Control of disease in aquaculture proceedings of the Nineteenth U.S.-Japan Meeting on Aquaculture, Ise, Mie Prefecture, Japan 29-30 October 1990 ; Satellite Symposium : 2 November

[From the U.S. Government Printing Office, www.gpo.gov]

\@ Id

NOAA Technical Report NMFS I I I October 1992

Control of Disease in Aquaculture

Proceedings of the Nineteenth
U.S. -Japan Meeting on Aquaculture
Ise, Mie Prefecture, Japan
29-30 October 1-990-

Ralph S. Svrjcek (editor)

U.S. Department of Commerce

SHII
A44672
no - 3IL11

NOAA Technical Reports NMFS
The major responsibi litics of the National Marine Fish- continuing programs of NMFS; intensive scientific reports
eries Service (NMFS) are to monitor and assess the abun- on studies of restricted scope; papers on applied fishery
dance and geographic distribution of fishery resources, to problems; technical reports of general interest intended to
understand and predict fluctuations in the quantity and aid conservation and management; reports that review, in
distribution of these resources, and to establish levels for considerable detail and at a high technical level, certain
their optimum use. NMFS is also charged with the devel- broad areas of research; and technical papers originating
opment and implementation of policies for managing na- in economics studies and in management investigations.
tional fishing grounds, with the development and Since this is a formal series, all submitted papers, except
enforcement of domestic fisheries regulations, with the sur- those of the U.S.-Japan series on aquaculture, receive peer
veillance of foreign fishing off U.S. coastal waters, and review and all papers, once accepted, receive professional
with the development and enforcement of international editing before publication.
fishery agreements and policies. NMFS also assists the
fishing industry through marketing services and economic Copies of NOAA Technical Reports NMFS are avail-
analysis programs and through mortgage insurance and able free in limited numbers to government agencies, both
vessel construction subsidies. It collects, analyzes, and federal and state. They are also available in exchange for
publishes statistics on various phases of the industry. other scientific and technical publications in the marine
sciences. Individual copies may be obtained from the
The NOAA Technical Report NMFS series was estab- U.S. Department of Commerce, National Technical Infor-
lished in 1983 to replace two subcategories of the Tech- mation Service, 5285 Port Royal Road, Springfield, VA
nical Report series: "Special Scientific Report -Fisheries" 22161. Although the contents of these reports have not
and "Circular." The series contains the following types of been copyrighted and may be reprinted entirely, reference
reports: scientific investigations that document long-term to source is appreciated.

Recently Atblished NOAA Technical Reports NMFS

98. Marine mammal strandings in the United States. the North Pacific, 1973-88, by Michael A. Perez and
proceedings of the second marine mammal Thomas R. Loughlin. December 1991, 57 p.
stranding workshop; Miami, Florida, 3-5 Decem-
ber, 1987, edited by John E. Reynolds III and Daniel 105. Biology, oceanography, and fisheries of the North
K. Odell. January 1991, 157 p. Pacific transition zone and subarctic frontal zone,
edited byJerry A. Wetherall. December 1991, 92 p.
99. Marine flora and fauna of the northeastern United
States: erect Bryozoa, by John S. Ryland and Peter 106. Marine ranching: proceedings of the eighteenth
J. Hayward, February 1991, 48 p. U.S.-Japan meeting on aquaculture; Port.Ludlow,
Washington, 18-19 September 1989, edited by Ralph
100. Marine flora and fauna of the eastern United States: S. Svijcck. February 1992, 136 p.
Dicyerridda, by Robert B. Short. February 1991, 16 p.
101. Larvae of nearshore fishes in oceanic waters near 107. Field guide to the searobins (P@ionotus and Bel-
lator) in the western North Atlantic, by Mike Russell,
Oahu, Hawaii, by Thomas A. Clarke. March 1991, Mark Grace, and Elmerj. Gutherz. March 1992, 26 p.
19 P.
102. Marine ranching: proceedings of the seventeenth 108. Marine debris survey manual, by Christine A. Ribic,
U.S.-Japan meeting on aquaculture; Ise, Mie Trevor R. Dixon, and Ivan Vining. April 1992, 92 p.
Prefecture, Japan, 16-18 October 1988, edited by
Ralph S. Svdcek. May 1991, 180 p. 109. Seasonal climatologies and variability of eastern
tropical Pacific surface waters, by Paul C. Fiedler.
103. Benthic macrofauna of the New York Bight, April 1992, 65 p.
1979-89, by Robert N. Reid, David J. Radosh, Ann B.
Frame, and Steven A. Fromm. December 1991, 50 p. 110. The distribution of Kemp's ridley sea turtles
(Lepidockelys kempt) along the Texas coast: an
104. Incidental catch of marine mammals by foreign adas, by Sharon A. Manzella and Jo A. Williams. May
and joint venture trawl vessels in the U.S. EEZ of 1992, 52 p.

NQAA Technical Report NMFS 111

Control of Disease in Aquaculture

Proceedings of the Nineteenth
U.S. -Japan Meeting on Aquaculture
Ise, Mie Prefecture, Japan
29-30 October 1990
Satellite Symposium: 2 November

Ralph S. Svijcek
Publications Unit
Northwest and Alaska Fisheries Science Centers

Panel Chairmen:
Conrad Mahnken, United States
Seiji Sakaguchi, Japan

Under the U.S. -japan Cooperative Progam.
in Natural Resources (UJNR)

October 1992

U.S. DEPARTMENT OF COMMERCE
Barbara Hackman Franklin, Secretary
National Oceanic and Atmospheric Administration
John A. Knauss, Under Secretary for Oceans and Atmosphere
National Marine Fisheries Service
William W. Foxjr., Assistant Administrator for Fisheries
Q

C"
C"
V-
LIB-RAPY
NOAA/CCEH
1990 HOBSON AVE.
CHA-3- SC 29408-2623

PREFACE

The United States and Japanese counterpart panels on aquaculture were formed in 1969 under the
United States-Japan Cooperative Program in Natural Resources (UJNR). The panels currently in-
clude specialists drawn from the federal departments most concerned with aquaculture. Charged
with exploring and developing bilateral cooperation, the panels have focused their efforts on ex-
changing information related to aquaculture which could be of benefit to both countries.
The LJNR was begun during the Third Cabinet-Level Meeting of the joint United States-Japan
Committee on Trade and Economic Affairs in January 1964. In addition to aquaculture, current
subjects in the program include desalination of seawater, toxic microorganisms, air pollution, energy,
forage crops, national park management, my@oplasmosis, wind and seismic effects, protein resources,
forestry, and several joint panels and committees in marine resources research, development, and
utilization.
Accomplishments include increased communication and cooperation among technical special-
ists; exchanges of information, data, and research findings; annual meetings of the panels, a policy-
coordinative body; administrative staff meetings; exchanges of equipment, materials, and samples;
several major technical conferences; and beneficial effects on international relations.

Conrad Mahnken-United States
Seiji Sakaguchi-Japan

The National Marine Fisheries Service (NMFS) does not approve, rec-
ommend or endone any proprietary product or proprietary Material
mentioned in this publication. No reference shall be made to NMTS,
or to this publication furnished by NMFS, in any advertising or sales
promotion which would indicate or imply that NMFS approves, recom-
mends or endorses any proprietary product or proprietary material
mentioned herein, or which has as its purpose an intent to cause
directly or indirectly the advertised product to be used or purchased
because of this NMFS publication. The U.S.-Japan subseries of NOAA
Technical Reports on aquaculture is used to communicate preliminary
results, interim reports, and similar timely information. It is not subject
to formal peer review.

CONTENTS

0. BERGH Studies on diseases of cultured Atlantic halibut 1
G. H. HANSEN
1. HUSE
A.JELMERT

T. AKIYAMA Scoliosis of fishes caused by tryptophan deficiency 7

T. MEYERS Control of IHN virus in sockeye salmon culture 13

R.J. BARRIE Identification of a conserved antigenic domain in the major 15
C. L. "ON capsid protein of infectious pancreatic necrosis virus
J. C. LEONG

T. AOKI Cloning of hemolysin genes of aeromonads 21
I. HIRONO

T. HONJO Harmful red tides of Heteros@gma akashiwo 27

J. L. BARTHOLOMEW Impact of the myxosporean parasite Ceratomyxa shasta on survival 33
J. L. FRYER of migrating Columbia River Basin salmonids
J. S. ROHOVEC

M. YOSHIMIZU Viral infections of cultured fish injapan 43
1 KIMURA

K. MOMOYAMA. Some important infectious diseases of kuruma shrimp, 49
Penaeusjaponicus, injapan

J. R- WINTON The application of molecular biology to the detection of 53
infectious hematopoietic necrosis virus

K. MUROGA Bacterial and viral diseases of marine fish during seed production 57

T. YOSHINAGA An ecological study of the, parasitic nematode Hysterothylacium haze in theJapanese 63
common goby Acanthogubiusfiat4manus, in a brackish inlet

H. ISHIOKA Epidemiology of marine fish diseases in the warm waters along the Kuroshio Current 69

P. W. RENO Characterization of hematic neoplasia in the softshell clam Mya arenafia 85
A. ILLINGWORTH
M. DORITY

M.OTOTAKE Kinetics of bovine serum albumin administered by the immersion method 95
T. NAKANISHI in fishes acclimatized to seawater and to fresh water

T. NOMURA The epidemiological study of furunculosis in salmon propagation 101
M. YOSHIMIZU
T. KIMURA

I SUZUKI Functions of hemocytes during the wound healing process in the pearl oyster 109

Y MAENO Skeletal abnormalities of fishes caused by parasitism of Myxosporea 113
M. SORIMACHI

E. READ-CONNOLE Presence of oncogenes in fish tissues and in fish cell lines 119
C. A. SMITH
R M. HETRICK

H. SAKO Streptococcal infection in cultured yellowtail 125

G. MOBERG Stress induced pathologies in fish: the cost of stress 131

A. MURATA Control of fish disease injapan 135

Studies on Diseases of Cultured juvenile Atlantic Halibut

01VIND BERGH-, GEIR HOVIK HANSEN' INGVAR HUSE* and ANDERS JELMERT*
* Institute of Marine Research
Austevoll Aquarulture Research Station
N-5392 Storebo, Norway
Department of Micro&iology and Plant Physiology
University of Bergen, Jahnebakken 5
N-5007 Bergen, Norway

ABSTRACT

Bacterial infections by Flexibact& and Vibrio species are major causes of mortalities in
Atlantic halibut (Hippoglossus hippoglossus L.) larviculture. Egg surface disinfection is a
possible prophylactic treatment. This article summarizes and reviews several experiments
concerning causes of mortality of Atlantic halibut eggs and larvae.

Introduction groups was infected with 200 @LL of a suspension of
an accenic culture of one of the following bacteria:
Cultivation of* Atlantic halibut (Hippoglossus Flexibacter sp. strains NCIMB 13128 and NCIMB
hippoglossus L.) is presently at the verge of a commer- 13127" (National Collection of Industrial and Ma-
cial breakthrough in Norway. However, as is the case rine Bacteria, Aberdeen, Scotland) which were
with all cultivated species, there are problems emerg- isolated from two different groups of halibut eggs
ing concerning diseases related to opportunistic and otherwise seemingly identical (Hansen and
microorganisms (Bergh et al. 1992; Bergh and Bergh et al. 1992), Vibrio strain HI-10448 (Institute of
Jelmert 1990; Opstad and Bergh 1990; Pittman et al. Marine Research, Bergen, Norway) and Vibrio
1990). The purpose of this work is to summarize sev- anguillarum NCMB 6 (National Collection of Marine
eral experiments by studying the effects of microorganisms Bacteria, Aberdeen, Scotland); and Vibrio fischeri
on mortalities of halibut eggs and yolk sac larvae, strain ATCC 7744 (American Type Culture Collec-
possible prophylactic treatment procedures, and ef- tion, Rockville, MD). Final total counts of bacteria in
fects of some physical stressors. the wells were measured by staining with DAPI (Por-
ter and Feig 1980) and counting in a Nikon
epiflourescence microscope at 60OX to be in the or-
Methods and Materials der of 2-3 X 101 bacteria x mU'. One group of 60
eggs was not infected, serving as the control. Within
Eggs from one female were artificially stripped and 24 hours after hatching, visible remnants of the egg-
fertilized with sperm from two males and reared in shell were removed along with 10 mL of the water,
250-L upstream incubators between 6 and 7' C until and 10 mL of sterile seawater were immediately
further processing. , added. Mortality was recorded until Day 37 after
The eggs were transferred to polystyrene muldwell hatching. For a further description of the infection
dishes (NUNC, Roskilde, Denmark) for the disinfec- experiment, see Bergh et al. (1992).
tion trials and the infection experiment. Each well For the disinfection experiment another egg group
contained one egg and 11 mL of sterilized seawater. was disinfected one day before hatching. The follow-
The dishes were "incubated in darkness between 5 ing procedure was followed. Eggs were divided into
and 6' C for the duration of the experiment. four groups, which were exposed to different concen-
For the infection experiment, eggs were divided trations of the iodophor disinfectant Buffodine
into 6 groups, each containing 60 eggs. Four days (Evans Vanodine, Preston, England): 0.5, 0.05, and
before hatching,. each well of each of the 5 treatment 0.005%, plus one untreated control group. Applica-

2 NOAA Technical Report NMEFS I I I

Figure I
Halibut egg showing
large surface wounds.
This appearance is
typical of eggs in-
fected with FL-xibacter
sp. Egg diameter is
approximately 3 mm.
The photograph was
't taken from a Wild
zoom binocular mi-
croscope operated in
the dark field mode.
Photo by Guri Grung
and Vibeke Valkner.

tion time was 10 minutes. Immediately following the The infection experiment revealed three different
disinfection, the solution was carefully pipetted off types of mortality patterns:
the-eggs, and more sterile seawater was added. This
procedure was repeated three times. The control 1 The uninfected control group showed very low
group was washed the same way as the other groups. mortality throughout the experiment, as only 5
Thereafter, 60 randomly chosen eggs from each out of 60 larvae died.
group were incubated in polystyrene multiwell dishes 2 The two groups infected with Flexibacter sp.
as previously described. Within 24 hours after hatch- showed high mortalities at hatching; out of 60 lar-
ing, visible remnants of the eggshell were removed vae per group, 40 and 49 had died in the NUMB
along with 10 mL of water, and 10 mL of sterile sea- 13127 T and NUMB 13128 groups, respectively. At
water was added. Mortality was recorded until Day 37 Day 18, these groups -were terminated in order to
after hatching. The remaining living larvae were ex- gain material for, re-isolation of bacteria. Eighty
amined microscopically under a dissection and 93% of the larvae were dead in Groups
microscope for developmental, disorders. A further NUMB 13127 T and NUMB 13128, respectively.
description of this experiment is given by Bergh and
jelmert (1990). 3 The groups infected with V anguillarum strains or
with V fische7i showed an intermediate mortality
Results pattern. Only 1-4 larvae died per group at hatch-
ing, but high mortalities occurred throughout the
rest of the experiment. When the experiment was
Figure I shows a halibut egg with severe surface dam- terminated at Day 37, mortalities had risen to
age, an appearance typical for eggs infected with 95% in the V anguillarum NCMB 6 group, 78% in
Flexibacter sp. Figure 2 demonstrates a normal egg, the V fischeri ATCC 7744 group, and 67% in the
without visible damage. group infected with V anguillarum 651.
Scanning electron microscopy of infected eggs re-
vealed the chorion to be completely dissolved over In the disinfection experiment, 7 out of 56 remain-
large areas (up to 206 @Lrn in diameter), whereas the ing larvae in the group treated with 0.5% Buff6dine
zona radiata was severely damaged. Isolation of were dead when the experiment was terminated, 9
epibiotic bacteria from this egg group revealed an out of 59 were dead in the 0.05% Buffodine group,
epiflora totally dominated (99% of colony-forming 24 out of 59 were dead in the 0.005% Buffodine
units) by Flexibacter sp. (Bergh et al. 1992) group, and 19 out of 58 in the untreated group.

Bergh et al.: Studies on Diseases of Cultured Atlantic, Halibut 3

Figure 2
A normal halibut
egg. Diameter is ap-
proximately 3 mm.
The photograph is
taken from a Wild
zoom binocular mi-
croscope operated in
the dark field mode.
Photo by Guri
Grung and Vibeke
Valkner.

The groups that were disinfected with 0.5% and been shown that bacteria are able to cause at least
0.05% Buffodine could not be significantly distin- minor destruction to the chorion (Hansen and
guished for cumulative mortality at the end of the Olafsen 1989), but we are not aware of other reports
experiment (P>0.05, t-test with arcus sinus transfor- showing that bacteria are able to dissolve the chorion
mation of proportions). The two remaining groups completely. It could not be deduced from the scan-
were not statistically separable, but they both had sig- ning electron micrographs that the zona radiata was
nificantly higher mortalities than the two groups that completely destroyed; however, evidence of severe de-
were treated with the highest Buffodine concentra- struction was clear. This biotype is commonly isolated
tions (P<0.05%, t-test with arcus sinus transformation from Atlantic halibut eggs and could be considered a
of proportions). major problem in halibut larviculture.
With respect to developmental disorders, 5 out of With the two Vibyio species that were used in the
49 living larvae treated with 0.5% Buffodine were infection experiment, the situation is different.
found to posess at least one kind of disorder when These bacteria seemed to be harmless to the egg
the experiment was terminated in the group. In the stage, causing no significant mortality. However, the
other groups, scores were 16 out of 50 (0.05% profound mortality of the infected groups through-
Buffodine), 22 out of 36 (0.005% Buffodine) and 23 out the yolk-sac stage indicates that Vibfio-infections
out of 39 (control). The. most common disorder was during the yolk-sac stage may help to explain the
the presence of black, probably necrotic, tissue in the high mortality rates experienced so far.
gill, heart, or frontal yolk-sac region. Work is in . Recent observations indicate that infectious pan-
progress to characterize the ultrastructural changes creatic necrosis virus (IPNV), serotype Nl, is
associated with these kinds of disorders. prevalent in juvenile stages of the Atlantic halibut
(Mortensen et al. 1990). The IPNV is associated with
high mortalities, but it is not verified whether the
Discussion virus is the principal lethal agent. The time of infec-
tion has so far not been shown. Also, nematodes
We have presented initial results showing that bacte- could be present in relatively large amounts (G.
rial infections. may be closely involved in the Bristow, Zoological Laboratory, University of Bergen,
mortality of cultured halibut eggs and yolk-sac larvae. pers. commun. 1991); these are probably introduced
The infection experiment proves that Flexibacter sp. is by feeding the larvae with collected zooplankton.
a causative agent of mortality at the egg stage, at The disinfection experiment demonstrated that
hatching, and at the early yolk-sac stage. It has earlier the bacterial epiflora on halibut eggs is at least partly

4 NOAA Technical Report NKIN I I I

responsible for many of the develo pmental disorders although four critical periods of high mortality were
commonly occuring in halibut yolk-sac larvae. Sur- identified for aquaculture systems at temperatures
face disinfection of eggs with iodophors should be an between 3 and 9' C: hatching, 10-14, 25-35, and
adequate prophylactic treatment, as they have good 45-60 days after hatching. The latter was probably
pathogen/host differential of toxicity (Amend and due to starvation (I. Opstad and A.B. Skiftesvik,
Pietsch 1972; Ross and Smith 1972; Amend 1974). Austevoll Aquaculture Research Station, pers.
However, more work is needed to establish reliable commun. 1991). For the three other critical periods
disinfection procedures. of the yolk-sac stage, evidence presented here indi-
Mortality rates could be augmented by several sub- cates that effects of bacteria could not be ruled out.
lethal factors (Rosenthal and Alderdice 1976). There are not yet any data available ranking the
Sublethal physical stressors might increase egg and quantitative importance of the different causes of
larvae sensitivity to infectious microorganisms, rather death of Atlantic halibut eggs and larvae, although
than per se be the causative agent of death. The effects early-life stage mortalities are still a major factor lim-
of sublethal stressors to early life stages of halibut iting the commercial success of halibut aquaculture.
have been investigated in several studies. The data presented here, however, give evidence that
Jelmert and Naas (1990) reported that lowered 02 pathogenic or opportunistic microorganisms are
concentrations, exposure to H2S and exposure to closely involved in some typical mortality and devel-
high light levels led to a higher prevalence of de- opmental disorder patterns of eggs and yolk-sac
formed yolk-sac larvae. larvae.
Sensitivity of halibut eggs to physical shocks was
investigated by Holmeflord and Bolla (1988), who Acknowledgments
found that eggs were most sensitive before the clo-
sure of the blastopore. In a more extensive study,
eggs of halibut were compared with several other ma- This work has been supported by the Royal Norwe-
rine fishes, Opstad (L Opstad, Austevoll Aquaculture gian Council for Scientific and Industrial Research
Research Station, pers. commun. 1991) found similar (NTNF), by the Norwegian Council for Fisheries Re-
results, with the addition that eggs during the hatch- search (NFFR), and by Mowi a/s. We are grateful to
ing period were highly sensitive to physical stress. Guri Grung and Vibeke Valkner who kindly let us use
Effects of water flow on yolk-sac larvae were studied their photographs.
by Opstad and Bergh (1990), who concluded that
high rates of water exchange in upstream incubators Citations
significantly increased mortality. Yolk-sac utilization
was inversely related to rate of flow.
Absence of flow, however, caused rapid increase in Amend, D. F.
1974. Comparative toxicity of two iodophors to rainbow
the amount of bacteria in the incubators (Opstad trout eggs. Trans Am. Fish. Soc. 103:73-78.
and Bergh 1990; Skiftesvik et al. 1990), and subse- Amend, D. F., andj. P. Pietsch.
quent larval mortality. Thus, these two effects must 1972. Virucidal activity of two iodophors to salmonid
be carefully weighed against each other. Although viruses. J. Fish. Res. Board Can. 29:6165.
normally not regarded as a sublethal stressor or caus- Bergh, 0., and A. Jelmert.
ative agent of diseases, extreme light regimes have 1990. Antibacterial treatment procedures of eggs of halibut
(Hippoglossus hippoglossus L.). Presented at council meeting,
been shown to induce reduced yolk-sac utilization International Council for the Exploration of the Sea. ICES-
and increased mortality of halibut yolk-sac larvae CM-1990/17:39, 6 p.
(Skiftesvik et al. 1990). Bergh, 0., G. H. Hansen, and R. E. Taxt.
Studying development and mortality of Atlantic 1992. Experimental infection of eggs and yolk sac larvae of
halibut eggs and larvae at different temperatures, halibut, Hippoglossus hippoglossus L. J. Fish Dis. 15. (In
Pittman et al. (1990) concluded that 3' C is near the press.)
Hansen, G. H., andj. Olafsen.
lower limit for development of halibut eggs and lar- 1989. Bacterial colonization of cod (Gadus morhua L.) and
vae. At this temperature, the larvae often showed halibut (Hippoglossus hippoglossus) eggs in marine
incomplete caudal development and suffered higher aquaculture. Appl. Environ. Microbiol. 55(6):1435-1446.
mortality than those reared at 6' C. The groups Hansen, G. H., 0 Bergh, J. Michaelsen, and D. Knappskog.
reared at 9' C had high egg mortality and quickly 1992. Flexibacter ovolyticus sp. nov., a pathogen of eggs and
larvae of Atlantic halibut, HippoglossushippoglossusL. Int.J.
,developed abnormalities, such as small hearts and System. Bacteriol. 42(3). (In press.)
livers, and large peritoneal and pericardial spaces, Holmeflord, I., and S. Bolla'
indicating that this temperature was sublethal. No 1988. Effects of mechanical stress on Atlantic halibut eggs at
primary cause of larval death could be identified, different times after fertilization. Aquaculture68:369-371.

Bergh et aL: Studies on Diseases of Cultured Atlantic Halibut 5

jelmert, A., and K E. Naas. Porter, K G., and Y S. Feig.
1990. Induced deformities of the Atlantic halibut 1980. The use of DAPI for identifying and counting aquatic
(Hippoglossus hippoglossus L.) yolk sac larvae. A new experi- microflora. Limnol. Oceanogr. 25:943-948.
mental approach. Presented at council meeting, Interna- Rosenthal, H., and D. F. Alderdice.
tional Council for the Exploration of the Sea. ICES-CM- 1976. Sublethal effects of environmental stressors, natural
1990/F:45, 8 p. and pollutional on marine fish eggs and larvae. J. Fish Res.
Mortensen, S. H., B. Hjeltnes, 0. Rodseth, J. Krogsrud, and K. E. Board Can. 33:2047-2065.
Christie. Ross, A. J., and C. A. Smith.
1990. Infectious pancreatic necrosis virus, serotype NI, iso- 1972. Effect of two iodophors on bacterial and fungal fish
lated from Norwegian halibut (Hippoglossus hippoglossus), pathogens. J. Fish Res. Board Can. 29:1359-1361.
turbot (Scophthalmus maximus), and scallops (Pecten Skiftesvik, A. B., 1. Opstad, 0. Bergh, K Pittman, and L. Skjoldclal.
maximus). Bull. Eur. Assoc. Fish. Pathol. 10(2):42. 1990. Effects of light on the development, activity and mor-
Opstad, L, and 0. Bergh. tality of halibut (Hippoglossus hippoglossus L.) yolk sac
1990. Effects of continuous flow rate on development and larvae. Presented at council meeting, International Coun-
mortality of halibut yolk sac larvae. Presented at council cil for the Exploration of the Sea. ICES-CM-1990/F:43, 16 p.
meeting, International Council for the Exploration of the
Sea. ICES-CM-1990/F:41, 11 p.
Pittman, K_ 0. Bergh, 1. Opstad, A. B. Skiftesvik, L. Skjolddal, and
H. Strand.
1990. Development of eggs and yolk sac larvae of halibut
(Hippoglossus hippogiossus L.). J. Appl. Ichthyol. 6:142-160.

