilized by fish to combat myxosporean infections (Lom and Dykova 1992). However, this response is usually not observed until the parasite achieves sporogony and the plasmodium contains mature myxospores. It is possible that later developmental stages express an increased antigenicity (Lom et al. 1983), and only then elicit the host's response. Once inflammation occurs, myxospores are ingested by macrophages and subsequently PLATE IX

Figure 3.5. Inflammatory responses of fish hosts to infections by Kudoa spp. (f -

fibroblast layer; m - host musculature).

a) Final stages of inflammatory response to Kudoa sp. in Pacific hake

(Merluccius productus). Bar = 250 Fm.

b) TEM preparation of host response to Kudoa sp. in Pacific hake. Arrow

indicates polar capsule of degenerated myxospore. Bar = 1 Pm.

c) Inflanunatory response of Atlantic salmon to Kudoa thyrsiies infection.

Bar = 100 Pm.

d) Increased magnifcation of the inflarnmatory reponse in Atlantic salmon

to K. rhyrsites idection. Arrow indicates myxospore within a macrophage.

Bar = IO Pm.

e) inflammatory response in lingcod (Ophiodon elongatus) infected with K.

rhyrsifes. Bar = 100 Pm. destroyed (Fig. 3 -5). The inflammatory response may be followed by complete repair of the af5ected site (Fig. 3.Sa). Ultrastmctud examinations of Kudoa spp. by Lom et al.

(1 983) and Voelker et al. (1 978) detemined that the host-parasite interaction is species- specific and is dependent upon the developmental stage of the parasite.

Histozoic phsrnodia (e.g., Kudoa and Henneguya species) ofien elicit the formation of an envelope either fiom the host's comective tissue or as the compression of adjacent host cells. This does not typically occur until the plasmodium is full of mature myxospores. For several species of histozoic myxosporeans that reside within muscle fibers. a "pseudocyst" is developed (Fig. 3.6). Lom and Dykova (1995) define the term

"pseudocystt as a cyst-like formation, or a cavity-like lesion surrounded by a dense fibrous capsule. They compared the pseudocyst with a typical cyst, by specifjing that the pseudocyst lacks an inner lining formed by the parasite, its formation follows necrotic changes. and the pseudocyst wall is formed by reparative inflammation. Anderson (1985) used the term pseudocyst in reference to those species that localize within the host's muscle fiber and lack a host ce11 envelope. h the context of this discussion, the definition of pseudocyst as used by Anderson (1 985) is more appropriate- In intracellular stages of

Kudoa spp., the oniy boundary between the early developmental stages and the muscle ce11 is the parasite's ce11 membrane (Lom et ai. 1983) (Fig. 3.4).

Morado and Sparks (1 986), using light rnicroscopy to investigate the host response of Pacific hake to K. thyrsites and K. paniformis infections, found that the host was unable to detect the parasites developing within the muscle fibers and therefore, did not initiate a host response. Only when the muscle fiber was replaced fully by the developing plasmodium was there an inflammatory response. This response was PLATE X

Figure 3.6. Histological preparations of Kudoa spp. in various fish species (H&E).

a) Kudoa pantfiormis in Pacific hake. intact pseudocyst pnor to host response

(white arrow). Parasite after host response was initiated (black arrow). Bar =

100 Fm.

b) Co-infect ion of Kudoa thyrsites (white arrow) and Kudoa paniformis (Mac k

mow) in Pacific hake. Bar = IO0 Pm-

c) Kudoa thyrsites pseudocyst in Atlantic salmon (XS). Bar = 10 Pm.

d) Early K- thyrsites infection in Atlantic salmon (LS). Bar = 10 Pm.

e) Kudoa thyrsires pseudocyst in lingcod (Ophiodon elongatus). Bar = 100

Cim.

f) Kudoa thyrsites in cardiac muscle of coho salmon (Oncorhynchus kisutch).

Bar = 1 O pm.

g) Kudoa ciliatae in intestinal musculature of Indo-Pacific whiting (Sillago

sp.). Bar = 100 Fm (courtesy of S. Hallett).

h) Unidentified Kudoa infection in lingcod (0.elongatus). Bar = 100 Pm. characterized by phagocytic infiltration, followed by granuloma and capsule formation

(Morado and Sparks 1986). This delay in host response was also observed using transmission electron microscopy (Stehr and Whitaker 1986). Host response, observed as encapsulation of the parasite by fibroblasts, occurred afier the parasite developed to the point where the plasmodium filled the muscle fiber. Originally, it was believed that the dafkening of infected muscle fibers was the result of the deposition of melanin (Kabata and Whitaker 198 1 ; Tsuyuki et al. 1982). However. Stehr and Whitaker (1986) used ultrastructuraI evidence to detemine that melanin was not present and concluded that the cyst's dark appearance was probably the result of an accumulation of breakdown products

(Fig. 3.5b).

Host response to K. thyrsites infections in Atlantic salmon, on the other hand, is characterized by multifocal, chronic inflammation between muscle fibers (Moran et al.

1998) (Fig. 3 .Sc). There was no inflarnrnatory response observed directIy related to muscle fibers that contained intact pseudocysts. However, myxospores that had been released fiom ruptured pseudocysts were actively engulfed by macrophages (Fig. 3.5d).

We have observed a comparable host response in lingcod (Ophiodon elongatus) infected with K. rhyrsires (Fig. 3.5e).

6. Post-morîem myoliquefrrfive autolysis

Post-mortem myoliquefactive autolysis, cornmoniy referred to as 'sofi flesh7, is a phenornenon that is associated with infections by Kudoa spp. as well as with other multivalvulids (Moser and Kent 1994; Lom and Dykova 1995). Infections by Unicapsula seriolae in yellowtail (Seriola lalandi) fiom Australia, Unicapsula muscularis in Pacific hali but (Hippoglossus srenolepis). and Hexacapsula neorhunni in Neorhunnus rnacroprerus fiom Japan result in enzymatic degradation of host musculature. However, the myxosporean genus most recognized for its detrimentai effect on muscle texture and therefore product quality is Kudoa. This sofi flesh phenornenon has been associated with infections of K. clupeidae in Atlantic hemng Kudoa crucijormum in Japanese seaperch

(Lateolabrax japonicus), Kudoa funduli in mwnmichog (Fundulus hereroclirus), Kudoa hisrolytica in Atlantic mackerel (Scomber scornbrus), Kudoa mirabilis in ribbonfish

(Trichiurus haume lu), K. musculoliquefaciens in swordfish K. paniformis in Pacific hake. Kudoa peruvianus in Chi lean hake (1Merluccius gayi), and K. thyrsites in several fish species (e-g., Atlantic and coho salmon). Those of concern to the international commercial fishery are K. clupeidae, K. histoiytica, K. musculoliquefaciens, K. panrormis, and K. peruvianus. Presently. only K. rhyrsites appears to have a detrimental effect on the aquaculture industry as the result of its association with sofi flesh in cultured

Atlantic (Figs. 3.1 b, c) and coho salmon (Whitaker and Kent 199 1; Whitaker et al. 1 994).

Kudoa amamiensis is also considered an important parasite of culhued buri; however, the effect on the product is the result of unsightly cysts in the musculature, as opposed to post-rnortem myoliquefaction (Fig. 3.1 e).

Egusa (1986) summarized the effect of Kudoa infections on living fish, and found that even though there were localized pathologicai changes, there was no observable effect on physiology, behavior, or Iife span of the fish host. Lom and Dykova (1 992) briefly mention a single exception in which enzymatic degradation before death resulted fiom an infection by an unidentified species of Kudoa in a live Karnchatka flounder

(Afheresrhesevermannr') fiom the Bering Sea. With post-mortem changes, there may be an accumulation or diffusion of enzymes that cause the odorless softening. Willis (1949)

believed that &er the death of the fish, enzymes produced by the parasite accumulate due

to the cessation of the blood circulation. in contrast, Patashnik et al. (1982) described the

localization of the enzymes within the pseudocyst, therefore resulting in only a localized

efkct by enzyme activity. Patashnik et al. (1 982) concluded that the proteolytic enzymes

responsible for muscle degradation diffuse out of the pseudocyst after the death of the

host and are distributed to other sites as host cells break down. These proteolytic enzymes

appear not to be associated with the myxospores, but are instead produced by the non-

sporous components of the plasmodium (Stehr and Whitaker 1986).

Krrdoa paniformis infections in Pacific hake and the associated enzymatic

degradation have been well docurnented over the past two decades (Kabata and Whitaker

198 1). Tsuyuki et al. (1982) investigated the effect of Kudoa infections (K. paniformis

and K. thyrsites) on the muscle texture of Pacific hake collected within the Strait of

Georgia and off the West Coast of Vancouver Island, British Columbia. They found that in

Pacific hake infected with K. thyrsites, proteolytic activity within the acid range increased

significantly; however, the enzyme was described as heat labile and not responsible for

the degradation of structural proteins. Therefore, K. thyrsites infections in Pacific hake

did not result in unacceptable sottness. Unacceptable muscle texture of Pacific hake due

to proteolytic activity within the neutral pH range was found associated with the K. panformis infections. The enzyme identified as responsible for this degradation was

found to be heat stable with optimal activity between the temperatures of 55-60°C.

Within the aquaculture industry, K. rhyrsites infections of Atlantic saimon and

coho salmon are recognized as a significant problem due to decreased marketability of infected fillets within a few days after harvest. This association of parasite and soft flesh

has resulted in a great deal of concern within the salmon farming industry worldwide due

to the parasite's ptential effect on the market's perception of fmed saimon. These

infections are apparently not associated with mortality, with the exception of the report by

Harrell and Scott (1 985). However, subsequent analysis of data has indicated that the K.

thyrsites infections were not responsible for the mortalities as originally believed (T.M.

Scott to D.J. Whitaker. pers. comm,).

Stressors affect the general health of fish (Schreck 1996). Under stressfiil

conditions such as seawater entry and maturation, high prevalences of K. rhyrsires

infections have been observed in Atlantic salmon (St-Hilaire et al. 1998; Moran and Kent

1999). The prevalence of infection reaches levels above 60% in pst-smolt Atlantic

salmon, probably the result of the stress associated with their initial transfer to seawater

and physiological changes while undergoing srnoltification. The majority of Atlantic

salmon smolts recover fiom infections (Moran and Kent 1999). However, we have seen

the prevalence of infection exceed 40% in a population of immature market-size Atlantic

salmon that had been maintained on a subsistence diet, suggesting that fish subjected to

suboptimal rearing conditions are more prone to infection (unpublished data). Blazer

( 1991 ) investigated the effect of nutrition on fish immune systems, and determined that poorly fed fish are often immunocompromised, which may explain this observation. St-

Hilaire et al. (1998) exarnined the prevalence of K. rhyrsiies infections in farmed Atlantic salmon in several harvests during the late winter and early spring of 1995. infections were relativeiy low in immature salmon; however, there was a sharp increase in prevalence (up to 70%) in sexually mature salmon and reconditioned grilses. They suggested that to reduce the prevalence of infections within the premium quality saimon sent to market, sexualIy reconditioned fish should be removed fiom the population before they are harvested (St-Hilaire et al. 1998).

The intensity of infection is positively correlated with the severity of soft flesh in

Atlantic salmon (St-Hilaire et al. 1997a). Lightly infected salmon (<20 000 myxospores per gram of host muscle) showed no observable signs of K. thyrsites infection, whereas heavy infections always resulted in unacceptable flesh quality. A similar correlation was observed between the intensity of K. panz~ormisinfections in Pacific hake and flesh quality (Kudo et al. 1987).

In addition to soft flesh. Kudoa infections have a detrimental impact on product quality due to the formation of macroscopic, wightly cysts in the musculature. This is a problem of cultured buri in Japan infected with K. amamiensis (Egusa 1983) (Fig. 3. le) and pseudocysts of K. thyrsites are readily observed in smoked Atlantic (Fig. 3.1 b) and coho salmon products.

7. Detection methods and contrd strategies

Sakanari (1 994) briefly mentioned four potential techniques that may be used either to detennine the prevalence of K. thyrsites infections within netpen Atlantic salmon populations or to estimate the intensity of infection in an individual host. These methods involved gross examination, rnicroscopic examination, PCR tests, and immunological techniques. 7.1. Gross examination

Gross examination for K. rhyrsires is impracticai due to the inability to observe

the infection macroscopically in fiesh Atlantic salmon. In contras, for Kudoa species that

produce larger visible cysts (e-g., K. amamiensis in buri (Fig. 3. le)), gross examination

allows the removal of heavily infected fish prior to king sent to market.

7.2. Microscopic examination

Historically, the detection of Kudoa infections in fish has ken by the exarnination

of wet mount preparations to detect the myxospore stage (Fig. 3.2). An experimental

protocol for wet mount screening for K thyrsites in Atlantic salmon is provided by St-

Hilaire et al. (1997b). The successfbl detection of infection requires that the parasite has

sporulated and contains the easily recognized, species-specific myxospore stage. A

priority of the salmon fming industry is the development of a rapid and inexpensive

method to detect K. thyrsifes infections in Atlantic salrnon shortly afier harvest, which

does not have a detrimental effect on the product. St-Hilaire et al. (1 99%) evaluated a

non-lethal sampling technique in which a sample of the hyohyoideus ventrulis muscle was removed fiom the opercullm and prepared for standard wet mount exarnination. They concluded the test was a reliable indicator of the presence or absence of infection within the Atlantic salmon fillets. The test was found to be extremely sensitive (93%) and highly specific (>94%) when used to detect heavily infected fish (St-Hilaire et al. 1997b).

Sensitivity decreased to 79% when light infections were included. This procedure not only leaves no observable sign of sampling damage on the salmon product, but also permits paired sampling over time for statistical analyses as it is a non-lethal technique. Unfortunately, this technique is time consuming and labor-intensive, and thus is only

practicd as a research tool.

7.3. PCR assays

As discussed previously, Hervio et al. (1997) developed PCR tests that are highly

specific and sensitive for some Kudoa species. The PCR test for K. thyrsites showed a

greater sensitivity for detecting infections in host musculature than standard wet mount

techniques (92% vs. 38%, respectively). This is possibly a result of the test's ability to

detect al1 stages of the parasite, whereas only rnyxospores are consistently detectable in

wet mounts. The primer pair developed specifically for K. thyrsites amplified a 909-bp

region of K. thyrsites SSU rDNA and did not ampli@ the SSU rDNA region of other

my?tosporeans (e-g., K. amamiensis, K. rniniauriculata, K. paniformis, H. sahinicola,

and the PKX myxosporean) or microsporidians (e-g., L. salmonae and N- salmonis) that

were tested for cross-reactivity (Hervio et al. 1997). Sakanari (1 994) raised specific

concems regarding the requirement for trained personnel, expensive equipment, and the

delay in achieving PCR results. Therefore, this would be very useful for research

purposes or pre-screening a netpen population pior to harvest, but would probably be

impractical for use as a rapid test at processing plants.

7.4. Immunological techniques

A possible screening technique may involve the detection of parasite antigens

using immunological methods (e.g., a dipstick test). An immunological test would provide the aquaculture industry with an inexpensive, rapid and quantitative (Le., related to intensity of infection) method of detecting K. thyrsites in processed Atlantic salrnon.

The ability of a test to quanti@ K. thyrsites infections is critical as light infections do not cause soft flesh. Such a test is yet to be developed.

7.5. Control strategies

Shulman (1 966) discussed three basic methods for controlling diseases induced by m yxos poreans: decrease the number of infective stages wi thin the environment; control the development of the vegetative stages within the fish host; and proper fish husbandry.

Due to the nature of myxosporean infections in the marine environment, attempting to control a parasite by interrupting its life cycle is probably impractical. Hofian and Putz

( 1969) have demonstrated that myxosporean spores have a high level of tolerance to environmental extremes (e-g., extended periods of fieezing). Avoiding exposwe to the infection in marine netpens at this point appears impossible. However, if the alternate host for the parasite occurs within the biofouling fiequently associated with netpens, then frequent changing of the nets may reduce accumulation of potential alternate hosts; therefore, decreasing the risk of exposure to the infective stage of the parasite.

Moran and Kent (1999) demonstrated that there is a seasonality in the presence of the infective stage of K. thyrsites. The wiknown infective stage is most common during the summer months, as indicated in expenments that exposed Atlantic salmon for various

8-week periods throughout the year. Infections were readily contracted through the summer and fdl months, and were absent during winter and early spring. However, considering that salmon are reared more than a year in seawater, the tirne of transfer to seawater probably would not affect the prevalence of the infection in market-size fish. At present, there are no approved chemotherapeutic treatments for myxosporean

infections in fish to be sold as human food. However, Fumagillin DCH has proven effective against some saimon diseases induced by myxosporean infections such as whirling disease and proliferative kidney disease (Hedrick et al. 1988; Wishkovsky et al.

1 990; El-Matbouli and Hoffinann 1991 ; Higgins and Kent 1996). The efficacy of this dmg and related products (e.g., the Fumagillin analog TNP-470)for Kudoa infections should be investigated.

Handling procedures of the product afler harvest may reduce the effects of sofi flesh. Porter et ai. (1 993) compared the eficacy of potato extract, egg white, and bovine plasma protein as inhibitors of the protease enzymes found in infected Pacific hake and mowtooth flounder (Atheresthes stomias). They concluded that al1 were effective at inhibiting the proteases responsible for the degradation of the Pacific hake flesh; however. the potato extract was a more effective inhibitor. in the arrowtooth flounder, bovine plasma was the most effective inhibitor of enzyme activity, although al1 three showed greater than 85% inhibition. Presently, potato extract has the greatest potential for use as an inhibitor of protease activity in surimi. as a result of the high cost of egg white and the questionable suitability of bovine serum as an ingredient for surimi. Enzyme inhibitors may contain specific proteins that compte for the enzyme's active site, which inactivates it (Laskowski and Kato 1980). Unfortunately, the procedure requires that the product be minced to expose the proteolytic enzyme to the inhibitors. This is not practical for the fiesh salmon market as salmon are usually sold in the round.

Currently, the best strategy for the aquaculture industry to avoid soft flesh problems resulting from K. thyrsites infections is to remove Atlantic saimon that have undergone maturation prior to harvesting. Al1 grilse should be sorted fiom the harvest population pior to slaughter while at the fmas it is difficult to differentiate a fuily reconditioned salmon fiom an immature salmon afier the viscera are removed, which has lead processors to inadvertently misclassiQ the former as premium fish (St-Hilaire et al.

1998).

8. Synopsis of descriôed Kudoa speck

The majority of the Kudoa species descriptions were published prior to the guide for myxosporean species descriptions by Lom and Arthur (1989).As a result, there is much confiision with regards to the measurements of myxospore dimensions. For this reason, myxospore dimensions are not included as part of this synopsis. For myxospore dimensions and schematics of myxospore morphology, see the original species descriptions and reviews by Egusa (1986) and Lom and Dykova (1992).

8.1. Kudoa species infecting the somatic musculature:

Kudoa alliaria Shulman and Kovaleva, in Kovaleva et al., 1979

Detrimental effects: Forms macroscopic pseudocysts.

Description: Myxospores quadrate in apical view. Polar capsules pyrifonn, equal.

Pseudocysts up to 6 x 1 mm in Notothenia spp. and up to 10 x 5 mm in southem blue whiting (Micromesistius australis).

Hosts: Southem blue whiting (M. australis); Notothenia spp. (N.conina and N. ramzay);

Argentine straptail (Macruronus magellanicus).

Locality: Coast of Argentina (western South Atlantic Ocean). Remarks: Zawistowski et al. (1986a) investigated the effect of K alliaria Uifections on the metabolic processes of blue whiting skeletal muscles and concluded that no detectable change in these processes occurred. Attempts to transmit the parasite to rodents were unsuccessful (Zawistowski et al. 1986b; Romano et al. 1988).

References: Egusa (1986); Kovaleva et al. (1 979); Romano et al. (1 988); Zawistowski et al. (1986a, b).

Kudoa amamiensis Egusa and Nakajima, 1980

Detrimenta1 effects: Fonns macroscopic cysts (Fig. 3.1 e).

Description: Myxospores quacirate in apical view, with rounded edges. Valves with anterior projections and several papillae. Polar capsules pyriform, equal. Cysts spherical to ellipsoidal. opaque' crearny-white, up to 5.0 mm in length.

Hosts: Buri (Seriola quinqueradiata); damselfishes (Abudefduf spp., Chrornis spp., and

Chrysiptera s p.).

Localities: Coasts of Okinawa and Amami-Ohshima, Japan (western North Pacific

Ocean).

Rernarks: In addition to the somatic musculature, the cysts were also obsewed in the heart, serosa, skin, and fins of heavily infected buri. Host response involving fibrous connective tissue was O bserved surrounding the cyst. The parasi te reduces the market value of the fish product as unsightly cysts are distributed throughout the skeletal muscle

(Egusa and Nakajima 1978). Egusa and Nakajima (1 978) and Egusa (1983) referred to K amamiensis infections in buri as muscular kudoasis.

References: Egusa (1 983); Egusa (1986); Egusa md Nakajirna (1 978); Egusa and Nakajima ( 1980); Hervio et al. ( 1997); Lom and Dykova ( 1992); Nakajima (1 984).

Kudoa bengalensis Sarkar and Manimder, 1983

Detrimentai effects: None reported.

Description: Myxospores stellate in apical view, with bluntly pointed edges, no ornamentation. Polar capsules nibular, equal.

Hos t : Tachysztrus plaiy-srornus.

Locality: Bay of Bengal (Indian Ocean).

Remarks: Deveiopmental stages were not observed.

References: Egusa (1 986); Sarkar and Mazumder (1 983).

Krrdoa bora (Fujita, 1930)

Detrimental effects: Forms macroscopic cysts.

Description: Myxospores sphencal in apical view. Polar capsules club-shaped, equal.

Cysts spherical to oblong, up to 2 mm in length.

Hos ts : Mu1 lets (Mugil carinatus, M. cephalus, M. japonica).

Locality: Coast of Taiwan (western North Pacific Ocean).

Remarks: This species was originally described as belonging to the genus Chloromyxum.

There have been no other reports of this species since its description.

References: Egusa (1 986); Fujita (1 930). Kudoa caudata Kovaleva and Gaevskaya, 1983

Detrimental effects: None reported.

Description: Myxospores quadrate in apical view.

Hos t : Chub mackerel (Scomber japonicus).

Locality: Coast of Peru (eastern South Pacific Ocean).

Remarks: Myxospores are characterized by having stripes on the lower lateral surface of each valve-

References: Egusa (2986); Kovaleva and Gaevskaya (1983).

Krrdoa clrrpeidae (Hahn, 19 1 7)

Detrimental effects: Foms macroscopic pseudocysts. Heavy infections may cause mortality in young Atlantic herring (Lom and Dykova 1992).

Description: Myxospores quadrate in apical view. Valves rounded, and with apical projections. Trophozoites spindle-shaped, up to 5 mm.

Hosts: Alewife (Alosa pseudoharengur), Atlantic herring (Clupea harengus), Atlantic menhaden (Brevoortia tyrannus), blueback hemng (Alosa aestivalis), hickory shad (Alosa mediocris), and ocean pout (~Macrozoarcesamericanus).