Scoliosis of Fishes Caused by Tryptophan Deficiency

TOSHIO AKIYAMA

National Research Institute of Aquaculture
Fisheries Agency
Thmaki, Mie 519-04, Japan

ABSTRACT

Scoliosis, caused by dietary tryptophan (Trp) deficiency, has been reported mainly in
salmonids. Neither abnormality in the vertebra per se nor microscopically visible damage
in the surrounding tissues was detected in the scoliotic fish, most of which returned to
normal shape within a short period of time after restoration of Trp to the diet. There-
fore, serotonin (5-HT), which is one of the Trp metabolites and a known
neurotransmitter, was suspected as a key substance responsible for the symptom. This
paper reviews several feeding studies where purified diets containing various combina-
tions of L-Trp, 5-hydroxy-L- tryptophan (5-HTP, direct precursor of 5-HT), MK486 (local
inhibitor of 5-HT synthesis only in periphery) and DL-p-chlorophenylalanine (PCPA,
general inhibitor of 5-HT synthesis) were fed to chum salmon fry (Oncorhynchus keta).
The findings indicate that occurrence of the spinal deformity is related to depletion of 5-
HT in the central nervous system. In addition, the relationship between water
temperature during rearing period and incidence of the scoliosis is also discussed.

Introduction Characteristics of Spinal Deformity

Although it is known that 10 amino acids are essen- The spinal deformity caused by Trp deficiency is
tial for normal fish growth, all of the quantitative mainly scoliotic or slightly lorcloscoliotic, and neither
requirements for essential amino acids have been lordosis or kyphosis has been noted. Scoliosis occurs
determined only for chinook salmon, Oncorhynchus after 1-2 weeks of feeding a Trp-deficient diet in
tschawytscha, coho salmon, 0. kisutsch, carp, rainbow trout (Kitamura 1969; Kloppel and Post
Cyprinus carpio, channel catfish, Ictalurus punctatus, 1975) and chum salmon (Akiyama et al. 1986a). Most
Japanese eel, Anguilla japonica, and Nile tilapia, scoliotic fish return to normal shape after restoration
Oreochromis niloticus. The author conducted a series of Trp to their diet; therefore, this symptom is revers-
of dietary studies to determine the amino acid re- ible. However, most spinal deformities caused by
quirements for the fry of chum salmon, 0. keta, nutrient deficiency are not reversed even by restoring
which is one of the most important species in the nutrients to their optimum level in the diet. For ex-
salmon enhancement project in Japan. In these ex- ample, ascorbic acid-deficient fish form thermally
periments, spinal deformity was observed in many labile underhydroxylated collagen which is dena-
of the fish fed a, tryptophan (Trp) -deficient diet tured and digested at higher temperatures; this
(Fig.1; Akiyama et al. 1985). Si 'nce the abnormality results in connective tissues with a low collagen con-
was first attributed to Tip, deficiency in sockeye tent and in the development of a fragile bone
salmon by Halver and Shanks (1960), the same de- structure, which finally results in irreversible symp-
ficiency symptom has been reported in rainbow toms of scurvy, such as lorclosis and scoliosis (Sato et
trout (Shanks et al. 1962; Kitamura 1969; Kloppel al. 1983; Ikeda et al. 1983). Although Kloppel and
and Post 1975; Poston and Rumsey 1983; Walton et Post (1975) observed some minor abnormalities such
al. 1984) and coho salmon (Ogata and Arai 1981). as protrusions of the fibrous matrix sheath investing
So far, the biochemical pathway resulting in the the notochord of scoliotic rainbow trout that were
occurrence of spinal deformity dule to Trp defi- caused by Trp deficiency, serious lesions of the verte-
ciency has not been elucidated. brae and microscopically visible damage in the

8 NOAA Technical Report NAUS I I I

100

0
U

@N 50
0

Z
PQ

150

100

50 Requirement
0.29%
Figure I
0 Relationships between tryptophan level
0.2 0.4 0.6 in diet and weight gain or incidence of
TRYPTOPHAN % IN DIET scoliosis; average value of duplicate tanks
of 35 fish, each group fed for 4 weeks at
16.0' C (Akiyama et al. 1985).

surrounding tissues have not been recognized in vous system. Many reports on spinal deformity caused
rainbow trout (Kitamura 1969) and chum salmon by a metabolic disfunction or lesion in the nervous
(Akiyama et al. 1986b). system are available. In mammals, scoliosis develops
in bipedal rats with brain-stem lesions (Tamura 1974)
and in rabbits whose dorsal root in the spinal cord
Construction of Hypothesis was removed (MacEwen 1973). In fish, yellowtail
parasitized by cysts of Myxobolus in the 4th ventricle
Trptophan is not only an essential structural element of the brain (Egusa 1985; Sakaguchi et al. 1987)
of protein but also the precursor of nicotinamide ad- showed scoliosis. Vertebral deformity was reported to
enine dinucleotide (NAD) and niacin in higher occur in yellowtail whose brain was infected by beta-
vertebrates (Fig.2). Therefore, attention was focused hemolytic streptococcaI bacterium (Shiomitsu 1982;
on the role of dietary niacin in the early studies of Kaige et al. 1984). Spinal deformities caused by an
Trp metabolism. Poston and Combs (1980), however, abnormality in the peripheral nervous system are
reported that dietary Trp is not an efficient precursor well known in fishes exposed to pesticides such as
of niacin in salmonids. Moreover, Poston and Rumsey diazinon, Which develop severe spinal curvature to-
(1983) showed that the deletion of dietary niacin did gether with fracturing (Hirose and Kitsukawa 1976;
not significantly increase the incidence of scoliosis in Hirose et al. 1979). It is speculated that these symp-
rainbow trout fed a diet containing a low level of Trp. toms are induced by excess accumulati6n of
It is possible that the symptom of scoliosis may be acetylcholine in the neuromuscular junction. Thus,
I I I
/ X;74@6

induced by an abnormal and involuntary contraction abnormality in the nervous system is one of the most
or relaxation of muscle due to a defect in the net- important factors for occurrence of spinal deformity.

Akiyama: Scoliosis of Fishes Caused by Tryptophan Deficiency 9

Protein 4 Tryptophan 05- ydroxy-ltryptophan Serotonin
(5-M) (5-HT)

Kynurenine Melatonin

Xanthurenic aci
3-Hydroxyanthranic acid

Guinolinic acid

C02,'H20 D Figure 2
Map of tryptophan metabolism.

Among the various Trp metabolites, serotonin (5- neuron. From these facts, I hypothesized that
HT) is known to function as a brain neurotransmitter scoliosis caused by Trp deficiency would be induced
or modulator and is involved in the regulation of by an imbalance of muscular tension due to a de-
sleep, body heat, sexual behavior, appetite, pain rec- creased 5-HT level in the nervous system.
ognition, secretion of growth hormone and
prolactin, besides classical functions such as the con-
traction of smooth muscle of blood vessels, the 5-HT Involvement in the Occurrence
uterus, and the digestive tract. In addition, it is of Scoliosis
known that torticollis and abnormal posture can be
induced by the destruction of rat midbrain in which In the first experiment which tested the involvement
both serotoninergic and dopaminergic neurons are of 5-HT in the occurrence of scoliosis an oral admin-
located (Tanaka and Kimura'1984), and that 5-HT istration of 5-HT to Trp-deficient chum salmon fry
modulates.the central pattern generator for locomo- decreased the incidence of scoliosis, but did not com-
tion in the spinal cords of the lampreys Ichthyomyzon pletely inhibit its occurrence (Akiyama et al. 1986a).
unicuspis and Petramyzon ma-yinus (Harris-Warrich et Therefore, we fed fry a Trp-deficient diet supple-
al. 1985). These reports suggest that muscular func- mented with 5-hydroxy@tryptophan (5-HTP, 100-130
tion can be partly controlled by a serotoninergic mg/100 g diet), which is a direct precursor of 5-HT

(Z) Trp 0.051
60 Trp 0.05% + Kynurenine
0-0 Trp 0.05Z + 5-HTP
0-0 Trp 0.29Z
Figure 3
Trp 0.29Z + PCPA
Effects of oral administra-
tion of tryptophan(Trp)
40
metabolites to Trp-defi-
Cn
cient or sufficient chum
salmon fry. Administra-
U tion of 5-hydroxy-L-
0 1 HTP) to Trp-
PA tryptophan(5-
20
deficient fish completely
prevented scoliosis, where-
as kynurenine failed to in-
hibit the occurrence.
DL-p-Chlorophenylala-
0 nine (PCPA) developed
2 4 scoliosis even in the fish
FEEDING PERIOD (WEEK) fed Trp-sufficient diet
(Akiyama et al. 1986b).

io NOAA Technical Report NIM I I I

Figure 4
Functions of adminis-
trated drugs on serotonin
(5-HT) pathway. DL-p-
Tryptophan hydroxylase Arowatic L-amino acid Chlorophenylalanine
decarboxylase (PCPA) is an inhibitor of
trypto'phan-hydroxylase
PCPA NK486 (TrpOHase) and inhibits
a biosynthesis of 5-hy-
droxy-L-tryptophan
(5-HTP) from Trp. L-2-
COOH OH,-t:::)@@ COOH hydrazino-alfa-methyl-
"@Y .1 NH OH N b e t a- (3,4- d i h y d r o xy-
n2 N 2 N H2
N phenylpropionic acid);
H H H
Tryptophan 5-Hydroxy-L-tryptophan (MK486) is an inhibitor
Serotonin of L-amino acid decar-
(Trp) (5-HTP) (5-HT) boxylase only in periphery
and consequently induces
a conversion of exogenous
5-HTP to 5-HT in central
nervous system.

and can easily pass through a blood-brain barrier in 5-HT and 5-hydroxyindoleacetic acid than did adding
contrast to 5-HT (Akiyama et al. 1986b, 1989). The 5-HTP alone (Akiyama and Murai, unpubl. data).
treatment completely prevented the occurrence of The experiment suggests that the deficiency of 5-HT
scoliosis and increased the brain 5-HT level in Trp- in the central nervous system was related to the oc-
deficient fish (Fig.3). Kynurenine, a precursor of currence of scoliosis.
niacin, NAD, and xanthurenic acid was fed at 120 mg
(as L-kynurenine)/100 g diet to Trp-deficient fish,
but failed to prevent scoliosis. Moreover, both Effect of Temperature on the Occurrence
scoliosis and decreased brain 5-HT levels were ob- of Scoliosis
served in fish fed a Trp-sufficient diet supplemented
with DL-p-chlorophenylalanine (PCPA) (Akiyama et Scoliosis caused by Trp deficiency has been reported
al. 1986a; 1986b), which is an inhibitor of trypto- only in salmonids such as sockeye salmon, rainbow
phan-hydroxylase (TrpOHase) and a potent depletor trout, coho salmon and chum salmon, although it
of both brain and peripheral stores of 5-HT (Fig. 4). has also been studied in chinook salmon (Halver et
TrpOHase is a rate-limiting enzyme on the 5-HT al. 1957), channel catfish (Dupree and Halver 1970),
pathway. These findings indicated a relationship be- eel (Arai et al. 1972), carp (Nose et al. 1974; Nose
tween the occurrence of scoliosis and 5-HT levels. 1979), red sea bream, Pagrus major (Yone 1976) and
The reduced ability of orally administrated 5-HT to tilapia, Tilapia zilhi (Mazid et al. 1978). At first I con-
inhibit the development of scoliosis compared with sidered. scoliosis to be a characteristic Trp deficiency
complete prevention with the use of 5-HTP suggested symptom of all salmonids except chinook salmon.
involvement of the serotoninergic neuron in the cen- Arai et al. (1986), however, reported development of
tral nervous system. Therefore, the author prepared scoliosis in Trp-deficient Ayu fish (Plecoglossidae)
Trp-deficient diets (0.05%) containing various com- reared at 16' C, although the incidence was low. I
binations of 5-HTP (10 or 50 mg/100 g diet) with or also found one scoliotic fish when a Trp-deficient
without L-2-hydrazino-alpha-methyl-beta-(3,4- diet was fed to yellowtail at 20' C (Akiyama, unpubl.
dihydroxyphenylpropionic acid) (MK486, I or 5 mg/ data), even though the brain 'was not infected by
100 g diet). MK486 functions hs an inhibitor of aro- Streptococcus and not parasitized by cysts of Myxobolus.
matic L-amino acid decarboxylase only in periphery These facts suggest that scoliosis due to Trp defi-
and inhibits 5-HT biosynthesis from 5-HTP. Thus, it ciency is unlikely to be peculiar to salmonids.
consequently functions to promote the conversion of Because salmonids are coldwater fish, and because
exogenous 5-HTP to 5-HT in the central nervous sys- most of the fishes in which scoliosis was not observed
tem. Feeding a Trp-deficient diet supplemented with as a symptom of Trp deficiency are warmwater fish
both 5-HTP and MK486 resulted in a significantly (except chinook salmon), I focused my attention on
lower incidence of scoliosis and higher levels of brain the influence of environmental temperature. It is

Akiyaxna: Scoliosis of Fishes Caused by Tryptophan Deficiency I I

likely that among the fishes developing scoliosis, inci- HalverjE., and W.E. Shanks.
dences of scoliosis decrease as the optimum 1960. Nutrition of salmonoid fishes. VIII. Indispensable
temperature for each species rises. Moreover, spinal amino acids for sockeye salmon. J. Nutr. 72:340-346.
deformity has not been observed to be an external HalverjE., D.C. Delong, and E.T. Mertz.
1957. Nutrition of salmonoid fishes. V. Classification of es-
symptom of dietary Trp deficiency in mammals and sential amino acids for chinook salmon. J. Nutr. 63:95-
birds, which are warm-blooded animals. In fact, the 105.
author presumed that the occurrence and incidence Harris-Warrick, R.M.,J.C. McPhee, andj.A. Filler.
of scoliosis might be influenced by rearing tempera- 1985. Distribution of serotonergic neurons and processes in
tures, and therefore fed the Trp-deficient diet to the lamprey spinal cord. Neuroscience 14(4):Il 127-1140.
chum salmon fry at three different temperatures: 10, Hirose, K., and M. Kitsukawa.
1976. Acute toxicity of agricultural chemicals to seawater te-
16,* and 200 C. The experiment showed that as the leosts, with special respect to TLM and the vertebral
rearing temperature was lowered, the incidence of abnormality. Bull. Tokai Reg. Fish. Res. Lab. 84:11-20. (In
scoliotic fish increased and brain 5-HT levels de- Japanese; English abstr.)
creased (Akiyama and Murai, unpubl. data). It is still Hirose, K_ M. Kitsukawa, and A. Ishikawa.
unknown why the brain 5-HT level in Trp-deficient 1979. Effects of water temperature on median lethal concen-
fish varied under different temperature conditions. trations (LC50) of a few pesticides to seawater teleosts. Bull.
Tokai Reg. Fish. Res. Lab. 98:45-53. (In Japanese; English
abstr.)
Ikeda, S., M. Sato, and R. Yoshinaka.
Conclusion 1983. Role of vitamin C in collagen formation of fish. Vita-
mins Uapan) 57(8):433-449. (Injapanese, English abstr.)
All of these findings described above indicate that Kaige, N., T Miyazaki., and S. Kubota.
scoliosis caused by Trp deficiency is related to the 1984. The pathogen and the histopathology of vertebral de-
formity in cultured yellowtail. Fish. Pathol. 19(3):173-179
level of 5-HT in the central nervous system. In fishes, (Injapanese, English abstr.)
hereafter, the central nervous system, especially the Kitamura, S.
5-HT neuron, should be considered as one of the 1969. Summary on the hypovitaminosis C of rainbow trout,
important factors in an occurrence of idiopathic spi- Salmo gairdneri. Fish Pathol. 3:73-85 (In Japanese.)
nal deformity. KJoppel, T.M., and G. Post.
1975. Histological alterations in tryptophan-deficient rain-
bow trout. J. Nutr. 105:861-866.
MacEwen, G.D.
Citations 1973.. Experimental scoliosis. Clin. Ortfiop. 93:69-74.
Mazid, M.A., Y. Tanaka, T. Katayama, K.L. Simpson, and C.O.
Akiyama, T., S. Arai, T. Murai, and T. Nose. Chichester.
.1985. Tryptophan requirement of chum salmon fry. Bull. 1978. Metabolism of amino acids in aquatic animals-Ill Indis-
Jpn. Soc. Sci. Fish. 51(6):1005-1008. pensable amino acids for Tilapia zillii. Bull. Jpn. Soc. Sci.
Akiyama, T., T. Murai, and T. Nose. Fish. 44(7):739-742.
1986a. Oral administration of serotonin against spinal defor- Nose, T.
mity of chum salmon fry induced by tryptophan 1979. Summary report on the requiremenis of essential
deficiency. Bull. Jpn. Soc. Sci. Fish. 52(7):1249-1254. amino acids for carp. In Finfish nutrition and fishfeed
Akiyama, T., T. Murai, K. Mori.. technology, Vol. I (J. E. Halver and K. Tiews, eds.), p. 146-
1986b. Role of tryptophan metabolites in inhibition of spi- 156. Heenemann Verlagsgesellschaft GmbH, Berlin.
nal deformity of chum salmon fry caused by Tryptophan Nose, T., S. Arai, D. Lee, and Y. Hashimoto.
deficiency. Bull. Jpn. Soc. Sci. Fish. 52(7):1255-1259. 1974. A note on amino acids essential for growth of young
Akiyama, T., H. Kabuto, M. Hiramatsu, T. Murai, and K. Mori. carp. Bull.jpn. Soc. Sci. Fish. 40(9):903-908.
1989. Effect of dietary 5-hydroxy-L-tryptophan for preven- Ogata, H., and S. Arai.
tion of scoliosis in tryptophan-deficient chum salmon 1981. Essential amino acid requirements for coho salmon 11 .
fry. Nippon Suisan Gakkaishi 55(1):99-104. Leucine, isoleucine, tryptophan and histidine
Arai, S., T. Nose, and Y. Hashimoto. requirements. Abstract of the Spring Meeting of Jpn. Soc.
1972. Amino acids essential for the growth of eels, Anguilla Sci. Fish., p.44. (Injapanese).
anguillaa-ndA.japonica. Bull. jpn. Soc. Sci. Fish. 38(7):753- Poston, H.A., and G.F. Combs, Jr.
759. 1980. Nutritional implications of tryptophan catabolizing en-
Arai, S., A. Nakazawa, and Y Deguchi. zymes in several species of trout and salmon. Proc. Soc.
1986. Effects of each essential and non-essential amino acids Exp. Biol. Med. 163:452-454.
on free amino acids in whole body of ayu fish. Abstr. of the Poston, H.A., and G.L. Rumsey.
Autumn Meeting ofJpn. Soc. Sci. Fish., p.147. (Injapanese.) 1983. Factors affecting dietary requirement and deficiency
Dupree, H.K., andj.E. Halver.
1970. Amino acids essential for the growth of channel cat- signs of L-tryptophan in rainbow trout. J. Nutr. 113:2568-
fish, Ictaluruspunctatus. Trans. Am. Fish. Soc. 99(1):90-@92. 2577.
Egusa, S. 4 Sakaguchi, S., T. Hara, T. Matsusato, T. Shibahara, Y. Yamagata, H.
1985. Myxobolus buii sp. n. (Myxosporea: Bivaivulida) parasitic Kawai, and Y. Maeno.
in the brain of Seriola quinqueradiata TEMMINCK et 1987. Scoliosis of cultured yellowtail caused by parasitic
SCHLEGEL. Fish Pathol. 19 (4):239-244. (In Japanese; En- Myxobolus buri. Bull. Natl. Res. Inst. Aquaculture 12:79-86.
glish abstr.) (Injapanese; English abstr.)

12 NOAA Technical Report NNEM I I I

Sato, M., T. Kondo, R. Yoshinaka, and S. Ikeda. Tanaka, C., and M. Kimura.
1983. Effect of water temperature on the skeletal deformity 1984. Serotonin(5-hydroxytryptamin). In Neuro transmitters
in ascorbic acid-deficient rainbow trout. Bull. Jpn. Soc. (G. Takagaki and T. Nagatsu, eds.), p.156-191. Kodansha,
Sci. Fish. 49(3):443-446. Tokyo. (Injapanese.)
Shanks, W.E., G.D. Gahimer, andj.E. Halver. Walton, MJ., R.M. Coloso, C.B. CoweyJ.W. Adron, and D. Knox.
1962. The indispensable amino acids for rainbow trout. Prog. 1984. The effects of dietary tryptophan levels on growth and
Fish-Cult. 24:68-73. metabolism of rainbow trout (Salmo gairdnen). Br. J. Nutr.
Shiomitsu, K. 51:279-287.
1982. Isolation of Streptococrus sp. from the brain of cultured Yone, Y.
yellowtail. Fish Pathol. 17(l):27-31. (In Japanese; English 1976. Nutritional studies of red sea bream. Rep. Fish. Res.
abstr.) Lab. Kyushu Univ., 3:87-101.
Tamura, T.
1974. An experimental study on scoliosis in bipedal rats with
brainstern lesion. J. Jpn. Orthop. Assoc. 48(3):137-158.
Injapanese. English abstr.)

Control of IHN Virus in Alaskan Sockeye Salmon Culture

THEODORE MEYERS

Alaska Department ofFish and Game
ERE.D. Division
RO. Box 25526
Juneau, AK 99802-5526

A recent review of trends in the prevalence and risk management of Infectious Hemato-
poietic Necrosis Virus (IHNV) in Alaskan sockeye salmon Oncorhynchus net*a culture has
been reported by Meyers et al. (1990). The reader is referred to this report for further
details, discussion, and references. 0
Prior to 1980, IHNV prevented successful culture of sockeye salmon in Alaska. This led
the Fisheries Rehabilitation, Enhancement, and Development Division (FRED) of the
Alaska Department of Fish and Game to develop a policy to control the negative effects
of the virus in sockeye salmon culture. This policy included procedures for the collection
and incubation of eggs and for the rearing of fry that were based upon the known and
suspected biological characteristics of the virus-host relationship. Many of these criteria
are common sense approaches such as: use of a virus-free water supply; disinfection of
utensils, facilities, and external surfaces of broodfish; separate fertilization of eggs from
each female using 1 or 2 males; separate water hardening of each family of eggs in a 100
ppm iodine disinfectant for I hour; compartmentalization of families into Kitoi Box
incubators or into stacks of Nopad trays at egg densities of 250,000-300,000 (80-100
females), or into modified Bams Boxes used at one facility that are each loaded with
500,000 eggs; physical isolation of each sockeye stock and isolation of all sockeye stocks
from any nonsockeye species; and release of fry unfed or after short-term rearing (4-6
weeks) with pooling of fry in raceways or start tanks according to the date of eggtake.
These criteria nearly eliminate opportunities for horizontal virus transmission from the
parents to offspring or from the water supply. They also further reduce the rare occur-
rence of vertical transmission of the virus within the egg and allow for
compartmentalization of eggs and fry so that the occasional incubators or raceways of
fish developing IHN can be destroyed and the virus contained to protect the remaining
fish inventory. This "sockeye culture policy" has allowed Alaskan hatcheries a great mea-
sure of success in controlling IHNV @t several different facilities around the state for
nearly 10 years. Based on these guidelines, an average of 2-3 million. sockeye salmon eggs
can be spawned in a day and totals of 20-36 million eggs may be taken at certain
facilities. Although vertical transmission of the virus generally occurs in fry almost every
year at certain facilities, total losses are minimized to between I and 3% of the statewide
fry production. In 1990 only 1% of the sockeye fry were destroyed owing to IHN of 68
million healthy fish that were released. Production data from various Alaskan sockeye
salmon hatcheries suggest that vertical transmission of IHNV is more dependent upon
the proportion of high virus-titered fernale fish rather than total virus prevalence. Also,
as one would. expect, the risk of vertical virus transmission increases with increasing
numbers of eggs taken from females of a high-titered stock. Hence, IHN outbreaks are
more common at the larger eggtake facilities that have greater prevalences of high-
titered broodfish.
During the past 14 years, yearly monitoring of sockeye salmon stocks by.FRED has
resulted in a data base summarizing IHNV occurrence in over 96 wild and hatchery
sockeye salmon stocks in Alaska. Yearly prevalence of IHNV ranges from 0 to 100% in
both ripe and postspawned females with as many as 92% within a stock having high titers
(@tlO% Repeated yearly sampling has shown that all anadromous sockeye salmon tested
in Alaska are positive for IHNV. The data base has been useful for examining general
trends and has shown some differences from previously reported IHNV-sockeye salmon
relationships. For example, no significant differences in mean IHNV prevalence were

14 NOAA Technical Report NXM I I I

found in ovarian fluids from postspawned female sockeye salmon vs. those from ripe
females. Furthermore there were no significant differences between geometric mean
virus titers of postspawned vs. ripe female fish, but postspawners did have a significantly
greater mean proportion of high-titered fish. The log value of 10' was selected as the
breakpoint for high virus titers owing to the tendency for bimodality of lHNV titers to
occur at this level in most stocks of Alaskan sockeye salmon. The significance of this
phenomenon needs further investigation. As found by other investigators, the mean virus
prevalence in male fish from 27 stocks of sockeye salmon was significantly less (9%) than
in female fish (40.1
This data base is a useful tool for examining general trends for IHNV within a geo-
graphic area or statewide. However, these 'trends may not always be true for certain
individual sockeye salmon stocks that may be unique due to genetic reasons, the strain
of the indigenous virus, or environmental factors affecting natural virus exposure and
transmission.

Citation

Meyers, T.R.J.B. ThomasJE. Follett, and R.R. Saft.
1990. Infectious hernatopoietic necrosis virus: trends in
pievalence and the risk management approach in Alaskan
sockeye salmon culture. J. Aquat. Anim. Health 2:85-98.

Identification of a Conserved Antigenic Domain in the
Major Capsid Protein of Infectious Pancreatic Necrosis Virus

RJ. BARRIE, C.L. "ON, andj.C. LEONG*
Department of Microbiology
Oregon State University
Corvallis, Oregon 97331-3804

ABSTRACT

The gene for the major capsid protein, VP2, of the Sp serotype of infectious pancreatic
necrosis virus (IPW) was cloned and expressed in Escherichia coli. Nonoverlapping frag-
ments of the VP2 gene were recloned in trpE fusion vectors of the pATH series and the
expressed fusion proteins were characterized for reactivity with antisera to three different
serotypes of IPNV. One clone, pBI0, which contained an insert encoding amino acids 99
to 206 of the VP2 protein, produced a fusion protein recognized by antisera for all three
serotypes. In contrast, the pA43 clone, which contained an adjacent region on the VP2
gene encoding for amino acids 207 to 315, produced a fusion protein that was only
recognized by homologous antisera in Western immunoblots. A comparison of the de-
rived amino acid sequence for each clone with that reported for two other IPNV clones
indicates that the pB 10 region is conserved and the pA43 region is very heterogeneous.

Introduction recognized only three major serotypes characterized
by the following virus isolates: VR299, a North Ameri-
Infectious pancreatic necrosis virus (IPNV) is a can strain; Sp, a European strain which is pathogenic
birnavirus that causes one of the most serious dis- for trout; and AB, a European strain which is
eases in trout and salmon farms in North America, nonpathogenic for trout (Wolf 1988). In our paper
Europe, and Asia. It can also kill a number of we review recent efforts to unravel the mechanisms
nonsalmonid fish species including striped bass that biologically distinguish the different IPNV iso-
(Morone saxatilis), turbot, menhaden, and eels (Wolf lates by characterizing the immunoreactive regions of
1988), and it has been isolated from a variety of ma- the major capsid protein of the Sp serotype of 1PNV.
rine fish and molluscs. The ubiquitous nature of this
birnavirus and its ability to infect such a wide variety
of hosts make this virus'important for scientific study. Methods
Most IPN-V isolates are closely related antigenically
and yet, they exhibit marked differences in host- The viral genome is composed of two double-
range in vivo and in vitro, pathogenicity, and stranded RNA segments, A and B. The B segment
temperature of replication. encodes the viral RNA polymerase, VPI. The A seg-
The most extensive study of the antigenic relation- ment encodes the virion proteins, VP2 and VP3. The
ships of the aquatic birnaviruses compared 175 virus major capsid protein, VP2, is responsible for the in-
isolates from 44 fish and shellfish species from 11 cluction of neutralizing antibodies (Lipipun et al.
countries by reciprocal plaque reduction tests using 1989 ). In addition, there is a nonstructural protein,
polyclonal sera (Hill and Way 1983). From these re- NS, which is an autocatalytic protease responsible for
sults, it was proposed that there are 2 major cleavage of the polyprotein, VP2-NS-VP3, encoded by
serogroups: I containing 9 serotypes which includes the viral genome A segment (Duncan et al. 1987;
171 isolates from fish, and the other containing I Manning et al. 1990).
serotype which includes those viruses isolated from
molluscs. These virus isolates also contain common
immunoreactive determinant(s). Other studies have *Send correspondence to this author

16 NOAA Technical Report NNM I I I

smal
TrPE p Bamm
KW XbW
son
NS Pal
Pst VP2 Arm R TV HIndill
A
Lao P pUC19/A+SAK PATH 1, 2, or 3
4.1 Kb 3.78 Kb

JV Pstvpni BaMHYCIP

ATG-1
Ndel

1,5 Kb

Pat
6MEMMIMTEI SaU3A fregmeM
225 105 321 227 183 235
27 81 37

GEL ISOLATION OF
EACH FRAGMENT
I T4 DNA Ugese Figure I
COLONYIMMUNOBLOT Construction of trp&VP2 gene fusions.
The 1.5 Kb cloned insert containing
T@pE P the IPNV-Sp VP2 gene was restricted
TtPE Sau3A with Sau3Al; the fragments were puri-
froWnerd fied and subcloned into the trpE
fusion expression vectors, pATH 1, 2,
or 3. Recombinants containing frag-
ments in the correct orientation and
.in the appropriate reading frame were
selected by colony immunoblot as pre-
viously described (Gilmore et al. 1988).

Since the entire VP2 gene has been expressed in Results and Discussion
Escheyichia coli as part of a trpE fusion protein (Man-
ning and Leong 1990), it was possible to examine Two recombinant plasmids were identified from the
different regions of the VP2 gene for immunoreactiv- clones derived from the pATH1 vector/insert ligation
ity with a panel of rabbit antisera and monoclo nal mixture, and the trpFVP2 fusion proteins expressed by
antibodies to different serotypes of IPNV. The VP2 these plasmids were characterized by Western
gene was excised from the plasmid pUC19/A+SAK immunoblot analysis. The antisera used to detect the
and cut with the restriction enzyme Sau3A1 which VP2-specific protein in the bacteria had been pre-
generated seven DNA fragments (Fig. 1). These frag- pared against purified virions of IPNV-Sp. The
ments were inserted in-frame with the trpE gene in recombinant plasmid, pB10, was found to produce a
one of the three pATH vectors, pATH 1, pATH2, or 47 kDa fusion protein and the recombinant plasmid,
pATH3, which put the resulting trpFVP2 fusion gene pA43, was found to produce a 52 kDa fusion protein
under the control of the tryptophan operator and (Fig. 2). The VP2 derivation of the IPNV sequence in.
promoter (Dieckmann and Tzagaloff 1985). Recom- the fusion proteins was verified by Western
binants expressing a portion of the VP2 gene were immunoblot using anti-VP2-specific antisera prepared
detected by direct colony immunoblot with anti- against purified VP2 virion protein from the IPNV-Sp
IPNV-Sp sera (Gilmore et al. 1988). strain (Huang et al. 1986). The cell lysate prepared

Barrie et al.: Antigenic Domain in Protein of Infectious Pancreatic Necrosis Virus 17

Figure 2
1 2 3 4 5 6 7 8 9 Immunoblot showing reactivity of 17p&
VP2 fusion proteins with anti-IPNV-Sp
sera. E. coli cells containing the recombi-
M nant plasmids (pB10 or pA43) were
75 kDa --- Mw
O-Vpl
grown to mid-log phase before induction
with 15 @ig/ml indoleacrylic acid. The
cultures were then grown to stationary
VP2
phase before the cells were harvested by
5 0
centrifugation for protein analysis. The
cells were lysed and the proteins sepa-
rated on an SDS-polyacrylamide gel as
39 ---W -4-VP3
previously described (Gilmore et al.
VP3a
27 0 1988). The proteins were transferred to
nitrocellulose and then exposed to anti-
17 IPNV-Sp antisera. Lane i contains
IPNV-Buhl infected fish tissue culture cell
lysate; lane 2, prestained low molecular
weight markers from BioRad at 75 kDa,
50 kDa, 39 kDa, 27 kDa, and 17 kDa;
lanes 3 and 4, pB10 induced bacterial cell
from induced cells containing pBlO or pA43 was lysate; lane 5, pA43 induced bacterial cell
found to contain protein bands that were strongly re- lysate; lane 6, bacterial cells containing
active with the anti-IPNV-Sp/VP2 sera (Fig. 3). the pATH vector with no insert; lane 7,
The DNA sequence of the viral insert in pB10 and bacterial cells without a plasmid; lane,8,
in pA43 was determined after subcloning of the in- low molecular weight markers; and lane
9, purified IPNV-Buhl. The arrow in lane
sert into the M13 sequencing vectors mp18 and 5 indicates the position of the trpFVP2
mp19. Sequence analysis was performed by the modi- fusion protein encoded by the recombi-
fied Sanger dideoxy chain termination method nant plasmid, pA43. The computed
(Davis et al. 1986). The pB10 insert comprised 323 molecular weight of the typ&VP2 fusion
nucleotides encoding 108 amino acids and mapped protein was 56.5 kDa.
to amino acid number 99 to 206 of the VP2 protein
of IPNV-Sp (Mason and Leong, unpubl. data). The
pA43 insert c6mprised two neighboring Sau3A frag- proteins produced by each plasmid were noted. Care-
ments of 297 and 27 nucleoticles. This insert ful analysis of the 3' terminal sequence of both
probably originated from a partial cleavage product plasmids by direct sequence analysis from the recom-
(Fig. 1) and it was mapped to the adjacent region of binant pATH fusion plasmids themselves (Wang et al.
the VP2 protein at amino acid number 207 to 315 of 1988) indicated that the translational termination
the VP2 protein. Although both p1310 and pA43 con- codon, TAG, was present immediately after the end
tained inserts encoding 108 amino acids, striking of both VP2 cDNA inserts. The calculated isoelectric
differences in the observed migration of the fusion points for both fusion proteins was 6.6, and there was

1 2 3 4 5 6 7

Figure 3
Immunoblot of trp&VP2 fusion proteins with antisera
of 75 kDa to the VP2 protein of IPNV-Sp. E. coli cells containing
the recombinant plasmid, pBlO or pA43, were grown
and prepared as described in Figure 2. Lane I con-
56.5kDa 10*0 50 tains purified IPNV-Sp; lanes 2 and 3, lysates from
'T uninduced and induced cells containing pBlO; lane
4, lysate from uninduced cells containing the pATH1
44 39 vector without any VP2 insert; lane 5, prestained low
molecular weight markers from BioRad at 75 kDa, 50
kDa, and 39 kDa; lanes 6 and 7, lysates from
U I U U I uninduced and induced cells containing pA43. The
photograph taken for lanes 6 and 7 was a lighter ex-
> CO
z Itr posure of the immunoblot and these lanes contained
ra < Y) <
LL 0. CL 0. CL -five times as much bacterial lysate as that used in
lanes 2, 3, and 4.

18 NOAA Technical Report NN[FS I I I

p810 insert

SP ISRKYDIQSSTLPAGLYALNGTLNAATFEGSLSEVESLTYNSIMSLTTNPODKA
jA ISRKYDIQSSTLPAGLYALNGTLNAATFEGSLSEVESLTYNSLMSLTTNPQDKV
mi ISRKYDIQSSTLPAGLYALNGTLNAATFEGSLSEVESLTYt4LMSLTTNPQDKV

sp NNQLVTKGVTVLNLPTGFDKPYVRLEDETPQGLQSMNGARMRCTAAIAPRRYEI
ix UNQLVTKGITVLNLPTGFDKPYVRLEDETPQGPQSMNGAPMRCTAAIAPRRYEI
Ni NNQLVTKGVTVLNLPTGFDKPYVRLEDETPQGLOSMNGAKMRCTAAIAPRRYEI

pA43 insert

Figure 4
sp DLPSQSLPPVPATGTLTTLYEGNADIVNSTTVTGDINFSLAEQPAMTRFDFQL
aA DLPSERLPTVAATGTPTTIYEGNADIVNSTAVTGDITFQLEAEPVNETRFDFIL A comparison of the derived amino acid se-
al DLPSQRLPPVPATGTLTTLYEGNADIVNSTTVTGDINFSLMQPANETKFDFQL quence of IPNV-Sp, IPNV- jasper, and IPNV-NI
* * * * * * cDNA inserts present in pBIO and in pA43. The
sp DLMGIJ)NDVPVVTVVS SVLAnMNYRGVSAK4TQS IPTENI TKP ITRVKLSYKI
jx QFLGIJ)NDVPVV7VTSSTLVTADNYRGASAKFTQSIPTFMITKPITRVKLAYQL asterisks indicate differences in amino acids.
KI DFMGLDLDVPVVTVVSSVIATNDNYRGASAKMTQSIPTENITKPITRVKLSYKI The IPNV-Jasper Ua) sequence was taken from
Duncan and Dobos (1986). The IPNV-NI se-
quence was taken from Havarstein et al. (1990).

A B
1 2 3 4 5 6 7 8 9 10 1 2 3 4

130 kDa Iw
75 --ill 50 kDa

mow,-4- 56.5 kDa
50 -10, -4-39

39 -0,

U U I U I U 1 U U I

C) 0 > M
Z
im CO M
CL M a CL CL CL

Figure 5
Immunoblots of trpFVP2 fusion proteins with antisera to the heterologous IPNV strains, IPNV-Buhl and IPNV-EVE. E. coli
cells containing trpFVP2 fusion proteins encoded by the recombinant plasmids, pBlO or pA43, were analyzed for reactiv-
ity with antisera prepared to purified virus of the two heterologous IPNV strains. (A) Reactivity with antisera to IPNV-Buhl.
Lanes I and 2 contain lysates from uninduced and induced cells containing the pATHI expression vector with no viral
insert; lanes 3 and 4, lysates from uninduced and induced cells containing pA43; lanes 5 and 6, lysates from uninduced
and induced cells containing pBlO; lane 7, prestained low molecular weight markers from BioRad; lanes 8 and 9, lysates
from uninduced and induced cells containing pBlO; lane 10, purified IPNV. (B) Reactivity with antisera to IPNV-EVE.
Lanes 1 and 2 contain lysates from uninduced and induced cells containing pBIO; lanes 3 and 4, lysates from uninduced
and induced cells containing pA43.

no dramatic difference in the amino acid composi- gous VP2 proteins in Western immunoblots (R.
tion. Thus far, the only possible explanation for the Barrie andJ. Leong, unpubl. data). Thus, there are
slower migration of the pA43 fusion protein might be conserved linear epitopes among the IPNV strains.
the series of four prolines found towards the amino When the expressed trpE-VP2 fusion proteins were
terminus of this insert (Fig. 4). examined for reactivity with polyclonal anti-IPNV
Polyclonal rabbit antisera prepared to the different sera prepared to three different lPNV serotypes, only
serotypes of lPNV will crossreact with the heterolo- pBlO reacted with the heterologous antisera in West-

Barrie et al.: Antigenic Domain in Protein of Infectious Pancreatic Necm@is Virus 19

ern immunoblots (Fig. 5, A and B). The anti-IPNV- Acknowledgments
Buhl sera was prepared against purified virions of
the Buhl virus isolate which had been previously This publication is the result of research sponsored
characterized as a member of the IPNV-VR299 sero- by the Oregon Sea Grant with funds from the Na-
type found in rainbow trout (Oncorhynchus mykiss) tional Oceanic and Atmospheric Administration,
in North America (Hill and Way 1983). The IPNV- Office of Sea Grant, Department of Commerce, un-
EVE isolate was obtained from Japanese eels der grant no. NA85AA-D-SGO95 (project no. R/
(Anguilla japonica) suffering from branchio- FSD-11) and the United States Department of Agri-
nephritis in Japan (Sano et al. 1981); it has been culture to the Western Regional Aquaculture
antigenically grouped with the AB serotype of Consortium under grant nos. 87-CRSR-2-2319 and 88-
IPNV, which is nonpathogenic for rainbow trout. 38500-4027. Oregon Agricultural Experiment Station
The fusion protein encoded by pA43 was com- Technical Paper No. 9327. We thank L. Bootland for
pletely nonreactive with the heterologous antisera. reviewing the manuscript.
Thus, it appears that the VP2 gene region encoded
by pBI0 contains an antigenic determinant(s)
which is conserved among the IPNV strains exam- Citations
ined and the insert in pA43 encodes a region
which is highly variable. Christie, K.E., L.S. Havarstein, H.O. Djupvik, S. Ness, and C.
A comparison of the, nucleotide sequence and its Endresen.
derived amino acid sequence of 'each insert. with 1988. Characterization of a new serotype of infectious pan-
creatic necrosis virus isolated from Atlantic salmon. . Arch.
that of other published sequences of the VP2 gene Virol. 103:167-177.
showed that the pBI0 region was highly conserved Davis, L.C., M.D. Dibner, andj.F. Batty (eds.),
at the nucleotide and amino acid level (Fig. 4; 1986.. Basic methods in molecular biology. Elsevier Science
Christie et al. 1988; Havarstein et al. 1990). There Pub. Co., NY, 377 p.
were three amino acid differences between the Sp Dieckmarm, C.L., and A. Tzagaloff.
and the jasper isolate (a member of the VR299 sero- 1985. Assembly of the mitochondrial membrane system.
Biol. Chem. 260:1513-1520.
type), and there were only two amino acid Duncan, R., and P. Dobos.
differences between Sp and NI isolates, the latter of 1986. The nucleotide sequence of in fectious pancreatic ne-
which is another IPNV isolate from the Sp serotype crosis virus (IPNV) dsRNA segment A reveals one large ORF
(Havarstein et al. 1990). In contrast, 27 amino acid encoding a precursor protein. Nucl. Acids Res, 14:5934-
5935.
changes between the Sp and jasper isolates were ob- Duncan, R., E. Nagy, P.J. Krell, and P. Dobos.
served for the pA43 gene fragment. Only five amino 1987. Synthesis of the infectious paricreatic necrosis virus
acid changes were found between isolates Sp and polyprotein, detection of virus-encoded protease, and fine
NI. The similarity between the Sp and NI genomes structure mapping of genome segment A coding regions.
indicates that these two isolates are highly related, a J. Virol. 61:3655-3644.
Gilmore, R.D.Jr, H.M. Engelking, D.S. Manning, andj.C. Leong.
finding that is consistent with the findings of 1988. Expression in Eschefichia coli of an epitope of the
Christie et al. (1988). glycoprotein of infectious hematopoietic necrosis virus
protects against viral challenge. Bio/Technology 6:295-
300.
Havarstein, L.S., K.H. Kalland, KE. Christie, and C. Endresen.
Conclusion 1990. Sequence of the large double-stranded RNA segment
of the NI strain of infectious pancreatic necrosis virus: A
In summary, two immunoreactive regions of the viral comparison with other Birnaviridae. J. Gen. Virol. 71:299-
major capsid protein, VP2, have been identified. One 308.
region from amino acids 99 to 206 contains a very Hill, BJ., and K Way. -
1983. Serological classification of fish and shellfish
conserved epitope(s) which was recognized by neu- birnaviruses. Abstract, First international conference of
tralizing antisera to three different IPNV serotypes. european association of fish pathology, Plymouth, England,
Another region from amino acids 207 to 315 contains October, 1983.
a highly divergent epitope(s) that may encode the Huang, M.T.F., D.S. Manning, M. Warner, E.B. Stephens, andj.C.
serotype-specific epitope(s) of an IPNV strain. A sug- Leong.
1986. A physical map of the viral genome for infectious pan-
gestion that the amino acid region from 206 to 350 creatic necrosis virus Sp: Analysis of cell-free translation
encoded the serotype-specific epitope(s) of IPNV was products derived from viral cDNA clones. J. Virol.
made by Havarstein et al. (1990) when a comparison 60(3):1002-1011.
of the deduced amino acid sequence of IPNV-Nl and Lipipun, V., P. Caswell-Reno, Y-L. Hsu, J.L. Wu, M.-C. Tung, P.W.
IPNV-Jasper capsid proteins revealed that this region Reno, W. Wattanarijarn, and B.L. Nicholson.
1989. Antigenic analysis of Asian aquatic birnavirus isolates us-
was very heterogeneous. ing monoclonal antibodies. Fish Pathology 24(3):155-160.

20 NOAA Technical Report NNM I I I

Manning, D.S., andj.C. Leong. Wang, L.M., DX Weber, T. Johnson, and AX Sakaguchi.
1990. Expression in Escherichia coli of the large genomic seg- 1988. Supercoil sequencing using unpurified templates pro-
ment of infectious pancreatic necrosis virus. J. Virol. duced by rapid boiling. Biotechniques 6:839-843.
179:16-25. Wolf, K
Manning, D.S., C.L. Mason, andj.C. Leong. 1988. Infectious pancreatic necrosis virus. In Fish viruses
1990. Cell-free translational analysis of the processing of in- and fish viral diseases (K. Wolf, ed), p. 115-157. Cornell
fectious pancreatic necrosis virus polyprotein. 1. Virol. Univ. Press, NY.
179:9-15.
Sano, T., N. Okamoto, and T. Nishimura.
1981. Anew viral epizootic of Anguilla japonica. J.Fish.Dis.
4:127-139.

Cloning of Hemolysin Genes of Aeromonads

TAKASHI AOKI and IKU0 HIRONO

Department of Biological Resources
Faculty of Agriculture
Miyazaki University
Miyazaki 889-21, Japan

ABSTRACT

The role of extracellular products is critical in the pathogenic mechanisms of bacterial
infections. In Aeromonas spp., hemolysins may be the most important of these products in
establishing and maintaining infections. This report reviews our knowledge of the struc-
ture and expression of hemolysin genes in Aeromonas and discusses preliminary results on
gene homology and ancestry among various Aeromonas spp.