Locality: western North Atlantic Ocean

Remarks: Type species of Kudoa. This species was originally described as Chloromyxum clicpeidae. Meglitsch (1 947) prepared a revised description of this species fiom parasites

CO 1lected fiom Atlantic menhaden (Brevoortia îyrannus), and subsequently tram ferred the species to the genus Kudoa.

References: Di Antonio and Cenci Goga (1 993); Egusa (1 986); Hahn (1 9 17); Kovaleva et al. (1 979); Lom and Dykova (1992); McDonald and Margolis (1 995); Meglitsch

(1 947); Nigrelli (1946); Shuhan (I 966); Sindennann (1 959).

Kudoa crzrciformum (Matsumoto, 1954)

Detrimental effects: Post-mortem myoliquefaction.

Description: Myxospores stellate in apical view. Valves subequal, no ornamentation

Polar capsules ovoid, unequal.

Host: Japanese seaperch (Lateolabrmjaponicus)

Locality: Coast of Japan (western North Pacific Ocean).

Remarks: This species was ongindly described as NeochZoromyxum cruc~ormum.

References: Egusa ( 1986); Matsumoto (1 954).

Krrdoa crtrmena Iversen and Van Meter, 1967

Detrimental effects: Forrns macroscopic cysts.

Description: Myxospores quadrate in apical view, no omamentaûon. Polar capsules equal. Cysts whitish, polysporic, ellipsoidal, up to 1.7 x 2.6 mm.

Hos t : Spanish mackerel (Scomberomorus macularus).

Locality: Coast of Flonda, USA (eastem North Atlantic Ocean).

Remarks: Cysts are situated within the connective tissue Layer surrounding the muscle fibers. We have observed the infection in yellowfui tuna (Thunnus albacares) collected off North Carolina, USA.

References: Egusa (1 986); Iversen and Van Meter (1 967); Kovaleva et al. (1 979); Lom and Dykova (1 992). Kudoa cynoglossi ObiekeUe and Lick, 1994

Detrimental effects: Forms rnacroscopic cysts.

Description: Myxospores stellate in apical view, with apicd projections. Valve tips

uplified. Polar capsules pyriform, equal. Cysts spindle-shaped, whitish, up to 14 x 4 mm

in size.

Host : Senegalese tonguesole (Cynoglossus senegalensis) .

Locality: Gulf of Guinea, western Afiica (eastern North Atlantic Ocean).

Remarks: A maximum of 5 cysts was observed in a single host. Primarily infects fishes

less than 30 cm in length.

References: Obiekezie and Lick (1 994); Obiekezie et al. (1 987).

Kudoa funduli (Hahn, 19 15)

Detrimental effects: Post-mortem myoliquefaction.

Description: Myxospores quadrate in apical view, with apical projections. Polar capsules pyriform. Pseudocysts opaque, whitish and elongate, up to 3 x 0.5 mm.

Host: Mummichog (Fundulus heteroclitus).

Locality: Coasts of New Jersey and Massachusetts, USA (western North Atlantic Ocean).

Remarks: Originally described as Chloromyxumfinduli, and is found in both the musculature and fins.

References: Egusa ( 1 986); Hahn ( 1 9 1 5); Kovaleva et al. (1 979); Meglitsch (1 947, 1948). Kudoa histoiyiica (Pérard, 1928)

Detrimental effects: Post-mortem myoliquefaction.

Description: Myxospores stellate in apical view, no ornamentation. Polar capsules

pyriform, unequal.

Host : Atlantic mackerel (Scornber scombrus).

Localities: Bay of Biscay (eastern North Atlantic Ocean), Mediterranean Sea.

Remark: Originally described as Chioromyxurn histoiyticurn. However, with the

establishment of the genus Kudoo by Meglitsch (1947), and its subsequent transfer to this

genus, the species name was modified to agree in gender with the generic name. See

Article 34b of the International Code of Zoological Nomenclature.

References: Egusa (1 986); Kovaleva et al. (1979); Lom and Dykova (1992); Meglitsch

(1 947); Pérard (1 928).

Kudoa insolita Shulman and Kovaleva, in Kovaleva et al., 1979

Detrimental effects: None reported.

Description: Myxospores quadrate in apical view, no omamentation. Polar capsules

pyri forrn, equd.

Host: Greater ambejack (Seriola dumeriii).

Locality: Coast of Portugal (eastem North Atlantic Ocean).

Remarks: Original drawings show slightly elevated apex associated with some spores.

References: Egusa (1 986); Kovaleva et al. (1 979). Kudoa iwatai Egusa and Shiomitsu, 1983

Detrimental effects: Forms macroscopic cysts.

Description: Myxospores subquadrate in apical view, with small apical projections. Polar capsules spherical, subequal. Cysts whitish to crearny, spherical to ellipsoidal, up to 1.5 mm in size.

Hosts: Madai (Pagrus major) and Oplegnathuspunctatus.

Locality: Coast of Kyushu, f apan (western North Pacific Ocean).

Remarks: Cysts situated prirnarily between the muscle fibers, with a few in the subcutaneous and interrnuscuiar adipose tissues.

References: Egusa (1 986); Egusa and Shiomitsu (1983); Lom and Dykova (1992).

Kudoa kabatai Shdman and Kovaleva, in Kovaleva et al., 1979

Detrimental effects: Forms macroscopic pseudocysts.

Description: Myxospores stellate in apical view, with tips uplifled and apical projections.

Postero-lateral processes present. Polar capsules equal. Pseudocysts oblong, up to 3 mm in size.

Hos t : topknot (Zeugopterus punctatus).

Locality: North Sea (eastern North Atlantic Ocean).

Remarks: Kudoa kabatai was first observed by Kabata (1960),but was describecl as a new species by Shulman and Kovaleva, in Kovaleva et al. (1979).

References: Egusa (1 986);Kabata (1 960); Kovaleva et al. (1 979); Lom and Dykova

( 1 992). Kudoa leiostomi Dykovi, Lom. and Overstreet, 1994

Detrimental effects: None reported.

Description: Myxospores quadrate in apical view, with a slightly raised apex. Polar

capsules elongate, eqd.

Host: Spot (Leiostomus xanthurus).

Locality: Coast of Mississippi, USA (Gulf of Mexico, western North Atlantic Ocean).

References: Dykova et al. (1994).

Krrdoa lunara Lom, Dykovi, and Lhotakov& 1983

Detrimental effects: Forms macroscopic pseudocysts.

Description: Myxospores stellate in apical view, with blunt tips uplifted and apical

projections. Polar capsules equal. Pseudocysts off-white to yellowish, spindle-shaped and

up to 0.7 x 3 mm.

Host: scaldfishes (Arnoglossus imperialis, A. laterna, A. thori).

Locality: Mediterranean Sea.

Remarks: The prevalence of infection reached 100% in some samples.

References: Egusa (1 986); Lom and Dykova (1 988); Lom and Dykova (1992); Lom et aI.

( 1983).

Kudoa miniauriculara Whitaker, Kent, and Sakanari, 1996

Detrimental effects: Forms macroscopic pseudocysts.

Description: Myxospores stellate in apical view, with valve tips upfifted and apical projections (Figs. 3.2c, 3-3 b). Polar capsules pyriform, equai. Pseudocysts either sofi, creamy white or hard and yellow, up to 20 x 2 mm.

Hosts: Bocaccio (Sebastespaucispinis); greenstriped rockfish (Sebastes elongatus).

Locality: Coast of California, USA (eastem North Pacific Ocean).

Remarks: Previously identified as K. clupeidae by Moser et al. (1 976) and Heckmann and Jensen (1 978).

References: Heckmann and Jensen (1 978); Hervio et al. (1 997); Moser et al. (1976);

Whitaker et al. (1 996).

Kudoa rnirabilis Naidenova and Gaevskaya, 199 1

Detrimental effects: Forms macroscopic pseudocysts or pst-mortem myoliquefaction.

Description: Myxospores stellate in apical view. valves unequal. Polar capsules ovoid, unequal. Pseudocysts ovoid or spherical, up to 20 x 5 mm in size.

Host: Ribbonfish (Trichiurus haurnela).

Locality: Coast of Yemen (Red Sea).

References: Naidenova and Gaevskaya (1 99 1).

Kudoa mzrsculoliquefaciens (Matsumoto, 1954)

Detrimental effects: Post-mortem myoliquefaction.

Description: Myxospores quadrate in apical view, no ornamentation. Polar capsules subspherical to ovoid, equal. Pseudocysts up to 1.3 mm in diameter.

Host: Swordfish (Xiphias gladius).

Locality: Coast of Japan (western Pacific Ocean).

Rerna rks : Onginall y described as Chloromyxum musculoliquefaciens. References: Egusa (1 986); Lom and Dykova ( 1 992); Matsumoto ( 1 954).

Ktrdoa nova Naidenova, in Naidenova et al., 1975

Detrimental effects: Forms macroscopic pseudocysts.

Description: Myxospores quadrate in apical view. Polar capsules pyriform, equal.

Pseudocysts spindle-shaped, up to 7 mm.

Hosts : Blue fish (Pornotomus salfatrix).Denfex macrophfhalmus,gobies (Neogobius spp.,

Gobius spp.), Knipowifschia longicaudata, little tunny (Euthynnus alletterat us),

Pomatoschistus microps leopardinus, Proterirhinus mamaratus, scads and horse

rnac kerels ( Trachunrs spp.), Spanish seabream (Pagellus acarne), and Thunnus obesus.

Localities: Atlantic Ocean, Black Sea, Mediterranean Sea, and Sea of Azov.

Remarks: Several previous reports had identified infections by this species as infections

by K-quadratum and KIclupeidae.

References: Egusa (1 986); Gaevskaya and Kovaleva (198 1); Iskov (1 989); Lom and

Dykova (1 992); Kovaleva et al. (1979); Naidenova et al. (1975).

Kudoa paniformis Kabata and Whitaker, 198 1

Detrimental effects: Post-mortem myoliquefaction.

Description: Myxospores quadrate in apical view, with valve tips rounded and no omamentation (Figs. 3.2d, 3.3~).Polar capsules pyriform, subequal.

Hos t : Paci fic hake (Merluccius producrus).

Locality: Offshore waters of British Columbia, Canada (northeast Pacific Ocean).

Remarks: The pseudocysts, observed within the muscle fibers of the host, apparently evoke no host reaction involving connective tissue (Fig. 3.6a). Co-infections of offshore

Pacific hake by K. thyrsires and K. paniformis are relatively cornmon (Figs, 3.1 a, 3.6b).

References: Egusa ( 1 986); Escos et al. ( 1 995); Hervio et al. ( 1997); Kabata and Whitaker

( i 98 1, 1986); Kent et al. (1 994b); Kudo et al. (1987); Lom and Dykova (1992);

McDonaid and Margolis (1 995); Morado and Sparks (1 986); Patashnik et al. (1982);

Stehr (1 986); Stehr and Whitaker (1986); Tsuyuki et al. (1 982); Whitaker (1 986);

Whitaker et ai. (1 994).

Kudoa pemianus Mateo, 1972

Detrimental effects: Post-mortem myoliquefaction.

Description: Myxospores spherical in apical view. Polar capsules ovoid, equal.

Host: C hilean hake (Merluccius gayi).

Locality: Coast of Peru (eastem South Pacific Ocean).

Remarks: Erroneously referred to as K. hallado by Egusa (1986).

References: Egusa (1986); Mateo (1972).

Kudou quudratum (Thélohan, 1 895)

Detrimental effects: None reported.

Description: Myxospores quadrate in apical view. Polar capsules pyriform.

Hos ts : Shorthom scul pin (Myoxocephalus scorpius), pipefish (Syngnathus acus),

Euro pean horse mac kerel (Trachurus trachurus), Entelurus olguerus, Callionymus lyra, and Julis vulgaris.

Locality: eastern North Atlantic Ocean, Mediterranean and White seas. Rema rks : Original1y described as Chloromyxum quadratum.

References: Egusa (1986); Iskov (1 989); Kovaieva et al. (1 979); Lom and Dykova

( 1992); Meglitsch ( 1947); Shulman (1966); Thélohan (1895).

Kzrdoa rosenbuschi (Gelonnini. 1 944)

Detrimental effects: Post-mortem myoliquefaction.

Description: Myxospores quadrate in apicai view. Polar capsules equal.

Host: Argentine hake (Merluccius hubbsi).

Locality: Coast of Argentina (western South Atlantic Ocean).

Remarks: Originally described as Chloromyxum rosenbdi. The priority for this species is often erroneously identified as K. rosenbuschi (Gelonnini, 1943). However, the copyright date of the publication containing the original description is 1944; therefore, the correct priority is as listed above according to Article 22 of the international Code of

Zoological Nomenclature.

References: Egusa (1086); Gelormini (1943); Kovaleva et al. (1979); Meglitsch (1 947);

Romano et al. (1 988); Sardella (1 988a, b); Sardella and Roldiin (1 989); Sardella et al.

(1 987); Szidat (1 966).

Kudoa sciaenae Terin, Llicin, and Luque, 1990

Detrimentai effects: Foms macroscopic pseudocysts.

Description: Myxospores quadrate in apicai view, with rounded edges and no omamentation. Polar capsules pyrifonn.

Host: Sciaenid fishes (Paralonchurus peruanus, Sciaena fasciata and S. deliciosa, and Stellijèr minor).

Locality: Coast of Peni (eastem South Pacific Ocean).

References: Llich et al. (1991);Oliva et al. (1 992);Téran et al. (1990).

Kudoa shkae Dykovi, Lom, and Overstreet, 1994

Detrimental effects: None reported.

Description: Myxospores spherical in apical view. Polar capsules sphericai, equal.

Plasmodia spindle-shaped. polyspric, 0.2 x 0.06 mm in size.

Hos t: Hardhead catfish (Ariusfelis).

Locality: Coast of Mississippi, USA (Gulf of Mexico, western North Atlantic Ocean).

References: Dykova et al. ( 1 994).

Kudoa thyrsites (Gilchrist, f 924)

Detrimental effects: Post-mortem myoliquefaction (Figs. 3.1 b, c).

Description: Myxospores stellate in apical view, no ornamentation (Figs. 3.2e. f; 3.3d).

Polar capsules pyri form, unequal.

Hosts: Arrowtooth flounder (Aiheresthes stomias), Atlantic salmon (Salmo salar),

Aus tralian anchovy (Engraulis australis), Australian pilchard (Sardinops sagm

neopilchardus), blue sprat (Spratelloides delicatulus), cape hake (Meriuccius capensis),

Cypselurus sp., dolphin (Coryphaena hippurus), Dover sole (Microstomus paczficus),

Japanese anchovy (Engraulisjaponicus), Lepidopus caudatus, lingcod (Ophiodon

eiongatus), Paci fic hake (Merlucciusproducfus), Paci fic hali but (Hippoglossus stenolepis), Pacific salmon (Oncorhynchus spp. ), rock sole (Pleuronectes bilineafus), scaly mackerel (Sardinella lemuru), snoek (Thyrsites atun), South Afiican pilchard

(Sardinops saga ocellatus), thread fin scul pin (Icelinusfilamentosus), tube-snout

(Aulorhynchus jlavidus), walleye pollock ( Theragra chalcogramma), Zeus spp-

Localities: North and South Pacific Oceans, North and South Atlantic Oceans.

Remarks: Ot-iginally described as Chlorompwn thyrsites fiom the snoek off the Coast of

South Africa. The parasite is primarily a parasite of the somatic musculature (Figs. 3.6~- e), but it has been reported fiom the hem muscle of several of the wild Pacific salmon

(Oncorhynchus spp.) and cultured Atlantic (Salmo salar) and coho salmon

(Oncorhynchus kisutch) (Fig. 3.6f).

References: Baja and Toranzo (1993);Castro and Burgos (1996); Egusa (1986);Escos et al. (1995); Gilchrist (1924); Harrell and Scott (1985); Hervio et al. (1997);Holliman

( 1994): Kabata and Whitaker (198 1. 1986, 1989); Kabata et al. (1 986); Kent (1 992); Kent et al. (I994b);Kovaieva et al. (1979);Kudo et al. (1987);Langdon (1991);Langdon et al.

(1 992): Lom and Dykova (1992);Marsal ( 1994); McDonald and Margolis (1995);

Meglitsch (1947); Morado and Sparks (1 986); Moran and Kent (1999);Moran et al.

(1 998, 1999a, b); Patashnik et al. (1982); Shaw et al. (1997);Stehr and Whitaker (1986);

St-Hilaire et ai. (1997a, b; 1998); Tsuyuh et al. (1982);Whitaker (1986);Whitaker and

Kabata (1987); Whitaker and Kent (199 1 ): Whitaker et al. (1994); Willis (1949).

8.2. Kudoa species from sites other than the somatic musculature:

Kudoa atropi Sandeep, Kalavati, and Narasimhamurti, 1986

Detrimental effects: Forms macroscopic cysts.

Description: Myxospores quadrate in apical view, with deep notches, no omamentation. Polar capsules pyriform, equal. Cysts opaque, whitish, up to 1.O mm in diameter.

Site: Gills.

Host: Cleftbelly trevally (Afropusafropus).

Locality : Bay of Bengal (Indian Ocean).

Remarks: No apparent changes in host tissue were observed as a resdt of infection.

References: Sandeep et al. (1986).

Kudoa brunchiafa Joy, 1972

Detrimental effects: Forrns macroscopic cysts.

Description: Myxospores quadrate in apical view, with tips uplified. Polar capsules pyriform, equal. Cysts white, up to 1.1 x 0.4 mm.

Site: Gills.

Hos t : Spot (Leiostomus xanthurus).

Locality: Clear Lake, Texas, USA.

References: Egusa (1 986);Joy (1972); Kovaleva et al. (1 979); Lom and Dykova (1 992).

Kudoa cascasia Sarkar and C haudhury, 1996

Detrimental effects: None reported.

Description: Myxospores quadrate in apical view, no ornamentation. Polar capsules pyriform, equal. Cysts creamy, up to 1.5 mm in diameter.

Site: Intestinal mesentery.

Host: Sicamugil cascasia.

Locality: Bay of Bengal (Indian Ocean). References: Sarkar and Chaudhury ( 1996).

Kztdoa cerebralis Paperna and Zwemer, 1974

Detrimental effects: None reported.

Description: Myospores quadrate in apical view. Cysts up to 2.2 mm in diameter.

Site: Brain.

Hos t : Striped bass (Morone smatilis).

Locality: Chesapeake Bay, USA (western North Atlantic Ocean).

Remarks: Cysts are situated in the connective tissue associated with the brain and spinal cord.

References: Egusa ( 1986) Kovaleva et al. ( 1 979);Lom and Dykova ( 1992); Paperna and

Zwerner ( 1974).

Kudoa chilkaensis Tripathi, 1 953

Detrimental effects: None reported.

Description: Myxospores quadrate in apical view. Polar capsules pyrifonn, equd. Cysts up to 0.8 mm in diameter.

Site: Muscles and peritoneum in the esophageal region.

Hos t : Needlefish (Srrongyfurasfrongylura).

Locality: Chilka Lake, India.

Remarks: The priority for this species is often erroneously identified as K. chilkaensis

Tripathi, 195 1. However, the copyright date of the publication containing the original description is 1953; therefore, the correct priority is as listed above according to Article 22 of the International Code of Zoological Nomenclature.

References: Egusa (1 986); Tripathi (195 1 ).

Kudoa ciliatae Lom, Rohde, and Dykovi, 1992

Detrimental effects: Forms macroscopic cysts in the smooth muscle layer of the intestine

(Figs. 3.1 d, 3.6g).

Description: Myxospores quadrate in apical view. Polar capsules pyriform, subequal.

Plasmodia ellipsoidal, polysporic, up to 0.25 mm in diameter.

Site: Intestinal musculature.

Host: Sand sillago (Sillago ciliata).

Locality: Coast of Australia.

Remarks: Kudoa ciliatae infections are not associated with clinical disease.

References: HalIett et ai. (1997);Lom et al. (1992).

Kzrdoa eleotrisi Siau, 197 1

Detrimental effects: Forms macroscopic cysts.

Description: Myxospores quadrate in apicai view.

Site: Gills.

Host: EIeotris kribensis.

Locality: West Coast of Afnca (Gulf of Guinea , Atlantic Ocean).

Remarks: Sac-Iike membrane surrounds spores.

References: Egusa (1 986); Kovaleva et al. (1979); Siau (1971). Kudoa haridasae Sarkar and Ghosh, 1991

Detrimental effects: None reported.

Description: Myxospores stellate in apical view. with lateral inflations. Polar capsules pyri form. equal. Plasmodia not observed.

Site: Gallbladder.

Hos t : Mugil persina.

Locality: Bay of Bengal (Indian Ocean).

References: Sarkar and Ghosh (1 99 1 ).

Kudoa intestinalis Maeno, Nagasawa, and Sorimachi, 1993

Detrimental effects: Fonns macroscopic cysts in the intestinal musculature.

Description: Myxospores quadrate in apical view (Fig. 3.2a). Valves with rounded edges and apical projections (Figs. 3.2b, 3.3a). Polar capsules elliptical, equal- Plasmodia sphencal or ellipsoidai, opaque, whîtish, up to 0.5 mm in diameter.

Site: Intestinal musculature.

Host: Striped rnullet (Mugil cephalus).

Locality: Gokasho Bay, Japan (western North Pacific Ocean).

Remarks: No host response involving connective tissue was observed surrounding the plasmodia.

References: Maeno et ai. (1993). Kudoa pericurdialis Nakaj ima and Egusa, 1978

Detrimental effects: None reported.

Description: Myxospores quadrate in apical view. Polar capsules elongate. Cysts elliptical or oval, up to 2.7 x 1.2 mm.

Site: Pericardial cavity.

Host: Buri (Seriola quinqueradiafa).

Locality : Aj iro Bay, Japan (western North Pacific Ocean).

Remarks: Egusa (1 983) referred to the disease as pericardial kudoasis, and mentioned that the disease typically occurs among the bwi during the first year of culture. The encapsulated tmphozoites are not considered harmful to the fish as no significant difference in condition factor between infected and non-infected fish was observed

(Egusa 1983). The effect on the market price is not of concem as the parasite is localized within the pericardial cavity, and therefore, inconspicuous.

References: Egusa (1983, 1986); Lom and Dykova (1992);Nakajima and Egusa (1978).

Kudoa shiornitsui Egusa and Shiomitsu, 1983

Detrimental effects: None reported.

Description: Myxospores subquadrate in apical view, with rounded tips. Polar capsules pyriform. Cyst ellipsoidal, creamy-white, up to 3 x 0.6 mm.

Sites: Pericardial cavity, heart.

Host: Puffer (Takifgu rubripes).

Locality : Coast of Kyushu, Japan (western North Pacific Ocean).

References: Egusa (1986); Egusa and Shiornitsu (1983); Lom and Dykova (1992). Kudoa sphyraeni Narasimhamwti and Kalavati, 1979

Detrimental effects: None reported.

Description: Myxospores quadrate in apical view, no ornamentation. Polar capsules club-

shaped, equal. Cysts pedunculate, opaque, white, up to 2.0 mm in diameter.

Site: Intestinal musculature.

Hos t : Pickhandle barracuda (Sphyraenajeilo) .

Locality: Bay of Bengal (Indian Ocean).

Remarks: Host response involving co~ectivetissue was observed surrounding the

pseudocyst.

References: Egusa (1986); Narasimharnurti and Kalavati (1979a).

Ktrdoa sfellula Yurakhno, 1991

Detrimental effects: None reported.

Description: Myxospores stellate in apical view. Polar capsules unequal.

Site: Kidney.

Host: Atherina hepsetus.

Locality: Black Sea.

References: Yurakhno (1 99 1).

Kudoa fachysurae Sarkar and Mazurnder, 1983

Detrimental effects: None reported.

Description: M yxospores quadrate in apical view, with rounded valves. Polar capsules pyriform, unequal. Site: Gallbladder.

Hos t : Tachysurus renrtispinis.

Locality: Bay of Bengal (Indian Ocean).

References: Egusa (1 986); Sarkar and Marwnder (1 983).

Kudoa retraspora Narasimhamurti and Kalavati, 1979

Detrimental effects: None reported.

Description: Myrtospores quadrate in apical view. with deep notches and no ornamentation. Polar capsules club-shaped, equai. Cyst up to 1.5 mm in diameter.

Site: Brain.

Host: Striped mukt (Mugil cephalus).

Locality: Coast of India (Lndian Ocean).

References: Egusa (1986); Narasimhamurti and Kalavati (1 979b).

Kudoa valamugili Kalavati and Anuradha, 1993

Detrimental effects: None reported.

Description: Myxospores quadrate in apical view, no ornamentation. Valves unequal with rounded edges, deep notches. Polar capsules pyrifom, unequai. Plasmodia sphericai or oval, opaque, white, up to 0.7 mm in diameter.

Site: Intestinal musculature.

Host : Mullets ( Vafarnugilcunnesius).

Locality : Bay of Bengal (Indian Ocean).

Remarks: Two uninucleate sporoplasms are situated in the larger valve. References: Kalavati and Anuradha ( 1993).

Kudoa vesica Kovaleva and Gaevskaya, 1 984

Detrimental effects: None reported.

Description: Myxospores stellate to quadrate in apical view. no ornamentation. Polar capsules pyrifonn, equal.

Site: Urinary bladder.

Host : Pseudoicichthys ausrrulis.

Locality: Coast of Antarctica.

References: Egusa (1 986); Kovaleva and Gaevskaya (1 984).

8.3. Reports of undescribed Kudoa spp.:

There have ken several worldwide reports of unidentified species of Kudoa since the establishment of the genus. Whether these are new species or additional host records should be investigated.

Krasin (1 976) described infections of Kudoa spp. fiom the musculature of several fish species in the northeast Pacific Ocean. The fish species identified as harboring infections included ocean whitefish (Caulolurilus princeps), northern anchovy (Engraulis rnordax), white croaker (Genyonemus linearus), Paci fic hake, and rockfishes (Sebastes spp.). Paperna (1 982) collected the rounded myxospores of what were probably two species of Kudoa fiom the kidney, mesentery, and peritoneurn of gilthead bream (Sparus oztruius) fiom the Bay of Eilat (Mediterranean Sea). Langdon (1990) described the stellate spores of a Kudoa sp. fiom the brain of barramundi perch (Lotes calcarifer) fiom off north Queensland, Australia.

Bunton and Poynton (1991) described the subquadrangular myxospores of a

Kudoa sp. fkom the musculature of the white perch (Morone antericana) from the

Choptank River, Maryland, USA. Dykova et al. (1 994) observed Kudoa spp. in several fish species collected fiom the waters of the Gulf of Mexico. They observed one species in the musculature of the sheepshead mimow (Cyprinodon variegarus), the gulf killifish

(Fundulus grandis), and the golden topminnow (Fundul~(schrysotu~). The other two species were fiom the musculature of the western mosquitofish (Gambusia aflnis) and the inland silverside (Menidia beryllina). Heupel and Bennett (1996) collected a species of Kudoa from the musculature of the epaulette shark (Hemiscylliurn ocellatum), caught off the Great Barrier Reef, Australia.

More recently, Davies et al. (1998)observed what may be two Kudoa spp., resembling K. nova superficially. in the musculature of gobies (Gobius paganellus and

Poma~oschistusrnicrops) fiom two localities near Devon. United Kingdom. Also, Moran and Kent (1999) have observed infections by an unidentified species of Kudoa brobably

K. paniformis) in the somatic musculature of lingcod collected off the West Coast of

Vancouver Island, British Columbia, Canada (Fig. 3.6h).

9. Conclusion

With the development of genus-specific and species-specific SSU rDNA primers designed for use with the PCR, the ability to ver@ the taxonomic validity of Kudoa species that are presently in doubt, is attainable. With sequence information for the ribosomal DNA gene, we should also be able to investigate the host and geographic ranges of those Kudoa species that presentty exhibit wide host specificity or are distributed throughout the world. This wiH permit us to determine if, in fact, these species are single species or an assemblage of morphologically simitar species. CHAPTER 4

PROGRESSION OF NATORAL INFECTIONS

"Reprinted fiorn Journal of Aquatic Animal Heaith (in press), Moran, J.D.W., and Kent, M.L., Kudoa thyrsites (Myxozoa: Myxosporea) infections in pen-reared Atlantic salmon in the eastern North Pacific Ocean, with a survey of potential non-salmonid reservoir hosts, Copyright 1999, with permission from The Arnencan Fisheries Society". 1. Introduction

The genus Kudoa (Myxozoa: Myxosporea) is comprised of more than 40 species

of parasites that affect marine and estuarine fishes around the world (see Moran et al.

I999a). Kudoa thyrsites (Gilchrist, 1924) has ken reported fiom the sornatic and cardiac

musculature of wild and aquaculture-reared marine fishes worldwide, with severe

infections resulting in soft fiesh (Harrell and Scott 1985; Kent et al. 1994b; Whitaker et

al. 1994; Moran et al. 1999a). The marketability of infected fish products decreases

drarnatically within days afier harvest, and losses result not only because of this

occurrence of soft flesh, but also as the result of an adverse market perception to the

industry responsible for supplying the infected products. Kudoa hyrsifesinfections of

pen-reared Atlantic salmon are recognized as a senous problem of the aquaculture

industries of Canada, the United States, and Ireland (Elston 1994; Palmer 1994; Whitaker

et al. 1994). There is also a potential impact on the rapidly expanding Atlantic salmon

aquaculture industry in Chile as reports of infections in wild fish (e.g., Paralichthys

adspersus) have been published (Castro and Burgos 1996).

Reports detailing the seasonai patterns of myxosporean infections in salmonid fish

(e.g., PKX,ceratomyxosis) have been published previously (Ching and Munday 1984;

Foott and Hedrick 1987). in these infections, the infective stage has been shown to be

present throughout the surnmer and fa11 months, and basically absent throughout the

winter. Yokoyarna et al. (1 993) have dso investigated the seasonality of various types of

actinosporeans in the ol igochaete Branchiura sowerbyi and found that the maximum prevalence of infection (5%) was observed in the spring and summer, and subsequently decreased through the winter months. The progression of myxosporean infections is closely related to environmental

temperat ure. For exarnple, C. shasta Sections in various salmonids (Oncorhynchus spp.)

develop at a much faster rate as water temperatures increase (Udey et ai. 1975;

Y amarnoto and Sanders 1979; Bartholomew et al. 1 989). Similarly, the rate at which the

PKX myxosporean developed in rainbow trout increased as water temperatures were

raised fiom 12°C to 1 8OC (Clifton-Hadley et al. 1986).

The objectives of this midy were to investigate the seasonality of the K. thyrsites

infective stage in the marine environment and to follow the progression of the infections

within a netpen population. in addition, we include a survey of non-saimonid fishes to

identifi potential reservoirs for the infection. The terms myxospore and actinospore are

used to distinguish between the mature spore stages observed in fish and the aitemate

amelid hosts, respectively, as suggested by Lom et al. (1997).

2. Methods

Al1 experiments were conducted at the PBS, in Nanaimo, BC. Hatchery-reared

Atlantic sdmon smolts were held in fieshwater tanks until transferred to the PBS experimental seawater netpens in Departure Bay, Nanaimo, for natural exposures to K. rhyrsires. Experimental fish were either held at the seawater netpens for the duration of the experiment or retumed to either seawater or fieshwater tanks at the PBS culture facility after their netpen exposure. Netpen (at 4 m depth) and tank water temperatures were recorded daily and biweekiy mean water temperatures were calculated fiom these data. The time to sporulation following initial exposure was determined by calculating degree-days, with the assumption that fish imrnediately become infected upon transfer to seawater. Degreedays PD) were calculated as the sum of alI biweekly mean water temperatures (BMT) multiplied by the number of days in each respective period (D) for the duration of the holding penod (e-g., DD = Z [IBMTxD]). Al1 experimental fish were fed an artificial commercial diet.

Wild fishes were collected fiom the vicinity of seawater netpen sites using either hook and line or fiom within the netpens during the saimon harvest. Offshore fishes were collected using a bottom trawl during research cniises of the Fisheries Research Vesse1

(FRV) W.E. Ricker off the West Coast of Vancouver Island, BC (Fig. 4.1). Infections by

K. rhyrsires were detected using the wet mount techniques described by St-Hilaire et al.

( 1997b) with the modifications described by Moran et al. ( 1999b).

The cornmon and scientific names of fishes used in this paper are as recommended by Robins et al. (1 99 1a, b).

2.1 Seasonality of the infective stage

Atlantic salmon were held at the PBS experimental seawater netpen site for 8- week penods between 25 April 1996 and 2 1 May 1997 to determine if the infective stage of K. thyrsires was present in the netpen environment. The 8-week perîods were as follows: 1 ) 25 April - 19 June 1996; 2) 19 June - 16 August 1996; 3) 16 August - 09

October 1996; 4) 09 October - 04 December 1996; 5) 04 December 1996 - 29 January

1997; 6) 29 January - 26 March 1997; and 7) 26 March - 2 1 May 1997. After each respective exposure period, the Atlantic Amon were transferred to 400-L tanks receiving flow-through fiesh water to prevent re-exposure to the parasite's infective stage. Moran et al. (1 999b) have detennined that the PBS sand-filtration system does not remove the PLATE XI

Figure 4.1. Map of Vancouver Island, British Columbia, showing sampling sites where

wiId-caught non-salrnonid fishes were collected fiom the coastal waters

offshore or near netpens. Key to localities is: A = Nootka Sound; B = Barkley

Sound; 1 = Clayoquot Sound; 2 = Philips Am; 3 = Quadra Island; 4 =

Departme Bay; 5 = Maple Bay. Locaiities identified with a letter indicate

offshore collection and those with a number indicate netpen collections. parasite's infective stage fiom the seawater that is supplied to the tanks. Previous studies also show that the parasite cm successfully sporulate in Atlantic salmon returned to fiesh water afier exposure in seawater (Moran et al. 1999b).

2.2 Progression of natural infections

This experiment was designed to follow the prevalence of K. thyrsires infections on a monthly basis in Atlantic salmon held at either the PBS experimentd seawater netpens or in seawater tanks. Seawater temperatures ranged fiom 6 - 18OC at the netpens and 8 - 1 2°C in the tanks throughout the year (Fig. 4.2).

2.2.1 . Seawater netpens

The first attempt was initiated with the transfer of Atlantic saimon (N = 400) srnolts to the seawater netpens on 09 June 1995. Twenty-five fish were exarnined montMy thereafter until 13 October 1995 when the experiment was terminated due to mortalities resulting fiom several blooms of the toxic alga Heterosigma carterae during the surnrner exposure period. The experiment was reinitiated on 04 June 1996, with the transfer of 450 Atlantic salmon smolts to the seawater netpens. The somatic musculature from 25 fish was collected monthly between 08 July 1996 and 05 June 1997 with additional samples of 25 and 24 Atlantic salmon on 26 August and 05 December 1997, respectively. PLATE XII

Figure 4.2. Biweekly mean water temperatures at the Pacific Biological Station

experimental seawater netpens (at a depth of 4 m) and in the seawater tanks

between Janwry 1995 and May 1997. Solid line indicates the seawater netpen

temperatures and the dashed line indicates the seawater tank temperature. Sea

water supplied to the tanks is brought into the PBS culture facility fiom a

depth of 22 m. (a,,) sa~n~e~aduiajJelem ueaw Aty aawg 2.2.2. Seawater tanks

Atlantic salmon smolts were initially exposed to the infective stage in the

seawater netpens on 26 May 1995 for 2 weeks and thereafier du~gtheir subsequent

holding in the seawater tanks at PBS. The ambient seawater that is supplied to the PBS

tanks is brought in fiom a depth of 22 m and the temperature typically ranges fiom a low

of 8°C in the winter months to a high of 12°C through the sumrner months (Fig. 4.2).

Muscle sarnples of 25 salmon were taken at 3,4,6,8, 12, 16, and 20 months post-

exposure (p.e.). Al1 samples were screened using wet mount preparations, with the

exception of the 3 and 4 months p.e. samples, which were screened using histological

examination.

2.3 Survey of potential reservoir hosts

Thirty-one non-salmonid fish species from 27 genera representing 15 families

were collected fiom both the vicinity of seawater netpens and during research cruises of the FRV W.E. Ricker off the Coast of Vancouver Island, BC (Fig. 4.1). Of the four

Sebasres spp. exarnined, two were collected offshore (canary rockfish pinniger and yetloweye rockfish S. ruberrimus) and two from the vicinity of netpens (copper rockfish

S. caurinzcs and quillback rockfish S. mafiger).

3. Resulis

3.1 Seasonality of the infective stage

Infections were contracted dwing the first four 8-week exposures fiom 25 April to

04 December 1996 (Table 4.1). AAer approximately 4 months, the infections could be Table 4.1 . Seasonality of infection by Kirdm ijtyrsires in Departurc Bay, near Nanainio, British Col unibia. Atlantic salnlon (Srrlrno salar) were held at the experimental seawater netpens for each Il-week exposure period and transferred to freshwater tanks until approxirnately 2000 degree-days, at which thethcy were screened for infection using wet mount preparations.

Exposure period (ddlmmlyy) Date Examined Kudoa thvrsites infections Mean wcight (* S.D.) No, InfectedINo. Examined % lnfected 25/04 - 1 9/06/96 20/09/96 712 5 28% 174.0k50.3g detected using wet mount preparations to detect myxospores in muscle tissue. The highest prevaience of infection (44%) was achieved during the exposure piod of 19 June - 16

August 1996. The remaining three 8-week exposure groups covering the period fiom 04

December 1996 to 2 1 May 1997 showed no signs of K. thyrsires infections.

3.2 Progression of natural infections

in the 1995 netpen study lasting 4 months, K. rhyrsites infections were not detected until2 months p.e. (1000 degree-days), and at this time only 4% (1/25) of the fish were positive for the infection (Fig. 4.3). At 3 and 4 months p.e. (1 500 and 2000 degree-days), the prevalence of infection reached 8% (2/25) and 1 1% (2/18), respectively.

A similar pattern was observed in the 1996 netpen study, with the first infections detected at 2 months p.e. (1000 degree-days) and the prevaience reached 56% at 4 months p.e.

(2000 degree-days). In the tank exposure, where surnmer temperatures were lower, the infection was not observed until3 months p.e. (1000 degree-days) and reached a maximum prevalence of 60% by 6 months p.e. (2000 degree-days) (Fig. 4.3). In both the netpen and tank studies, the prevalence of infection declined substantiaily arnong fish between 9 and 12 months p.e.

3.3 Survey of potential reservoir hosts

Seventeen species of fishes representing 16 genera were collected fiom the vicinity of fmedAtlantic and chinook salmon seawater netpens at various points dong the Coast of Vancouver Island, BC (Fig. 4.1). Of those fishes collected near the seawater netpens, only three species were found to be positive for K. thyrsites infection (Table PLATE XII1

Figure 4.3. Progression of Kudoa thyrsites infections in Atlantic salmon within two

seawater netpen populations (1995 and 1996) and one seawater tank

population (1 995). Exposures were initiated late May or early June.

Table 4.2. Prevalence of Kudoa thyrsites Sections (No. positiveNo. examined) in wild-

caught nonsalmonid fishes collected fiom the coastal waters of Vancouver Island, British

Columbia, offshore or near netpens (Fig. 4.1 ). Key to Iocalities is A = Nootka Sound, B =

Barkley Sound, 1 = Clayoquot Sound, 2 = Philips Am, 3 = Quadra Island, 4 = Departure

Bay, and 5 = Maple Bay. Localities identifed with a letter indicate offshore collection and

those with a number indicate netpen collections. Prevalences greater than zero are in bold.

Hos t Locality Kudorr th vrsiies infections (No. positive/No. examined) Offshore Netpens A losa sapidissirna American shad jlnoplopomafimbria Blackcod Atheresthes stomias Arrowtooth flounder Azilorhynchusj7avidus Tube-snout Citharichthyes sordidus Pac i fic sanddab Clupea pullasi Pacific hemng Cymatogaster aggregata Shiner perch Embiotoca lateralis Striped seaperch Eopserta jordani Petrale sole Gadzis macrocephalus Paci fic greycod Gasterosteus aculeatus Threespine stickleback Glyptocephalus zachirus Rex sole Hippoglosso ides elassodon F 1athead sole Hippoglossus stenolepis Pacific halibut Ice linus filamentosus Thread fin sculpin Leprocortus arrnatus Pacific staghom sculpin Lyopsetta exilis Slender sole Microgadus proximus Pacific tomcod Ophiodon elongatus Lingcod Pholis ornata Saddleback gunnel Platicht hys stellatus Starry flounder Pleuronectes bilineafus Rock sole Porichthyes notatus Midshipman Scomber japonicus Pacific mackerel Se bastes spp. Rockfishes Sqrralus acanthias Spiny dogfish Syngnathus griseolineatus Pipe fish Theragra chalcogramma Walleye plloc k 4.2). Seven of 10 tube-snout. a fish that is cornmon near netpens, were identified as

positive. Sixteen of 32 lingcod were also positive for K. thyrsites infections and lastly,

one of three rock sole.

During the researc h cruise, 19 species of non-salmonid fis hes fiom 17 genera

collected fiom various points around the Coast of Vancouver Island were exarnined for K.

rhyrsires infections (Table 4.2). Kudoa thyrsites infections were detected in two species of

non-salmonid fish including arrowtooth flounder and threadfin sculpin. The prevalences

of infection in both non-salmonid species were 1 00% (2/2 and 1/ 1).

4. Discussion

The seasonal nature of infections has been demonstrated with several freshwater

myxosporeans (see Clifion-Hadley et al. 1986; Foott and Hedrick 1987; Mitchell 1989;

Duarte et al. 1993; Bellerud et ai. 1995). In contrat, little is known about the seasonality

of marine rnyxosporean infections but we presume that seasonality results fiom physical

factors (e-g., water temperature) directly afTecting developmental cycles of the parasites

or hosts (Le., fish or annelids), including the immune response (Foott and Hedrick 1987;

Hedrick 1998).

In our study, there was a distinct seasonality to the K. thyrsites infective stage in

the marine environment, as infections were readily contracted fiom late May to early

December, and not at al1 fiom late December to early May (Table 4.1). Although the

seasonality study was tenninated in late May 1997, K. thyrsites infections were contracted

by salmon that were transferred to Our seawater netpens in June 1997. This demonstrates that the dedine in infections through the winter months of 1996 and 1997 was in fact seasonal. as the prevalence of infection in these salmon exposed in June 1997 reached

high levels.

Myxosporean infections may take a few weeks (e-g., Sphaerospora renicola in

common carp fry Cyprinus carpio) to a few months (e.g., M. cerebralis in rainbow trout)

to develop (see Lom and Dykova 1992). In other myxosporean idections, such as

Mjxidium salvelini, the parasite may undergo a period of arrested development (Higgins et al. 1993). The rate of development and time to spodation of a myxosporean parasite is

in part dependent upon the temperature of the environment in which the host fish resides.

As a result, the calculation of degree-days is imperative in determining the time required

for a myxosporean to sporulate.

in field exposures of rainbow trout to M. cerebralis, the cause of salmonid whirling disease, approximately 1800-2000 degree-days were required for parasite spomtation (R.B. Nehring, Colorado Division of Wildlife, pers. comm.). Assuming that the Atlantic salmon become infected with K. thyrsites immediately upon their transfer to seawater, Mly developed myxospores may be detected as early as 1000 degree-days p.e.

However, the majority of K. thyrsites infections will require approximatei y 2000 degree- days (in salmon held in either fiesh or seawater) to produce mature myxospores.

Although seawater temperatures in the experimental netpens and tanks varied considerably between seasons, the overall range was more limited in tanks compared to that typically observed in seawater netpens (Fig. 4.2). These temperature differences would account for the 2 month delay in sponilation seen in fish held in the seawater tanks compared to the seawater netpens.

A majority of smolts become idected with K. thyrsites shortly after their transfer to the seawater netpens (Fig. 4.3). These infections may be readily contracted because salmon are undergoing the process of smoltification at the time they are introduced into the netpens. Plasma cortisol levels, used as indicators of stress and immune compromise, have been shown to increase significantly in salmon that are undergoing smoltification or sexual maturation (Made et al. 1987; Barton and Iwama 1991 ; Clarke and Hirano 1995;

Fagerlund et al. 1995). As demonstrated in Figure 4.3, the Atlantic sdmon srnolts generally recovered fiom the infections within a year. Recovery fkom these infections is associated with a host response characterized by multifocal, chronic inflammation between muscle fibers that is triggered by niptured pseudocysts. As salmon undergo sexual maturation in the netpens, the prevalence of K. thyrsites infections has been observed to increase dramatically (St-Hilaire et al. 1998). We believe that the compromised immunity of the host, both at smoltification and sexual maturation, permits the parasite to proliferate andlor reinfect the fish. Alternately, and as suggested by Moran et al. (1999b), a cryptic stage of the parasite rnay reside within the host perhaps for periods up to 2 years after initial seawater exposure. The reactivation and then uncontrolled proliferation of this cryptic stage may result in severe infections, and subsequently, soft flesh syndrome which is particufarly problematic in grilse and reconditioned grilse (St-Hilaire et al. 1998).

With the uncontrolled exchange of seawater in netpens, wild marine fishes may act as reservoirs for several parasites that affect cultured salmon (e.g., the cestode

Eubothrium sp., sea lice L. salnonis, the microsporidian L. salmonae) (cf. Bristow and

Berland 1991 ; Hastein and Lindstad 1991 ; Hedrick 1998; Kent et al. 1998). Although yet to be demonstrated, many marine myxosporeans probably require an altemate host for completion of their life cycles, as has ken described for several fieshwater myxosporeans

(see El-Matbouli et al. 1992; Kent et al. 1994a). However, Diamant (1 997, L 998) reported

that ;M. feei can be directly transmitted to gilthead bream and red dm(Sciaenops ocellarus) using both cohabitation and efnuent exposure experiments. We have been unable to transmit K. rhyrsites to Atlantic salmon by intubation of fksh myxospores into susceptible Atlantic salmon (Moran et al. 1999b). The potential for transmission between wild fishes and fanned salmon via an altemate invertebrate host should therefore be investigated fiuther.