The related species Aeromonas hydrophila (Ljungh erythrocytes (T-lysin) (Rockey et al. 1988). H-lysin
and Wadstr�m 1982) and A. salmonicida (Titball and contains GCAT. Nomura et al. (1988) purified
Munn 1985a) produce several extracellular proteins salmolysin, an extracellular hemo lytic toxin from A.
that are virulence factors. In the study of pathogenic salmonicida. Salmolysin was lethal to rainbow trout
mechanisms.of these bacteria, there has been interest Oncorhynchus mykiss when it was injected intramuscu-
in the role of these extracellular substances as toxins. larly.
A. hydrophila produces a variety of extracellular prod- Almost -all isolates of A. hydrophila and A.
ucts, including a protease, glycerophospholipid salmonicida produce aerolysin, a substance with he-
cholesterol acyltransferase (GCAT), cytotoxin, an en- molytic activity. The level of aerolysin production is
terotoxin, acetylcholinesterase (Nieto et al. 1991), known to vary under different growth conditions,
and hemolysins (Ljungh and Wadstr6m 1982). Extra- and individual isolates can alternate between high
cellular products of A. salmonicida include hemolytic, and low level phases of production. When this hemo-
leukocytolytic, proteolytic, and GCAT activities, (Ellis lysin gene is cloned into E. coli, the gene's
et al. 1981, 1988). The virulence of A. hydrophila and characteristics can be more easily studied.
A. salmonicida is significantly enhanced by their abil- For this reason, we cloned the two hemolysin
ity to secrete hemolysin. Hemolysin may be the most genes from A. hydrophila and the one hemolysin
important of these products in causing tissue damage gene from A. salmonicida to study their structure
and in establishing and maintaining infections with and expression (Aoki and Hirono 1991; Hirono and
Aeromonas. Aoki 1992, a and b).
There are many reports describing the number We have previously reported the cloning of two he-
and nature of hemolysins found in A. hydrophila molysin genes (for aerolysins AHH-1 and AHH-2)
(Ljungh et al. 1981; Thune and Johnson 1986; Asao from A. hydraphila ATCC7966 into a plasmid vector in
et al. 1986) and A. salmonicida (Titball and Munn E. coli K-12 (Aoki and Hirono 1991; Hirono and Aoki
1985a; Rockey et al. 1988). Asao et al. (1986) purified 1991, Table 1). Open reading frames (ORF) of the
and characterized two hemolysins from A. hydrophila AHH-1 and AHH-2 genes were 1,734 and 981 base
which were biologically similar but immunologically pairs (bp), respectively. The sequences included the
distinct. Both hemolysins caused fluid accumulation -10 region and the -35 region of a promoter and a
in infant mouse intestines 'and rabbit ileal loops and ribosome binding site (Shine-Dalgarno sequences)
elicited a cytotoxic effect on Vero cells. Two distinct upstream from the ORF. Two palindromic sequences
hemolysins have also been found in A. salmonicida. were found immediately following thetermination
One is a broad-spectrum hemolysin with maximum site. Analysis of the deduced amino acid sequences
activity against horse erythrocytes (H-lysin) (Titball indicated a highly hydrophobic N-terminal region in
et al. 1985b), and the other is active against trout the AHH-1 gene with the characteristics of a leader

22 NOAA Technical Report NWS 111

10 20 so 40 50 60 70 so so 100
OCATGCMAATCATC(AWTTAGATOOTATMMATccnwTemchAuATAAATVATTCDCMCCACACTGTtAATT('ACGMGAATAATGAATGTCA

Ito 120 130 Ito 150 150 170 180 190 too 210
TGACAPGCAAGCAGAAUACGC=AAATATAA ffTM=ATTMTTCATTOMAAATAGCMCMMr=C&AGOAGAT.AA
45
220 230 240 250 260 270 290 290 Soo 310
WAAAAAUAAAAACCACGCAAATTrATCAMCAAG=CCAM7MTCTCGMCG=WMMZCAGGCAGMTGCAAGOCOAAGATATTOGOGA
-to
320 330 340 $50 260 370 $80 390 400 410 420
ACGTACCGACCMCAM=7G=MGCCrGCAATCCOAACAMMMTTTA=CAATG===MA&MCAGGGIMMACG=CT
so MetLeuAlaSerLeuGInSerG 10MG lyLeu I leTYrLetL&snAlaAspYalTrpLedysGIrGinGlyAI aThrProLe

430 440 450 460 470 480 190 500 510 520
CATGACCMGGATCACCrGMMAGMGGTG=CAOGGGGTGAGMTCMTT(ATOGATrUGC=TCADMACrAGATMAGMACAACAGOOCAGAAA
uMetThrAr&AspGlnLeuArgGluArgValLauAlaArgGlyGlukrW.wPhaI IeAspPheSerA1aValThrAspG1nI leGlaAr&G1nGInAlaArgLy

530 540 550 560 570 580 590 goo 610 $20 630
GGCCATOWCAGCTGGCMGGATCrCMTOATCOGGACTOGGTGCTWMCMGCTAUAGOOV@AGMCMTTCAC=MGGAGOMTOGATGACCC
&A)&NetGluGlnLonAlaGlylleSerPbeAspA]aAspTrpYalLeuVa]SerGlyTyrLysG]yG]uLeuLeuPboThyPrdeuGlyGlyValAspAspPr

640 650 Soo STO Soo 690 100 710 120 730
=MCTATCAGCMATMMGGTMAGCMGAAGOGUMGCAACDGCCACAAGMCrCGCTGACMAGOMOMCMCCGAGGCCGGTCTGCM
oAlaPheTyrOInLeuMtGluArgYalGluSerLeW luGlyGInGlYA8nGlyRieLyaArgSarLwThrGInProProAlaAlaOluAlaGlyLeuProilI

740 750 780 770 780 790 800 $to 820 830 840
TGTGG=MACCTCAAMTUA=CAAGATCAGMATCCMAGTOCACM=M=CGTACMMCCMMA=GMITMCGATrCOCC
&V&IAIaPheTyrLaaAenValAsnArgLyalleSerAspAlaOluCyaThrPhaProArgSorArgThrtrpSerArgGlyAsWgLouPheCYaAspSerPr

850 860 870 980 890 goo 910 920 930 940
GMCAT=CTCOTCTACDOWCAAM7MGCOCTCWrWMTTTGGCAACAMOMMMAOGCCOGATOML4uTAGTOMOAT=CTGGAMA
oAsalleSerLeuVallyrArgValAsriLau0luArgSarLeuGInPheOlyAsnThrAlaSerAlaThrProAspAIaLysileValArglieSerLeuAspGI

950 goo 270 980 990 1000 1010 1020 1030 1040 1050
AGAOTCGGCMGTGC=ATOCAOCTUACGAGGATCMAGCTOGAGMAAAACATMMCTATCTGCTW=CM=G=MACTATGOCACMA
tiGiuSerAlaGlyAlaGlylleGInLeubnGluAspLeuSerTrpSerGluAsulleAlaAspTyrLeaLsuLeuAspGly7rpAlaArgAepTyrAIRThrAs

1060 1070 1080 1090 1100 1110 1120 1130 1140 1150
TGOCATCGCCCAGGATTATMCMCOACMAGGMTMAACACUMGCMMGMCTCAAGAGOMMMOCAAOCTCAAuGcuffAcoAcCATOOMA
pAlaileAlaGInAspTyrArsPhoThrThrOluAlaSerAenThrLysAlaAlaVelLeuLyeSerLeuProThrA&nLwAgnSerLYsTyrOluBigAr&iGI

1160 1170 1180 1190 1200 1210 1220 1230 1240 1250 1200
GATCTM'MMGMMGTCA=MGGGTMAGGTGAACAACOATGGCCCAAGGCUAGCMAGWCGGCCAAMCAGMAGCAGMCCAGCTC
alleSerGlyPheGluValGlyValThrGlyGlyValGluValAenLyeAspOlyProArsProSerTrpArgAr&ArsProSerSerAlaSerSerAlaSerSe

. 1270 1280 1290 1800 1310 1320 1330 1340 1356 logo
GCCTACAACACCrAGGATTACCGGGTrGAACGC==AGMMGAAGGMAOTTT(AGC=MMMATCAATAOGMACCGUGAMWWCT
rProThrThrPr*Argl leThrOlyLeuAsnAlaProMaCluArgProOluGlyGluPheGtrLLeuDlYAlaArgSerI leArgAsPArgAr&ValProAlaLe

1370 1390 1390 1400 1410 1420 1430 1440 1450 1460 1470
OCTCCAACAMOCCAMTCMCATGGCTAMAOffGGATCACAACMCATMW=CTCAGCTACAAGGGGTTMTGCMMTCMAMTCATCTACAAGOC
uLeuGInAspGlyHi8ValTrpAlaTrpLeuArgArgOlySerGInProHisProAlaAlaGlnLeuGIrLGIyVaICY&AIsGiuSerAspVaI I I e7yrLysA I

1480 1490 1500 1610 . 1520 15SO 1540 1550 1560 1570
GGCGCMGAMAGAMGGCAGCACCOAGTTCAAGATCOACTCCMTCAACATCOOCCCCATCrAcAa=UTLTACAAGCACTACtAMMMGGGC=
aAl droAspOluThrGlySerThrOluPhdyalleAspUrSerValAorLIleArgProlleTYrThrGlYlleTyrLysH[sTyrTyrValValGlyAlaUt

1580 1590 1600 1610 1620 1630 1640 1650 1660 1670 1680
TG7CMCTrWAGGGMMMOATACCGAUMCGCAr,AMGGTGGCGGCGTCUCCAGCrTUAGGTGOACTMUCCACCMGMMACMGC(;GTCO=
sValSerPhaOIROlyPhoOluAspThrAspLy&ArgArgAr&ValAlaAlaSerThrSer?heLysValAepTrpAsngleProValPhoThrGlyGlyArgPr

1690 1700 1710 1720 1730 1740 1750 1760 1770 1780
COMA 'OMCAGCMGGGUMACMCMMCCMAGMMOATOCCAACCATMTCrGAGOGCGGTGACCMMAGAMM=CMAGTCM
oValAinLeuGinLouG[YGIYPheAspAoMrgCysLeuSerA[aAW laAanHisGlyLeu$erAlaValThrPheAspGluThrSerAlaAlaGInSerSe

1790 1800 1810 1820 logo 1840 1950 1860 1870 1880 1890
CAlWATGAMMAMGCMCrAMUGCGCGCAGGATA=CMCMWMATGGUACAAWnMMGMCAGAGCTGCAGCCMAG=M=
rileTyrAspOluTyrOlyArgryrValSerAlaGInAaMrLauArgCnLeuAspGlyAsMsnLeuGlYGlnLeuGInSerCysSarLouSerLeuGlyGI

1900 1910 1920 1930 1940 1950 1960 1970 1980 logo
GOG=GAffOGAMGMGACAGCGATGCGCTCAGCAAWrGAGTGCMAMAGCTGmGmcAmAcAACUaAGMGGGCOMMGCTCTACOACGACAA
nArgTrPGIUTrPLY&AlaAspSerAspAiaLeuSerAonLeaSerAlaNiSCInLeuLeuVallisAspLyaGInSerOlYAtaLeuDlyLeuTyrAspOluAs Figure I
2000 2010 2020 2080 2040 2050 2060 2070 2090 2090 2100 Nucleotide sequence and deduced
=UAT=CWAA7MOMTACGGACCMA==ArAC=CATCnMMccAmm=MUC7r,ACAGAGCAMGWMMMAGACAACC
nGlyAsnProG]nAsnValSerVa]ArgThrLeuThrSerTyrThrArgl]ePbeGlyProProAlaSerRi9 amino acid sequence of hemolysin
2110 2120 2130 2140 2150 2160 2170 2180 2190 2200 gene AHH-1 from Aeromonas hydrophila
ACDACGCCAAGMCMGMTTTMATGAGGGGOGGATCAGGI)CMTTTTCrAMAGMMGOGAGCACOTCM=GCMAGAAGGTCTGCAC==G ATCC7966. The deduced amino acid
PWWMM BUWUG sequence is given under the nucleo-
2210 2220 2230 2240 2250 2260 2270 2280 2290 2300 2310
MCMGGCGCCACACCAG=ATCTMCGGTAGTGAMMCMGCAGMMATMACCAGAT=MTMTMAGGATC=CATCGATMCrAICTGe tide sequence. A putative promoter is
2S20 2830 2340 2350 2360 2S70 2380 2390 2400 2410 indicated by the areas marked -35 and
GGCAGGMCGTMMAG=GCGCMACCATMCACCAGGMMCAGOCTGMMC,WMGGTMATCMMMTCA=CAG=CAGGGC -10, and ribosome binding site is indi-
2420 2430 2440 2450 2460 2470 2480 2490 NO 2510 2520 cated by SD. The palindromic structure
rACCGWMn=GTGAGAGAGTCGTCCCOCTOCAOCAGGAAGATACTCfGGTMCAGCTMTMACTMAAM=mTr,cArGrMG(rAmo.Tc4w,G sequence of a terminator-like region is
also indicated downstream from the
AHH-1 coding region.

Aoki and Hirono: Cloning of Heniolysin Genes of Aeronionads 23

of E. coli and the plasmid-encoded proteins examined
Table I by polyacrylamide gel electrophoresis (Sancar et al.
Maximum matching comparison of amino acid se-. 1979). Strains containing pAHH-1 and pASH-1 pro-
cluences of cloned hcmolysin genes of Aeromonas duced proteins of 62 kDa and 56 kDa, respectively
hydrophila and A. salmonicida. (Hirono and Aoki 1991, 1992a). These results are in
AHH-1 AHH-2 ASH-1 aer agreement with the size of the proteins predicted by
the DNA-sequenced ORK However, the molecular size
AHH-1 of novel protein synthesized by cells harboring the
AHH-2 17.4% pAHH-2 plasmid was 26 kDa., The molecular size of
ASH-1 17.0 17.0 the protein was clearly different from the size pre-
aer* 17.1 16.9 18.5 dicted by the hemolysin AHH-2 gene sequence which
Howard and Buckley 1986. inferred a protein with a molecular weight of 37.7
kDa. The products of transcrip6on or translation in
AHH-2 may be either greatly truncated or reduced in
quantity. As mentioned above, analysis of the deduced
amino acid sequences indicated that the N-terminal
peptide (Figure 1). However, the N-terminal region. region of the AHH-2 gene was not hydrophobic. As
of the AHH-2 gene was not hydrophobic. yet, it is difficult to explain the difference in the mo-
Two aerolysin genes were also cloned from A. lecular size of the final protein.
hydrophila, Ah65 (Howard and.Buckley 1986; Howard Hemolysin was released by the E. coli cells contain-
et al. 1987) and AH2 (Chakraborty et al. 1986). The ing the AHH-1 gene and those containing the ASH-1
nucleoticle sequence of the Ah65 aerolysin gene gene. The hemolytic activity in a supernatant was in-
was 1,458 bp. There was very low homology between activated by heating to 70' C for 10 minutes.
the abrolysin gene from Ah65 and each of the However, the hemolysin was expressed but was not
ATCC7966 genes, and there were no indications of secreted from E. coli cells carrying the recombinant
similarity in the predicted amino acid sequences plasmid containing the AHH-2 gene. The aerolysin
(Table 1). cloned by Howard and Buckley (1986) similarly was
We also cloned one hernolysin gene (ASH-1) from not released from the E. coli cells in which they were
A. salnionicida ATCC14174 (Hirono and Aoki 1992, cultured. The E. coli hemolysin requires four cistrons
a and b) (Table 2) which had an ORF of 1,716 bp. It encoded with the hemolysin structured protein for
had the -10 region and the -35 region of a putative full expr .ession of active protein to be achieved
promoter and a ribosome-bin ding site up stream (Felmlee et al. 1985). These proteins are associated
from the CIRF, and the termination codon and with the cell envelope and are involved in transport-
palindromic sequences downstream from the ORK ing the hemolysins out of the cells. This complex
The N-terminal region of the ASH-1 gene was highly system is unlike the one producing the extracellular
hydrophobic. Comparative analysis of the fundamen- hemolysins in A. hydrophila and A. salmonicida. It is
tal molecular structures of our cloned AHH-I, very interesting that the hemolysin release mecha-
AHH-2, and ASH-1 genes, and the previously re- nism produced by E. coli cells containing the hemo-
ported aerolysin gene from Ah65 (Howard et al. lysin gene is different from Aeromonas spp.
1987) suggests that they have not descended from a Using colony hybridization analysis, the cloned he-
common ancestor. molysin genes were tested for presence of
The recombinant plasmids pAHH-1, pAHH-2, and homologous regions in isolates A. hydrophila and A.
pASH-I were introduced into a maxicell strain CSR603 salmonicida from humans and fish in Japan (Hirono

Table 2
Cloned hemolysin genes from Aeromonas hydrophila and A. salmonicida.

Hemolysin Open reading predicted molecular size Molecular size
genes Sources frame (bp) from DNA sequences (Da) from Maxicell (kDa)

AHH-1 A. hydrophila 1,734 63,658 62 (Hirono and Aoki 1991)
AHH-2 A. hydrophila 981 37,797 26 (Aoki and Hirono 1991)
ASH-1 A. salmonicida 1,716 64,780 56 (Hirono and Aoki 1992a)

24 NOAA Technical Report NXIFS I I I

and Aoki 1992, a and b). Ten of 15 strains of A. of A. hydrophila and A. salmonicida isolated in the
hydraphila isolated from humans and 14 of 33 strains United States for cloning the hemolysin gene and
from fish possessed sequences homologous to the he- tested for homology with strains isolated in Japan.
molysin gene of AHH-1 (Table 3). The AHH-1 gene Homology of the tested strains with the cloned hemo-
was detected in all 38 strains of A. salmonicida. The lysin was low. We shall attempt to survey the
AHH-2 gene was detected only in the original strain distribution of our cloned hemolysin genes in more
ATCC7966 and not in the other strains of A. isolates of aeromonads. The presented data, however,
hydrophila and A. salmonicida tested. The ASH-1 gene indicate that the hemolysin genes having different
was detected.in two strains of A. hydrophila from fish structures occur in A. hydrophila and A. salmonicida
and the original ATCC14174 strain. We used strains isolates from different geographical locations.

Table 3
Detection of DNA sequences homologous to the hemolysin genes from Aeromonas hydrophila and A. salmonicida by
colony hybridization. (Hirono and Aoki 1991, 1992b.)

Strains Sources AHH-I AHH-2 ASH-1

A. hydrophila Human 10/15* 0/4 0/15,
Fish 14/33 0/5 2/33
ATCC7966 1/1 1/1 0/1
A. salmonicida Fish 38/38 0/19 0/38
ATCC14174 1/1 0/1 1/1

Number of strains containing hemolysin genes/Number of tested strains.
Hemolysin genes of the AHH-1 and AHH-2 of A. hydrophila ATCC7966 and the ASH-I gene of A. salmonicida ATCC14174 were
cloned into plasmid vectors in E. coli K-1 2.

Citations ings of the Symposium on Diseases in Asian Aquaculture.
Asian Fisheries Society@ (In press.)
Aoki, 1, and 1. Hirono. 1992b. Nucleotide sequence, expression, and characteriza-
1991. Cloning and characterization of the hemolysin deter- tion of a hemolysin gene from Aeromonas salmonicida.
minants from Aeromonas hydrophila. J. Fish Dis. 14:303-312. Submitted to infection Immun.
Asao, T., S. Kozaki, K. Kato, Y. Kinoshita, K_ Otsu, T. Umemura, Howard, S.P., andj.T. Buckley.
and G. Sakaguchi. 1986. Molecular cloning and expression in Escherichia coli of
1986. Purification and characterization of an Aeromonas the structural gene for the hemolytic toxin aerolysin from
hydrophila hemolysin. J. Clin. Microbiol. 24:228-232. Aeromonas hydrophila. Mol. & Gen. Genet. 204:289-295.
Chakraborty, T., B. Huhle, H. Bergbauer, and W. Goebel. Howard, S.P., W.J. Garland, M.J. Green, andj.T. Buckley.
1986. Cloning, expression, and mapping of the Aeromonas 1987. Nucleotide sequence of the gene for the hole-forming
toxin aerolysin of Aeromonas hydrophila. J. Bacteriol.
hydraphila aerolysin gene determinant in Escherichia coli K- 169:2869-2871.
12. J. Bacteriol. 167:368-374. LJungh A., and T. Wadstr6m.
Ellis, A.E., T.S. Hastings, and A.L.S. Munro. 1982. Aeromonas toxins. Pharmacol. & Ther. 15:339-354.
1981. The role of Aeromonas salmonicida extracellular prod- LJungh, A.,-B. Wretlind, and R. M611by.
ucts in the pathology of furunculosis. J. Fish Dis. 4:41-51. 1981. Separation and characterization of enterotoxin and
Ellis, A.E., A.S. Burrows, and K-J. Stapleton. two haemolysins from Aermnonas hydrophila. Acta Pathol.
1988. Lack of relationship between virulence of Aermonas Microbiol. Scand. Sect. B. Microbiol. Immunol. 89:387-397.
salmonicida and the putative virulence factors: A-layer, extra- Nieto, T.P., Y Santos, L.A. Rodriguez, and A.E. Ellis.
cellular proteases and extracellular haemolysins. J. Fish 1991. An extracellular ace tylcholinesterase produced by
Dis. 11:309-323. Aeromonas hydrophila is a major lethal toxin for fish. Microb.
Felmlee, T., S. Pellett, and R.A. Welch. Pa thog. 11:101-110.
1985. Nucleotide sequence of an Escherichia coli chromo- Nomura, S., Fujino, M., Yamakawa, M., and E. Kawahara.
somal hemolysin. J. Bacteriol. 163:94-105. 1988. Purification and characterization of salmolysin, and
Hirono, L, and T. Aoki. extracellular hemolysin * toxin from Aeromonas
1991. Nucleotide sequence and expression of a hemolysin gene sabnonicida. J. Bacteriol. 170:3694-3702.
from Aeromonas hydrophila. Microb. Pathog. 11:189-197. Rockey, D.D.J.L. Fryer, andj.S. Rohovec.
Hirono, I., and T. Aoki. 1988. Separation and in vivo analysis of two extracelluar
1992a. Molecular cloning, expression, and nucleotide sequence proteases and the T-hemolysin from Aeromonas
of a hemolysin gene from Aeramonas salmanicida. The Proceed- salmonicida. Dis. Aquat. Org. 5:197-204.

Aoki and Hirono: Cloning of Hemolysin Genes of Aerontonads 25

Sancar, A., A.M. Hack, and W.D. Rupp. Titball, R.W., and C.B. Munn.
1979. Simple method for identification of plasmid-coded 1985a. Interrelationships of extracellular products from
proteins. J. Bacteriol. 137:692-693. Aeromonas salmonicida. In Fish and shellfish pathology, (A.E.
Thune, R.L., and M.C. Johnson. Ellis, ed.), p. 61-68. Acad. Press, NY
1986. Aeromonas hydrophila B-haemolysin: purification and ex- Titball, R.W., and C.B. Munn.
amination of its role in virulence in O-group channel cat- 1985b. The purification and some properties of H-lysin from
fish, Ictaluruspunctatus (Rafinesque). J. Fish Dis. 9:55-61. Aeromonassalmonicida. J. Gen. Microbiol..131:1603-1609.

Harmful Red Tides. of Heterosigma akashiwo

TSUNEO HONJO

Nansei National Fisheries Research Institute
Fisheries Agency
Ohno, Saeki, Hiroshima'739-04, Japan

ABSTRACT

The raphidophyte Heterosigma akashiwo is one of several species of flagellates that
cause harmful red tides. This paper reviews the distribution of H. akashiwo, the relation-
ship between eutrophication and the occurrence of red tides in the Seto Inland Sea, and,
environmental and biological features of red tide development (life cycle, growth dy-
namics, and allelopathy).

Introduction Distribution

The raphidophyte Heterosig7na akashiwo is one of sev- Heterosigma akashiwo occurs in the temperate and sub-
eral species of flagellates causing prodigious. red tropical embayments in Japan (Hara and Chihara
tides. The total damage by H. akashiwo red tides has 1987), Singapore (Taylor 1990), New Zealand (Taylor
amounted to about 2 billion yen over a period of 16 1990), England (Lackey and Lackey 1963), and Bel-
years (1972-87) in the Seto Inland Sea of Japan. gium (Conrad and Kufferath 1954), and in the
Most of this damage affected fish culture operations. eastern (Tomas 1982) and western areas (Lackey and
This paper seeks to advance research on red tides by Clendenning 1965; Taylor 1990) of North America,
revie*wing the known ecological features of Hetero- Bermuda (Tomas 1982), and Chile (Taylor 1990)
sigma red tides. (Fig. 1). Damages to fish have been recorded for the
Seto Inland Sea (yellowtail, Seriola quinqueradiata and
red sea brearn, Pagrus'major), New Zealand (salmon),
Morphology British Columbia (salmon), and Chile (salmon) (Tay@
lor 1990).
The cells of Heterosigma akashiwo (HADA) Hara et
Chihara are yellow-brown'and ovoid, and slightly
compressed dorso-ventrally. The cell size is 8-25 X Eutrophication and Red Tides
6-15 pm. This organism is thought to lack a cell In the Seto Inland Sea, nitrate, inorganic nitrogen
wall and be limited only by a single membrane. The (ammonia + nitrite + nitrate), and phosphorous con-
delicate structure of the cell has hindered the study centrations have increased rapidly since the
of its surface Morphology, although the existence of mid-1960s (Fig. 2). Nutrient concentrations reached
an external structure, the glycocalyx, has been dem- a maximum in the mid-1970s then gradually de-
onstrated (Yokote et al. 1985). In Japan, H. akashiwo creased thereafter. The total number of red tide
has been referred to as Entornosigma sp. and H. occurrence .s in the Seto Inland Sea before 1965 was
inlandica; in many other countries it has been con- less than 50 cases per year. Beginning in 1968, the
fused with the chrysophyte Olisthodiscus luteus number increased year by year, reaching a peak of
CARTER. Recently, Hara and Chihara (1987) re- 299 cases in 1976. After 1976, the number decreased
ported that Entornosigma sp. and H. inlandica are to 160-170 cases per year. The increase in the 1-970s
synonymous with H. akashiwo, and that most red coincided with a rapid development in the Japanese
tides ascribed to 0. Iuteus actually involve H. economy, with a time lag of 2-3 years. The decrease
akashiwo. in frequency after the first oil shock (197a-1974) cor-

28 NOAA Technical Report NNIFS 111

cog

Figure I
Distribution of Heterosig7na akashiwo red tide (dotted areas). Fish indicate places where cultured fish have sustained damage.

300
\Z
WZ+ z
E C A@
ff E
=4 !E + -k Nitrate, nitrite and
- arnrnonia in Bisan Strait
4- 8- 2 -0.8

0 Phosphate in Bisan
200- 3- 6- ral
0
0
.0
0 0/
2 4 tO - 0.4
a te. Figure 2
inHi hima Bay Total number of
E red tide cases (open
1- 2 *5 - 02
z 0 Phosphate ....... 6-1 barsl and of Hetero-
100 - , @-O-- - -*-* ---- in Hiuchi Nada .... sigma akashiwo red
0 L o@-O 0 10 0 tides (black bars)
and concentrations
of nitrogen (open
circles and solid
lines) and phospho-
lad rous (black circles
and dashed lines)
0 n
1960 WX IM in the Seto Inland
Sea (1950-89).