Infections in non-salmonid fishes represent a reservoir for the parasite near the seawater netpens. If an annelid host is involved in the life cycle of K. thyrsites, non- salmonid fish resident in the area could keep the prevalence of infection within the altemate host at a greater level. The result would be significant increases in the nurnber of infective stages (actinospores) for the farmed salmon in the nearby seawater netpens.

Kudoa thyrsites was originally described fiom snoek collected off the Coast of

South Afiica (Gilchrist 1924). McDonald and Margolis (1 995) provide a surnmary of hosts for K. thyrsites in Canadian waters, which include wild and farmed fish species.

The wild host records included arrowtooth flounder, Pacific halibut, Pacific hake, Dover sole, Pacific salmon (Oncorhynchus spp.), rock sole, and waileye pollock. Their host records also included two of the salmonid species reared in aquaculture in British

Columbia: Atlantic and coho salmon. Using SSU rDNA sequence, Hervio et al. (1997) suggested that the K. rhyrsites in pen-reared Atlantic salmon was undistinguishable fiom those observed in tube-snout and Pacific hake. In our survey, five species of non- salmonid fish fiom four families were found to be infected with K. thyrsites (Table 4.2). Any of these susceptible non-salmonids may act as potential reservoirs for the infection

as some of these species are known to reside near seawater netpens. A second unidentified Kudoa sp. was also detected during the survey in the musculature of lingcod collected in Barkley Sound- Its spore morphology was sirnilar to that of K. paniformis, a common myxosporean parasite of Pacific hake off the West Coast of Vancouver Island.

The results of this study demonstrate the seasonal nature of K. thyrsites infections.

Atlantic salmon smolts contract the infections in the spring through early fall, exhibit a high prevalence of infection at 4 to 6 months p.e., and then recover fiom the infections.

The rate at which the infections progress is dependent upon the water temperature. We have detennined that the tirne to sporulation of the parasite in the majority of the infections afier their initial exposure to seawater is approximately 2000 degree-days, comparable to that of the fieshwater myxosporean M. cerebralis. Of the 5 non-salmonid species identified as reservoir hosts for K. rhyrsites infections, two (lingcod and threadfin sculpin) represent new host records. Other important aspects of the biology of K. rhyrsites in Atlantic salmon remain to be elucidated including: 1) the mechanism of recovery; 2) the route of infection; and 3) the possible existence of an aitemate host. CHAPTER 5

TRANSMISSION

"Reprinted fiom Journal of Aquatic Animal Health (in press), Moran, J.D.W., Whitaker, D.J., and Kent, M.L.,Natural ar,d laboratory transmission of the marine myxozoan parasite Kudoa thyrsites (Gilchrist, 1924) to Atlantic salmon, Copyright 1999, with permission fiom The American Fisheries Society". 1. Introduction

Kudoa thysites (Gilchrist, 1924) is a histozoic myxozoan parasite that infects the

somatic and cardiac musculature of several wild and cultured marine fish species

throughout the world (Harrell and Scott 1985; Baja and Toranzo 1993; Kent et al.

1994b; Whitaker et ai. 1994; Castro and Burgos 1996; Moran et al. 1999a). Severe

infections in cultured Atlantic and coho salmon ofien result in pst-mortem

myoliquefaction, which is commonly referred to as soft flesh syndrome. Within the aquaculture industry, K. ihyrsites infections of Atlantic and CO ho salmon are recognized as a significant problem due to decreased marketability of infected fish products within days after harvest (Whitaker and Kent 1991 ; Whitaker et al. 1994; Moran et al. 1999a).

Experimental transmission studies have show that oligochaetes (see reviews by

El-Matbouli et al. 1992; Kent et al. 1994a) or polychaetes (Bartholornew et al. 1997) serve as required alternate hosts in the life cycle of several fieshwater myxozoans.

Actinosporean stages, sirnilar to those found in fieshwater annelids, have been observed in marine oligochaetes (Marques 1984; Hallett et al. 19%). However, they have yet to be united with their corresponding marine myxosporeans. Diamant (1 997, 1998) has suggested that certain marine myxozoans may possess direct life cycles. He found that M. leei was transmitted directly fiom fish to fish in cohabitation studies that should have excluded any possible contribution of an altemate host.

Kudoa thyrsires infections in Atlantic salmon provide an excellent mode1 to study marine myxozoan life cycles as parasite-free fish are easily obtained, infections are easily contracted with exposure to seawater, and numerous myxospores can be harvested from the muscle of infected fish. The objective of this study was therefore to examine possible routes of transmission and the development of K. thyrsires in Atlantic salmon following

naturd and experimentai exposures to the parasite. In addition, we studied the fate of

infections in Atlantic salrnon transferred back to fiesh water fier a brief seawater exposure.

The terrns myxospore and actinospore are used to distinguish between the spore stages observed in fish and the aitemate annelid hosts, respectively, as suggested by Lom et al. (1 997).

2. Methoh

Al1 expenments were conducted at the PBS, in Nanaimo, BC. Hatchery-reared

Atlantic salmon were held in fresh water before being subjected to the various exposures.

Therefore, they were unequivocaily free of the infection as K. rhyrsires is strictly a marine parasi te, and negative controls were not necessary with each expriment. Atlantic salmon smolts were naturally exposed to K. thyrsires either at the PBS exprimental seawater netpens in Departure Bay or while held in fiberglass tanks supplied with seawater within the PBS culture facility. Seawater was pumped into the PBS culture facility from a mean depth of 22 m in Departue Bay and filtered by iarge, industrial sand filters. Al1 fish were fed an artificial commercial diet.

Kudoa thyrsites infections in the somatic musculature were detected using the wet mount techniques described by St-Hilaire et al. (1997b) with some minor modifications.

Briefly, muscle samples were removed from the left abdominal wall of the Atlantic salmon. Approximately 3-4 g of muscle was removed and minced in saline (0.65%). This minced material was pressed between two Plexiglas plates and the resulting extract collected. A wet mount preparation was prepared by adding one drop of the extract to a

glas slide containing one drop of saline and a coverslip was added. Wet mount screening

was completed by searching randomly for 5 min using 320x magnification and phase

contrast illumination. Al1 dissection tools and materials were thoroughiy cleaned to

prevent cross-contamination or transfer of myxospores between samples.

Moran and Kent (1 999) have detennined that K. thyrsites infections in Atlantic

salmon can be detected as early as 1000 degree-days p-e., with most fish showing fùlly-

fomed myxospores by 2000 degree-days p.e. Sampling times for al1 experiments were

therefore based upon degree-day calculations to increase the probability of detecting Mly

developed myxospores in the somatic musculature of infected fish. Degree-days are calculated as described in Moran and Kent (1999).

2.1. Exposure to filtered seawater

To compare the effectiveness of removing the infective stage of K. thyrsites fiom the seawater supplied to the tanks, Atlantic salmon (N=50) smolts were transferred to each of three 400-L fiberglass tanks on 14 June 1996. Al1 seawater that is supplied to the tanks at PBS is filtered by sand filtration. Each tank was supplied with flow-through seawater filtered by one of three different systems. The control tank received only seawater passed through the sand filtration system. The first experimental tank received seawater passed through three stainless steel filter housings (Mode1 FI -50, Tate Andale

Canada Inc., Concord, Ontario, Canada) fitted with a 25,5, or 1 pm bag filter in succession. The second experimental tank received seawater treated by both the bag filters and an additional in-line cartridge designed to remove al1 particles greater than 1 pm in size. Somatic muscle samples were collected from 25 fish at approximately 6 months p-e. and wet mounts were screened tu determine the prevalence of K. thyrsites infections.

2.2. Development in fresh water

This experiment was conducted to determine if a short exposure in seawater followed by transfer to fiesh water would result in detectable levels of infection. Atlantic salmon (N = 900) smolts were transferred to the PBS seawater netpens on 17 June 1996 for 2 weeks and returned to a 4500-L semi-recirculating fieshwater tank receiving 20% continuous exchange (temperature ranged from 6-1 8°C; mean = 12°C) for the duration of the experiment. Somatic muscle samples of Atlantic salmon (IV = 25) were collected at 3,

4, 5,6, and 23 months p.e. and placed in individual plastic bags and stored at 4°C until microscopie examination using wet mount preparations. Another group of Atlantic salmon (N = 50) smolts were transferred to the PBS seawater netpen on 19 June 1996 and held for 8 weeks at the seawater netpens (temperature ranged from 16- 1 8"; mean = 1 7°C) before they were returned to a 400-L tank receiving flow-through fiesh water

(temperature ranged from 1 1- 18OC; mean = 1 5°C) on 16 August 1996. These Atlantic salmon (IV = 25) were sampled on 07 November 1996, approximately 20 weeks p.e. 2.3. Transmission by injection of blood

Blood was collected on 27 November 1995 using 10 rnl heparinized vacutainer tubes from four coho sdmon (Oncorhynchus kisuich) which were held in netpens for over i year and subsequently identified as positive for K. ihyrsites sections. The blood was refkigerated oveniight at 4°C and on the following day, 0.4 ml of the blood was injected via the intraperitoneal route into Atlantic salmon (N = 3 1) held in a 400-L tank receiving fresh water (temperature ranged from 6- 10°C; rnean = 8°C). Screening of the surviving 23 fish for K. thyrsiies infections was initiated on 09 April 1996 (1 8.5 weeks p.e.). Individual samples were collected and screened on 15 April, 17 April, 24 April, and

14 May 1996. The remaining 17 fish were examined on 16 May 1996 (24 weeks p-e.).

2.4. Transmission by intubation of myxospores

Muscle tissue heavily infected with K. thyrsites was collected fiom an Atlantic salmon that was removed fkom a 4000-L tank receiving flow-through seawater. A cmde homogenate of the muscle was intubated directly into the stomachs of 28 Atlantic salmon

(0.6 ml/fish) held in a 400-L tank receiving flow-through fresh water (temperature ranged from 8- 18°C; mean = 15°C). This inoculurn contained approxirnately 1.8 million K. thyrsites myxospores. Two fish died during the expriment, and the remaining fish were sampled at 4.5 months p.e. (N= 10) and 6 months p.e. (N = 16). 3. Results

3.1. Exposure to filtered seawater

Seawater temperatures ranged from 8-1 2°C (mean = 1 1OC) for the duration of this

experiment. Of the Atlantic salmon that were held in the 400-L tanks receiving flow-

through seawater, 13 of 25 (52%) were positive for K. thyrsites infections. Thirteen of 25

(52%) fish held in the tanks supplied with seawater filtered through the successive bag

filters were also positive for the infection. Fifteen of 25 (60%) fish he1d in the tanks

supplied with seawater filtered with the in-line cartridge, in addition to the bag filters, were positive for the infection.

3.2. Development in fresh water

Arnong the Atlantic sairnon exposed to seawater for 2 weeks, the infection was detected in 2 of 25 (8%) fish sampled at both 3 and 4 months p.e. Only 1 of 25 (4%) fish was positive at 5 months p-e., and no infections were detected at 6 months p.e. On 21

May 1998 (23 months p-e.), 1 of 10 Atlantic salmon was positive for the infection.

Among the Atlantic salmon exposed at the PBS seawater netpens for the 8-week period (1 9 June - 16 August 1996), K. thyrsires infections were detected in 1 1 of 25

(44%) fish sampled approximately 4.5 months p.e.

3.3. Transmission by injection of blood

Eight fish died before screening for K. thyrsires was initiated. Two of the remaining 23 fish (exarnined 15 and 17 April 1996) injected with infected blood exhibited K. rhyrsites infections.

3.4. Transmission by intubation of myxospores

None of the 26 fish examined that were exposed per os exhibited K. *sites infections.

4. Discussion

Aside fiom the direct transmission studies of Diamant (1 997, 1998), our study represents the first expenmental transmission of a marine myxozoan. The results of these studies have provided important information on the transmission and subsequent development of K. fhyrsites in Atlantic salmon, which should provide some new uisights on the transmission mechanism of marine myxozoans. The scarcity of information currently avaiiable on the transmission of marine myxozoans is, undoubtedly, in part due to the difficulties with obtaining parasite-free water supplies and marine fishes. The

Atlantic salmon-K. thyrsites mode1 is excellent for these investigations (Le., the availability of unequivocally parasite-free experimental animals and the abiiity to transfer fish between fresh water and seawater (and vice versa)).

Pnor to the initiation of these experiments, we had anecdotal evidence that the infective stage of K. thyrsites was able to bypass the PBS sand filtration system. As a result, additionai filters were installed in an attempt to remove particles dom to 1 pm in size fiom seawater but this was proven ineffective at removing the infective stage of the parasite frorn the incoming seawater supply. This does not suggest that the infective stage is srnaller than 1 Pm. but rather that the filters are ineffrcient at removing al1 rnicroscopic organisms and particles. Nevertheless, the system was effective at removing macroinvertebrates particularly annelids (oligochaetes or polychaetes) that rnight sewe as alternate hosts for K. thyrsites. Of those studied, transmission via water containing actinospores released fiom annelids is the principal route by which freshwater myxozoan infections are contracted (El-Matbouli et al. 1992; Bartholomew et al. 1997). It is likely that the seawater netpen environrnent provides a suitable habitat and ample nutrients for annelid populations that may be associated with the biofouling fiequently occwring on seawater netpens. If these annelids are infected with the actinosporean stage of K. thyrsiles, they may shed actinospores periodicalty into the seawater netpen environrnent, increasing the risk of exposure of pen-reared fish to the infective stage. These same infective stages are those that passed through the filtration system to infect fish in the seawater tanks at the PBS.

Presently, natural exposure of Atlantic salmon in seawater netpens to infective stages within the environment is the most effective way to infect a salmon population. in

Atlantic salmon that were returned to fiesh water after the brief seawater exposure, the parasite continued to develop and successfüily sporulated within the muscle fibers.

Cornparhg a 2-week to an 8-week natural exposure, it appears that longer exposures result in a higher prevalence of infection (and apparently a higher intensity of infection).

The results of this study and our other experiments (Moran and Kent 1999) indicate that a significant proportion of Atlantic salmon smolts become infected wiùlln 2 months of their transfer to seawater. The prevalence of infection is high arnong pst-smolt Atlantic salmon in their first swnmer in seawater, but fish recover &er about one year (Moran

and Kent 1999). The prevalence and severity of infections rise again among market-size

fish, particularly grilse and reconditioned grilse (St. Hilaire et al. 1998), suggesting either

reinfection or reactivation of a cryptic, persistent stage. Reuifection or reactivation may

be triggered by the immunosuppression associated with sexual maturation. The

occurrence of the infection in fish held in fkesh water (Le., K. thyrsites-fiee water) almost

2 years (23 months p.e.) afier seawater exposure supports the latter possibility.

We were unable to induce infections among Atlantic salmon by the intubation of

fieshly harvested myxospores of K. thyrsites. The use of either fiesh or aged myxospores

in direct transmission experiments have consistently failed to transmit myxozoan

infections in previous studies (Schafer 1968; Fryer and Sanders 1970; Wyatt 1978;

Molniir 1979; Seenappa and Manohar 198 1 ;Markiw and Wolf 1983; Boyce et al. 1985;

El-Matbouli and Hoffmann 1989; El-Matbouli et al. 1992). Our results are consistent with previous findings of the inability to transmit myxosporean infections directly from fish to fish (e.g., the PKX myxosporean (Clifion-Hadley et al. 1984; D'Silva et al. 1984);

C. shasra (cf. Bower 1985); and M. arcticus (cf. Kent et al. 1993)). In contrast, Diamant

(1 997, 1998) reported that M. leei could be directly transrnitted to gilthead bream and red drum during cohabitation expenments and when exposed to the runoff of tanks holding infected sea bream. This myxozoan is a parasite of the intestinal mucosa and several different stages may be released in fecal casts shed fiom infected fish. Therefore, the possibility that M. leei was transmitted by vegetative stages rather than the myxospores should be considered.

Kudoa thyrsites was transmitted using intraperitoneal injection of the blood of an infected salmon. This suggests that an extrasporogonic blood stage is involved in the development of this parasite. The low prevalence of infection in fish receiving blood injections may be the result of their inactivation by holding at 4°C for several hours or the paucity of extrasporogonic stages in the blood when collected. Blood stages may be prevalent oniy in early infections and the donor fish in this experiment had advanced infections. Aithough myxospores of rnost species are apparently not ciirectly infective to fish. transmission of mycozoan infections (e.g., the PKX myxosporean. C. shasra, S. renicola) by intraperitoneal injections of pre-spore stages has been successful (Clifion-

Hadley et al. 1984; D'Silva et al. 1984; Bower 1985: Molnhr and Kovacs-Gayer 1986).

Furthemore, the PKX myxosporean and S. renicola have also been transrnitted by injecting blood containing extrasporogonic stages (Kent and Hedrick 1985: Molnir and

Kovacs-Gayer 1986).

In conclusion, the results of our investigations into the mechanism of transmission of K. thyrsites infections are significant in that they are comparable to those results achieved with fieshwater myxozoans. Our inability to transmit K. thyrsifes to Atlantic salmon using fkesh rnyxospores leads us to believe that there is an alternate host required for completion of the life cycle, as has been demonstrated in several fieshwater rn-y-xozoans.The seawater netpen environment provides an ideal habitat. not oniy for the invertebrate altemate hosts, but also for the numerous wild fish species that are known to harbor K. thyrsites infections (Moran and Kent 1999). It is impossible to avoid exposure to the infection in the netpen environment. However. if the alternate host is prevalent in the biota that accumulates on the nets, then fiequent changing of nets may reduce the intensity of infections. We are presently screening annelid Worms collected fkom the seawater netpen environment using a K. thyrsires specific PCR test developed by Hervio et al. (1997) in an attempt to identifjr the potential alternate host. CHAPTER 6

DEVELOPMENT

"Reprinted fiom Diseases of Aquatic Organisms (in press), Moran, J.D.W., Margolis, L., Webster, J.M., and Kent, M.L., Development of Kudoa thyrsites (Myxozoa: Myxosporea) in netpen-reared Atlantic salmon detennined by light microscopy and a polymerase chah reaction test, Copyright 1999, with permission fiom inter-Research". 1. introduction

Several species of myxosporeans have their sporogonic phase in the musculature of freshwater and marine fishes. These histozoic species are represented in genera such as

Myxo bolus, Unicapsula, Henneguya, and Kudoa. Some of the most documented species are members of the multivalvuiid genus Kudoa, which is presently comprised of 44 species of parasites (Moran et al. 1999a). Many species within this genus are assoçiated with post-mortem myoliquefaction. which is commonly referred to as "sofi flesh syndrome" (see reviews by Egusa 1986, Moran et ai. 1999a). Kudoa thyrsites has been implicated as a cause of sofi flesh syndrome in fmed Atlantic and coho salmon reared in

BC seawater netpens (Whitaker and Kent 1991 ; Kent et al. 1994b; Kent and Poppe 1998).

However, concem about K. thyrsites infections is not limited to BC aquaculturalists as this parasite has both a worldwide distribution and wide host range (Moran and Kent

1999; Moran et al. 1999a).

The early development of several rnyxosporean genera (e.g., Myxobolus,

Sphaerospora, Sphaeromyxa, Myxidiurn, Hoferellus, Ceratomyxa, and Kudoa), as well as the enigrnatic PKX myxosporean has been described (see reviews by Lom and Dykova

1992; Moser and Kent 1994). Myxosporean development may include a proliferative phase in a site other than that of sporuiation, and is referred to as an extrasporogonic phase (Lom and Dykova 1992). This phase has been reported among rnembers of the

Sphaerospora (cf. Csaba 1976) and Myxobolus (cf. El-Matbouli et al. 1995) genera. As well, the PKX trophozoite represents the extrasporogonic stage of an undescribed species

(Kent and Hedrick 1986). Although extrasporogonic development has not been reported for any mernber of the genus Kudoa, it was recently reported that K. thyrsifes can be experimentally transmitted by blood injections, which suggests that extrasporogonic development rnay occur in this species (Moran et al. 1999b).

Sporogenesis in myxozoans may occur by either development in pansporoblasts or direct spore morphogenesis within plasmodia (Lom and Dykova 1992). Within the genus

Kudoa. sporogenesis has been investigated using electron microscopy in both K. Zunata

(cf. Lom and Dykova 1988) and K. paniformis (cf. Stehr 1986). in both K. Iunata and K. panif~rrnis~sporogenesis occurs without the formation of pansporoblasts (hmami

Dykova 1992).

In this study, I describe the results of experiments using a combination of histological and PCR screening that provide further details on the development of K. thyrsites within the Atlantic salmon host and observations of the host response to these infections.

2. Methods

2.1. 1995-97 Experiment

A11 experiments were conducted at the PBS, in Nanaimo, BC. Atlantic sahon

(hr=450) reared at a fieshwater hatchery were naturally exposed to K. fhyrsites at the PBS expenmental seawater netpens situated near the research station. They were held at the site from 25 May to 08 June 1995 to permit natural exposure to K. thyrsites. After this 2- week natural exposure period, the fish were returned to three, 400-L tanks supplied with flow-through, sand-filtered seawater for the remainder of the experiment. Samples of 25 fish were collected at 4,8, 13, 1 7, and 26 weeks p.e. The remaining sarnples collected at

35, 52, 70, and 87 weeks p.e. varied in size at 10, 15, 15, and 17 fish, respectively. Mean seawater temperature was 10°C (range=8-15°C) for the experiment's duration (i.e., 25

May 1995 to 23 January 1997). Al1 fish were maintained on an artificiai, commercial diet.

Detection of Kudoa infections was by examination of wet mount preparations, as

described in Moran et al. (1 999b), by histological preparation, or by a combination

thereof. Sarnples of the gill, muscle, stomach, pyloric caeca, postenor intestine, kidney,

gonad. and spleen were collected and fixed in Davidson's solution for histological

processing. Sections were stained using H&E. The presence of blood stages was also

assessed using blood smears. Blood was collected fiom the severed caudal vesse1 using

heparinized capillary tubes. The thin blood smears were stained using either LeukoStatTM

(Fisher Scientific, Piasburg, Pemsylvania, USA) or ~iff-~uik~(Baxter Healthcare

Corporation, McGaw Park, Illinois, USA) stains.

2.2. 1997 Experiment

Hatchery-reared, Atlantic salmon (N450)were transferred to the PBS

experimental seawater netpens for the 26-week exposure period of 09 June to 09

December 1997. Mean seawater temperature was 14°C (range=8-22°C) for the duration of the natural exposure period. Samples of 25 salmon were coIlected at 3,6, and 9 weeks p.e. The remaining sample sizes collected at 12, 15, and 26 weeks p.e. were 24, 22, and 9 fish, respectively. Organs, as listed above, were collected and prepared for histoIogical examination in the manner previously described.

Blood, gill, muscle, skin, and intestine samples were collected fiom the first IO fish at each sample date (with the exception of the 26-week sample of 9 fish) for the 1997 study only, and were fixed in 100Y0 ethanol. DNA preparations made fiom each sample were analysed with the K. thyrsiles-specific PCR test described by Hervio et al. (1997).

Samples of 25-50 mg of tissue or 50-100 pi of whoie blood were lysed in 10 mM Tris, 1

mM EDTA, pH 8.0, 1% SDS containing proteinase K (200 pg/ml) for at least 4 h at 37°C

with agitation. Samples were extracted twice using the standard phenol: chloroform:

isûamyl alcohol(50:50: 1) mixture, precipitated with chilled isopropyl alcohol and sodium

acetate and centrifbged at 10 000 rpm. The pelleted DNA was air-dried before king

resuspended in 100 pl of Tris EDTA (TE) buffer (10 mM Tris, 1 mM EDTA, pH 8.0) and stored at 4OC. The amount of DNA collected fiom each sample was quantified using a spectrophotometer. Each solution was diluted in TE to prepare the PCR template working solution (300-500 ng of DNA). PCR reactions (50 pl) included the following constituents: 10X PCR buffer (5 pl); 2mM dNTPs (5 pl); 50mM MgCl2 (1 -5 pl); 20 pMoUp1 Kt 18s1 r (1 -25 pl) (reverse primer); 20 pMoYp1 Kt 18S6f (1 -25 pl) (forward primer); 5 unitdpl Taq (0.25 pl); double distilled Hz0 (32.75 pl); and DNA template (3 pl). The PCR cycles included a 3 min denaturation at 95"C, followed by 30-35 cycles

(94°C for 1 min. 60°C for 1 min, 72°C for 1.5 min). Afier al1 cycles were completed, the process concluded with a 10 min extension at 72OC. Agarose gels (1.25%) were nui at lOOV for 1 hour. Positive results were indicated by a product equivalent to 909 base pairs in size.

3. Resulis

3.1. Ligh t microscope observations

Plasmodia were first detected at 13 weeks p.e. in 1995 and at 9 weeks p.e. in f 997. The earliest stage of development observed was in the somatic musculature, detected at 9 weeks p.e. (Fig. 6.1a). This plasmodium was approxirnately 7 pm in diameter and contained about four interna1 cells (or nuclei). A slightly larger plasmodial stage (1 1 prn) containing several cells also was observed in the somatic musculature (Fig.

6.1 b). Developing plasmodia were detected within the cardiac muscle of Atlantic salmon at 1 7 weeks p.e. (Figs. 6.1 c, d). The plasmodia in the muscle fibers continued to increase in size as the sporogenesis occurred (Fig. 6.2). Polysporic plasmodia contained Mly- formed myxospores in the center and developing myxospores were situated around the periphery of the plasmodia (Fig. 6.2d).

No inflamrnatory response was associated with somatic or cardiac muscle fibers that contained developing plasmodia or intact plasmodia with fully-formed myxospores

(Figs. 6.1.6.2). The host response was evident ody after muscle fibers and the plasmodial wall were lysed, releasing the myxospores and other cytoplasmic constituents into the endomysium. This host response to infection was charactenzed by multifocal, chronic inflammation between the muscle fibers. As well, macrophages engulfed myxospores that were released fiom ruptured plasmodia (Fig. 6.3). Inflammation was first detected at 8 weeks p.e. in 1995, and at 6 weeks p.e. in 1997 (Figs. 6.4,6.5).

3.2. 1995-97 Experiment

The results of this experiment are shown in Figure 6.4. A single presporogonic plasmodium was first detected at 13 weeks p.e. in the somatic musculature. Aggregates of trophozoites and plasmodia containing developing myxospores were observed in the cardiac muscle at 1 7 weeks p.e. (Figs. 6.1c, d). Presporogonic plasmodia reached a PLATE XIV

Figure 6.1. Developing plasmodia of Kudoa thyrsites within the musculature of Atlantic

salmon.

a) Presporogonic plasmodium within the somatic musculature. Scale bar = 30

Cr*-

b) Presporogonic plasmodium containing very early developmental stages of

K. thyrsites. Scale bar = 20 Pm.

C) Early plasmodial stage (arrow) developing within the cardiac musculature.

Scale bar = 70 @m.

d) Increased magnification of Figure 6. lc. Scale bar = 20 Pm.

PLATE XV

Figure 6.2. S porogonic plasmodia of Kudoa rhyrsiies within the somatic musculature of

Atlantic saimon.

a) and b) Early sporogonic plasmodia within the somatic musculature with

developing myxospores. Scaie bars = 40 Pm.

C) Apparent double infection of a muscle fiber by plasmodia. Scaie bar = 70

Pm-

d) Plasmodium with fully-forrned myxospores of K. rhyrsiies. Scale bar = 60

Pm- lllb PLATE XVI

Figure 6.3. Mammation associated with Kudoa fhyrsites infections in Atlantic salmon.

a) Intact polysporic plasmodium within a muscle fiber. The inflammation

within the endomysiwn is not associated with this plasmodium. Scale bar = 80

Pm-

b) Granuloma with myxospores of K. thyrsites between the muscle fibers.

Scale bar = 80 Fm.

C) Chronic, extensive inflammation between the muscle fibers as seen in

cross-section. Scale bar = 200 Pm.

d) Myxospores of K. fhyrsites (arrows) within phagocytes. Scale bar = 20 Fm.

e) Regenerating muscle fibers dwing resolution of infection. Scale bar = 80

Pm-

PLATE XVII

Figure 6.4. Prevaience of Kudoa thymires stages and associated infiammation in the

musculaiure of Atlantic salmon using a combination of histology and wet

mount preparations in fish first exposed 25 May 1995.

PLATE XVIII

Figure 6.5. Prevalence of Kudoa thyrsites stages and associated inflammation in the

musculature of Atlantic salmon using histological examination of fish first

exposed 09 June 1997. The prevalence of infection was also determined by

screening various tissues (Le., blood, gill, muscle, skin, and intestine) using

the K. thyrsites-specific PCR test developed by Hervio et al. (1997).

maximum prevalence of 40% at 26 weeks p.e., and were not observed after this sample.

Infections containing fully-formed myxospores reached a maximum prevalence of 64% at

26 weeks p.e. and fish recovered fiom these infections within 1 year. The pattern of the

inflammatory response showed complete resolution (ive.,was undetectable) by the end of

the experiment (Fig. 6.4).

No stages suggestive of myxosporeans were observed in the blood smears.

3.3.1997 Experiment

Kudoa rhyrsires infections were not detected histologically in the somatic

musculature until9 weeks p-e. (Fig. 6.5). At this time, a single presporogonic

plasmodium was observed (Fig. 6.1 a). The prevalence of inflammation coincided with the

infections (Fig. 6.5). Compared with the previous study, both the prevalence and intensity

of the infections were Iess.

No K. thyrsifes infections were detected in the first 3 weeks p.e. using the PCR

test. The prevalence of infection reached a maximum of 80% (8/10) at 6 weeks p-e., and

a11 of the tissues sarnpled (i.e., blood, gill, muscle, skin, and intestine) were positive for

infection (Fig. 6.5). infections were detected in the muscle and skin at 9 weeks p.e., and

in only the muscle at 12 weeks p-e. At 15 weeks p-e., the infections were detected in the

somatic muscle and skin of one fish, and in ody the intestine of al1 other fish determined

to be positive by PCR. At 26 weeks p.e., the positive signal was from the gill (Fig. 6.6).

4. Discussion

The genus Woais compnsed of species that are described pnmarily as parasites PLATE XIX

Figure 6.6. PCR of DNA from Atlantic salmon tissues screened using the Kudoa

thyrsites-specific PCR test. Atlantic salmon were naturally exposed at the

experimental seawater netpens located at the Pacific Biological Station, in

Nanaimo, British Columbia, beginning 09 June 1997. The parasite DNA was

amplified using the pnmers Kt 18S6f and Kt 18s 1r (1.25% agarose gel,

ethidiurn-bromided stained). Positive control used K. rhyrsires fiom snoek

(Thyrsires alun) collected off South Afica. A positive signal fiom gill tissue,

equivalent to 909 base pairs (bp), indicating a K. thyrsites infection is seen in

lane 5. Negative tissue sarnples are demonstrated in lanes 2-4 and 6-1 1.

Molecular weight markers (bp) are shown in the lefi lane.

of the skeletal musculature (see reviews by Egusa 1986; Lorn and Dykova 1992; Moran et

al. 1999a). Several of the species (e-g., K. lunata, K. miniauriculata, K. paniformis, K.

thyrsires) are intracellular parasites as their site of infection is within the muscle fiber

(Gilchrist 1924; Kabata and Whitaker 198 1 ; Lorn et al. 1983; Whitaker et al. 1996). in

intracellular stages of K. thyrsires, the only boundary between the early developrnental

stages and the muscle fiber was the parasite's ce11 membrane (Figs. 6.1,6.2).

Consequently. K. thyrsites can be regarded as an intracellular parasite of Atlantic salmon.

Plasmodia of histozoic myxosporeans produce spores either following division in

the pansporoblasts or by direct spore morphogenesis (Lom and Dykova 1992).

Pansporoblaçts, which originate by the union of two generative cells, usually develop to

produce two spores (Lom and Dykova 1992). Pansporoblast formation is seen in

myxosporean species that produce large plasmodial trophozoites, such as Sphaeromyxa,

n/@xobolirs,and A4yxidium. Based upon the results of electron microscopic studies of K.

/mata (cf. Lorn and Dykova 1988) and K. paniformis (cf. Stehr 1986), it has been

suggested that Kudoa spp. may not produce pansporoblasts but instead, undergo direct spore morphogenesis. In this process, sporogonic cells produce sporoblasts that give rise to valvogenic, capsulogenic, and sporoplasrnic cells. Each ce11 then assumes its predetermined role in the formation of spores (Lom and Dykova 1992). Direct spore rnorphogenesis is seen in pseudoplasmodia of Sphaerospora and Ceratomyxa, and in some histozoic multivalvulid species (Lom and Dykova 1988). In K. *sites infections of Atlantic salmon, once the parasite has becorne established within the muscle fiber, the polysporic plasmodium apparently does not undergo division, but instead increases greatly in size. The sporoblasts are produced fiom the sporogonic cells within the plasmodia without also fonning pansporoblasts.

Using histology and blood smears, we were unable to observe developrnental stages of K. thyrsifes in sites other than the sornatic and cardiac musculture. The development of the trophozoite stage of myxosporeans within the fish host may include a proliferative phase in tissues or organs different fiom the final site (i.e., extrasporogonic phase). Extrasporogonic stages of Sphaerospora spp. have been observed in the blood, swimbladder, and rete mirabile of the eye, and within the cerebral axons and epidennis for various Myxobolus spp. (cf. Lom and Dykova 1992). This proliferative phase may also induce 'kenorna" formation, as has been observed with M. lieberkuehni infections in pike (Esox lucius) (cf. Lom et al. 1989). Such a proliferative cycle has yet to be described for Kudoa species. However, it has been determined that a stage of K. rhyrsites occurs in the blood and may be transmitted to naive salmon by intraperitoneal injection, albeit with low success (Moran et al. 1999b). Our attempts to observe this stage in blood smears were unsuccessfül which may signiQ that it is transient and undergoes Iittle, if any, proliferation within the blood.

There is brief mention of the presence of early stages of either K. rhyrsites or K. paniformis in the muscle fibers of Pacific hake (Morado and Sparks 1986). However, the

"putative infective stage" observed by Morado and Sparks (1986) appears to be a myxospore engulfed by a macrophage. Gifchrist (1924) provided some details on the early developmental stages of K. thyrsites, but we did not observe the long chahs of cells that he described. Stehr (1 986) provided ultrastnictural details of the generative cells in plasmodia containing fully-formed myxospores of K. paniformis.

In the 1995-97 expriment, presporogonic plasmodia were first detected in the somatic musculature of 4% of the fish at 13 weeks p.e., reached a maximum prevalence of 40% at 26 weeks p-e., and were not obsewed in any of the later samples (Fig. 6.4).

Plasmodia with hilly-formed myxospores were not detected until26 weeks p-e., at which time the prevaience was 64%, using histology. The prevalence of these piasmodia decreased in later samples, but increased again in the 87 weeks p.e. sarnple. This slight increase in prevalence over the previous two samples rnay be the result of a second exposure to K. rhyrsires during the summer of 1996. The higher prevalence of infection observed in later sarnples using wet mount preparations, compared with that using histology, is probably because more tissue was used in the former. As well, the wet mount preparation enables visual detection of myxospores within both plasmodia and in inflamrnatory lesions.

In the 1995-97 experiment, inflammation in the somatic musculature was first observed in salmon at 8 weeks p.e., which was pnor to detection of the parasite in the muscle by microscopy (Fig. 6.4). This may have been associated with a different pathogen, as these fish were exposed in seawater netpens. Conversely, it is possible that this was a mild inflamrnatory response to cryptic, extrasporogonic stages migrating to the muscle fibers. In the 1997 experiment (Fig. 6.5), early stages of the parasite were present in many fish (as demonstrated by the PCR test) that were undetectable using light microscopy.

The general pattern of inflammation coincided with the K. rhyrsites infections in the Atlantic saimon until the host response was completely resolved by 87 weeks p.e.

(Fig. 6.4). The host response was one of chronic inflammation surrounding the parasites that were released fiom disrupted plasmodia. No infiammatory response was observed in the muscle fibers of Atlantic saimon that contained developing plasmodia and plasmodia containing fdly-formed myxospores (Figs. 6.1,6.2). Even when the plasmodia were in close proximity to the sarcolernma, no host response was evident (Fig. 6.2d). Presumably, only after the muscle fiber is ruptured, releasing the rnyxospores and other cytoplasmic constituents of the plasmodia into the endomysium, does the host response occur (Fig.

6.3). In our study, myxospores that were released fiom ruptured plasmodia were actively engulfed by macrophages. Lom and Dykova (1995) described host responses to myxosporean infections as ce11 and tissue reactions that do not usually target the early developmental stages, and significant inflammation is usually not observed until fklly- formed myxospores are observed. Lom and Dykova ( 1992) suggested that this host response may be the result of an increased antigenicity associated with the later developmental stages. Once inflammation occurs in K. thyrsires infections, the myxospores are ingested by macrophages and subsequently destroyed. The inflammatory response is apparently followed by cornplete resolution of the infection and associated lesions (Fig. 6.4). Both Morado and Sparks (1 986) and Stehr and Whitaker (1 986) investigated the host-parasite relationship between Pacific hake and Kudoa infections.

The infections were described as deep within the muscle fibers with plasmodiai growth along the length of the fiber. There was no hypertrophy in the infected fiber, and adjacent, uninfected muscle fibers were not affected. Morado and Sparks (1986) concluded that the host was not able to recognize the parasites developing within the muscle fibers untiI the fiber was either replaced or necrotic. They also observed asynchronous myxospore development and ptasmodial branching in K. paniformis infections of Pacific hake. This branching of K. paniformis plasmodia may give the appearance of multiple infections in some muscle fibers and may explain the appearance, in our experiments, of multiple K. rhyrsifes infections within a muscle fiber (Fig. 6.2~).

Resistance to reinfection has been observed in other myxosporean infections such as PKX and C. shastu (cf. Ferguson 198 1, Banholomew 1998). Ferguson (1 98 1) and

Clifion-Hadley et al. (1986) investigated this resistance in PKX infections in rainbow trout and determined that those fish that survived the initial PKX infections did not develop clinical disease at cooler water temperatures (e-g.,

(1 998) investigated the prevalence of these infections in sexually mature and immature netpen-reared Atlantic salmon and found that sexually mature fish were 13 times more likely to be infected. They suggested that this difference may be the result of irnmunosuppression associated with sexual maturation, therefore predisposïng mature fish to these infections. With the present data, it is impossible to determine whether fish that are infected as smolts are immune to re-infection (i.e., it is unknown whether grilse and reconditioned grilse with K. thyrsifes infections are fish that have recovered fkom previous infections dwing their first year in seawater). This aspect of resistance must be investigated in greater detail with K. thyrsites to detemine whether vaccine development could be a viable alternative to chemotherapeutic treatment.

Histology and wet mount preparations are the two most common methods used for detecting K. thyrsites infections, and they appeared to be equally effective when myxospores are present (Fig. 6.4). However, wet mount preparations are dificult for use in detecting presporogonic plasmodia. This method only detects Mly-forrned myxospores, as the early developmental stages are destroyed during the preparation.

Hervio et al. (1 997) demonstrated the increased sensitivity of their PCR test for detecting

K. thyrsites infections in host tissues when compared with the standard wet mount techniques. The 1997 experiment (Fig. 6.5) followed a similar pattern as the 1995-97 experiment. Unfortunately, the apparent level of natural exposure was too low for conclusive results as both the prevalence and the intensities of infections remained low throughout the expriment. However, the data are important in demonstrating the increased sensitivity of the PCR test in detecting K. thrysites infections when compared with standard screening techniques (i.e., wet mount preparations and histology). An anomalous drop in the prevalence, as detected by the PCR test, occwred at 12 weeks p.e.

This was unlikely the result of testing problems (e-g., contaminated primers) as both the negative and positive controls worked properly, and may actually reflect the problems inherent in small sample sizes. The PCR results that we obtained indicated that there were parasite stages in al1 of the tissues that were screened (i.e., blood, giii, muscle, skin, and intestine). It is likeiy that the positive PCR signals resulted fiom detecting the circulating extrasporogonic stages in the blwd in these organs. With respect to screening for K. thyrsites infections, the PCR assay is more effective in that it not only detects al1 stages of the parasite (when compared to wet mount preparations), but does so by utilizing a greater amount of tissue (25-50 pg versus a 5 pm section). Therefore, PCR increases the possibility of detecting sparsely occuring stages in light to moderately infected fish as compared with histological screening.

The infective stage of M. cerebralis has been identified and cmbe propagated in the labomtory. Using this model, El-Matbouli et al. (1995) provided the most detailed description to date of the early development of myxosporeans within a fish host. The presurnptive aiternate host and infective stage of K. thyrsires has not been identified, thus at present, we can infect fish with K. thyrsites only by natural exposure. Presurnably when this infective stage is found, it will be possible to expenmentally-induce massive infections, which wouid ailow us to more precisely identiQ and describe the extrasporogonic stages and presporogonic plasrnodia of K. ihyrsites. DISCUSSION This dissertation reports on investigations into various aspects of the development and host-parasite interrelationships of the marine myxozoan K. thyrsires in Atlantic salmon, the primary salrnonid species reared by the BC aquaculture industry. Significant advances have ken made in our understanding of: 1) the seasonality of the infective stage; 2) the development in the fish host; 3) the modes of transmission; 4) the host response to infections; and 5) the identity of non-salmonid reservoir hosts.

Probtems may aise when a fish species is introduced into an area outside of its histonc geographical range. For example, such exotic species rnay be highly susceptible to parasites that are endernic to the area into which the fish has ken relocated. Such may be the case with Atlantic salmon culture in BC. Not only are these fish king reared in waters outside of their typical range, but they are reared in a confined environment near the surface of the ocean in relatively shallow water. Kudoa thyrsires infections may be an exampie of this phenomenon, and the associated soft flesh is one of the. if not the, most important health problems in pen-reared Atlantic salmon in BC. This high profile of concem is because the primary site of K. thyrsires infection is the musculature. After the fish are harvested, heavy infections in Atiantic salmon may result in sofi flesh. As this species accounts for 69% of the fmed salmon produced in BC, the potential impact of the disease on the industry is very large. With the expansion of Atlantic salmon fming in BC, the severity of the problem appears to be increasing and will become a serious problem untii control methods are implemented. Exposure to the infection cannot be avoided. Therefore, the most likely methods for control include the development of an effective vaccine, eficacious and environrnentally acceptable dmgs, and a practical method for identifjhg fish at risk for sofi flesh (Le., moderately or heavily infected fish). 1 determined that the K. rhyrsites infective stage could easily bypass the seawater filtration systems at the PBS. Before this project was initiated, there was anecdotal evidence that the K. thyrsites infective stage was prevalent in Departure Bay and that it couid bypass the PBS sand filtration system. Consequently, additional bag and carûidge filters were installed to remove al1 particles larger than 1 prn in size fiom the incoming seawater. During the investigation, it became evident that these additional measures were ineffective at excluding the infective stage fiom the PBS seawater supply. It is unlikely that the infective stage is less than 1 pm in size, but rather, the filters were unable to effectively remove dl microscopic particles greater than this size. This failure to remove the K. fhyrsites infective stage using standard, physical, seawater filtration systems should serve as a warning of the limitations of such filtration to other marine researchers that require uncontaminated, filtered seawater for experimental purposes.

Sources of infective stages of the parasite are dinicult to identi@ when searching for marine myxozoans fiom open bodies of water. The marine myxozoan infective stages may enter the region of the netpens with the uncontrolled exchange of seawater. On the other hand, seawater netpens provide suitable habitats and ample nutrients for iarge annelid populations, the altemate hosts of several myxozoan species. These annelids may be associated with either the benthic or pelagic zones, or with the biofouling of seawater netpens. If an altemate annelid host is required in the life cycle of K. thyrsites, as has been demonstrated in freshwater myxozoans, these annelids may shed significant numbers of actinospores into the seawater netpen environment. Such an increased risk of exposure to an adjacent high density of actinosporeans could explain the high prevalences and intensities of infections reported in mature sahon harvested fiom several of the BC aquaculture industry's seawater netpen sites (St-Hilaire et al. 1998). The seawater intakes of the PBS are located near the experimental netpens (Fig. 2.2), and it is possible that the annelids inhabiting the pens and nearby area act as the source of very large numbers of the infective stages that pass into the filtration system. The filters are unable to cope with such nurnbers and some subsequently "escape" through the filtration system and infect fish in the PBS seawater tanks.

The inability to provide consistently, parasite-free seawater to the experimental tanks at the PBS was a crucial obstacle in these biological investigations. Any experiment that required the use of parasite-fiee seawater could not be performed as the infective stage could not be filtered from the seawater with any degree of confidence. For example, experiments designed to identie the source of the infective stage either could not be performed or had to be terminated prernaturely once the futility of filtration was established. These types of experirnents (e.g., exposure of naive fish to marine biota collected fiom the seawater netpens) require c~~rrned,parasite-fiee seawater.

Ultirnately, the source of the infective stage could not be confidently identified (i.e., the stages could have been released fiom either the marine annelids in an exposure trial or were present within the incoming seawater).

A heteroxenous life cycle has yet to be demonstrated in marine myxozoans.

However, it is widely believed that marine myxozoans require alternate hosts in thek life cycles, as they are required by fieshwater myxozoans (see El-Matbouli et al. 1992; Kent et al. f 994a). Recently, there have been claims of direct transmission of the marine myxozoan, M. leei, to gilthead seabrearn and red dnun (Diamant 1997, 1998). However, this myxozoan is an intestinal parasite and the various developrnental stages may be released in fecal casts. Therefore, it is possible that the parasite was transrnitted by vegetative stages rather than by myxospores.

My study represents the only other successful experimental transmission of a marine myxozoan. However, attempts to transmit the infection among Atlantic salmon by intubating Fieshly harvested K. thyrsites myxospores into naive fish were unsuccessful.