Honjo: Red Tides of Hekirosigma akashiwo 29

Red Tide -summer ----O@Autumn

0
Affl
0 .............
t
iatoms le
Ivertical migrat@ion
0 0 j div day
1-2div/dqy 1-2div/da
0 (exponential)
0 0 2-5div/day 0 0
apid lysis
overwintering 0
lysis
-5 vegetative cells et ama 0
CL - ----- - -----
0 7m 0 0 ------ 0 -------------- --- Pycnocline 0
C3
Utilization of nutrients 0
0 and growth-promoting \ 0 Water mass with
0 germination substances Veand org nics low oxygen content 0
150C
_,Release from bottom sediment
esting spores
15m- Bottom
resting spores

Figure 3
Schematic diagram of ecological features during the period of red tide development.

responds with the switchover of the Japanese economy middle layer (Fig. 3). Oxygen conce 'ntrations i n
to a lower rate of developryient. The patterns of bottom water decrease rapidly after the formation
change-in the total number of tides and the number of the pycnocline. The pH of the anoxic bottom
of Heterosigma red tides were similar to that of nutrient water decreases to about 7. As a result, nutritive
concentrations. Thus, the frequency of red tide occur- substances (inorganic nutrients, metals, and or-
rences is closely related to eutrophication. ganic constituents including growth-promoting
substances) are released from the bottom sedi-
ments and these substances accumulate to high
Environmental Features During the Period levels in anoxic bottom waters (Honjo 1974). The
of Red Tide Development Heterosigma population has access to these sub-
stances in the bottom water through diurnal
Terriperatures suitable for the growth of H. akashiwo vertical migration at night.
have been reported fr@rri culture experiments to be
in the range of 15-30' C (Tomas 1978a; Mori et al.
1982; Yamochi 1989). This helps to explain why The Development of Heterosigma
Heterosig7na red tides tend to occur in coastal waters Red Tides
from May through late June. However, the range of
suitable salinity differs among culture strains: 30%o
for a Fukuyama strain (Iwasaki et al. 1968); 10%o for Growth originates from cell stocks that overwintered
a Gokasho strain (Iwasaki and Sasada 1969); 27-28%o as motile forms (Yamochi 1989) and/or that germi-
for a Hakata strain (Honjo, and Hanaoka 1973); 12- nated from resting cells (Imai et al. Nansei National
40%o for a Narragansett strain (Tomas 1978a); and Fisheries Research Institute pers. commun. 15 Octo-
12-28%o for a Tanig4wa strain (Mori et al. 1982). ber, 1990). Tomas (1978b) and Yamochi (1989)
This suggests that these strains have developed a found that motile cells aggregate and change to
physiological acclimation to the range of salinity in nonmotile cells. Population growth can be divided
each habitat. into two phases: rapid and exponential (Fig. 3; Honjo
0

0
0

verwi ntering
ge tive cell
ve 5ta
s
0

g Or
15

Growth of H. akashiwo is usually initiated in early and Tabata 1985). The rapid phase is defined as the
summer when a pycnocline is formed in the period when cells increase more than four-fold each

30 NOAA Technical Report NWN I I I

day and the exponential phase as that period. when ARelopathy
the correlation coefficient between log cell number
and time is greater than or equal to 0.95 and there The study of allelopathy is important for elucidating
is an obvious increase in the number of cells for the mechanisms of temporal succession of phyto-
more than 4 days. The duration of Heterosig7na red plankton and of monospecific bloom events.
tides is shorter than for other flagellates. The Allelopathic interactions in which organic metabo-
abrupt disappearance of Heterosigma cells requires lites of one plant or microorganism suppress or
further study. enhance the growth of other plants or microorgan-
isms are different from. competitive interactions
which involve the removal or reduction of certain fac-
tors such as water, minerals, foods, and light. The
Growth Rate and Division Periodicity interactions among marine and freshwater phyto-
plankton have been well studied and have been
Blooms of thi .s organism develop so rapidly and dra- reviewed by Smayda (1980), Maestrini and Bonin
matically that many workers have studied the growth (1981), and Rice (1984). Some of the most dramatic
of H. akashiwo in vitro and in situ as an a proach to changes of species composition in marine phyto-
p plankton have been observed during Heterosigma red
understanding the enigma of red tide formation. tide blooms, and allelopathic interactions have been
Their results, however, have conflicted (Honjo and observed between H. akashiwo and the centric diatom
Hanaoka 1973; Tomas 1978a). Skeletonema costatum during in vitro experiments
Honjo and Tabata (1985) studied Heterosigma (Pratt 1966; Honjo et al. 1978). Their studies imply
growth dynamics and division periodicity in outdoor that an allelopathic relationship between these or-
tanks. They found that H. akashiwo has a potential for ganisms occurs during high cell densities of H.
high growth rates (2-5 divisions per day). There was akashiwo. Stuart (1972) and Sakshaug (1977) exam-
a strong tendency for large cells to dominate at the ined physical features of the allelopathic substances
beginning of this fast growth phase and to be re- but experienced difficulty in determining their mo-
placed by small cells toward the end of the phase. lecular weight.
During culture experiments, a portion of a H. Recently Honjo et al. (unpubl. data) investigated
akashiwo patch in a tank was cultured in a I-liter volu-
metric flask hung in the tank. The growth rate in this the biological and chemical features of allelopathic
experiment was 2.3 divisions per day. Other H. substances from H. akashiwo (Fig. 4). During a bloom
akashiwo cells were collected from the tank and indi- of H. akashiwo in an outdoor tank, an abrupt de-
crease in cell numbers of centric diatoms (dominant
vidually cultured in small tissue chambers. In these species, S. costatum) and an increase in the dinophyte
chambers the highest growth rate was 3.3 divisions Prorocentrum triestinum occurred when high concen-
per day; the growth rates of large cells inoculated trations of dissolved carbohydrates were detected in
into these chambers were much higher than those of the water. Crude polysaccharide extracts from
small cells. Heterosigma cells and from filtrates of the bloom sea-
In the tank environment, H. akashiwo prefers dark- water were both separated into two main fractions. by
associated cell division. Cell division first occurred a gel chromatography. In bioassay experiments, a mac-
little before sunset and continued through the night romolecular fraction from Heterosigma cells greatly
until the next morning. Similarly, in the small cham- suppressed the growth of S. costatum at concentra-
bers cell division began just before the onset of the tions above 1.0 jig glucose per mL, whereas this same
dark period and continued for 4-5 hours into the fraction enhanced the growth of P triestinum and
next light period. Seven of 24 H. akashiwo cells di- H. akashiwo and had no effect on the growth of
vided three times during the night in the small Phaeodactylum sp. The other fraction caused moder-
chambers, with a cell division interval of about 6 ate suppression of S. costatum growth. Results of
hours. Puiseux-Dao (1981) comprehensively reviewed bioassays using S. costatum and the two fractions from
the events of the cell cycle and Chisholm (1981) de- H. akashiwo bloom water were similar to those using
scribed the chronobiology of cell division in the fractions from Heterosigma cells. Histochernical
unicellular algae. In their reviews, the cell division analysis of the crude polysaccharide extracts from
cycle was discussed in terms of the length of the cir- Heterosigma cells indicated a polysaccharide-protein
cadian period (I division per day), but no complex with features analogous to the glycocalyx on
consideration was given to the rapid division cycles of the cell surface of the organism. Results suggest that
H. akashiwo. a polysaccharide-protein complex exfoliated from

Honjo: Red Tides of Hekmsigma akashiwo 31

fn
growth
cell suppresAion

lycocalyx

(seawater)
Heterosi ma growth
excreti n polysaccha ride - --a, cell enhancement
exfoliation protein complex

Heterosigme
red tide

Pro ocentrum growth
enhancement

Figure 4
General scheme of allelopathic relationships between Heterosigma akashiwo and other phytoplankton.

and/or excreted by H. akashiwo is a species-specific Honjo, T., and T. Hanaoka
1973. Studies on the mechanisms of red tide occurrence in
allelopathic substance and plays an important role in Hakata Bay. 11. General features of the red tide flagellate,
causing dramatic changes in cell numbers of other Heterosigma sp. Bull. Plankton Soc. Jpn. 19:75-81.(In
phytoplankton during Heterosigma blooms. Japanese.)
Honjo, T., and I- Tabata.
1985. Growth dynamics of 01isthodiscus luteus in outdoor
tanks with flowing coastal water and in small vessels.
Citations Limnol. Oceanogr. 30:653-664.
Honjo, T., T. Shimouse, N. Ueda, and T. Hanaoka.
Chisholm, S.W. 1978. Changes of phytoplankton composition and its charac-
1981. Temporal patterns of cell division in unicellular teristics during red tide season. Bull. Plankton Soc. Jpn.
algae. In Physiological bases of phytoplankton ecology (T. 25:13-19. (Injapanese.)
Platt, ed.) p.150-181. Can. Bull. Fish. Aquat. Sci. 210. Iwasaki, H. and K. Sasada.
Conrad, W., and H. Kufferath. 1969. Studies on the red tide dinoflagellites .11. On Hetero-
1954. Recherches sur les eau saumatres des environs de sig7na inlandica appeared in Gokasho Bay, Shima Penin-
Lilloo. Perti 11. Descrictive. Inst. R. Sci. Nat. Belg. Mem. sula. Bull.jpn. Soc. Sci. Fish. 35:943-947. (Injapanese.)
127:1-346. Iwasaki, H., T. Fujiyama, and E. Yamashita.
Hara, Y., and M. Chihara. 1968. Studies on the red tide dinoflagellates I. On
1987. Morphology, ultrastructure and taxonomy of the Entomosigma sp. appeared in coastal area of Fukuyama. J.
raphidophycean alga Heterosi&a ahashiwo. Bot. Mag. To- Fac. Fish. Anim. Husb. Hiroshima Univ. 7:259-267. (In Japa-
kyo 100:151-163. nese.)
Honjo, T. Lackey, J.B., and E. Lackey.
1974. Studies on the mechanisms of red tide occurrence in 1963. Microscopic algae and protozoa in the waters near Ply-
Hakata Bay. IV. Environmental conditions during the mouth in August 1962. J. Mar. Biol. Assoc. UK 43:797-805.
blooming season and essential factors of red tide Lackey, J.B. and K-A. Clendenning.
cell
geteros', m2
cell

occurrence. Bull. Tokai Reg, Fish. Res. Lab. 79: 77-121. 1965. Ecology of the microbiota of San Diego Bay,
(Injapanese.) California. Trans. San Diego Soc. Nat. Hist. 14:9-40.

32 NOAA Technical Report NMB I I I

Maestrini, S.Y, and Dj. Bonin. Stuart, M.
1981. Allelopathic relationships between phytoplankton spe- 1972. The effect of 01isthodiscus luteus Carter upon the
cies. In Physiological bases of phytoplankton ecology (T. growth of SkeLetonema costatum (Grev.) Cleve. M.S. thesis,
Platt, ed.), p.323-338. Can. Bull. Fish. Aquat. Sci. 210. Univ. Rhode Island, Kingston, RI, 82p.
Mori, S., Y. Nakamura, M. Watanabe, S. Yamochi, and M. Taylor, F.J.R.
Watanabe. 1990. Red tides, brown tides and other harmful algal
1982. The effect of various environmental factors on the blooms: the view into the 1990s. In Toxic marine phyto-
growth yield of red tide algae. 11. 01isthodiscus luteus. Res. plankton (E. Grameli, B. Sundstr6m, L. Edler, and D.M.
Rep. Nad. Inst. Environ. Stud. 30:71-86. (Injapanese.) Anderson, eds.), p.527-533. Elsevier, NY.
Pratt, D.M. Tomas, C.R.
1966. Competition between Skeletonema costatum and 1978a. 01isthodiscus luteus (Chrysophyceae) 1. Effects of salin-
01isthodiscus luteus in Narragansett Bay and in cul- ity and temperature on growth, motility and survival. J.
ture. Limnol. Oceanogr. 11:447-455. Phycol. 14:309-313.
Puiseux-Dao, S. 1978b. 01isthodiscus luteus (Chrysophyceae) II. Formation
1981. Cell-cycle events in unicellular algae. In Physiological and survival of a benthic stage. J. Phycol, 14:314-319.
bases of phytoplankton ecology (Platt, T. ed.), p.130- 1982. 01isthodiscus luteus (Chrysophyceae) V. Its occurrence,
149.. Can. Bull. Fish. Aquat. Sci. 210. abundance and dynamics in Narragansett Bay, Rhode
Rice, E. L. Island. J. Phycol. 16:157-166.
1984. Allelopathy. Acad. Press. London, 422p. Yamochi, S.
Sakshaug, E. 1989. Mechanisms for outbreak of Heterosigina akashiwo red
1977. Limiting nutrients and maximum growth rates for dia- tide in Osaka Bay, Japan. Bull. Osaka Pref. Fish. Exp. Stat.
toms in Narragansett Bay. J. Exp. Mar. Biol. Ecol. 28:109-123. 8:1-110. (In Japanese.)
Smayda, T. J. Yokote, M., T. Honjo, and M. Asakawa.
1980. Phytoplankton species succession. In The'physiologi- 1985. Histochernical demonstration of a glycocalyx on the
cal ecology of phytoplankton (I. Morris, ed.), p.493- cell surface of Heterosigma ahashiwo. Mar. Biol. 88:295-299.
570. Blackwell, London.

Impact of the Myxosporean Parasite Ceratomyxa shasta on Survival
of Migrating Columbia River Basin Salmonids

J.L. BARTHOLOMEW, J.L. FRYER, andj.S. ROHOVEC*
Department of Microbiology
Oregon State University
Corvallis, Oregon 97331-3804

ABSTRACT

Columbia River Basin salmonids are exposed to the parasite Ceratomyxa shasta during
both their seaward and return migrations. The impact of ceratomyxosis on the survival of
migrating fish is difficult to assess because there are few data on causes of fish mortality
once they are released from the hatchery. This study examines the impact of this disease
on juvenile salmonids by 1) sampling the downstream migrants to determine what per-
cent of fish leaving the basin are infected with the parasite and by 2) determining the
effects of entering salt water on the progress of the infection. Results of comparisons of
methods for diagnosing a C. shasta infection indicate that serological techniques using
monoclonal antibodies are more sensitive than techniques for examination of spores.

Introduction distribution Uohnson et al. 1979; Hoffmaster et al.
1988) indicate that the parasite has spread within the
The Columbia River Basin has long supported an im- Columbia River Basin, but the extent and cause of
portant Pacific salmon fishery for Oregon, expansion is not yet known. The degree of resistance
Washington, and Idaho. However, the resource has to infection among salmonid fish within the basin has
steadily declined even with supplementation of fish also been examined and compared with resistance
from hatcheries. As research efforts are directed to- among fish populations from watersheds where the
ward determining the causes behind the depletion of parasite is not found (Zinn et al. 1977; Buchanan et
Columbia River Basin salmonid stocks, the impact of al. 1983). All groups of fish tested from the Columbia
diseases must be considered. One disease that is enzo- River Basin were relatively resistant. However, the
otic to the basin is caused by the myxosporean parasite ability of many wild and upriver stocks to resist infec-
Ceratomyxa shasta. Although this parasite has devas- tion has not been examined.
tated certain hatchery productions, it is not normally a Further investigations of the impact of
disease of hatchery fish; therefore, its impact on sur- ceratomyxosis on migrating Columbia River salmo-
vival of fish in the wild is difficult to evaluate. The nids were designed to answer 1) how many Columbia
parasite first infects and multiplies in the intestinal River salmonids become infected during their down-
tract of the fish and from that site spreads to other stream migration, 2) how the disease progresses after
tissues. The infection results in tissue necrosis accom- the fish enter salt water, and 3) how effective are the
panied by a severe host inflammatory reaction. Signs methods currently used to diagnose ceratomyxosis.
of infection may include lethargy, darkening of the
body surface, abdominal distension, and hemorrhag-
ing in the area of the vent (Bartholomew et al. 1989a). Materials and Methods
In addition, the disease has a prolonged incubation
period and current diagnostic methods (Amos 1985)
detect the parasite only in heavily infected fish. Examination of Downstream Migrants for
Assessing the impact of ceratomyxosis requires ex- Infection
amining a variety of factors. Studies of geographic
To estimate how many salmonids become infected
Send correspondence to this author during their downstream migration, outmigrants

34 NOAA Technical Report NWS I I I

Cross section
Rock Sand
Cliff Puget Ship beach
Island channel
A 7 m 7 rn 9M 11ml4mlim 7m B

-*-600 rn-i-I 1200 rn I
950 rn ------------- V,-

X.
C11

A
WASHINGTON
Puget Island
River
flow
Cape
-.Horn'
A-
coo oll
Jetty
v e r
C 0 U M
0.0 cy,
14mdepth 200rnwidth
7i-C. p in
- - - - - - - - - - - - - -

n
e: U
10
e
C
W a
'Fish proceisin
'facility
U.

OREGON
0 1
I
Kitometers

Figure I
Jones Beach sampling site. The beach and purse seining areas are indicated by the two asterisks on the main map. From
Dawley et al. (1984a).

were collected just prior to entering the estuary and tember in 1984. In 1984, fish were collected by beach
were maintained in fresh water to monitor disease seine only. All fish were transported to the Round
development. Outmigrating juvenile salmonids of dif- Butte Hatchery Isolation Facility on the Deschutes
ferent species and year classes were obtained by River, a facility operated by the Oregon Department
beach and purse seine from a collection facility oper- of Fish and Wildlife. Holding tanks were I-m
ated by the National Marine Fisheries Service at circulars with a 375-L capacity. The water supply was
Jones Beach (75 river km, measured upstream from free of pathogens and the water temperature was
the mouth of the Columbia River) on the Oregon 10' C. All groups of fish were fed an Oregon Moist
side of the Columbia River (Fig. 1). Seining proce- Pellet diet containing 3% terramycin in the form of
dures were those described by Dawley et al. (1984a). TM50 (Pfizer) as a prophylactic measure against bac-
A purse seine 206 m long and 11 m deep with a mesh terial fish pathogens (Udey et al. 1975). Groups were
of 1-2 cm was used in water about 9 m deep. The held for at least 150 days and observed for develop-
seine was set drifting with the current, then towed ment of ceratomyxosis. Fish deaths occurring within
upstream for 5 minutes before closing and pursing. 10 days aft@r transportation were attributed to han-
Beach seining with a net 95 m long, 5 m deep, with a dling mortality and were not included in the results.
mesh of 1-2 cm, was done in water about 6 m deep at In all experiments, dead fish were collected daily,
the outer end of the net sweep. Fish were collected and either immediately necropsied or frozen for later
on 12 occasions between 20 May and 8 September in examination. All fish held in tanks that remained at
1983 and on six occasions between 5 July and 20 Sep- the end of the observation period were killed and

Bartholomew et al: Impact of Cera"xa shasto on Salmonid Survival 35

examined for C. shasta. Wet mounts of intestinal tract supplied with 126 C pathogen-free water until they
scrapings were examined microscopically (as detailed reached smolt stage. Fish were exposed to the infec-
later) for up to 5 minutes, and samples containing tious stage of C shasta in 'the Willamette Ri@,er at
spores of C shasta were considered infected. Corvallis, Oregon, on three occasions. On the first
occasion, two groups of 50 Alsea steelhead trout
were exposed: one group for 3 days and the other
Determining Saltwater Effects for 5 days. On the second occasion, groups of 50
Alsea steelhead trout and 50 Big Creek coho salmon'
Three strains of salmonid smolts were used to deter- were held for 5 days. On the third *occasion, 100
mine the effects of salt water on the progress of Alsea steelhead trout and 100 Round Bu tte chinook
ceratomyxosis. Big Creek coho salmon salmon were exposed for 3 days. Control groups of
(Oncorhynchus kisutch) were from Big Creek Hatch- an equal number of each species were not exposed
ery located on the lower Columbia River and Round in the river. After exposure, the groups were divided
Butte chinook salmon (0. tshawytscha) Were from and half of the fish were transported to freshwater
the Round Butte Hatchery on the Deschutes River, a holding facilities at the OSU-FDL and the other half
Columbia River tributary. Both strains migrate to ultraviolet-treated saltwater facilities at the Mark
through waters enzootic for C. shasta and are rela- 0. Hatfield Marine Science Center in Newport, Or-
tively resistant to the parasite (Zinn et al. 1977; egon. The fish were fed Oregon Moist Pellets
Johnson e.t al. 1979). Alsea steelhead trout (0. containing 3% terramycin in the form of TM 50
mykiss) were taken from the Alsea Hatchery. The (Pfizer). All groups were held for at least 100 days.
Alsea River empties directly into the Pacific Ocean
and does not harbor the infectious stage of C
shasta. Salmonids from this system have not devel- Comparison of Detection Methods
oped resistance. All fish were held at the Oregon
State University-Fish Disease Laboratory (OSU-FDL) The sensitivity of the standard detection method of
in 1-m diameter tanks of 375 L capacity and were examining wet mounts of material from the lower

Table I
Prevalence of Ceratomyxa shasta in chinook salmon smolts beach seined from the Columbia River at Jones Beach,
Oregon.

Percent of
Date Mortalities fish collected
collected Number Holding infected with infected with
(1983) collected, mortalities C shasta C shasta

May 20 91 3 3 3
27 81 6 1 1

June 3 75 8 2 3
10 53 1 1 2
17 65 36 3 5
'24 130 21 3 2

July 1 113 24 17 15
15 141 46 17 12
29 68 33 2 3

Aug. 12 109 36 13 12
26 34 20 8 24

Sept. 8 112 39 25 22

Total 1072 273 95 9

aNumber collected minus holding mortality during the first 10 days.

36 NOAA Technical Report NNM I I I

Table 2
Prevalence of Ceratomyxa shasta in chinook salmon smolts purse seined from the Columbia River at Jones Beach,
Oregon.

Percent of
Date Mortalities fish collected
collected Number Holding infected with infected with
(1983) collecteda mortalities C shasta C shasta

May .20 9 5 3 33
27 37 6 2 5

June 10 58 17 0 0
17 47 18 4 9
24 98 25 8 8

July 1 128 46 21 16
15 43 38 5 12
29 47 33 11 23

Aug. 12 47 45 4 9
26 21 18 1 5

Sept. 8 14 12 2 14

Total 549 263 61 11

Number collected minus holding mortality during the first 10 days.

Table 3
Prevalence of Ceralomyxa shasta in coho salmon smolts purse seined from the Columbia River atJones Beach, Oregon.

Percent of
Date Mortalities fish collected
collected Number Holding infected with infected with
(1983) collected" mortalities G shasta C. shasta

May 20 82 13 4 5
27 38 4 3 8

June 3 23 1 0 0
10 5 4 1 20
17 1 2 0 0
24 12 2 1 8

July 1 1 0 0 0
15 1 0 0 0

Aug. 26 2 0 0 0

Total 179 26 9 5

Number collected minus holding mortality during the first 10 days.

intestinal wall (Amos 1985) was compared with detec- fish. The first smear was examined for spores as a
tion of C. shasta by serological methods (Bartholo- wet mount at 40OX magnification for a maximum of
mew et al. 1989b). The lower portion of the intestine 5 minutes. The second smear was air-dried, fixed in
was excised from 121 adult salmon that had died 1:1 acetone:xylene solution, and incubated for 15
prior to spawning. Two smears were made from each minutes with a monoclonal antibody specific for

Bartholomew et al: Impact of Ceratomyza shasta on Salmonid Survival 37

Table 4
Prevalence of Ceratomyxa shasta in steelhead trout smolts purse seined from the Columbia River at Jones Beach,
Oregon.

Percent of
Date Mortalities fish collected
collected Number Holding infected with infected with
(1983) collecteda mortalities C shasta C shasia

May 20 11 5 2 18
27 13 4 0 0

June 3 45 13 6 13
10 4 2 0 0
24 1 1 1 100

July I I 1 0 0

Total 75 26 9 12

'Number collected minus holding mortality during the first 10 days.

Table 5
Prevalence of Ceratomyxa shasta in chinook salmon smolts beach seined from the Columbia River at Jones B,Zch,
Oregon.

Percent of
Date Mortalities fish collected
collected Number Holding infected with infected with
(1984) collected' mortalities C shasta C shasta

July 5 88 76 6 1 7
26 75 22 12 16
31 84 31. 14 17

Aug. 16 82 34 11 13
23 87 74 13 15

Sept. 20 47 29 9 19

Total 463 266 65 14

'Number collected minus holding mortality during the first 10 days.

prespore stages of C shasta. Specific antibodies were Results
detected using biotinylated horse anti-mouse IgG
and fluorescein-conjugated avidin D (Vector Labora-
tories, Burlingame, CA). Methyl green dye (1% in Infections in Downstream Migrants
distilled water) was used as a counterstain. Smears
were examined with a Zeiss standard microscope Chinook salmon were the largest group collected in
containing an IV epifluorescence condensor at 1983; 1072 subyearling smolts were captured in
250X magnification until prespore or spore stages beach seines and 549 yearling chinook salmon were
of C shasta were detected or for three minutes. All taken in purse seines. Smaller numbers of coho
microscopic examinations were made by the same salmon (179) and steelhead trout (75) were also cap-
individual. tured in purse seines. Ceratomy'pca shasta was present in
A

38 NOAA Technical Report NWS I I I

Table 6
Effects of salt water on steelhead trout and coho and chinook salmon exposed to the infectious stage of Ceratomyxa
shasta.