The use of both fiesh and aged myxospores in direct transmission experiments have consistently failed to transmit fieshwater myxozoan infections (Lom and Dykova 1992), and thus, the present intubation results are consistent with the historic inability of researchers to transmit myxozoan infections directly, However, infections were transmitted successfully using blood injections (i.p.). This indicates that an extrasporogonic blood stage exists in the development of K. fhyrsites. inactivation of the blood stage (resulting fiom ovemight storage at 4OC) or the paucity of these stages at the time of collection (Le., blood stages may be prevalent only at certain times of the infection) may have caused the low rate of success. The experïmental transmission of other myxozoan infections (e-g., the PKX myxosporean, C. shasfa, and S. renicota) using blood injections (i.p.) also has ken successful (Clifion-Hadley et ai. 1984, D'Silva et al.

1984; Bower 1985; Kent and Hedrick 1985; Molnir and Kovacs-Gayer 1986).

The lack of research on the transmission of marine myxozoans is probably the result of the difficulties associated with obtaining both parasite-fiee, seawater sources and confimed, parasite-fiee, marine fishes for the expenments. In contrast, the Atlantic salmon-K. rhyrsifes mode1 has proven to be excellent for these investigations. This is the result of the salmon's anadromous physiology; Atlantic salmon reared solely in fiesh water are unequivocally K. rhyrsires-fiee experimental animals as the parasite is strictly marine. The ability to transfer these Amon fiom fiesh water to seawater, and vice versa,

is important in limiting the Iength of exposure to the infective stage. In these

investigations, naturally exposing Atlantic salmon in seawater netpens or tanks in BC proved to be the most effective and consistent way to infect the salmon population, and

longer exposures resulted in both higher prevalences and intensities of infection.

Whether marine myxozoans require alternate hosts in their life cycles, as has been demonstrated with the freshwater species, remains to be elucidated in the laboratory. Kent et al. (1994a) proposed that the actinosporean genera be considered as "collective groups" as these organisms cannot be assigned to any known myxosporean genus. This proposal was based solely upon the results of experiments on fieshwater myxozoan life cycles.

However. rny inability to transmit K. thyrsites infections directiy suggests that alternate hosts also are required for the completion of sorne, if not dl, marine myxozoan life cycles. Therefore, the marine actinosporeans descri bed thus far are pro babl y al temate li fe stages of myxosporeans from marine fish. This leads me to believe that the arguments set forth by Lester et al. (1998) in support of the continued use of the traditional binomial system for descnbing new actinosporean species are contras. to the recornmendations of the International Commission on ZooIogical Nomenclature (1985). Kent and Lom (1 999) explain, in detail, the contradictions within the proposal by Lester et al. (1998). 1 have no doubt that this taxonomic issue will remain a contentious point for the near fùture. A system for categorizing actinosporean stages, such as that proposed by Lom et al. (1997) which is accepted by the majority of rnyxozoan researchers, should be adopted without delay to avoid any Merconfision. Lom et al. (1997) recommended that actinosporean stages be reported in the vemacuiar and that the authority be quoted when ceferring to these stages to avoid confusion (e-g., triactinomyxon actinosporean stage of Marques

1984).

It has been established, without a doubt, that myxozoans are metazoans based on

ultrastructurai and molecular data (Smothers et al. 1994; Siddall et al. 1995; Schlegel et

al. 1996; Anderson et al. 1998). The majority of phylogenetic analyses suggest that the

myxozoans are most closely related to the bilateral animais (Smothers et al. 1994;

Schlegel et ai. 1996; Anderson et ai. 1998). However, the hallmark paper of Siddall et al.

(1 995) and the morphological similarities between myxozoan polar capsules and cnidarian nematocysts should not be ignored. The phylogenetic analyses of Siddall et al.

( 1995) suggest that the myxozoans are most closely related to the midarians, lending creedence to previous hypotheses on the origins of the rnyxozoans (Weill 1938). Until this issue is resolved, the suggestion of Siddall et al. (1995) to suppress the phylum

Myxozoa should be considered as premature.

The presence of the K. thyrsires infective stage in Departure Bay is seasonal. I found that infections of K. thyrsifes were readily contracted in fish exposed between late

May 1996 and eariy December 1996, and not at al1 during the period fiom late December

1996 to early May 1997 (Table 4.1). The prevalence of K. thyrsites infections in fish exposed in June 1997 reached 80%, using the PCR test. Although the seasonality study was concluded in May 1997 these earIier observations were proven by K fhyrsifes infections being contracted by salmon exposed in June 1997, as demonstrated by the PCR results (Chapter 6). These independent observations and the lack of infections during the late 1996 and early 1997 exposures confirm collectively that the infections are seasonal.

1 was able to show that the parasite could sporulate successfulIy in Atlantic salmon held in fiesh water afler the parasite was established in the fish. This finding

permitted the continuation of my seasonality investigations, aithough several months of

results had to be discarded upon confirmation that the infective stage was bypassing the

filtration. However, those experimental transmission investigations that involved the

direct administration of the parasite to the fish (e.g., biood injections (i-p.), intubation)

could be continued without concern of contamination as the fish were maintained in fiesh

water for the entire experiment.

Kudoa thyrsifes infections in Atlantic saimon required approximately 2000

degree-days to produce the fully-formed myxospores. The rate of development of

myxozoan infections may be fast (Le., 2-3 weeks) or relatively slow (Le., 5-6 months)

(Lom and Dykova 1992). This developmental rate (i.e., time fiom uptake of infective

stage to sporulation) of a myxozoan parasite is certainly af5ected by the temperature of the

hostosenvironment. Tlierefore, marine myxozoans in temperate regions tend to develop at

a slower rate than many freshwater temperate species. in temperate climates, the seawater

temperatures fluctuate significantly less than the fieshwater temperatures, and are much

cooler during the surnmer months. As a result, calculating degree-days is necessary to

determine the time required for a myxozoan to sporulate. 1 could not determine exactly

when the fish became infected during the natural exposures. Therefore, 1 had to make the

assumption that the fish contracted the infections immediately upon their transfer to

seawater. Unfortunately, this affected the accuracy of the calculation, but it proved to be of significant assistance in determining sampling times so that I could detect confidently the fully-developed myxospores.

Myxobolus cerebralis infections in rainbow trout require approximately 1800- 2000 degree-days for spodation (RB.Nehring, Colorado Division of Wildlife, pers.

comm.). Kudoa thyrsites myxospores were detec ted as early as 1O00 degree-days p.e.

However. most infections required approximately 2000 degree-days to produce the fully-

formed myxospores. Seawater temperature differs significantly between the PBS seawater

netpens and the tanks (Fig. 4.2), which probably accounts for the 2 month delay of sporulation in the salmon in these tanks supplied with the cooler seawater.

Most Atlantic salmon smolts contract K. thyrsites infections within a relatively short period post-transfer (Fig. 4.3). Stress is known to have a detrimentd effect on the competency of the fish's immune system. Fish are, therefore, more prone to these infections during penods of transition (e-g., smoltification, maturation). Plasma cortisol levels. which are used to monitor stress (and therefore immune compromise), increase significantly in salmon with the onset of smoltification and sexual maturaîion (Maule et al. 1 987; Barton and Iwama 199 1 ; Clarke and Hirano 1995; Fagerlund et al. 1995).

Spore formation in myxozoan plasmodia occurs either fotlowing division within pansporoblasts or by direct spore morphogenesis (Lom and Dykova 1992). Electron rnicroscopic studies of K. lunata and K. panl#!ormisrevealed that these two Kudoa spp. do not form pansporoblasts (Stehr 1986; Lom and Dykova 1988). Whether this is a consistent characteristic of species within this genus has yet to be detennined.

Developmental stages of K. thyrsites were observed using light microscopy, in only the somatic and cardiac musculture. in some myxozoans, the development of the trophozoite stage within the fish host includes an extrasporogonic phase. These extrasporogonic stages have been observed in the blood, swimbladder, eye, cerebral axons, and epidermis (Lom and Dykova 1992). To date, this proliferative cycle has not been descnbed in any Kudoa species. However, the successfiil transmission of K. ihyrsiies infections using blood injections (i-p.) has confumed the presence of a developmental blood stage. The failure to observe this blood stage suggests that the stage is either transient or undergoes only slight proliferation once inside the fish. It is, therefore, probable that the the wide variation in intensities of infections that was observed in these studies resulted fiom fish king infected by exposure to several actinospores in lieu of a single infection with a subsequent massive proliferative cycle.

My study demonstrated the biphasic nature of the infection (Figs. 4.3,6.4). Within a year post-tram fer, the initiai infections were resolved. The prevalence of K. thyrsites infections in maturing Atlantic salmon has ken observed to increase dramatically (St-

Hilaire et al. 1998). This second increase in the prevaience and intensity of the infections suggests that there is either a proliferation of a cryptic stage of the parasite within the fish or a reinfection by the parasite (or possibly a combination of the two). Although unconfirmed, the existence of a cryptic stage within "recovered" fish would explain the occurrence of the infection in a salmon held in fiesh water for 2 years after a brief, 2- week seawater exposure.

The host response to K. thyrsites infections is characterized as multifocal, chronic inflammation in response to parasites released from disrupted plasmodia. The inflammation of the tissues did not appear to be due to the infected muscle fiber containing a developing, intact plasmodium (Figs. 6.1,6.2), regardless of the proximity of the plasmodium to the sarcolemma of the muscle fiber (Fig. 6.2d). However, once an infected muscle fiber ruphwd (due to the increasing size of the plasmodium), and released the plasmodial contents into the endomysium, there was a profound host response (Fig. 6.3). After the onset of inflammation, the myxospores are destroyed by the

macrophages. Thus, this inflammatory response probably plays an important roie in the

resolution of the infection.

Resistance to reinfection has been observed in other myxosporean Uifections such

as PKX and C. shasta (cf. Ferguson 198 1; Bartholomew 1998). With the present data, it

is not possible to detennine whether Atlantic salmon are immune to subsequent K.

ttzyrsires infections. In other words, it is unknown whether the grilse and reconditioned

grilse with K. rhyrsites infections are fish that have recovered fkom previous infections

during their first year in seawater. As discussed previously, the high prevalence of

infection seen in pst-smolts is probably partly the result of the stress associated with

smoltification. and the subsequent immunocompromised state of the fish. The effect of

irnrnunocompromisation on K. rhyrsites infections has ken demonstrated in sexually

mature fish. which were 13 times more likely to be infected than immature fish (St-

Hilaire et al. 1998).

Histology and wet mount preparations are the two most comrnon methods used for detecting K. rhyrsires infections, and they appear to be equaily effective when myxospores are present (Fig. 6.4). Wet mount preparations are dependent upon the sponilation of the parasite which produces the identifiable myxospore and the presence of these myxospores can be detected in both the plasmodia and in the inflammed tissues.

This technique is highly effective in detecting K. rhyrsires infections, and is an effective field technique when screening for problematic infections (i-e., high intensity of infection which may lead to soft flesh).

With adequate training and suitable equipment, wet mount preparations may be used by fann site personnel to screen netpen populations to detennine the prevalence of

K. rhyrsires infections within theu fmed fish. This technique is an inexpensive and rapid

alternative to the histological screening of muscle tissue. However, wet mount

preparations are not effective for use in detecting presporogonic forms in the muscle fiber

because the delicate plasmodia are destroyed during the preparation. Therefore,

recognizing this limitation, periodic and non-lethal screening of the fish using wet mount

preparations should be suggested to farm site managers so that they may accurately

monitor and determine a baseline level of infections within the fish at problematic sites

(St-Hilaire et al. 1998).

Although a PCR test has been developed by Hervio et ai. (1997), the test is too

sensitive for practical. on-site use. The PCR test is so sensitive that the test will identiQ

any fish that is positive for the K. thyrsites infection, regardless of the level of the

infection. Consequently, if this test (in its current state) was used by the industry to screen

fish, it would be impossible to differentiate between severe infections that would result in

soft flesh and light to rnoderate infections that would have no noticable effect on the

visual muscle quaiity. This subsequently would result in the industry discarding salmon

unnecessarily. However, if the sensitivity of the PCR test could be adjusted to detect different infection levels, there would be potential for its use by the industry. For

exampte, it may be possible to adjust the sensitivity so that the number of PCR cycles is equivalent to only high intensity infections and to visualize this by a bright band on the agarose gel.

In this study, the PCR results indicated that there were parasite stages in ail of the tissues that were screened (i.e., blood, gill, muscle, skin, and intestine). However, the detection of parasitic stages outside of the musculature was unsuccessfûl using histology, which suggests that the positive PCR signais were caused by circulating extrasporogonic stages in the blood in these organs. Also, the efficacy of the PCR assay is higher when compared with histological screening. The PCR utilizes a greater amount of tissue (25-50 pg versus a 5 prn section), which increases the possibility of detecting sparsely occurring stages in light and moderate infections.

Presently, the K. thrysite.~-specificPCR test is used most cornmoniy for screening fish tissues to identim early and low-level infections which may be undetectable by routine screening techniques. This PCR test wilI aiso prove to be extremely usefid in the

K. thyrsires Iife cycle investigations. Not only cm it be used for rapidly screening large numbers of potential altemate hosts (Le., annelids) for the K- fhyrsifes actinospore stage, but also in confiming the identity of the actinosporean stage once it has been observed.

Of the 31 non-salmonid fish species belonging to 27 genera that were exarnined for K. ihyrsites infections, five species were identified as potential reservoir hosts for the parasite including arrowtooth flounder, lingcod, rock sole, threadfm sculpin, and tube- snout. Kudoa fhyrsites was originally described fiom snoek collected off the Coast of

South Afnca (Gilchrist 1924). Since this original description, the parasite has ken reported in severai marine fish species world-wide. In BC waters, it is a cornmon parasite of the musculature of wild fishes (primarily non-salmonids). McDonaid and Margolis

(1995) provide detaiied records of 11 host species for K. thyrsites. infections in wild

Pacific salmonids appear limited to the cardiac musculature, and the effect of these cardiac infections on the wild salrnon is not known. WiId marine fishes are known to act as reservoirs for various parasite species that affect fmed saimon (e.g., sea lice L. salmonis, the microsporidian Loma salrnonae) (Bristow and Berland 1991; Hhtein and

Lindstad 1991; Hedrick 1998; Kent et al. 1998). Two of the non-salmonid fishes identified as potential reservoirs of the parasite, namely lingcod and threadfm sculpin, are new host records for K. thyrsites. A second Kudoa sp. was encountered while screening lingcod collected in Barkley Sound, BC, and based upon the spore morphology, 1 concluded that it was K. paniformis.

Kudoa thyrsites in pen-reared Atlantic salmon is indistinguishable fiom those observed in tube-snout and Pacific hake, using SSU rDNA sequences (Hervio et ai.

1997). This suggests that several non-salmonid fish species may act as indirect reservoirs for the infection near the seawater netpens. Non-salmonid fishes tend to reside near the netpens for the protection and food supply offered by the sites. When K. thyrsites myxospores are released following death of an infected fish, there is a significant increase in the number of myxospores in the area immediately adjacent the netpens. This increases the likelihood of exposing altemate hosts (known and unknown species) to the myxospores, which subsequently produce the stages (Le., presumptive actinospores) that are infective to the saimon.

Only after the presumptive altemate host and infective stage of K. thyrsites is identified and the fish can be supplied consistently with parasite-fiee seawater, can the final details of the life cycle be delineated. When the altemate host is identified, it will be possible to induce massive infections experimentally, which would enable researchers to more precisely identify and describe the presporogonic development and ultrastructure of

K. thyrsites.

This study has demonstrated that K. thyrsites is not directly transmitted to fish, and that the irifective stage is seasonal. Atlantic salmon smolts, of the type used to start new culture in the BC aquaculture industry, are highly susceptible to the parasite during their first summer in seawater. Mer transition @ossibly with a lirnited proliferative stage) in the blood, the parasite migrates to the musculature, and forms polysporic phsrnodia within the muscte fibers. Host response to mptured plasmodia is characterized as chronic, multifocal inflammation between the muscle fibers, and infections typically are resolved within a year. Non-salmonid fishes, such as arrowtooth flounder, lingcod, rock sole, threadfin sculpin, and tube-snout, act as indirect reservoirs for the infection. CHAPTER 8

CONCLUSIONS Significant insights into the biology of the marine myxozoan, K. thymifes,afTecting seawater netpen-reared Atlantic salmon were achieved through investigating various aspects of its seasonality, the progression of natural infections, its modes of transmission, and its sequential development in Atlantic salmon hosts.

Physical filtration was not an effective rneans of removing the myxozoan infective

stage fiom seawater that was supplied to fish.

Attempts to transmit the infection arnong Atlantic salmon by intubating fieshly

hmested K. rhyrsires myxospores into naive fish were unsuccessful.

Kudoa thyrsiies was transrnitted successfully to naive fish using blood injections

(Lp.), suggesting that there is an extrasporogonic stage that circulates within the

blood.

Naturally exposing Atlantic salmon in seawater netpens or tanks proved to be the

most effective and consistent way to infect the salmon population with K. thyrsires,

and longer exposures resulted in both higher prevalences and intensities of infection.

The presence of the K. thyrsiles infective stage in Departure Bay, BC, is seasonal.

The parasite can develop and sporulate successfully in Atlantic sdmon held in fiesh

water after the fish has ken exposed to contaminated seawater.

The rate at which K. fhyrsites infections progress is dependent upon water

temperature. Kudoa rhyrsites myxospores were detected as early as 1000 degree-days

p-e.; however, most infections required approximately 2000 degree-days to produce

fùlly-formed myxospores.

Most Atlantic salmon smolts contract K. thyrsites infections within a short penod post-transfer to seawater.

9. Using Iight microscopy, developmentai stages of K. thyrsites were not observed in

tissues other than the somatic and cardiac musculature.

10. The failure to observe the blood stage suggests that the stage is eithcr transient or

undergoes only slight proliferation.

1 1. There is a biphasic nature to K. thyrsites infections in Atlantic salmon.

1 2. Host response to K. ihyrsites infections was c haracterized as multi focal, chronic

inflammation in response to parasites that were released fiom disrupted plasmodia.

The inflamrnatory response probably plays an important role in the resolution of the

infection as the released myxospores are destroyed by macrophages.

1 3. Histological preparations and wet mounts appeared to be equally effective for

detecting K. rhyrsifes infections when myxospores are present.

14. The PCR test is so sensitive that the test will identify any fish that is positive for K.

rhyrsites infection, regardless of the level of infection.

15. The PCR results indicated that there were parasite stages in al1 of the tissues that were

screened. although they were undetectable using histological methods.

16. Forty-four species of Kudoa have been described fiom marine and estuarine fishes

world-wide. The majority of Kudoa spp. are histozoic parasites of the musculature;

however, a few coelozoic species have been described.

17. Of the 37 species of non-salmonids examined for infection, five species were

identified as reservoir hosts for K. thyrsites including arrowtooth flounder, lingcod,

rock sole, threadfin sculpin, and tube-snout. The reports of lingcod and theadfin

sculpin represent new host records. CHAPTER 9

LIST OF REFERENCES Anderson, C.L., Canning, E.U., and Okamura, B. 1998. A triploblast origin for M yxozoa? Nature (Lond.) 392: 346-347.

Anderson, J.A. 1985. A morphological and histological study of Kudoa spp. (Myxozoa, Myxosporea) in four Pacific marine fishes. MSc. thesis, Oregon State University, Corvallis, Oregon.

Barja, J.L., and Toranzo, A.E., 1993. Myoliquefaction pst-mortem caused by the my;uosporean Kudoa thyrsites in reared Atlantic salmon in Spain. Bull. Eur. Assoc. Fish Pathol. 13: 86-88.

Bartholomew, J.L. 1998. Host resistance to infection by the myxosporean parasite Ceratomyxa shasra: a review. J. Aquat. Anim. Health 10: 1 12- 120.

Bartholomew, J.L., Smith, C.E., Rohovec, J.S., and Fryer, J.L. 1989. Characterization of the host response to the myxosporean parasite, Ceratonryxa shasta (Noble), by histology, scanning electron microscopy and immunological techniques. J. Fish Dis. 12: 509-522.

Bartholomew, J.L., Whipple, M.J., Stevens, D.G., and Fryer, J.L. 1997. The life cycle of Ceratomyxa shasta, a myxosporean parasite of salmonids, requires a fieshwater polychaete as an alternate host. J. Parasitol. 83: 859-868.

Barton, B.A., and Iwama, G.K. 1991. Physiological changes in fish from stress in aquaculture with ernphasis on the response and effects of corticosteroids. Annu. Rev. Fish Dis. 1: 3-26.

Bellenid, B.L., Pote, L.M., Lin, T.L., Johnson, M.J., and Boyle, C.R. 1995. Etiologicai and epizootological factors associated with outbreaks of proliferative gill disease in channel catfish. J. Aquat. Anim. Health 7: 124-1 3 1.

Blazer, V.S. 1991. Piscine macrophage fùnction and nutritional influences: a review. J. Aquat. Anim. Health 3: 77-86.

Bower, S.M. 1985. Ceraiomyxa shasta (Myxozoa: Myxosporea) in juvenile chinook salmon (Oncorhynchus tshawytscha) : experimental transmission and natural infections in the Fraser River, British Columbia. Can. J. 2001.63: 1 737- 1740.

Boyce, N.P., Kabata, Z., and Margolis, L. 1985. Investigations of the distribution, detection, and biology of Henneguya salminicola (Protozoa, Myxozoa), a parasite of the flesh of Pacific salmon. Can. Tech. Rep. Fish. Aquat. Sci. No. 1405.

Brackett, J., and Karreman, G. 1998. Disease treatrnent in netpen aquaculture. In Diseases of seawater netpen-reared saimonid fishes. Edited by M.L. Kent and T.T. Poppe. Quadra Printers Ltd., Nanaimo, British Columbia. pp. 9- 16. Bristow, G.A., and Berland, B. 199 1. A report on some metazoan parasites of wild marine salmon (Salmo salar L.) fiom the West coast of Norway with comments on theu interactions with farmed sahon. Aquaculture 98: 3 1 1-3 18.

British Columbia Environmental Assessment Office. 1997. Salmon aquaculture review. Report of the Environmental Assessment Office, Vol. 1. Queen's Printer for British Columbia Victoria, British Columbia.

Bunton, T.E., and Poynton, S.L. 199 1. Kudoa sp. (Myxosporea, Multivaivulida) infection in juveni le white perch, Morone americanu (Grnelin): histopathology and spore morphology. J. Fish Dis. 14: 589-594.

Canning, E.U., Okamura, B., and Curry, A. 1996. Development of a myxozoan parasite Terracapsula bryozoides gen.n. et sp-n. in Cristarella mucedo (Bryozoa: Phylactolaemata). Folia Parasitol. 43: 249-26 1.

Castro. R.R., and Burgos, R. 1996. Kudoa fhyrsires (Myxozoa, Multivalvulida) causing .'milky condition" in the musculature of Parulichrhys ahpersus (Neopterygii, Pleuronectifonnes, Paralichthyidae) fiom Chile. Mem. Inst. Oswaldo Cruz 91: 163-164.

Ching, H.L.. and Munday, D.R. 1984. Geographic and seasonal distribution of the infectious stage of Ceratomyxa shasfa Noble, 1 950, a myxozoan salmonid pathogen in the Fraser River system. Can. J. 2001.62: 1075- 1080.

Clarke, W.C., and Hirano, T. 1995. Osmoregulation. In Physiological ecology of Pacific salmon. Edited by C. Groot, L. Margolis, and W.C. Clarke. UBC Press, University of British Columbia, Vancouver, British Columbia. pp. 3 17-377.

Clifion-Hadley, R.S., Bucke, D., and Richards, R.H. 1984. Proliferative kidney disease of salmonid fish: a review. J. Fish Dis. 7: 363-377.

Clifion-Hadley, R.S.. Richards, R.H.,and Bucke, D. 1986. Proliferative kidney disease (PKD) in rainbow trout Salmo gairdneri: Merobservations on the effects of water temperature. Aquaculture 55: 165- 171.

Csaba, G. 1976. An unidentifiable extracellular sporozoan parasite fiom the blood of the carp. Parasitol. Hung. 9: 21-24.

Davies. A.J., Andrews, T., Upton, N.P.D.,and Matthews, R.A. 1998. Kudoa as a contributary cause of muscle necrosis in gobies. Bull. Ew. Assoc. Fish Pathol. 18: 62-66.

Diamant, A. 1997. Fish-to-fish transmission of a marine myxosporean. Dis. Aquat. Org. 30: 99-105. Diamant, A. 1998. Red dmSciaenops ocelfafus(Sciaenidae), a recent introduction to Mediterranean mariculature, is susceptible to Myxidium leei (Myxosporea). Aquaculture 162: 33-39.

Di Antonio. E.. and Cenci Goga, B.T. 1993. Reperti parassitari riferibili a cisti di Kudoa clupeide ne1 tessuto muscolare di merluzzi (Merluccius hubbsj] congelati. Boll. Soc. Ital. Patot. 11: 44-52.

D'Silva, J., Mulcahy, M.F.. and de Kinkelin, P. 1984. Experimental transmission of proliferative kidney disease in rainbow trout, Safrno gairdneri Richardson. J. Fish Dis. 7: 235-239.

Duarte, S.A., Masser, M.P.? and Plumb, J.A. 1993. Seasonal occurrence of diseases in cage-reared channel catfish, 1987- 1991. J. Aquat. Anim. Health 5: 223-229.

Dykova. I., Lom, J., and Overstreet, R.M. 1994. Myxosporean parasites of the genus Kudoa Meglitsch, 1947 fiorn some Gulf of Mexico fishes: description of two new species and notes on their ultrastructure. Eur. J. Protistol. 30: 3 16-323.

Egusa, S. 1 983. Disease problems in Japanese yellowtail, Seriola quinqueradiata, culture: a review. Rapp. P.-v. Réun. Cons. Int. Explor. Mer 182: 10-18.

Egusa, S. 1986. The order Multivalvulida Shulman, 1959 (Myxozoa; Myxosporea): a review. Fish Pathol. 21: 26 1 -274 [in Japanese].

Egusa. S., and Nakajima, K. 1978. Kudoasis of cultured yellowtail. Fish Pathol. 13: 1-7 [in Japanese] .

Egusa, S., and Nakajima, K. 1980. Kudoa amamiensis n-sp. (Myxosporea: Multivalvulida) found in cultwed yeIlowtails and wild damselfishes from Amami- Ohshirna and Okinawa, Japan. Bull. Jpn. Soc. Sci. Fish. 46: 1193-1 198.

Egusa, S., and Shiomitsu, T. 1983. Two new species of the genus Kudoa (Myxosporea: Multivalvulida) fiom marine cultured fishes in Japan. Fish Pathol. 18: 163- 1 7 1 [in Japanese].

El-Matbouli, M., and HofIinann, R. 1989. Experimental transmission of two Myxoboh spp. developing bisporogeny via tubifkid wonns. Parasitol. Res. 75: 461-464.

El-Matbouli, M., and Hohann, R. W. 199 1. Prevention of experimentally induced whirling disease in rainbow trout Oncorhynchus mykiss by Fumagillin. Dis. Aquat. Org. 10: 109-1 13.

El-Matbouli, M., Fischer-Scherl, T., and Hoffmann, R.W. 1992. Present knowledge on the life cycle, taxonomy, pathology, and therapy of some Myxosporea spp. important for fieshwater fish. Annu. Rev. Fish Dis. 2: 367-402. El-Matbouli, M., Hoffinann R. W., and Mandok, C. 1995. Light and electron rnicroscopic observations on the route of the triactinomyxon-sporoplasm of Myxobolus cerebralis from epidennis into rainbow trout cartilage. J. Fish Biol. 46: 919-935-

Elston, R. 1994. Kudoa in farmed Atlantic salmon in Washington. In Kudoa Workshop Proceedings. Edited by D.C. Conley. Province of British Columbia, Ministry of Agriculture, Fisheries and Food, Aquaculture Industry Development Report 94-0 1. pp. 12.

Escos, J., Alados, CL., Emlen, J.M., and Alderstein, S. 1995. Developmental instability in the Pacific hake parasitized by myxosporeans Kudoa spp. Trans. Am. Fish. Soc. 124: 943-945.

Eveiyn, T.P.T., Kent, M.L., Poppe, T.T., and Bustos, P. 1998. Bacterial diseases. In Diseases of seawater netpen-reared salmonid fishes. Edited by M.L. Kent and T.T. Poppe. Quadra Pnnters Ltd., Nanaimo, British Columbia. pp. 17-35.

Fagerlund, U.H.M., McBride, J.R., and Williams, I.V. 1995. Stress and tolerance. In Physiological ecology of Pacific salmon. Edited by C. Groot, L. Margolis, and W.C. Clarke. UBC Press, University of British Columbia, Vancouver, British Columbia. pp. 459-503.

Ferguson, H. W. 198 1. The effects of water temperature on the development of proli ferative kidney disease in rainbow trout, Salmo gairdneri Richardson. J. Fish Dis. 4: 175- 177.

Foott, J.S., and Hedrick, R.P. 1987. Seasonal occurrence of the infectious stage of proliferative kidney disease (PKD)and resistance of rainbow trout, Salmo gairdneri Richardson, to reinfection. J. Fish Biol. 30: 477-483.

Fryer, J.L., and Sanders, J.E. 1970. investigation of Ceratomyxa shasta, a protozoan parasite of salmonid fish. Abstracts Fom the Second International Congres of Parasitology, Washington, D.C. J. Parasitol. 56: 422-423.

Fuj ita, T. 1930. On a new Myxosporidia in the muscle of the gray-mullet Chloromyxum bora nov.sp. Dobutsugaku Zasshi 42: 45-48 [in Japanese].

Gaevskaya, A.V., and Kovaleva, A.A. 1981. Infestation of commercial species from the eastern central Atlantic by Kudoa nova. Rybn. Khoz. 4: 48 [in Russian].

Gelormini, N. 1943 (1 944). Un nuevo parisito de la merluza- Rev, Fac. Agron. Vet. Univ. B. Aires 10: 458-463.

Gilchrist, J.D.F. 1924. A protozoal parasite (Chlorornyxurn thyrsites sp-n.) of the Cape sea-fish, the "snoek" (Thyrsites atun, Euphr.). Trans. R. Soc. S. Afi. 11: 263-273. Hahn. C. W. 19 15. Sporozoon parasites of certain fishes in the vicinity of Woods Hole, Massachusetts. Fish. Bull. (Wash. D.C.) 33: 193-214.

Hahn, C. W. 1917. On the sporozoon parasites of the fishes of Woods Hole and vicinity. III. On the Chlorornyxurn clupeidae of Clupea harengus (Young), Pornolobus pseudoharengus (Young), and P. aestivalis (Young). J. Parasitol. 4: 13-20.

Hallett. S.L., O'Donoghue, P.J., and Lester, R.J.G. 1997. Infections by Kudoa ciliatae (Myxozoa: Myxosporea) in Indo-Pacific whiting Sillago spp. Dis. Aquat. Org. 30: 1 1- 16.

Hallett, S.L., O' Donoghue, P.J., and Lester, R.J.G. 1998. Structure and development of a marine actinosporean, Sphaeractinom~onersei n.sp. (Myxozoa). J. Eukaryot. Microbiol. 45: 142- 150.

Harrell, L.W., and Scott, T.M. 1985. Kudoa thyrsitis (Gilchnst) (Myxosporea: Multivalvulida) in Atlantic salmon. Salmo salar L. J. Fish Dis. 8: 329-332.

Hastein. T., and Lindstad, T. 1991. Diseases in wild and cultured saimon: possible interaction. Aquaculture 98: 277-288.

Heckmann, R.A., and Jensen, L.A. 1978. The histopathology and prevalence of Henneguya sebasta and Kudoa clupeidae in the rockfish, Sebastes paucispinis of southem California. J. Wildl. Dis. 14: 259-262.

Hedrick, R.P. 1998. Relationships of the host, pathogen, and environment: implications for diseases of cultured and wild fish populations. J. Aquat. Anim. Health 10: 107-1 11.

Hedrick, R.P., Groff, J.M., Foley, P., and McDowell, T. 1988. Oral administration of Fumagillin DCH protects chinook salmon Oncorhynchus fshawytscha from expenmentally-induced proliferative kidney disease. Dis. Aquat. Org. 4: 165- 168.

Hervio. D.M.L.. Kent, M.L., Khattra, J., Sakanari, J., Yokoyama, H., and Devlin, R.H. 1997. Taxonomy of Kudoa species (Myxosporea), using a small-subunit ribosornal DNA sequence. Cm. J. Zool. 75: 21 12-21 19.

Heupel, M.R., and Bennett, M.B. 1996. A myxosporean parasite (Myxosporea: Multivalvulida) in the skeletal muscle of epaulette sharks, Hemiscyllium ocellarum (Bornaterre), fiom the Great BherReef. J. Fish Dis. 19: 189-191.

Higgins, M.J., and Kent, M.L. 1996. Field trials with fwnagillin for the control of proliferative kidney disease in coho salmon. Prog. Fish-Cult. 58: 268-272.

Higgins, M.J., and Kent, M.L. 1998. TNP-470,the analogue of fûmagillin-DCH, controls PKX in naturaily infected sokeye salmon, Oncorhynchus nerka (Walbaum), underyearlings. J. Fish Dis. 21: 455-457. Higgins, MJ., Margolis, L., and Kent. M.L. 1993. Arrested development in a fieshwater myxosporean, Myxidium salvelini, following transfer of its host, the sockeye salmon (Oncorhynchus nerka), to sea water. J. Parasitol. 79: 403-407.

Higgins, M.J., Kent, M.L., Moran, J.D. W., Weiss, L.M.. and Dawe, S.C. 1998. Efficacy of the fùmagillin analog TNP-470 for Nucleospora sahonis and Loma salmonae infections in chinook salmon Oncorhynchus tshmvytscha. Dis. Aquat. Org. 34: 45-49.

Hoffman, G.L.. and Putz, R.E. 1969. Host susceptibility and the effect of ageing, freezing, heat, and chemicals on spores of Myxosorna cerebralis. Prog. Fish-Cult. 31: 35- 37.

Holliman, A. 1994. Sea trout. Salrno frufta,a new host for the myxosporean Kudoa thyrsites (Gilchrist)? Vet. Rec. 134: 524-525.

Humason, G.L. 1979. Animal tissue techniques. W.H. Freeman Co., San Francisco, California.

International Commission on Zoological Nomenclature. 1985. International Code of Zoological Nomenclahue, 3rd Ed. University of California Press, Berkeley.

Iskov. VA4. 1989. Myxosporea. No. 4 of Vol. 37. In Sporozoa, Cnidosporidia, Microsporidia. Edited &y A.P. Markevich and S.S. Shulman. In Fauna of the Ukraine. Edired by V.A. Topachevski. Naukova Dumka, Kiev, 209 pp. [in Russian].

Iversen, E.S., and Van Meter, N.N. 1967. A new myxosporidian (Sporozoa) infecting the Spanish mackerel. Bull. Mar. Sci. 17: 268-273.

Johnson, C.A. 1969. A redescription of Myxidium chelonarum Johnson, 1969 (Cnidospora: Myxidiidae) from various North Arnerican turtles. J. Protozooi. 16: 700- 702.

Johnson, S.C. 1998. Crustacean parasites. In Diseases of seawater netpen-reared salmonid fishes. Edifed by M.L. Kent and T.T. Poppe. Quadra Printers Ltd., Nanaimo, British Columbia. pp. 80-90.

Johnson, S.C., and Albright, L.J. 1992. Effects of cortisol implants on the susceptibility and the histopathology of the responses of naive coho salmon Oncorhynchus kisutch to experimental infection with Lepeophtheirus sahonis (Copepoda: Caligidae). Dis. Aquat. Org. 14: 195-205.

Joy, LE. 1972. A new species of Kudoa (Myxosporidea: Chloromyxidae) from the spot, Leiostomus xanthurus Lacépède, in Clear Lake, Texas. J. Protozool . 19: 264-265.

Kabata, Z. 1960. On two myxosporidian parasites of marine fishes, including one new species (Ceratomyxa tenuispora). Am. Mag. Nat. Hist. Ser. 13.3: 305-308. Kabata, Z., and Whitaker, D.J. 198 1. Two species of Ku& (Myxosporea: Mu1tivalvulida) parasitic in the flesh of MerZuccius productus (Ayres, 18 55) (Pisces: Teleostei) in the Canadian Pacific. Can. J. 2001. 59: 2085-2091.

Kabata, Z., and Whitaker, D.J. 1986. Distribution of two species of Kudoa (Myxozoa: Multivalvulida) in the offshore population of the Pacific hake, Merluccius productus (Ayres, 1855). Can. J. 2001.64: 2 103-2 1 10.

Kabata, Z., and Whitaker, D.J. 1989. Kudoa thyrsites (Gilchrist, 1924) (Myxozoa) in the cardiac muscle of Pacific salmon (Oncorhynchus spp.) and steelhead trout (Salm0 gairdneri). Can. J. 2001.67: 341-342.

Kabata. Z., Whitaker, D.J., and Bagshaw, J.W. 1986. Kudoa rhyrsitis (Gilchrist) (Myxosporea: MuItivalvulida) in coho salmon, Oncorhynchus kisutch (Walbawn). Can. J. Zool. 64: 1038-1040.

Kalavati, C., and Anuradha, 1. 1993. Two new species of myxosporeans infecthg Valarnugil cunnesius in Visakhapatnarn harbour, east Coast of India. Uttar Pradesh J. Zool. 13: 148- 152.

Kent, M.L. 1992. Diseases of seawater netpen-reared salmonid fishes in the Pacific northwest. Can. Spec. Publ. Fish. Aquat. Sci. No. 116.

Kent, M.L., and Hedrick, R.P. 1985. Transmission of the causative agent of proliferative kidney disease (PKD)with the blood and spleen of infected fish; merevidence that the PKX parasite belongs to the phylum Myxozoa. Bull. Eur. Assoc. Fish Pathol. 5: 39-42.

Kent. M.L., and Hedrick, R.P. 1986. Development of the PKX myxosporean in rainbow trout Salmo gairdneri. Dis. Aquat. Org. 1: 169- 1 82.

Kent, M.L., and Palenzuela, 0. 1999. Phylum Myxozoa Grassé, 1970. In Encyclopedia of Life Sciences, MacMillan Reference Ltd. (Stockton Press), Basingstoke, Hampshire, England (in press).

Kent, M. L., and Poppe, T.T. 1998. Diseases of seawater netpen-reared salmonid fishes. Quadra Printers Ltd., Nanaimo, British Columbia.

Kent, M.L., and Lom, J. 1999. A rebuttal to the proposal of Lester et al. (1998) to describe new species of Myxozoa based solely on their actinosporean stage. Parasitol. Today (in press).

Kent, M.L., Whitaker, D.J., and Margolis, L. 1993. Transmission of Myxobolus arcticus Pugachev and Khokhlov, f 979, a myxosporean parasite of Pacific salmon, via a triactinomyxon from the aquatic oligochaete Stylodrilus heringianus (Lumbriculidae). Cm. J. 2001. 71: 1207-121 1. Kent. M.L., Margolis, L., and Corliss, J.O. 1994a. The demise of a class of protists: taxonornic and nomenclahual revisions proposed for the protist phylum Myxozoa Grassé, 1970. Can. J. 2001.72: 932-937.

Kent, M.L., Margolis, L., Whitaker, D.J., Hoskins, G.E., and McDonald, T.E. 1994b. Review of Myxosporea of importance in saimonid fisheries and aquaculture in British Columbia. Folia Parasitol. 41: 27-37.

Kent. M.L., Traxler, G.S., Kieser, D., Richard, J., Dawe, SC, Shaw, R.W., Prosperi- Porta. G., Ketcheson, J., and Evelyn, T.P.T. 1998. Survey of saimonid pathogens in ocean-caught fishes in British Columbia, Canada. J. Aquat. Anim. Health 10: 2 1 1-2 19.

Kovaleva. A.A., and Gaevskaya, A.V. 1983. The first report of fish Myxosporidia fiom the open waters of the southeastern Pacific Ocean. Vestn. 2001. 1: 6-1 1 [in Russian].

Kovaleva, A.A., and Gaevskaya, A.V. 1984. New species of the genera Kudoa and Pentacapsula (Myxosporidia, Multivalvulea) with unusual location. 2001. Zh. 63: 1090- I 092 [in Russian].

Kovaleva, A.A., Shulrnan, S.S., and Yakovlev, V.N. 1979. Myxosporidia of the genus Kudoa (Myxosporidia, Multivalvulea) of the Atlantic Ocean basin. In The Systematics and ecology of the Sporozoa and Cnidosporidia. Edited by M.V. Krylova. Tr. 2001. bt. .4kad. Nauk SSSR 87: 42-64 [in Russian].

Krasin, V.K. 1976. Myxosporidian and microsporidian infections of the musculature of fish in the north-eastem part of the Pacific Ocean. in Kratkie tezisy dokladov II Vseso yuznogo simpoziuma po parazitam i bde~yammorski kh zhivotnykh. Ministerstvo Rybn. Khoz. SSSR, AtlantNIRO, Kaliningrad, U.S.S .R. pp. 35-36 [in Russian].

Kudo. G., Bamett, H.J., and Nelson, R.W. 1987. Factors afTecting cooked texture quaiity of Paci fic whiting, Merlirccius producrus, fillets with particular emphasis on the effects of infection by the myxosporeans Kudoa panij0rmis and K. thyrsitis. Fish. Bull. (Wash. D.C.)85: 745-756.

Kudo, R.R. 19 19 (1 920). Studies on Myxosporidia. 111. Biol. Monogr. 5: 24 1-503.

Langdon, J.S. 1990. Observations on new Myxobolus species and Kudoa species infecting the nervous system of Australian fishes. J. Appl. Ichthyol. 6: 107-1 16.

Langdon, J.S. 199 1. Myoliquefaction pst-mortern ('milky flesh') due to Kudoa thyrsites (Gilchrist) (Myxosporea: Multivalvulida) in mahi mahi, Coryphaena hippurus L. J. Fish Dis. 14: 45-54. Langdon, J.S.. Thome, T., and Fletcher, W.J. 1992. Reservoir hosts and new clupeoid host records for the myoliquefactive myxosporean parasite Kudoo thyrsites (Gilchrist). J. Fish Dis. 15: 459-47 1.

Laskowski, M., and Kato. 1. 1980. Protein inhibitors of proteinases. Annu. Rev. Biochem. 49: 593-626.

Lester, R.J.G., Hallen, S.L., El-Matbouli, M., and Canning E.U. 1998. The case for naming actinosporeans using the Zoological Code. Parasitol. Today 14: 476477.

Levine. N.D.. Corliss. J.O., Cox. F.E.G.. Derou,. G.. Grain, J.. Honigberg, B.M., Leedale, G.F.. Loeblich, A.R., Lom, J., Lynn, D., Merinfeld, E.G., Page, F.C., Poijansky, G.. Sprague. V., Vavra, J., and Wallace, F.G. 1980. A newly revised classification of the Protozoa. J. Protozool. 27: 37-58.

Llich, L.R., Teh, L.V., Luque, J.L, and Oliva, M.E. 1991. Aspectos cuantitativos de la infeccion por Kudoa sciaencre TerAn, Llich y Luque, 1990 (Myxozoa: Multivalvulidae) en la musculatura somatica de Stellijêr rninor (Pisces: Sciaenidae). Estud. Oceanol. 10: 123-125.

Lom? 1. 1987. Myxosporea: a new look at long-known parasites of fish. Parasitol. Today 3: 327-332.

Lom. J.. and Arthur, J.R. 1989. A guideline for the preparation of species descriptions in Myxosporea. J. Fish Dis. 12: 15 1-1 56.

Lom, J.. and Dykovk 1. 1988. Sporogenesis and spore structure in Kudoa lunata (Myxosporea, Multivalvulida). Parasitol. Res. 74: 52 1-530.

Lom, J., and Dykovh, 1. 1992. Protozoan parasites of fishes, Elsevier Science Publishers, Amsterdam.

Lom. J., and DykovA, 1. 1995. Myxosporea (phylum Myxozoa). in Fish diseases and disorders, Volume 1: Protozoan and metazoan infections. Edited by P.T.K.Woo. CAB International, Wallingford, United Kingdom. pp. 97- 148.

Lorn, J., Dykovk L, and Lhotiikovii, 9. 1983. Kudoa Iunata n-sp. (Myxozoa, Myxosporea) and notes on the nature of muscular "cysts" of the genus Kudoa. Arch. Protistenkd. 127: 387-397.

Lom , J., Dykovi, I., and Feist, S. 1989. Myxosporea-induced xenorna formation in pike (Esox lucius L.) renal corpuscles associated with Myxidium iieberkuehni infection. Eur. J. Protistol. 24: 27 1-280. Lom, J., Rohde, K., and Dykovi, 1. 1992. Studies on protozoan parasites of Austraiian fishes 1. New species of the genera Coccomyxa Léger et Hesse, 1907, Orthoiinea Shulman, 1962 and Kudoa Meglitsch, 1947 (Myxozoa, Myxosporea). Folia Parasitol. 39: 289-306.

Lom. J., McGeorge. J., Feist. S.W., Morris, D., and Adams. A. 1997. Guidelines for the unifonn characteridon of the actinosporean stages of parasites of the phylum Myxozoa. Dis. Aquat. Org. 30: 1-9.

Maeno, Y ., Nagasawa, K., and Sorimachi, M. 1993. Kudoa intestinalis n-sp. (Myxosporea: Multivalvulida) fiom the intestinal musculature of the striped mullet, Mugi2 cephah, from Japan. J. Parasitol. 79: 1 90- 1 92.

Margolis, L. 1982. Parasitology of Pacific salmon - an overview. In Aspects of parasitology - a Festschnfi dedicated to the fiftieth anniversary of the Institute of Parasitology of McGill University, 1932- 1982, Edized by E. Meerovitch. McGill University. Montreai, Quebec. pp. 1 35-226.

Margolis, L. 1993. A case of forensic parasitology. J. Parasitol. 79: 461 -462.

Markiw, M.E., and Wolf, K. 1983. Myxosoma cerebralis: (Myxozoa: Myxosporea) etiologic agent of sdmonid whirling disease requires tubificid worm (helida: Oligochaeta) in its Iife cycle. J. Protozool. 30: 56 1-564.