Fresh water Salt water
Exposure

length N6. offish No. of fish Percent No. of fish No. of fish Percent
Salmonid (days) recovered' infected infected recovered, infected infected

Alsea steelhead trout 36 21 21 100 6 3 50
31 24 24 100 37 37 100
controld 25 0 0 11 0 0
5b 23 23 100 13 7 54
5` 18 18 100 9 8 89
con trold 25 0 0 16 0 0

Big Creek coho salmon 5b 25 1 4 25 0 0
controld 25 0 0 25 0 0

Round Butte 31 27 0 0 30 0 0
chinook salmon controld 25 0 0 27 0 0

aNumber of fish exposed minus number of fish that died before spores were detected.
'Fifty fish exposed; 25 were transferred to fresh water and 25 to salt water.
One hundred fish exposed; 50 were transferred to fresh water and 50 to salt water.
d Control fish were not exposed to the infectious stage of C shasta.

1-24% of the individuals of the chinook salmon Effects of Salt Water
groups caught by beach seine (Table 1). The preva-
lence of infection was between 1-3% in groups All Alsea steelhead trout held in fresh water after ex-
collected from May'through June but increased to posure to C shasta died from ceratomyxosis (Table
12-24% from July to the final sample period in Sep- 6)'. In groups of Alsea trout transferred to salt water,
tember. The prevalence of infection among yearling, between 26 and 76% of the fish died prior to devel-
purse-seined chinook salmon also showed a tendency opment of the disease; losses of 36 to 56% of
to increase during the later collection periods (Table unexposed, control fish held in salt water indicated
2). Among the yearling chinook salmon, the preva- that these deaths were caused by inability.of the fish
lence of infection in groups collected from May to adjust to saltwater conditions. The prevalence of
through June was generally less than 10%, except for infection among fish surviving the prepatent losses in
a 20 May collection where three of. nine fish devel- salt water was lower in two of 'the four groups than
oped ceratomyxosis. Infection incidence after July among fish transferred to fresh water. Big Creek
averaged 15%. The total incidence of C. shasta infec- coho and Round Butte chinook salmon were resis-
tion among chinook salmon was 9% for beach-seined tant to infection when held in either fresh or salt
groups and 11% for purse-seined groups, Smaller water after exposure to C. shasta.
numbers of coho salmon and steelhead trout were
collected in May and early June, during the peak of
their migration. The prevalence of infection for these Sensitivity of Detection Methods
species was 5 and 12%, respect 'ively (Tables 3 and 4).
In 1984, 463 chinook salmon smolts were caught in Detection of C shasta infections by standard wet
beach seines between 5 July and 20 September. Infec- mount examination of intestinal tract scrapings was
tion incidence among these subyearling salmon less sensitive than detection by indirect fluorescent
averaged 14% (Table 5). antibody techniques (IFATs) in which a monoclonal

Bartholomew et al: Impact of Ceratomyza shasta on Salmonid Survival 39

Table 7
Comparison of detection sensitivity between wet mount examination for spores and indirect fluorescent antibody
techniques (IFAT) using monoclonal antibodies against Ceratomyxa shasta prespore stages.

Wet mount IFAT
Sample origin' No. samples No. positive % positive No. positive % positive

Spring chinook salmon
Willamette Hatchery 20 8 40 19 95
Clackamas Hatchery 27 16 59 25 93

Fall chinook salmon
Bonneville Hatchery 40 7 17 14 35

Coho salmon
Big Creek Hatchery 14 14 100 14 100
Cascade Hatchery 20 17 85 20 100

Total 121 62 51 92 76

Adult prespawning mortalities.

antibody was used. Of 121 intestinal samples exam- among Columbia River strains of chinook salmon
ined where both methods were used, 51% of the and steelhead trout was generally less than 5%, even
samples were diagnosed as positive for the presence when fish were exposed to the infectious stage for
of spores by the wet mount technique and 76% were 120 days. However, of 2084 chinook salmon smolts
diagnosed positive for the presence of prespore and collected ationes Beach during 1983-84, 221 fish or
spore stages using IFATs (Table 7). 10.6% died from ceratomyxosis. The number of fish
infected did not vary significantly between year
classes, as the prevalence in yearling purse-seined
Discussion chinook salmon paralleled that of the subyearling
chinook salmon collected in beach seines. The preva-
Evaluating the impact of ceratomyxosis on salmonid lence also did not vary significantly between the 2
populations requires consideration of a variety.of fac- years of this study, although the number and type of
tors. Two of these, geographic distribution of the samples collected were limited during 1984, but
parasite and resistance to infection of resident salmo- there was a trend toward increasing pumbers of in-
nids, have been examined in previous studies (Zinn fected fish from the later collection dates.
et al. 1977; Johnson et al. 1979; Buchanan et al. 1983; The dates of peak migration at Jones Beach were
Hoffmaster et al. 1988). Past studies have demon- determined by Dawley et al. (1984b) for each species.
strated the presence of the infectious stage of C. Coho salmon and yearling chinook salmon and steel-
shasta in the mainstem of the Columbia and Snake head trout migrated through the lower Columbia
Rivers and have shown that parasite presence and River during May and early June. Migration of
concentration vary between locations. There is also subyearling chinook salmon past Jones Beach began
evidence that its distribution has increased within the to increase during May but reached its peak from
basin. Because all salmonids migrating in the Colum- early June to mid-July. This means that the majority
bia River Basin are exposed to C. shasta, native strains of subyearling chinook salmon were migrating dur-
have developed some resistance to infection. How- ing periods of increasing water temperature when
ever, resistance is not complete in the' fish the impact from C shasta is greater (Udey et al.
populations studied and little is known about the sus- 1975). Dawley et al. (1984b) also determined that
ceptibility of feral and upper Columbia and Snake migration rates were 22, 18, 17, and 35 km/day for
River salmonids to infection by C. shasta. subyearling chinook, yearling chinook and coho
The prevalence of infection among outmigrating salmon, and steelhead trout, respectively. Therefore,
fish was higher than the incidence predicted from the duration of exposure to C. shasta for subyearling
susceptibility studies. In studies by Zinn et al. (1977) and yearling salmon and steelhead trout migrating
and Buchanan et al. (1983), the infection incidence from Oxbow Dam on the Snake River, a distance of

40 NOAA Technical Report NWS I I I

over 1000 river miles from the Columbia Estuary,. ity was 100% in both groups. Similar experiments
could be as long as 43, 53, and 27 days, respectively. were conducted in this study using shorter exposure
Coho salmon originate from the lower Columbia lengths and both susceptible and resistant strains of
River and therefore have a shorter time of exposure salmonids. Although survival was poor, in C. shasta-
to the parasite. Long periods of exposure to C. shasta susceptible steelhead trout transferred to salt water
combined with the stress of migration and after exposure, two of four groups had a lower inci-
smoltification may explain why the numbers of dence of infection than groups transferred to fresh
chinook salmon and steelhead.trout infected in our water. It appears that migration to salt water may re-
study are higher than numbers predicted from sus- duce the progress of ceratomyxosis if the fish are not
ceptibility studies. Also, it was not possible in this overwhelmed by a large infectious dose. Entry into
study to determine the strain origin of the infected salt water did not impair the ability of resistant fish to
fish so their resistance status cannot be assumed. block the infection.
Studies by Ratliff (1981) and Ching and Munday While all of the studies described here show the
(1984) demonstrate that fish originating from rivers impact of ceratomyxosis on captured fish held under
enzootic for the parasite can still be highly suscep- laboratory conditions, many field studies rely on sam-
tible to infection. Ching and Munday exposed pling, fish prior to overt signs of ceratomyxosis.
chinook salmon representing six stocks from the Reports by Yasutake et al. (1986) and Bartholomew et
Fraser River, British Columbia, to the infectious stage al. (1989a) indicate that spores do not form until late
of C shasta in the lower Fraser River. Between 87 and in the infection and that, in cases of acute
100% of the fish in each group died from ceratomyxosis, fish may die before spore develop-
ceratomyxosis. Ratliff (1981) further demonstrated ment. Therefore, diagnosis made on the basis of
that prevalence of infection may increase with in- identifying spores is likely to underestimate the
creasing length of exposure to C shasta and he prevalence of the parasite. To demonstrate this, an
calculated that 50-70% of all chinook salmon re- IFAT utilizing monoclonal antibodies specific for
leased into the Deschutes River may become prespore stages of C shasta was compared with ex-
infected. Prior to changes in the Columbia River Ba- amination by standard wet mount procedures. When
sin that resulted from the building of dams, upriver both methods were used to examine samples col-
salmonids may also have avoided ceratomyxosis by lected from prespawning adult salmon mortalities,
migrating through infectious areas before parasite there was a 25% increase in the number of infections
concentrations were high. However, the presence of detected by using the serological method. An even
dams and reservoirs has impeded outmigration, greater increase in sensitivity could be expected
raised water temperatures (Raymond 1979) and may when techniques are developed which can detect the
have created conditions favorable for proliferation of earlier stages of C. shasta infection. Monoclonal anti-
C. shasta (Ratliff 1981). These changes may have bodies and DNA probes that specifically recognize all
caused the range and numbers of the parasite to in- life stages of the parasite are necessary for an accu-
crease more rapidly in the upper portions of the rate evaluation of the numbers of fish infected with
basin than fish could adapt by developing resistance C shasta.
or avoidance strategies. Therefore, strains of salmo-
nids from the upper Columbia and Snake Rivers may
have lower or more variable levels of resistance than Acknowledgments
the strains that have been examined.
Because ceratomyxosis has a long incubation pe- The authors acknowledge the technical help of J.E.
riod, most salmonids that become infected during Sanders, C.A. Arakawa, W. Moynihan, D.E. Ratliff,
their downstream migration enter the ocean before and E.M. Dawley. This publication is the result of re-
the disease results in death. Acute ceratomyxosis has search sponsored by the Bonneville Power
been reported in juvenile chum salmon (0. keta) cap- Administration under contract No. DE-A179-83 BP
tured off the coast of British Columbia (Margolis and 11987; G.R. Bouck, Contracting Officer's Technical
Evelyn 1975). This finding indicates that the disease Representative; and, in part, by Oregon Sea Grant
is not attenuated when fish enter salt water. To dem- with funds from the National Oceanic and Atmo-
onstrate this under laboratory conditions, Ching and spheric Administration, Office of Sea 'Grant,
Munday (1984) exposed chinook salmon to the infec- Department of Commerce under grant No. NA89AA-
tious stage of C. shasta for 10 days, then held the fish D-SGIO8. This is Oregon Agricultural Experiment
in either fresh or salt water. They found that mortal- Station Technical Paper No. 9491.

Bartholomew et al- Impact of Ceratomyxa shasta on Salmonid Survival 41
r

Citations Hoffmaster, J.L., J.E. Sanders, J.S. Rohovec, J.L. Fryer, and D.G.
Stevens.
Amos, KH., (ed.). 1988. Geographic distribution of the myxosporean parasite,
1985. Procedures for the detection and identification of cer- Ceratomyxa shasta Noble, 1950, in the Columbia River basin,
tain fish pathogens, 3rd ed. Am. Fish. Soc. Fish Health U.S.A. J. Fish Dis. 11:97-100.
Johnson, ILA.,J.E. Sanders, and].L. Fryer.
Section, 114 p..
Bartholomew, J.L., C.E. Smith,J.S. Rohovec, andj.L. Fryer. 1979. Ceratomyxa shasta in salmonids. U.S. Fish Wildl. Serv.,
1989a. Characterization of a host response to the Fish Dis. Leafl. 58, 11 p.
Margolis, L., and T.P.T. Evelyn.
myxosporean parasite, Ceratomyxa shasta (Noble), by histol- 1975. Ceratonyxa shasta (Myxosporida) disease in chum
ogy, scanning electron microscopy and immunological
techniques. J. Fish Dis. 12:509-522. salmon (Oncorhynchus keta) in British Columbia. J. Fish.
Bartholomew, J.L.,J.S. Rohovec, andj.L. Fryer. Res. Board Can. 32:1640-1643.
1989b. Development, characterization, and use of mono- Ratliff, D.E.
clonal and polyclonal antibodies against the myxosporean, 1981. Ceralmnym shasta: epizootiology in chinook salmon of
Ceratomyxa shasta. J. Protozool. 36:397-401. centralOregon. Trans. Am. Fish. Soc. 110:507-513.
Buchanan, D.V.,J.E. Sanders,J.L. Zinn, andj.L. Fryer. Raymond, H.L.
1983. Relative susceptibility of four strains of summer steel- 19 .79. Effects of dams and impoundments on migrations of
head to infection by Ceratomyxa shasia. Trans. Am. Fish. juvenile chinook salmon and steelhead from the Snake
Soc. 112:541-543. River, 1966 to 1975. Trans. Am. Fish. Soc. 108:505-529.
Udey, L.R.,J.L. Fryer, and ILS. Pilcher.
Ching, H.L., and D.R. Munday.
1984. Susceptibility of six Fraser chinook salmon stocks to 1975. Relation of water temperature to ceratomyxosis in
Ceratom xa shasta and the effects of salinity on rainbow trout (Salmo gairdneri) and coho salmon
Y (Oncorhynchus kisutch). J. Fish. Res. Board Can. 32:1545-
ceratomyxosis. Can.J. Zool. 62:1081-1083. 1551.
Dawley, E.M., R.D. Ledgerwood, T.H. Blahm, R.A. Kirn, A.E. Yasutake, W.T., J.D. McIntyre, and A.R. Hemmingsen.
Rankis, and F.J. Ossiander. 1986. Parasite burdens in experimental families of coho
1984a. Migrational characteristics and survival of juvenile salmon. Trans. Am. Fish. Soc. 115:636-640.
salmonids entering the Columbia River estuary during Zinn,J.L., ICA. Johnson, J.E. Sanders, andl.L. Fryer.
1982. Annual report of research by the Coastal Zone and 1977. Susceptibilty of salmonid species and hatchery strains
Estuarine Studies Division, NMFS, January 1984, 49 p. of chinook salmon (Oncorhynchus tshauytscha) to infections
Dawley, E.M., R.D. Ledgerwood, TH. Blahm, R.A. Kirn, and A.E. by Geratomyxa shasta. J. Fish. Res. Board Can. 34:933-936.
Rankis.
1984b. Migrational characteristics and survival of juvenile
salmonids entering,the Columbia River estuary during
1983. Annual report of research by the Coastal Zone and
Estuarine Studies Division, NMFS, July 1984, 88 p.

Viral Infections of Cultured Fish in Japan

MAMORU YOSHIMIZU and TAKALHISA KIMURA

Laboratory of Microbiology
Fa cu Ity of Fisheries
Hokkaido University, Minato 3-1-1
Hakodate, Hokkaido 041,japan

ABSTRACT

Since infectious pancreatic necrosis virus and infectious hematopoietic necrosis virus
were first isolated in the 1970s, more than 20 fish viruses have been isolated and at least 5
viruses have been observed by electron microscopy. Viral diseases are major problems
and cause economic losses among cultured fishes in Japan and other countries. This
paper reports our current understanding of the extent of viral infection in the cultured
fishes of Japan.

Introduction and masu salmon (0. masou). Recently, the fry of
rainbow trout were observed to be less susceptible to
A virological study of cultured fishes in Japan was IPN-V and, consequently, damage attributed to IPN
initiated when an unknown disease occurred. among has decreased (Okamoto et al. 1987).
rainbow trout Oncorhynchus mykiss in the 1960s. The
causative agent was identified as infectious pancreatic
necrosis virus (IPNV) (Sano 1971). Subsequently, in- Infectious Hematopoietic Necrosis
fectious hematopoietic necrosis virus .(IHNV) was
isolated from kokanee salmon 0. nerka (Kimura and Infectious hematopoietic necrosis is an acute sys-
Awakura 1977). Since then, various viral infections temic disease which mainly affects the fry of rainbow
of fish have been reported. At present, more than 20 trout and masu and kokanee salmon but which has
fish viruses have been isolated and at least 5 viruses also been isolated from moribund ayu Plecoglossus
have been observed by electron microscopy studies altivelis (Yoshimizu et al. 1987c). The characteristic
(Table 1). sign of IHNV infection is V-shaped hemorrhages
located in muscle tissue. Recently, a large rainbow
trout, with a body weight of 50-80 g was found to be
Viral Diseases of Salmonid Fishes infected with IHNV and subsequently died (Mori et
al. 1987). In this case, petechiae were observed in the
fatty tissues and on the wall of the body cavity. This
Infectious Pancreatic Necrosis virus is widespread and especially prevalent in the
central part of Honshu, the Japanese mainland (Sano
Infectious pancreatic necrosis is an acute systemic et al. 1977). In several districts, river waters have
disease affecting the fry and fingerlings of rainbow been contaminated with IHNV and are now unsuit-
trout. Its occurrence is widespread in Japan. Suscep- able for rainbow trout culture. Although the vertical
tibility of fish to IPNV depends on body weight; transmission of IHNV is doubtful (Yoshimizu et al.
smaller fry are more susceptible. The signs of this 1988, a and b), it can be controlled by disinfecting
disease are darkening of body coloring, moderate eggs with iodine during the early eyed stage. Fish at
exophthalmia, and abdominal distention. Internally, the fry stage are very. susceptible to IHNV. They
the spleen, heart, liver, and kidneys are pale and the should be reared in either well water or ultra-violet
digestive tract is almost always devoid of food. Rain- irradiated river water. When fish are past this sensi-
bow trout is the fish most affected by IPNV, but the tive stage, they can be transferred to the usual
virus has also been isolated from amago (0. rhodurus) rearing ponds.

44 NOAA Technical Report NWS I I I

Table I showed that OMV is enzootic in the northern part,of
Viral infection in cultured fishes in Japan. Japan (Yoshimizu et al. 1988b) and that the charac-
terist.ics of these three herpesviruses are similar
Isolated virus Host except that NeVTA lacks oncogenicity (Hedrick et al.
1987; Sano et al. 1988). In 1983, we recommended
DNA virus the disinfection of fish eggs with iodine at the early
Nerkavirus in Towada Lake, eyed stage in Hokkaido. Now OMV cannot be de-
Aomori and Akita tected in most of the hatcheries in this area
Prefecture (NeVTA) Kokanee salmon (Yoshimizu et al. 1988b). Although the host species
Oncorhynchus masou virus (OMV) Masu salmon of this virus is primarily masu salmon, OMV has also
Yarname tumor virus (M) Yamame (masu salmon)
lcosahedral cytoplasmic been isolated from the tumor tissues of pen-cultured
deoxyribovirus (ICDV) Japanese eel coho salmon, 0. kisutch.
He?pesvims typyini Fancy carp
Herpesvirus Japanese eel
Unidentified small virus Tiger puffer Chum Salmon Virus Infection
RNA virus
Infectious pancreatic necrosis
virus (IPNV) Salmonid fish In 1978, a reovirus was isolated from an apparently
Infectious hematopoietic normal adult chum salmon, 0. keta returning to its
necrosis virus (IHNV) Salmonid fish hatchery in Hokkaido (Winton et al. 1981). After ini-
Chum salmon virus (CSV) Masu salmon
Yellowtail ascitic virus (YAV) Yellowtail tial isolation and characterization, it was named
Rhabdovirus olivaceus (HRV) Various marine fish chum salmon virus (CSV). This virus was not ob-
Eel virus from European eel (EVE) European eel served again until 1986, during an episode of mass
Eel virus of America (EVA) American eel mortalities of masu salmon fry for which it was re-
Eel virus of Europe X (EVEX) European eel sponsible. Since then, the virus has been detected in
Papovavirus Japanese eel stocks of adult masu salmon at new locations in
Birnavirus Yellowtail
Birnavirus Japanese flounder Hokkaido (Yoshimizu 1988). Artificial infection stud-
Birnavirus Red sea brearn ies of this virus showed no significant mortality in
Coronavirus cypTini virus (CACV) Common carp several species of salmonid fishes (Winton et al.
Picornavirus Japanese eel 1989).
Reovirus Common carp
Reovirus Japanese eel
Observed by electron microscopy
Viral erythrocytic necrosis virus Various marine fish Viral ]Fxythrocydc Necrosis
Lymphocystis virus Various marine fish
Paramyxovirus Black rockfish Inclusion bodies stained with Giemsa were observed
Herpesvirus Japanese flounder in the erythrocytes of churn and pink salmon, 0.
Picornavirus Ishidai
gorbuscha, collected in Okhotsuku and along the
north Pacific coast of Hokkaido. The causative agent
of viral erythrocytic necrosis (VEN), an iridovirus,
Herpesvirus Infection was subsequently observed by electron microscopy
(Yoshimizu et al. 1988b).
A herpesvirus, nerkavirus in Towada Lake Akita and
Aomori Prefecture (NeVTA), was first isolated from
diseased kokanee salmon in Towada Lake (Sano Viral Infections of Eels
1976). In 1978, another herpesvirus was isolated
from the ovarian fluid of apparently normal mature Many viruses have been isolated from cultured eels
masu salmon (Kimura et al. 1980). This virus was (Anguilla anguilla, A. japonicus, and A. rostrata) by
named Oncorhynchus masou virus (0MV) from the sci- . Sano (1976) and Sano and Fukuda (1987). They in-
entific name of the host fish. Oncorhynchus masou clude a birnavirus, eel virus from the European eel
virus was found to be pathogenic and significantly (EVE); the rhabdoviruses, eel virus of America (EVA)
more oncogenic in young masu salmon and several and eel virus of Europe X (EVEX); papovavirus; her-
other salmonid fish (Kimura et al. 1981, a and b; pesvirus; picornavirus; and a reovirus. These viruses
Yoshimizu. et al. 1987a). In 1983, a similar herpesvi- are not recognized as pathogenic against eel except
rus 'were isolated from tumor tissue.of yamame for EVE (Nishimura et al. 1981). Sorimachi (1982,
(landlocked 0. masou) and was named yamame tu- 1984) reported having isolated icosahedral. cytoplas-
m .or virus (Sano et al. 1983). Subsequent study mic deoxyribovirus (ICDV) from a diseased eel. This

Muroga and Yoshimizu: Viral Infections of Ctiltured Fish in Japan 45

virus was shown to be pathogenic against Japanese Sebastes inermis, and also for salmonid species, espe-
eels following artificial infection. Mortality was 40- cially rainbow trout and masu salmon (Yoshimizu et
75% at water temperatures of 14.5-18.50 C, 15% at al. J987b). Signs of HRV infection are gonadal con-
22.8' C, and 0% at 24.1' C. Infected fish showed gest@ion, focal hemorrhage of skeletal muscle and
signs of decoloration; congestion of the anal, pecto- fins, and accumulation of ascitic fluid. Hirame rhab-
ral, and dorsal fins; and an increase of mucus on the dovirus. is distributed widely from Hokkaido to
body surface. Honshu in Japan.

Viral Infections of Carp Ruchishiroshou of Tiger Puffer

Herpesvirus cyprini was isolated from papilloma tissue From cultured tiger puffer, Fugu rubripes, an uniden-
of cultured fancy carp (Cyprinus carpio, also called tified, small virus was isolated (Inoue et al. 1986).
common or asagi carp) and confirmed as the agent The epizootic period is from May to June during wa-
of infection by induction of epithelia] tumors by arti- ter temperatures of 18-22' C. Moribund fish had
ficial infection (Sano et al. 1985). Coronavirus cyprini, necrosis around the mouth and had been observed
carp coronavirus (CACV), was isolated from diseased to be fighting with each other. From the signs of this
common carp raised in the laboratory. Fish infected infection, the disease was named "Kuchishiroshou",
with CACV showed acute mortality showing no exter- from the Japanese words "kuchi", meaning mouth,
nal signs except erythematous skin on the abdomen. "shiro", meaning white, and "shou", meaning disease.
Experimentally, CACV was virulent for carp fry at Viral particles were observed in the brain by electron
20' C. Cumulative mortality for 3-week-old fry was microscopy. Kuchichiroshou occurs in southwest Ja-
72.5%. The affected fish manifested swollen and pan where tiger puffer are cultured.
hemorrhagic abdomens filled with ascites and even-
tually died. Reovirus was also isolated from common
carp (Sano and Fukuda 1987). Epidermal Hyperplasia of Japanese Flounder.