Marques, A. 1984. Contribution à la co~aissancedes Actinomyxidies: ultrastructure, cycle biologique, systématique. Ph.D. thesis, Université des Sciences et Techniques du Languedoc, Montpellier, France.

Marsal, R. 1994. The implications of Kudoa thyrsites for the British Columbia salmon fming industry. M-Aq. professional paper. Simon Fraser University, Burnaby, B.C.

~Mateo,ES. 1972. Investigacion parasitologica de la merluza (Merluccius gayii peruanus). Convenio entre el Instituto del mar y la Universidad Nacional Federico Villarreal, Inst. Mar Peru (Callao) Inf. No. 40.

Matsumoto, K. 1954. On the two new Myxospondia, Chlorornymrm musculoliquefaciens spnov. and Neochloromyxum crucifomurn gen. et sp.nov., fiom the jellied muscle of sword fish, Xiphias gladius Linné, and common Japanese sea-bas, Lateolabrmjaponicus (Temmink et Schlegel). Bull. Jpn. Soc. Sci. Fish. 20: 469-479.

Maule, A.G., Schreck, C.B., and Kaattari, S.L. 1987. Changes in the immune system of coho salmon (Oncorhynchus kisutch) diiiing the parr-to-smolt transformation and after implantation of cortisol. Can. 1. Fish. Aquat. Sci. 44: 16 1 - 166.

McAllister, C.T., and Trauth, S.E. 1995. New host records for Myxidium serotinum (Protozoa: Myxosporea) fiom North Amencan arnphibians. J. Parasitol. 81: 485-488. McDonald, T.E., and Margolis, L. 1995. Synopsis of the parasites of fishes of Canada: supplement (1978- 1993). Can- Spec. Publ. Fish. Aquat. Sci. No. 122.

Meglitsch, P. A. 1947. Studies on Myxosporidia fiom the Beaufort Region. II. Observations on Kudoa clupeidae (Hahn), gen. nov. J. Parasitol. 33: 271-277.

Meglitsch, P.A. 1948. On Kudoafinduli (Hahn). Trans. Am. Microsc. Soc. 67: 271-274.

Mitchell, L.G. 1989. Myxobolid parasites (Myxozoa: Myxobolidae) infecting fishes of western Montana, with notes on histopathology, seasonality, and intraspecific variation. Can. J. Zool. 67: 1915- 1922.

Molnhr, K. 1979. Myxobolus pmlovskii (Akhmerov, t 954) (M yxosporidia) - infection in the silver carp and bighead carp. Acta Vet. Acad. Sci. Hung. 27: 207-2 16.

Molnir, K., and Kovacs-Gayer, É. 1986. Experimental induction of Sphaerospora renicola (Myxosporea) infection in common carp (Cyprinus carpio) by transmission of SB-protozoans. J. Appl. Ichthyol. 2: 86-94.

Morado, J.F., and Sparks, A. 1986. Observations on the host-parasite relations of the Pacific whiting. Merlucciusproducrus (Ayres), and two myxosporean parasites, Kudoa thyrsiris (GiIchrist? 1924) and K. panryormis Kabata & Whitaker, 198 1. J. Fish Dis. 9: 345-455.

Moran, J.D. W ., and Kent, M.L. 1 999. Kudoa thyrsites (Myxozoa: Myxosporea) infections in pen-reared Atlantic salmon in the eastem North Pacific Ocean, with a survey of potential nonsalmonid fish reservoir hosts. J. Aquat. Anim. Health (in press).

Moran, J.D. W., Kent, M.L., Khattra, J., and Webster, J.M. 1998. Endogenous development of the rnposporean parasite associated with pst-mortem myoliquefaction. Abstracts of the 39th Annual Western Fish Disease Workshop, American Fisheries Society, 24-26 June 1998, Parksville, British Columbia, pp. 24.

Moran, J.D.W., Margolis, L., Webster, J.M., and Kent, M.L.1999. Development of Kztdoa thyrsiies (Myxozoa: Myxosporea) in netpen-reared Atlantic salmon determined by light microscopy and a polymerase chain reaction test. Dis. Aquat. Org. (in press).

Moran, J.D.W., Whitaker, D.J., and Kent, M.L. 1999a. A review of the myxosporean genus Kudoa Meglitsch, 1947, and its impact on the international aquaculture industry and commercial fisheries. Aquaculture 172: 163- 196.

Moran, J.D. W., Whitaker, D.J.,and Kent, M-L. 1999b. NaturaI and laboratory transmission of the marine myxozoan parasite Kudoa fhyrsifes(Gilchrist, 1924) to Atlantic salmon. J. Aquat. Anim. Health (in press). Moser, M., and Kent, M.L. 1994. Myxosporea In Parasitic Protozoa, 2ndedition, Vol. 8. Edited by J.P. Kreier. Academic Press Inc., San Diego, California. pp. 265-3 18.

Moser. M., Love. M.S., and Jensen, L.A. 1976. Myxosporida (Protozoa) in California rockfish, Sebastes spp. J. Parasitol. 62: 690-692.

Naidenova, N.N., and Gaevskaya, A.V. 1991. Kudoa mirabilis sp.n. (Mixosporidea, Multivalvulea) from ribbonfish of the Indian Ocean. Hydrobiol. J. 27: 66-68.

Naidenova, N.N., Shulman, S.S., and Donets, Z.S. 1975. Protozoa, Mastigophora, Sporozoa, Cnidosporidia, Plasmospondia. In Key to the parasites of vertebrates of the Btack and Azov seas. Naukova Durnka, Kiev. pp. 7-71 [in Russian].

Nakajima, K. 1 984. Kudoa amamiensis infection, or musculo-myxosporiasis in yellowtail. Nihon Juishikai Zasshi 37: 673-676 [in Japanese].

Nakaj ima, K., and Egusa, S. 1978. Kudoa pericardialis n.sp. (Myxosporidea: Chloromyxidae) fiom cultured yellowtail, Seriola quinqueradiata Temminck et Schlegel. Bull. Jpn. Soc. Sci. Fish. 44: 1 17-1 20.

Narasirnharnurti, C.C ., and Kaiavati, C. 1979a. Kudoa sphyraeni n.sp. (Myxosporidia: Protozoa) parasitic in the muscles of the gut of the marine fish, Sphyraena jello Cuv. Proc. Indian Acad. Sci. Sect. B. 88: 265-268.

Narasirnharnurti, C.C ., and Kalavati, C . 1979b. Kudoa tefraspora n.sp. (Myxosporidea: Protozoa) parasitic in the brain tissue of Mugil cephalus. Proc. Indian Acad. Sci. Sect. B. 88: 85-89.

Nigrel li, R.F. 1946. Parasites and diseases of the ocean put, Macrozoarces americanus (Bloch and Schneider). Bull. Bingharn Oceanogr. Colt. Yale Univ. 9: 187-221.

Obiekezie, A.I.. and Lick, R. 1994. Kudoa cynogiossi n-sp., a new species of Kudoa Meglitsch (Myxosporea: MuItivaIvulida) fiom the West African tongue sole, Cynoglossus senegalensis (Kaup) (Teleostei : C ynoglossidae). Arc h. Protistenkd. 144: 20 1-205.

Obiekezie, A.I., Moller, H., and Anders, K. 1987. Kudoa sp. infection in the musculature of wild tongue sole, Cynoglossus senegalensis (Kaup, 1858) fiom the Coast of west Afi-ica. Bull. Eur. Assoc. Fish Pathol. 7: 38-41.

Oliva, M., Luque, J.L., Teh, L., and Llich, L. 1992. Kudoa sciaenae (Myxozoa: Multivalvulidae) cysts distribution in the somatic muscles of Stellifér minor (Tschudi, 1 844) (Pisces: Sciaenidae). Mem. Inst. Oswaldo Cruz 87: 33-35.

Overstreet, R.M. 1976. Fabespora vermicola sp.n., the first myxosporidan fiom a platyhelrninth. J. Parasitol. 62: 680-684. Palmer, R. 1 994. Kudoa - the Irish experience. In Kudoa Workshop Proceedings. Edited by D.C. Conley. Province of British Columbia, Ministry of Agriculture, Fisheries and Food, Aquaculture Industry Development Report 94-01, pp. I 8-21.

Paperna. 1. 1982. Kudoa infection in the glomeruli, mesentery and peritonewn of cultured Sparus aurata L. J. Fish Dis. 5: 539-543.

Paperna. I., and Zwerner, D.E. 1974. Kudoa cerebralis sp-n. (Myxosporidea, Chloromyxidae) from the striped bass, Morone saxatilis (Wal baum). J. Protozool. 2 1: 1 5- 19.

Patashnik, M., Groninger, H.S., Barnett. H.. Kudo. G.. and Koury, B. 1982. Pacific whi t ing, Merluccius productus: 1. Abnormal muscle texture caused b y myxosporidian- induced prote01ysis. Mar. Fish. Rev. 44: 1- 12.

Pérard, M.Ch. 1928. Sur une maladie du maquereau (Scomber scomber L.) due à une Myxosporidie: Chforomyxum histofyticum nsp. C. R. Acad. Sci., Paris, 186: 108-1 IO.

Porter. R.W., Koury, B., and Kudo. G. 1993. Inhibition of protease activity in muscle extracts and surimi fiom Pacific whiting, Merlucciirs productus, and arrowtooth flounder, Atheresthes stomias. Mar. Fish. Rev. 55: 10-1 5.

Robins, C.R.,Bailey, R.M., Bond, C.E., Brooker, J.R.,Lachner, E.A., Lea, R.N., and Scott, W.B. 199 1 a. Common and scientific names of fishes fiom the United States and Canada. 5LhEdition. Am. Fish. Soc. Spec. Publ. No. 20.

Robins, C.R.?Bailey, R.M., Bond, C.E., Brooker, J.R.,Lachner, E.A., Lea, R.N., and Scott, W.B. 1991 b. World fishes important to North Arnericans. Am. Fish. Soc. Spec. Publ. No. 2 1.

Romano, LA., Sardella, N.H., and Russomando, F. 1988. Ausencia de evidencias de dafios histologicos en roedores alimentados con quistes mixosporidianos de1 genero Kudoa de filetes de Merluccius hubbsi (merluza) y de Micromesistius australis (polaca), su relacion con la salud humana, Rev. Investig. Desarro. Pesq. 8: 83-86.

Sakanari, J. A. 1994. Detecting parasites in fish tissue. In Kudoa Workshop Proceedings. Edired by D.C. Conley. Province of British Columbia, Ministry of Agriculture, Fisheries and Food, Aquaculture Industry Development Report 94-01,pp. 1 1.

Sandeep, B .V.,Kalavati, C., and Narasimharnurti, CC. 1986. Kudoa atropi sp-n. (Myxosporea: Multivalvulida) a myxosporidian parasite fiom the gills of Atropus arropus. Vestn. Cesk. Spol. ZooI. 50: 132-135. Sardella, N.H. 1988a. Description de la espora y de los estadios previos del mixosporidio Kudoa rosenbuschi, parasito muscular de Merluccius hubbsi. Ciclo de vida. Parasitol. al Dia 12: 8- 13.

Sardella, N.H. l988b. Secuencia de la infestacion y reacciones tisulares en la musculatura de hferluccius hubbsi Marini ("merluza") y de Micromesistius auslralis Norman ("polaca") por la presencia de mixosporidios del genero Kudoa Meglitsch. Parasitol. ai Dia 12: 13-1 8.

Sardella, N.H., and Roldh, M.I. 1989. Mixosporidiosis producida por Kudoa rosenbuschi, parasito muscular de merluza comun (Merluccius hubbsl) en la zona comun de pesca Argentino-Uruguaya. Frente Marit. 5: 83-85.

Sardella, N.H., Trïnchero, J.L., and Manca, E.A. 1 987. Efectos del deterioro producido por mixospondios de la especie Kudoa rosenbuschi en la musculatura de Merluccius hubbsi. Rev. Investig. Desarro. Pesq. 7: 105- 1 12.

Sarkar, N.K., and C haudhury, S.R. 1996. Kudoa cascasia sp-n. (Myxosporea: Kudoidae) parasitic in the mesentery of Sicamugil cascasia (Ham.) fiom Hooghly estuary of West Bengal. India. Acta Protozool. 35: 33 5-338.

Sarkar, N.K., and Ghosh, S. 1991. Two new coelozoic Myxosporida (Myxozoa: Myxosporea) from estuarine teleost fishes (Mugilidae) of West Bengal, India. Proc. 2001. Soc. (Calcutta) 44: 13 1- 135.

Sarkar. N.K.' and Mazumder, S.K. 1983. Studies on myxosporidian parasites (Myxozoa: Myxosporea) fiom marine fishes in West Bengal. India- I. Description of three new species fiom Tachysurus spp. Arch. Protistenkd. 127: 59-63.

Schafer, W.E. 1 968. Studies on the epizootiology of the m yxosporidan Cerafomyxa shasta Noble. Calif. Fish Game 54: 90-99.

Schlegel, M., Lom, J., Stechmann, A., Bernhard, D., Leipe, D., Dykovii, I., and Sogin, M.L. 1996. Phylogenetic evidence of complete small subunit ribosomal RNA coding region of Myxidium lieberkuehni: evidence that Myxozoa are Metazoa and related to the Bilateria. Arch. Protistenkd. 147: 1-9.

Schreck, C.B. 1996. Immunomodulation: endogenous factors. In The fish immune system: organism, pathogen, and environment. Edired by G. Iwama and T. Nakanishi. Academic Press, San Diego, California. pp. 3 11-337.

Scott, W.B., and Scott, M.G. 1988. Atlantic fishes of Canada. Can. Bull. Fish. Aquat. Sci. No. 2 19. Seenappa D., and Manohar, L. 198 1. Experimental infection of mrigd, Cirrhina rnrigala (Hamilton), with the spores of Myxo bolus vaniviiasae (Myxosporida: Protozoa) and contaminated mud. Mysore J. Agric. Sci. 15: 108-1 12.

Shaw. R.W., Hervio, D.M.L., Devlin, R.H., and Adamson, M.L.1997. Infection of Aulorhynchusflavidus (Gill) (Osteichthyes: Gasterosteiformes) by Kudoa thyrsites (Gilchnst) (My-xosporea: Multivalwlida). J. Parasitol. 83: 8 10-8 14.

Shulman, S.S. 1966. Myxosporidia of the USSR. 2001. Inst., Akad. Nauk U.S.S.R., Moscow and Leningrad (Translated fiom Russian by U.S. Department of the Interior, N.T.I. S., Springfield, Virginia), 504 pp.

Siau. Y. 197 1. Myxosporidies de Synodontis ansorgii Bouleng. Ann. et Mag . N.H. 19 11 et de Eleotris (fiibia) kribensis Boulenger t 964, poissons des eaux saumâtres de la lagune de Porto-Nova (Dahomey). Bull. Soc. 2001. Fr. 96: 563-570.

Siddall, M.E., Martin, D.S., Bridge, D., Desser, S.S., and Cone, D.K. 1995. The dernise of a phylum of protists: phylogeny of Myxozoa and other parasitic Cnidaria. J. Parasitol. 81: 961-967.

Sindermann, C.J. 1959. Zoogeography of sea hemng parasites. Program and abstracts of the 34" Annual Meeting, American Society of Parasitologists, University Park, Pennsylvania, 30 August - 2 September 1959. J. Parasitol. 45 (Supplement): 34.

Srnothers?J.F., von Dohlen, C.D., Smith, L.H., and Spall, R.D. 1994. Molecular evidence that the myxozoan protists are metazoans. Science (Wash. D.C.) 265: 17 19- 1721.

Stehr, C. 1986. Sporogenesis of the myxosporean Kudoa paniformis Kabata & Whitaker, 198 1 infecting the muscle of the Pacific whiting, Meriuccius productus (Ayres). J. Fish Dis. 9: 493-504.

Stehr, C., and Whitaker, D.J. 1986. Host-parasite interaction of the myxosporeans Kudoa paniformis Kabata & Whitaker, 198 1 and Kudoa thyrsites (Gilchrist, 1924) in the muscle of Pacific whiting, Merluccius productus (Ayres): an ultrastructural study. J. Fish Dis. 9: 505-5 17.

St-Hilaire, S., Hill, M., Kent, M.L., Whitaker, D.J., and Ribble, C. 1997a. A comparative study of muscle texture and intensity of Kudoa thyrsites infection in farm-reared Atlantic salmon Salmo salar on the Pacific Coast of Canada. Dis. Aquat. Org. 31: 22 1-225.

St-Hilaire, S., Ribble, C., Whitaker, D. J., and Kent, M.L. 1997b. Evaluation of a nondestructive diagnostic test for Kudoa thyrsites in farmed Atlantic salmon (S'alma salar). Aquacuihue 156: 139- 144. St-Hilaire, S., Ribble, C., Whitaker, DJ., and Kent, M. 1 998. Prevalence of Kudoa thyrsites in sexually mature and immature pen-reared Atlantic salmon (Safmo salar) in British Columbia, Canada. Aquaculture 162: 69-77.

Stoffiegen. D.A., and Anderson. W.I. 1990. A myxosporidian parasite in the skeletal muscle of a black-tip reef shark. Carcharhintrs melanopterus (Quoy and Gaimard, 1824). J. Fish Dis. 13: 549-552.

Stolc, A. 1899. Actinomyxidies, nouveau groupe de Mesozoaires parent des Myxosporidies. Bull. Int. Acad. Sci. Bohème 22: 1- 12.

Szidat, L. 1966. Estudios sobre el Desarrollo y la biologia de esporozoos parasitas de la merluza (Merluccius hubbsi Marini) Kudoa rosenbuschi (Ge 1orrnini) Szidat. Dir. Gen. Pesca Conservacion Fauna, pp. 1-7.

Terin. L., Llich, L., and Luque, J.L. 1990. Kudoa sciaenae n-sp. (Myxozoa, Multivalvulida), parisito muscular de peces de la farnilia Sciaenidae (Osteichthyes) en el mar peruano. Rev. Ibér. Parasitol. 50: 25-29.

Thélohan, P. 1895. Recherches sur les Myxosporidies. Bull. Sci. Fr. Belg. 26: 100-394.

Traxler. G.S., Roome, J.R., and Kent. ML. 1993. Transmission ofinfectious hematopoietic necrosis virus in seawater. Dis. Aquat. Org. 16: 1 1 1- 1 14.

Tripathi, Y.R. 195 1 (1 953). Studies on parasites of Indian fishes. 1. Protozoa Myxosporidia together with a check list of parasitic Protozoa described fiom Indian fishes. Rec. Indian Mus. 50: 63-89.

Tsuyuki. H., Williscrofi. S.N., Kabata, Z.. and Whitaker, D.J. 1982. The relationship between acid and neutral protease activities and the incidence of sofi cooked texture in the muscle tissue of Paci fic hake Merluccius producrus infected with Kudoa paniformis and/or K. thyrsitis, and held for varying times under different pre-fieeze chilled storage conditions. Cm. Tech. Rep. Fish. Aquat. Sci. No. 1130.

Udey, L.R., Fryer, J.L., and Pilcher, K.S. 1975. Relation of water temperature to ceratomyxosis in rainbow bout (Salmo gairdneri) and coho salmon (Oncorhynchus kisutch). J. Fish. Res. Board Can. 32: 1545-155 1.

Voelker. F.A., Kassel, S.H., Weinberg, H.D., and McKee, A.E. 1978. Microscopic and ultrastructural characteristics of Kudoa infection in butterfly fish (Chaetodon). Vet. PathoI. 15: 40-48.

Weill, R. 1938. L'interpretation des Cnidosporidies et la valeur taxonomique de leur cnidome. Leur cycle comparé a la phase larvaire des Narcomeduses Cuninides. Trav. Stn. Zool. Wimereaux 13: 727-744. Whitaker, D.J. 1986. Intensity of infection by two species of the myxosporean parasite Kudoa and its distribution in six areas of the flesh of Pacific hake (Merluccius productus). Can. Data Rep. Fish. Aquat. Sci. No. 598.

Whitaker, D.J., and Kabata, 2. 1987. Early infection of Merluccius productus (Ayres) (Pisces: Teleostei) with Kudoa thyrsites (Gilchrist) (Myxozoa). Can. J. 2001.65: 936- 939.

Whitaker, D.J., and Kent, M.L. 1991. Myxosporean Kudoa thyrsites: a cause of sofi flesh in farrn-reared Atlantic salmon. J. Aquat. Anim. Health 3: 291 -294-

Whitaker, D. J., Kent, M.L.. and Margolis, L. 1994. Myxosporean parasites and their potential impact on the aquaculture industry, with emphasis on Kudoa species. In Kudoa Workshop Proceedings. Edited by D.C. Conley. Province of British Columbia, Ministry of Agriculture, Fishenes and Food, Aquaculture Industry Development Report 94-01. pp. 2-7.

Whi taker. D. J., Kent, M.L., and Sakanari, J.A. 1 996. Kudoa miniauriculutcz n.sp. (Myxozoa, Myxosporea) from the musculature of bocaccio (Sebastes paucispinis) from California. J. Parasitol. 82: 3 12-3 15.

Willis, A.G. 1949. On the vegetative forms and life history of Chloromyxum thyrsites Gilchrist and its doubtful systematic position. Aust. J. Sci. Res. Ser. B: Biological Sciences 2: 379-398.

Wishkovsky, A., Groff, J.M., Lauren, D.J., Toth, R.J., and Hedrick, R.P. 1990. Efficacy of fumagillin against proliferative kidney disease and its toxic side effects in rainbow trout (Oncorhynchus mykiss) fingerlings. Fish Pathol. 25: 14 1- 146.

Wyatt, E.J. 1978. Studies on the epizootiology of Myxobolus insidiosus Wyatt and Pratt, 1963 (Protozoa: Myxosporida). J. Fish Dis. 1: 233-240.

Yamamoto, T., and Sanders, J.E. 1979. Light and electron microscopic observations of sporogenesis in the myxosporida, Ceratomyxa shasta (Noble, 1950). J. Fish Dis. 2: 4 1 1- 428.

Yasutake, W.T. 1970. Comparative histopathology of epizootic salmonid virus diseases. In A symposium on diseases of fishes and shellfishes. Edited by S.F. Snieszko. Am. Fish. Soc. Spec. Publ. No. 5, Washington, DC. pp. 341-350.

Yokoyama, H., Ogawa, K., and Wakabayashi, H. 1993. Some biological characteristics of actinosporeans fiom the oligochaete Branchiura sowerbyi. Dis. Aquat. Org. 17: 223- 228.

Yurakhno, V.M. 1991. New species of Myxosporidia from fishes of the Black Sea. Parazitologiya 25: 104- I 09 [in Russian]. Zawistowski, S., Czewzyk, T., and Ganowiak, 2. 1986a. Histochemical investigations of Falkland blue whiting muscles infested with parasitic protozoans of the genus Kudoa sp. Bull. Sea Fish. Inst. 17: 33-35 [in Polish].

Zawistowski, S., Czewzyk, T., and Ganowiak, Z. l986b. Histological examination of rats fed on southern blue whiting Micromesistius austraiis infested with parasitic protozoans of the genus Kudou. Bdl. Sea Fish. Inst. 17: 55-56 [in Polish].