Outbreaks of a disease resulting in mass mortalities
Viral Infection of Other Marine Fishes- of larval and juvenile Japanese flounder was reported
by Iida et al. (1989). Once the disease occurs in a
pond, the resident fish populations usually become
Viral Pancreatic-Hepatic Necrosis of Yellowtail extinct within one month. Affected fish are character-
ized by opaque fins. Histopathologically, hyperplasia
A yellowtail ascites virus (a birnavirus) was isolated is observed in the epidermal layer of the fins and
from the fry of yellowtail Seriola quinqueradiata skin. In the epidermal tissues of infected fish, hex-
(Sorimachi and Hara 1985). This epizootic is an agonal virus particles were observed by electron
acute viral infection of both naturally grown and microscopy. Japanese flounder larvae experimentally
hatchery-raised fry. The epizootic period occurs from exposed to the filtrate of infected tissue homogenate
May to June at water temperatures of 18 to 22' C. suffered 18-50% mortalities with 93-100% of the sur-
The moribund fry typically show anemic gills, hemor- vivors exhibiting epidermal hyperplasia. This virus
rhaging in the liver, and ascites and petechiae in the has not been isolated in any of the 33 fish cell lines in
.pyloric caea. The disease name, viral pancreatic-he- which culture has been attempted, including that of
patic necrosis, was proposed by Egusa and Sorimachi the host species.
(1986). Birnaviruses were also isolated from Japanese
flounder and red sea bream (Yoshimizu and Kimura,
unpubl.. data), These viruses were neutralized with
Rhabdovirus Infection of Japanese Flounder antibody against IPNV; the pathogenicities of these
birnaviruses have not been clarified.
The rhabdovirus, Rhabdovirus olivaceus-also referred
to as hirame rhabdovirus (HRV)-was isolated from
diseased hirame (Japanese flounder), Paralichthys Lyinphocystis Disease
olivaceus, and black sea bream, Milio macrocephalus
(Gorie et al. 1985; Kimura et al. 1986). This virus is In several species of marine fishes, suzuki, Lateolabrax
pathogenic for marine fish such as hirame, black sea japonicus, yellowtail, red sea bream, Japanese
bream, red sea bream Pagrus major and black rockfish flounder, and others, lymphocystis disease was re-

46 NOAA Technical Report MM I I I

ported and iridovirus was observed by electron mi- Kimura, T., M. Yoshimizu, and S. Gorie.
croscopy (Matsusato 1975; Miyazaki and Egusa 1972; 1986. A new rhabdovirus isolate in Japan from cultured
Tanaka et al. 1984). Seasonal variation in the preva- hirame (Japanese flounder) Paratichthys olivaceus and ayu
Plecoglossus altivelis. Dis. Aquat. Org. 1:209-217.
lence of lymphocystis was noted with increased Matsusato, T.
prevalence in summer. Lymphocystis cells were ob- 1975. On the lymphocystis disease in cultured yellow-
served mainly on the fins or body sur--ace. The virus tail. Fish Pathol. 10:90-93. (Injapanese.)
particles were polyhedral, presenting hexagonal or Miyazaki, T., and S. Egusa.
pentagonal profiles in tissue sections. They may be 1972. A histopathological observation on lymphocystis dis-
ease in sea bass Lateolabrax japonicus (Cuvier and
seen in a crystalline array and are always located in Valenciennes), Fish PaLhol. 6:83-87. (In Japanese.)
the cytoplasm. Miyazaki, T., K. Fujiwara,J. Kobara, and N. Matumoto.
1989. HistopaLhology associated with two viral diseases of lar-
val and juvenile fishes: epidermal necrosis of Japanese
flounder Paratichthys olivaceus epithelial necrosis of black sea
Other Diseases bream Acanthopagrus schlegeli. J. Aquatic Animal Health
1:85-93.
Yoshikoshi and Inoue (1988) reported picornavirus Mori, S., F. Iketani, T. Komatsu, and T. Nishimura.
1987. IHN of large size of rainbow trout. Proceedings of
in moribund fry of ishidai, Oplegnathus fasciatus, and the Annual Meeting of Japanese Society of Fish Pathology,
Miyazaki et al. (1989) reported herpesvirus in the p.9. (Injapanese.)
epidermal necrosis of Japanese flounder and Nishimura, T., M. Toba, F. Ban, N. Okamoto, and I Sano.
paramyxovirus in the epithelial necrosis of black sea 1981. Eel rhabdovirus, EVA, EVEX and their infectivity to
bream. To date, these viruses have not been isolated. fishes. Fish Pathol. 15:173-183. (In Japanese.)
Okamoto, N., T. Matsumoto, N. Kato, S. Tazaki, M. Tanaka, N. Ai,
H. Hanada, Y Suzuki, C. Takamatsu, T. Tayama, and T. Sano.
1987. Difference in susceptibility to IPN virus among rain-
bow trout populations from three hatcheries in
Citations Japan. Nippon Suisan Gakkaishi 58:1121-1124. (In
Japanese.)
Egusa, S., and M. Sorimachi. Sano, T.
1986. A histopathological study of yellowtail ascites virus 1971. Studies on viral diseases of Japanese fishes-1. Infectious
(YAV) infection of fingerling yellowtail, Seriola pancreatic necrosis of rainbow trout; first isolation from
quinqueradiata. Fish Pathol. 21:113-122. epizootics in Japan. Bull. Jpn. Soc. Sci. Fish. 37:405-408.
Gorie, S., K. Nakamoto, K. Katashima. (InJapanese.)
1985. Disease of cultured hirame Uapanese flounder, 1976. Viral diseases of cultured fishes in Japan. FishPathol.
Paralichthys olivaceus)-I. Preliminary report on a disease of 10:221-226.
marine pen cultured hirame may be caused by viral Sano, T., and H. Fukuda.
infection. Bull. Hyogo Pref. Fish. Exp. Sm. 23:66-68. (In 1987. Principal microbial diseases of mariculture in
Japanese.) I Japan. Aquaculture 67, Vol. 1:59-69.
Hedrick, R.P., T. McDowell, W.D. Eaton, T. Kimura, and T. Sano. Sano, T., H. Fukuda, and M. Furukawa.
1987. Serological relationships of five herpesviruses isolated 1985. Herpesvirus cyprini: Biological and oncogenic
from salmonid fishes. J. Appl. Ichthyol. 3:87-92. properties. Fish Pathol. 20:381-388.
lida, Y, IL Masumura, T. Nakai, M. Sorimachi, and H. Matsuda. Sano, T., N. Fukuda, N. Okamoto, and F. Kaneko.
1989. A viral disease occurred in larvae and juveniles ofJapa- 1983. Yamame tumor virus: lethality and oncogenicity@ Bull.
nese flounder, Paralichthys olivaceus. Aquatic Animal Jpn. Soc. Sci. Fish. 49:1159-1163.
Health 1: 7-12. Sano, T., H. Fukuda, T. Inokari, F. Tichiya, and H. Hosoya.
Inoue, K., S. Yasumoto, N. Yasunaga, and I. Takami. 1988. Comparison of three representative strains of salmo-
1986. Isolation of a virus from cultured tiger puffer, Tahifugu nid herpesvirus. Proceedings of the Annual Meeting of
ruhripes, infected with "Kuchishiro-sho" and it's patho- Japanese Society of Fish Pathol, p.21.
genicity. Fish Pathol. 21:129-130. Sano, T., T. Nishimura, N. Okamoto, T. Yamazaki, H. Hanada, and
Kimura, T., and T. Awakura. Y. Watanabe.
1977. Current status of disease of cultured salmonids in 1977. Studies on viral disease of Japanese fish. IV. Infectious
Hokkaido, Japan. In Proceedings from the international hematopoietic necrosis (IHN) of salmonids in the mainland
symposium on disease of cultured salmonids, sponsored by of Japan. J. Tokyo Univ. Fish. 63:81-85.
Tavolek Inc., Seattle, WA, p. 124-160. Sorimachi, M.
Kimura, T., M. Yoshimizu, and M. Tanaka. 1982. Characteristics and distribution 'of viruses isolated
1980. Salmonid viruses: a syncytium-forming herpesvirus from pond-cultured eels. Bull. Nad. Inst. Aquaculture
from landlocked Oncorhynchus masou. Fish Health News 3:97-105. (Injapanese.)
9:iii. 1984. Pathogenicity of ICD virus isolated from Japanese
1981a. Studies on a new virus (OMV) from Oncorhynchus eel. Bull. Natl. Inst. Aquaculture 6:71-75. (In Japanese.)
masou-1. Characteristics and pathogenicity. Fish Pathol. Sorimachi, M., and T. Hara.
.15:143-147. 19M. Characteristics and pathogenicity of a virus isolated
1981b. Studies on a new virus (OMV) from Oncorhynchus from yellowtail fingerling showing ascites. Fish Pathol.
masou-11. Oncogenic nature. Fish Pathol. 15:149-153. 19:231-238.

Muroga and Yoshimizu: Viral Infections of Cultured Fish in Japan 47

Tanaka, M., M. Yoshimizu, M. Kusakari, and T Kimura. Yoshimizu, M., M. Tanaka, and T. Kimura.
1984. Lymphocystis disease in kurosoi Sebastes schlegeli and 1987a. Oncorhynchus masou virus (OMV): incidence of tumor
hirame Paralichthys olivaceus in Hokkaido, Japan. Bull. Jpn. development among experimentally infected representative
Soc. Sci. Fish. 50:37-42. (Injapanese.) salmonid species. Fish Pathol. 22:7-10.
Winton, J. R., C. N. Lannan, J. L. Fryer, and T Kimura. Yoshimizu, M., N. Oseko, T Nishizawa, and T Kimura.
1981. Isolation of a new reovirus from chum salmon in 1987b. Rhabdovirus disease of Japanese flounder. Fish
Japan. Fish Pathol. 15:155-162. Pathol. 22:54-55.
Winton, J. R., C. N. Larman, M. Yoshimizu, and T Kimura. Yoshimizu, M., M. Sami, T. Kimura, and M. Ugazhin.
1989. Response of salmonid fish to artificial infection with 1987c. Characteristics of the virus isolated from cultured ayu
chum salmon virus. In Viruses of lower vertebrates (W. (PL-coglossus altivelis). Proceedings of the Annual Meeting
Ahne and E. Kurstak, eds.), p. 270-278. Springer-Verlag, of Japanese Society of Fish Pathology, p. 12.
Berlin. Yoshimizu, M., M. Sami, and T. Kimura.
Yoshikoshi, K., and K Inoue. 1988a. Survival and inactivation of infectious hernatopoietic
1�88. Viral neuron necrosis of ishidai. Proceedings of the necrosis virus (IHNV) in fertilized eggs of masu salmon
Annual Meeting of Nippon Suisan Gakkai, p.121. Oncorhynchus masou and chum salmon 0. keta. Nippon
Yoshimizu, M. Suisan Gakkaishi 54:2089-2097.
1988. Chum salmon virus isolated from masu salmon, Yoshimizu, M., T. Nomura, T Awakura, and I Kimura.
Oncorhynchus masou. Fish and Eggs, Techn. Rep. Hokkaido 1988b. Incidence of fish pathogenic viruses among anadro-
Salmon Hatchery 157:36-38. mous salmonid in northern part of Japan. Sci. Rep.
Hokkaido Salmon Hatchery 42:1-20.

Some Important Infectious Diseases of Kuruma Shrimp,
Penaeusjaponicus, in Japan

KAZUO MOMOYAMA

Yamaguchi Prefectural Naikai Fisheries Experimental Station
Yamaguchi 754, Japan

ABSTRACT

The kuruma shrimp, Penaeus japonicus, is the most widely cultured crustacean in Japan.
Shrimp farming operations for this species have been greatly affected by several infec-
tious diseases including baculoviral midgut gland necrosis (BMN), vibriosis, and black
gill disease. This paper discusses our knowledge of the BMN virus in detail, and brieflly
reviews the latter two diseases with respect to the culture of crustaceans.

Introduction al. 1987; Johnson and Lightner 1988). They exhibit
higher pathogenicity to larval or postlarval shrimp
It may safely be said that kuruma shrimp, Penaeus than to adults. Baculoviral midgut gland necrosis was
japonicus, is the only crustacean species cultured on an first noticed in 1971, and since then it has often
industrial scale in Japan. About six to seven hundred caused mortalities of over 190% during the mass pro-
million post-larval kuruma shrimp are produced at duction of kuruma shrimp larvae in Japan
public and private hatcheries every year for stocking (Momoyama 1981; Sano et al. 1981). Heavy losses
and farming (Japan Sea Farming Assoc. 1990). Annual due to baculovirus infection during the larval pro-
production of cultured kuruma shrimp reached 3,020 duction of P. aztecus (Couch 1978) and R monodon
metric tons in 1988 (DSI 1990). (Lightner et al. 1983) have also been reported in
The infectious diseases responsible for high mortali- other countries.
ties to cultured kuruma. shrimp injapan are baculoviral The results of our histological and epizootiological
midgut gland necrosis (BMN), vibriosis, and black gill studies on BMN in kuruma shrimp *are summarized
disease. Baculoviral midgut gland necrosis is a very se- as follows.
vere viral disease that occurs during the seed
production process which usually causes 90% or higher
mortalities in the population of the rearing tank. Re- Histopathology
cently, the frequency of BMN outbreaks has decreased
greatly, probably because of the preventive measures ex- Midgut glands of diseased shrimp become soft and
ecuted at hatcheries. Vibriosis is a bacterial disease develop a white turbid appearance at the advanced
affecting the lymphoid organ that has been responsible stage of infection, whereas those of healthy
for a great deal of damage to the shrimp farming indus- postlarvae appear brown or colorless during the early
try over the past several years. Black gill disease is developmental stages (Momoyama 1981). From histo-
caused by Fusarium solani, a ubiquitous pathogen of logical examinations, it was confirmed that only the
shrimp and lobster, invading the gills, appendages, and midgut gland and the intestine are affected by the
various parts of the exoskeleton. disease. Disarrangement and exfoliation of epithelial
In the present paper BMN is discussed in detail, and cells are remarkable in the midgut gland of the dis-
vibriosis and black gill disease are touched briefly. eased shrimp. Nuclear hypertrophy and chroma-
tolysis of infected epithelia] cells are the most
characteristic cytop ath o logical changes in BMN
Baculoviral Midgut Gland Necrosis (Momoyama 1981). In contrast with other penaeid
shrimp baculovirus infections, no inclusion bodies
To date, four baculoviruses have been recorded in are present in the hypertrophied nuclei (Lightner
penaeid shrimp worldwide (Lightner 1985; Lester et 1985; Lester et al. 1987; Johnson and Lightner 1988).

50 NOAA Technical Report NMB I 11

Electron microscopy observations of the hypertro- zoeae, mysis larvae, and postlarvae were exposed,
phied nuclei and the midgut gland lumen have some or all of the shrimp were diagnosed to be in-
revealed many rod-shaped particles having outer and fected on the final day of the test period (6 to 16
inner envelopes, which represent virions of the days). Susceptibility to infection tended to decrease
baculovirus group. The average length and diameter with the advance in developmental stages from zoea
of the virions are 310 nm and 72 nm, respectively to P-10 (10-day-old postlarva). Cumulative mortalities
(Sano et al. 1981). decreased from 100% in zoeal stage larvae to 0% in
Baculoviral midgut gland necrosis can be diag- P-6 larvae. The growth rates of test shrimp inoculated
nosed by detecting the hypertrophied nuclei of at stages P-2 through P-6 were lower than those of
affected midgut gland epithelial cells in both fresh controls' but there were no differences in mortality
and stained squash preparations. In stained squash or growth rates between test shrimp and controls in-
preparations homogeneous hypertrophied nuclei oculated at stages P-8 and P-10. These results indicate
about 20 to 30 @Lrn in diameter appear among the that kuruma shrimp from the zoeal to the P-6 stage
smaller normal ones which are about 10 @Lm in diam- are highly susceptible to BMN virus, but stage P-8 or
eter. The Feulgen reaction makes the difference older postlarvae become refractory to this disease.
clearer between hypertrophied and normal nuclei. The route of baculovirus infection in shrimp has
The diagnostic technique of using a dark field micro- generally been considered to be by oral ingestion of
scope with a wet-type condenser has the advantages virus contaminated sediments or by cannibalism of
of precision and rapidity. Hypertrophied nuclei are diseased shrimp (Lightner et al. 1983; Couch 1974).
clearly seen as white bodies of 10 to 30 pm diameter In the previously mentioned study on waterborne sus-
in fresh squash preparations from diseased samples ceptibility, kuruma shrimp were not administered any
under the dark field microscope. The reason why the food during the inoculation period. However, peri-
infected nuclei appear white is thought to result from staltic movements were frequently observed in the
an increased number of reflected or diffracted rays oesophageal part of the shrimp, suggesting that test
due to the numerous virus particles in the nucleus shrimp ingest the seawater containing the virus Par-
(Momoyama 1983). ticles through the mouth. This hypothesis was
supported by an observation that azocarmine G accu-
mulated in the stomach and midgut gland lumen of
Source of Infection shrimp when they were dipped in seawater contain-
ing this dye (Momoyama and Sano 1989).
Epizootiological investigations indicate that mature
female kuruma shrimp spawners with latent BMN-vi-
rus infections and cultured young kuruma shrimp Inactivation and Survival of BMN Virus
that have recovered from infection with the virus are
the main source of infection in hatchery epizootics The effects of disinfectants, heating, and ultraviolet
(Momoyama 1988). Histological examinations reveal irradiation on BMN virus, and the survival time of
nuclear hypertrophy of the midgut gland epithelial the virus in seawater at different temperatures were
cells in both spawners and young cultured survivors examined by waterborne infectivity experiments us-
of the disease, Fluorescent antibody techniques have ing larval and postlarval kuruma shrimp. The virus
been used to reveal the presence of BMN-specific vi- was inactivated by 10-minute exposure at 25' C to any
rus antigen in the hypertrophied nuclei of spawners of the following chemicals: 5 ppm chlorine, 25 ppm
(Momoyama 1988). iodine, 100 ppm benzalkonium chloride and
benzethonium chloride, 30% ethyl alcohol, and 0.5%
formalin (Momoyama 1989a). The virus was still ac-
The Effect of Developmental Stage of the Host tive after 2 hours of heat treatment at 35 and 40' C,
on the Susceptibility to BMN Virus but was inactivated within 2 hours at 45' C, 30 min-
utes at 50 and 55' C, and 5 minutes at 60' C
The susceptibility of kuruma shrimp to BMN virus by (Momoyama 1989b). The virus'was also inactivated
waterborne infection (Momoyama and Sano 1988) within 20 minutes by ultraviolet irradiation with a 15-
was determined for fertilized eggs, nauplii, zoeae, watt ultraviolet lamp at a distance of 30 cm
mysis larvae, and postlarvae (2, 4, 6, 8, and 10 day (Momoyama 1989b). In seawater, the virus could not
old) (Momoyama and Sano 1989). When the fertil- survive longer than 4 days at 30' C, -7 days at 25' C,'
ized eggs and nauplii were exposed to the virus, they 12 days at 20' C, and 20 days at 15' C (Momoyama
showed no evidence of infection. But when the 1989c).

Momoyarna: Infectious Diseases of Penaeusjaponicus 51

Prevention Black gill disease is caused by Fusafium solani,
which is a member of the imperfect fungi. This fun-
The following two preventive measures are now used gus is a ubiquitous pathogen and infects penaeid
against this epizootic in some hatcheries. One pre- shrimp (Cook 1971; Lightner 1975) as well as lobster
vents vertical infection from spawners by rinsing the (Lightner and Fontaine 1975; Alderman 1981) and
fertilized eggs with virus-free seawater then transfer- freshwater shrimp (Burns et al. 1979). The fungus
ring them to a disinfected rearing tank. The other usually invades the gills, appendages, and various
prevents horizontal infection by adding chlorine in parts of the exoskeleton.
the rearing tank to kill infected populations. Since In kuruma shrimp, gills are most susceptible to the
1985, the measure to prevent vertical infection has fungus and infected gills always become black
been carried out on an industrial scale, and BMN has (Ishikawa'1968; Bian and Egusa 1981). In the degen-
never occurred in the hatcheries where this treat- erating gill filaments, hemal channels are 'found
ment has been practiced. extremely congested with hemocytes, encapsulated
hyphae, and tissue debris. The fungus also often pen-
etrates into the thoracic central nerve and sometimes
Vibriosis into the ventral thoracic artery (Momoyama 1987).
Tissue destruction, cellular inflammation, afid
Vibriosis has caused a great deal of damage to the hyphae encapsulated by multiple layers of hemocytes
shrimp farming industry over the last several years. are always observed in the lesion. Failure to ex-
Although some dead shrimp infected with vibriosis change gas in the gills and damage to the central
develop white turbid muscle at the 6th abdominal nerve and ventral thoracic artery are thought to be
segment, shrimp suffering from this disease do not responsible for death.
usually show any specific external clinical signs. As blacke ning of the gills in, shrimp is often i n-
The lymph6id organ (Oka 1969) is intensively in- duced. by various causes, detection of macro- and
vaded by the causative bacterium resulting in micro-conidiospores is necessary for diagnosis of this
extensive necrosis and nodule formation (Egusa et disease (Egusa and Ueda 1972).
al. 1988). The nodules are seen by the unassisted eye
as very small black spots and are composed of a bac-
terial colony in the center, a melanized zone around
the bacterial colony, and multiple layers of hemocytes Citations
encapsulating the melanized zone. Small nodules are
also frequently observed in other organs such as the Alderman, DJ.
gills, heart, midgut gland, and abdominal muscula- 1981. Fusarium solani causing an exoskeletal pathology in
ture, but extensive necrotic lesions can not be found cultured lobsters, Homarus vulga7is. Trans. Brit. Mycol.
in these organs. Soc. 76:25-27.
The Vib7io sp. isolated from kuruma shrimp has Bian, B.Z., and S. Egusa.
.1981. Histopa!hology of black gill disease caused byFusalium
been identified as a new species (Takahashi et al. solani (Ma*rtius) infection in the kuruma prawn, Penaeus
1985a), and was tentatively named Vib7io sp. PJ (PJ is japonicus Bate. J. Fish. Dis. 4:19@-201.
the abbreviation of the scientific name of kuruma Burns, C.D., M.E. Berrigan, and G.E. Henderson.
shrimp Penaeus japonicus). Vibyio sp. PJ has very high 1979. Fusmium sp. infections in the freshwater prawn Maao-
pathogenicity to kuruma shrimp. LD,, values ob- brachium rosenbergii (De Mann). Aquaculture 16:193-198.
Cook, H.L.
tained by intramascular injection were about 20 to 1971. Fungi parasitic on shrimp. FA0 Aquaculture Bull.
100 cells/g body weight of shrimp. 3:13.
Now, two antibiotics, oxytetracycline (Takahashi et Couch, J.A.
al. 1985b) and oxolinic acid, are on the market with 1974. An enzootic nuclear polyhedrosis virus of pink shrimp:
ultrastructure, prevalence and enhancement. J. Invertebr.
the government's approval. Although mortalities are Pathol. 24:311-331.
decreased significantly by administering these antibi- 1978. Diseases, parasites and toxic responses of . commercial
otics, disease often returns shortly after treatment. penaeid shrimps of the Gulf of Mexico and South Atlantic
coasts of North America. Fishery Bull, U.S. 76:1-44.
DSI (Department of Statistics Information Division).
Black Gill Disease 1990. Annual statistics on the fishery and culture production
1988. DSI, Ministry of Agriculture, Forestry, and Fisheries,
Tokyo, 296 p. (In Japanese.)
Black gill disease often occurs in the intensive culture Egusa, S., and T Ueda.
systems in the Okinawa and Kagoshima districts, bui 1972. A Fusafium sp. associated with black gill disease of the
rarely in those of the Chugoku district. kuruma prawn, Penaeus japonicus Bate. Bull. Jpn. Soc. Sci.
Fish. 38:1253-1260.

52 NOAA Technical Report NMFS I I I

Egusa, S., Y. Takahashi, T. Itami, and K Momoyama. 1983. Studies on baculoviral mid-gut gland necrosis of
1988. Histopathology of vibriosis in the kuruma prawn, kuruma shrimp (Penaeus japonicus)-III Presumptive diag-
Penaeus japonicus Bate. Fish Pathol. 23:59-65. (In Japa- nostic techniques. Fish. Pathol. 17:263-268. (In Japanese;
nese; English abstr.) English abstr.)
Ishikawa, Y 1987. Distributions of the hyphae in kuruma shrimp, Penaeus
1968. Preliminary report on black gill disease of the kuruma japonicus, infected with Fusarium solani. Fish Pathol.
prawn, Penaeus japonicus Bate. Fish Pathol. 3:34-38. (In 22:15-23. (Injapanese; English abstr.)
Japanese.) 1988. Infection source of baculoviral mid-gut gland necrosis
Japan Sea Farming Association. in mass production of kuruma shrimp larvae, Penaeus
1990. Annual seed production and stocking record for sea japonicus. Fish Pathol. 23:105-110. (In Japanese; English
farming 1988, p421. (Injapanese.) abstr.)
Johnson, P.T., and D.V. Lightner. 1989a. Virucidal effect of some disinfectants on baculoviral
1988. Rod-shaped nuclear viruses of crustaceans: gut-infect- mid-gut gland necrosis (BMN) virus. Fish Pathol. 24:47-
ing species. Dis. Aquat. Org. 5:123-141. 49. (Injapanese; English abstr.)
Lester, R.G., A. Doubrovsky, J.L. Paynter, S.K. Sambhi, and 1989b. Inactivation of baculoviral mid-gut gland necrosis
J.G. Atherton. (BMN) virus by ultraviolet irradiation, sunlight exposure,
1987. Light and 0ectron microscope evidence of baculovirus heating and drying. Fish Pathol. 24:115-118. (In Japa-
infection in the prawn Penaeus plebejus. Dis. Aquat. nese; English abstr.)
Org.3:217-219. 1989c. Survival of baculoviral mid-gut gland necrosis virus
Lightner, D.V. (BMNV) in infected tissues and in seawater. Fish Pathol.
1975. Some potentially serious disease problems in the cul- 24:1797181. (Injapanese; English abstr.)
ture of penaeid shrimp in North America. In Proceedings Momoyama, YL, and T. Sano.
of the third U.S.japan meeting on aquaculture; 15-16 Oct. 1988. A method of experimental infection of kuruma shrimp
1974, Tokyo, Japan, p. 75-97. Special Publ. Fish. Agency larvae, Penaeus japonicus Bate, with baculoviral mid-gut
Japanese Government and Japan Sea Reg. Fish. Res. Lab., gland necrosis (BMN) virus. J. Fish Dis. 11:105-111.
Niigata. 1989. Developmental stages of kuruma shrimp, Penaeus
1985. A review of the diseases of cultured penaeid shrimps japonicus Bate, susceptible to baculoviral mid-gut gland ne-
and prawns with emphasis on recent discoveries and devel- crosis (BMN) virus. J. Fish Dis. 12:585-589.
opments. In Proceedings of the first international confer- Oka, M.
ence on the culture of penaeid prawns/shrimps; 4-7 De- 1969. Studies on Penaeus orientafts Kishinoue-VIII. Structure
cember 1984, Iloilo City, Philippines (Y. Taki, J.H. of the newly found lymphoid organ. Bull. Jpn. Soc. Sci.
Primavera, J.A. Llobera, eds.), p. 79-103. Fish. 35:245-250.
Lightner, D.V., and C.T. Fontaine. Sano, T., T. Nishimura, K_ Oguma, K Momoyama, and N. Takeno.
1975. A mycosis of the American lobster, Homarus 1981. Baculovirus infection of kuruma shrimp, Penaeus
ameficanus, caused by Fusafium sp. J. Invertebr. Pathol. japonicus in Japan. Fish Pathol. 15:185-19 1.
25:239-245. Takahashi, Y, Y. Shimoyama, and K. Momoyama.
Lightner, D.V., R.M. Redman, and T.A. Bell. 1985a. Pathogenicity and characteristics of Vib7io sp. isolated
1983. Observations on the geographic distribution, patho- from cultured kuruma prawn Penaeusjaponivus Bate. Bull.
genesis and morphology of the baculovirus from Penaeus Jpn. Soc. Sci. Fish. 51:721-730. (Injapanese; English abstr.)
monodon Fabricius. Aquaculture 32:209-233. Takahashi, Y, T. Itami, A. Nakagawa, H. Nishimura, and T. Abe.
Momoyama, K. 1985b. Therapeutic effects of oxytetracycline trial tablets
1981. Studies on infectious mid-gut gland necrosis of against vibriosis in cultured kuruma prawns Penaeus
kuruma shrimp (Penaeus japonicus)-I. Occurrences and japonicusBate. Bull. Jpn. Soc: Sci. Fish. 51:1639-1643.
symptoms. Bull. Yamaguchi Pref. Naikai Fish. Exper. St.
8:1-11. (Injapanese.)

The Application of Molecular Biology to the Detection of Infectious
Hematopoietic Necrosis Virus

JAMES R. WINTON
U.S. Fish and Wildlife Service
National Fisheries Research Center
Building 204, Naval Station
Seattle, Washington 98115

ABSTRACT

Traditionally, the detection of most fish viruses, including infectious hematopoietic
necrosis virus (IHNV), has relied upon isolation of the virus in cell culture and identifi-
cation by serum neutralization or fluorescent antibody assays using polyclonA rabbit
antisera. In recent years, the techniques of molecular biology have provided new strate-
gies for rapid and sensitive detection of antigens, antibodies, and nucleic acids. This
paper reviews the application of these powerful new methods for the detection of IHNV.
Included, are discussions on the creation of monoclonal antibodies specific for viral
proteins, development of enzyme-linlked immunoassays for detection of viral antigens
and antibodies, electrophoretic identification of radiolabeled viral proteins', a nonradio-
active DNA probe specific for the nucleoprotein (N) gene messenger RNA of IHNV, and
a polymerase chain reaction (PCR) based method for amplification of genomic or mes-
senger RNA of the virus. As these methods become more widely accepted, they will result
in significant improvements in the speed, precision, and sensitivity of IHNV detection.

Introduction Virological examinations are regularly conducted
on stocks of trout and salmon. These include routine
Infectious hematopoietic necrosis (IHN) is the most health checks, diagnostic examinations to identify
important viral disease of trout and salmon in west- causes of mortality, and pathogen-free certification
ern North America. Depending upon the species, examinations, which are required prior to moving
stock, and size of the fish, strain of the virus, and fish from one location to another. Traditional meth-
environmental conditions such as temperature, an ods for isolation and identifi@ation of IHNV rely
outbreak of IHN may result in losses approaching upon cell cultures for recovering the virus and serum
100% when fish are infected at a small size. Among neutralization or fluorescent antibody assays to iden-
groups of larger fish, mortality is reduced, often be- tify the agent, a time-consuming and expensive
coming chronic, with a typical loss of 25% or less process (Amos 1985). Recently, new techniques from
(Wolf 1988). Presently, no control measure for IHN is molecular biology utilizing monoclonal antibodies,
available other than avoidance of exposure to the enzyme-linked immunosorbent assays, nonradioactive
causative rhabdovirus, infectious hematopoietic ne- DNA probes, and the polymerase chain reaction,
crosis virus (IHNV). This has not proven practical at promise to provide new tools for detection and iden-
many large hatcheries with open water supplies or at tification of IHNV that are more rapid, sensitive,
commercial trout operations where. production of specific, and cost-effective.
fish occurs on a continuous basis. The chemical,
physical, and serological characteristics of IHNV, es-
sential features of the biology of the virus, and a Electrophoresis of Structural Proteins
description of the host and geographic range, have
been extensively reviewed (McAllister 1979; Pilcher Leong et al. (1981) described a method for identifi-
and Fryer 1980; Nicholson 1982; Wolf 1988). cation of different strains of lHNV using

54 NOAA Technical Re ort NWS I I I
I P

polyacryramide gel electrophoresis (PAGE) and auto- Another application of monoclonal antibodies for
radiography of the virion structural proteins that had detection of IHNV includes an immunohistochemical
been radioactively labeled with 'IS-methionine. The staining method developed by Yam 'amoto et al.
method required approximately 48 hours and pro- (1989). The, method proved useful for in situ detec-
vided a type of fingerprint that would not only tion of fish cells initially infected with IHNV
confirm the presence of IHNV, but yielded data following waterborne exposure to the virus
about the strain of the virus, giving researchers addi- (Yamamoto et al. 1990). Monoclonal antibodies have
tional epizootiological information. Hsu and Leong also been use to detect neutralization variants among
(1985) compared this electrophoretic method with strains of IHNV (Winton et al. 1988) and in deter-
two immunoblotting techniques using either 1211-la- mining important epizootiological information about
beled Protein A or peroxidase to detect the rabbit the distribution of IHNV strains (Ristow and Arnzen
anti-IHNV serum bound to structural proteins sepa- 1989).
rated by PAGE and transferred to nitrocellulose
membranes. When used to detect and confirm the Enzyme-Linked Immunosorbent Assay
presence of IHNV, the direct electrophoretic method
was the most sensitive and rapid of the three. Al- The enzyme-linked immunosorbent assay (ELISA) is
though the need to use radioactive materials has
limited the field application of this method, it has an important tool for the rapid and sensitive detec-
proven useful in analyzing the structural proteins of tion of antigens. An ELISA was used to detect IHNV
IHNV (Hsu et al. 1985) and in identifying strains of antigens present in infected cell cultures and fish tis-
the virus (Hsu et al. 1986). This electropherotypic sues (Dixon and Hill 1984; Way and Dixon 1988).
analysis (based upon variation in'the mdlecular Because the assays used polyclonal antisera, low level
weights of structural proteins) has become widely cross-reactions were observed with other fish rhab-
used in various epizootiological studies of IHNV. doviruses and with uninfected fish cell cultures or
fish tissue extracts. While differences between anti-
sera affected the results, the IHNV ELISA appeared
Monoclonal Antibodies to be quite specific and could detect infections in cell
culture as early as 48 hours after infection at the time
Monoclonal antibodies (MAb.s) are useful for both that cytopathic effect was first noted. The assay was
research and diagnostic applications as they are also used to detect IHNV in acutely infected fry.
highly reactive, very consistent, and do not cross-re- A modification of the ELISA, the dot blot, where
act with other a@tigens to any significant extent IHNV antigens were spotted onto nitrocellulose pa-
per and detected by labeled antisera, was reported by
(Harlow and Lane 1988). The first monoclonal anti- McAllister and Schill (1986). This assay was rapid and
body against IHNV was developed by Schultz et al. required no special instrumentation, but the
(1985). While the antibody lacked neutralizing activ- polyclonal IHNV antiserum required extensive ad-
ity, it could be biotinylated and used to develop an sorption with both uninfected cells and fetal bovine
immunoblot assay for IHNV (Schultz et al. 1989). For serum to remove cross-reacting antibodies before
maximum sensitivity, 'the assay required initi2il ampli- use. While the assay was not suitable for use with tis-
fication of the virus in cell culture where as few as sues or fluids because the filter matrix became
100 plaque forming units of IHNV could provide a clogged, it was able to detect I-HNV in cell culture
poIsitive diagnosis of IHNV as early as 36 hours after supernatant fluids at concentrations of 105 to 106
infection. plaque forming units (PFU)/mL when initial cyto-
A monoclonal antibody was compared with pathic effect was observed.
polyclonal rabbit serum in a rapid fluorescent antibody Although fish lack the complex immune system of
test (FAT) for IHNV by L 'aPatra et al. (1989). The FAT the higher vertebrates, it is possible to confirm past
used fish cell cultures grown on coverslips and infected infection with IHNV using methods that detect the
with IHNV for approximately 48 hours at which time presence of antibodies in serum. Amend and Smith
the test approached the plaque assay in sensitivity and (1974) reported that rainbow trout Oncorhynchus
offered a significant improvement in speed. One advan- mykiss developed good titers of neutralizing antibod-
tage of the MAb over the polyclonal serum was that it ies that could be detected at least 90 days after
did not require extensive adsorption before use. immunization with live virus. Techniques that have
Arnzen et al. (1991) 'Used a monoclonal antibody that been used to assay for antibodies to IHNV in fish
was conjugated with fluorescein isothiocyanate to pro- serum include neutralization, FAT, and ELISA
duce a direct FAT that detected IHNV antigens in cell (Amend and Smith 1974; Hattenberger-Baudouy et
cultures within 6-8 hours after infection. al. 1989; Jorgensen et al. 1991). Of these, the ELISA

Winton: Molecular Biological Methods for Detection of EON 55

appeared to have the best combination of sensitivity able to detect IHNV infections in rainbow trout for
and ease of processing the numerous samples needed up to 32 weeks post-infection.
for serological screening of fish populations.
Nonradioactive DNA Probe Discussion
A biotinylated DNA probe for rapid detection of Control of IHNV continues to rest upon avoidance of
IHNV was developed by Deering et al. (1991). This the virus through use of virus-free fish reared in vi-
probe was designed to detect the nucleoprotein (N) rus-free water supplies (Wolf 1988). Currently,
assurance of the virus-free status of salmonid fish re-
gene messenger RNA (mRNA) of IHNV because this quires a series of time-consuming, expensive, and
molecule is synthesized early and in high abundance labor-intensive procedures (Amos 1985). Many of the
during viral replication. A 30 nucleotide target site methods reviewed in this paper will be important im-
for the probe was chosen by computer search of the provements in the speed, sensitivity, and precision of
published sequence of the N gene of IHNV (Gilmore these assays. These novel assays and procedures will
and Leong 1088). A synthetic oligonucleotide, find increasing use in examination of fish and fish
complementary to this sequence, was made by auto- eggs to be moved to new locations and in providing
mated chemical synthesis, coupled with biotin, and early diagnosis of IHNV outbreaks allowing prompt
detected with a streptavidin-peroxidase conjugate. management actions.
The probe was shown to be specific for IHNV mRNA
One area of concern has been the ability to trans-
and detected as little as one picogram of target se- fer these new methods to field stations where much
quence. It did not react with mRNA of -viral of the fish diagnostic work is being performed. In
hemorrhagic septicemia virus or hirame rhabdovirus, this regard, the use of radioactive material, expensive
but it did recognize strains of IHNV representing equipment, or technically demanding procedures
each of the electropherotypes described by Hsu et al. have not proven attractive. While it seems likely that
(1986). The method required initial amplification of improvements in speed and sensitivity of diagnostic
the virus mRNA in fish cell cultures, but at high rnul- methods are still possible, they must be developed
tiplicities of infection, detectable levels of mRNA with these limitations in mind.
could be extracted from these cells after less than 24 In the future, additional methods will become
hours yielding a positive diagnosis in less than 48 available for control of IHNV. These include the use
hours. of antiviral drugs (Hasobe and Saneyoshi 1985;
Kimura et al. .1990), chemicals (Batts et al. 1991), or
modem vaccines based upon recombinant DNA tech-
Polymerase Chain Reaction nology (Leong et al. 1988). The combination of
newer detection methods and improved control strat-
The polymerase chain reaction (PCR) uses two DNA egies promises to provide - substantial reduction in the
primers and repeated cycles of DNA synthesis per- losses caused by this important virus.
formed by a thermostable polymerase to amplify low
copy numbers of a specific nucleic acid sequence to
levels that can be easily detected (Erlich 1989; Innis Acknowledgment
et al. 1990). Arakawa et al. (1990) used PCR to am-
plify a 252 nucleofide portion of the N gene of IHNV
that included the site for the DNA probe used by Some of the research appearing in this review was
Deering et al. (1991). The primers directed the syn- supported by the Cooperative State Research Service,
thesis of large amounts of DNA from all strains of U.S. Department of Agriculture to the Western Re-
IHNV tested. The DNA produced was confirmed to gional Aquaculture Consortium under Agreements
be of the appropriate sequence by Southern and dot 87-CSRS-2-3219, 88-38500-4027, 89-385004287, 90-
blot assays. Sufficient messenger RNA for cDNA syn- 38500-5025, or 91-38500-6078.
thesis using reverse transcriptase and subse quent
PCR amplification could be extracted from cell cul- Citations
tures infected for as little as 24 hours. The PCR.was
able to amplify target sequences from IHNV-infected Amend, D.F., and L.S. Smith.
rainbow trout to levels that were easily detected with 1974. Pathophysiology of infectious hematopoietic necrosis
a biotinylated probe. Recent tests using clinical mate- virus disease in rainbow trout (Salmo gairdneri): early
changes in blood and aspects of the immune response after
rial from IHNV-infected fish revealed that the injection of IHN virus. J. Fish. Res. Board Can. 21:1371-
method approached cell culture in sensitivity and was 1378.

56 NOAA Technical Report NMB I I I

Amos, K.H. LaPatra, S.E., K.A. Roberti, J.S. Rohovec, andj.L. Fryer.
1985. Procedures for the detection and identification of cer- 1989. Fluorescent antibody test for the rapid diagnosis of
tain fish pathogens, 3rd ed. Fish Health Section, American infectious hematopoietic necrosis. J. Aquat. Anim. Health
Fisheries Society@ Corvallis, Oregon, 114 p. 1:29-36.
Arakawa, C.K., R,E. Deering, K-H. Higman, K.H. Oshima, P.J. Leong, J., YL. Hsu, H.M. Engelking, and D. Mulcahy.
O'Hara, andj.R. Winton. 1981. Strains of infectious hematopoietic necrosis (IHN)
1990. Polymerase chain reaction (PCR) amplification of a virus may be identified by structural protein dif-
nucleoprotein gene sequence of infectious hematopoietic ferences. Dev. Biol. Stand. 49:43-55.
necrosis virus. Dis. Aquat, Org. 8:165-170. Leong,J.C.,J.L. Fryer, andj.R. Winton,
Arnzen,J.M., S.S. Ristow, C.P. Hesson, andj. Lientz. 1988. Vaccination against infectious hematopoietic necrosis.
1991. Rapid fluorescent antibody tests for infectious hemato- In Fish vaccination (A.E. Ellis, ed.), p. 193-203. Acad.
poietic necrosis virus (IHNV) utilizing monoclonal antibod- Press, London.
ies to the nucleoprotein and glycoprotein. J. Aquat. Anim. McAllister, P.E.
Health 3:109-113. 1979. Fish viruses and viral infections. In Comprehensive
Batts, W.N, M.L. Landolt, andj.R. Winton. virology, Vol. 14 (H. Fraenkel-Conrat and R.R. Wagner,
1991. Inactivation of infectious hematopoietic necrosis virus eds.), p. 401-470. Plenum, New York, NY.
by low levels of iodine, Appl. Environ. Microbiol. 57:1379- McAllister, P.E., and W.B. Schill.
1385. 1986. Immunoblot assay: a rapid and sensitive method for
Deering, R.E, C.K. Arakawa, K.H. Oshima, P.J. O'Hara, M.L. identification of salmonid viruses. J. Wildl. Dis. 22:468-
Landolt, andj.R. Winton. 474.
1991. Development of a biotinylated DNA probe for detec- Nicholson, B.L.
tion and identification of infectious hematopoietic necrosis 1982. Infectious hematopoietic necrosis (I.H.N.). In Anti-
virus. Dis. Aquat. Org. 11:57-65. gens of fish pathogens: Development and production for
Dixon, P.F., and BJ. Hill. vaccines and serodiagnostics (Anderson, D.P., M. Dorson,
1984. Rapid detection of fish rhabdoviruses by the enzyme- and P. Dubourget, eds.), p. 63-69. Merieux, Lyon.
linked immunosorbent assay (ELISA). Aquaculture. 42: Pitcher, K.S., andj.L. Fryer.
1-12. 1980. The viral diseases of fish: a review through 1978. Part
Erlich, H.A. (ed.). 1: Diseases of proven viral etiology. CRC Crit. Rev.
1989. PCR technology. Stockton Press, NY. Microbiol. 7:287-364.
Gilmore, R.D., andj.C. Leong. Ristow, S.S., andj.M. Arnzen.
1988. The nucleocapsid gene of infectious hematopoietic 1989. Development of monoclonal antibodies that recognize
necrosis virus, a fish rhabdovirus. Virology 167:644-648. a type-2 specific and a common epitope on the nucleopro-
Harlow, E., and D. Lane. tein of infectious hematopoietic necrosis virus. J. Aquat.
1988. Antibodies, a laboratory manual. Cold Spring Har- Anim. Health 1:119-125.
bor Laboratory, Cold Spring Harbor, NY. Schultz, C.L., B.C. Lidgerding, P.E. McAllister, and F.M. Hetrick.
Hasobe, M., and M. Saneyoshi. 1985. Production and characterization of monoclonal anti-
1985. On the approach to the viral chemotherapy against body against infectious hematopoictic necrosis virus. Fish
infectious hematopoittic necrosis virus (IHNV) in vitro and Pathol. 20:339-341.
in vivo on salmonid fishes. Fish Pathol. 20:343-351. Schultz, C.L., P.E. McAllister, W.B. Schill, B.C. Lidgerding, and
Hattenberger-Baudouy, A.-M., M. Danton, G. Merle, C. Torchy, and F.M. Hetrick.
P. de Kinkelin. 1989. Detection of infectious hernatopoietic necrosis virus in
1989. Serological evidence of infectious hematopoietic ne- cell culture fluid using immunoblot assay and biotinylated
crosis in rainbow trout from a French outbreak of monoclonal antibody. Dis. Aquat. Org. 7:31-37.
disease. J. Aquat. Anim. Health 1:126-134. Way, K., and RE Dixon.
Hsu, YL., andj.C. Leong. 1988. Rapid detection of VHS and IHN viruses by the en-
1985. A comparison of detection methods for infectious he- zyme-linked immunosorbetit assay (ELISA). j. Appl.
matopoietic necrosis virus. J. Fish Dis. 8:1-12. Ichthyol. 4:182-189.
Hsu, YL., H.M. Engelking, andj.C. Leong. WintonjR., CX Arakawa, C.N. Larman, andj.L. Fryer.
1985. Analysis of the quantity and synthesis of the virion pro- 198@8. Neutralizing monoclonal antibodies recognize anti-
teins of infectious hematopoietic necrosis virus. Fish genic variants among isolates of infectious hematopoietic
Pathol. 20:331-338. necrosis virus. Dis. Aquat. Org. 4:199-204.
1986. Occurrence of different types of infectious hematopoi- Wolf, K_
etic necrosis virus in fish. Appl. Environ. Microbiol. 1988. Infectious hematopoietic necrosis. In Fish viruses and
52:1353-1361. fish viral diseases, p. 83-114. Cornell Univ. Press, Ithaca,
Innis, M.A., D.H. GelfandjJ. Sninsky, and T.J. White (eds.). NY.
1990. PCR Protocols. Acad. Press, NY. Yamamoto, T., C.K. Arakawa, W.N. Batts, andj.R. Winton.
Jorgensen, P.E.V., NJ. Olesen, N. Lorenzen, J.R. Winton, and S. 1989. Comparison of infectious hematopoietic necrosis in
Ristow. natural and experimental infections of spawning salmonids
1991. Infectious hematopoietic necrosis virus: detection of by infectivity and immunohistochemistry. In Viruses of
humoral antibodies in rainbow trout by plaque neutraliza- lower vertebrates (W. Ahne and E. Kurstak, eds.), p. 411-
tion, immunofluorescence, and enzyme-linked immuno- 429. Springer, Berlin.
sorbent assays. J. Aquat. Anim. Health 3:100-108. Yamamoto, T., W.N. Batts, C.K. Arakawa, andj.R. Winton.
Kimura, T., M. Yoshimizu, Y Ezura, and Y. Kamei. 1990. Multiplication of infectious hematopoictic necrosis vi-
1990. An antiviral agent (46NW-04A) produced by Pseudo- rus in rainbow trout following immersion infection: Whole
monas sp. and its activity against fish viruses. J. Aquat. body assay and immunohistochemistry. J. Aquat. Anim.
Anim. Health 2:12-20. Health 2:271-280..

Bacterial and Viral Diseases of Marine Fish
During Seed Production

KIYOKUNI MUROGA

Faculty of Applied Biological Science
Hiroshima University
Higashi-Hiroshima 724, Japan

ABSTRACT

In Japan, seed production techniques have been developed for about 40 species of
marine fishes; however, mass mortalities due to infectious and noninfectious diseases
have often occurred. Among these problems are the bacterial and viral diseases reviewed
in the present paper. Vibriosis and other bacterial diseases have occurred in various
marine fishes during theirjuvenile stage. These diseases have essentially the same pathol-
ogy in both juvenile and adult fish in their tendency to terminate in systemic infection,
except for Flexibacter infections. On the other hand, larvae most frequently develop intes-
tinal infections, such as bacterial enteritis with Vibrio sp. INFL in Japanese flounder. It
has been suspected that live diets contaminated with pathogenic bacteria serve as an
important source of these intestinal infections. Recently, some new viral infections, such
as viral epidermal hyperplasia and viral nervous necrosis, have been reported in larval
and juvenile marine fishes. Although isolation of the causative viruses in cultured cells
has not been successful, except for yellowtail ascites virus (YAV), these diseases were
confirmed to occur only in the larval, or larval and juvenile, stages by infection experi-
ments and epidemiological surveys.

Introduction Bacterial Diseases

In Japan, demand for coastal fisheries and aquacul- A wide range of pathogenic bacteria have been iso-
ture products has been increasing drastically over the lated from cultured fishes during theirjuvenile stage
past, few decades, especially since the establishment (Table 1). Vibrio anguillarum infection has been re-
of the 200-mile economic zone. Consequently, gov- ported in red seabream, tiger puffer (Takifugu
ernmental programs have concentrated on the ruhripes), and Japanese flounder (Muroga and Tatani
improvement and development of coastal fisheries 1982; Muroga et al. 1987b ; and Yamanoi et al. 1988;
stocks, including mass release programs for impor- respectively) and Vibrio ardalii infection has been ob-
tant species. At present, more than 100 million served in juvenile rockfish (Sebastes schlegeli) (Muroga
juveniles of about 40 marine finfish species ate pro- et al. 1986). Pasteurella piscicida was isolated from ju-
duced annually at national, prefectural, and private veniles of black seabrearn and red grouper
hatcheries. Red seabrearn (PagTus major), Japanese (Epinephelus akaara) as a disease agent (Muroga et al.
flounder (Paralichthys olivaceus), and black seabream 1977; Ueki et al. 1990; respectively). Flexibacter
(Acanthopagrus schlegeli) are the representative species maritimus infection has been reported in red
produced in southern Japan. Fish production tech- seabream, black seabream, and flounder (Waka-
niques, particularly for the three species mentioned bayashi et al. 1984, 1986; Baxa et al. 1986). Except
above, are apparently well established (Fukuhara for Flexibacter infections, these bacterial infections ter-
1987). However, during fish production, difficulties minate in septicemi'a, or systemic infection in
are often encountered in controlling diseases of juvenile fishes as they do in adult fishes. On the other
known and unknown etiology. The present paper will hand, these systemic infections have rarely been re-
review bacterial and viral diseases of marine fishes in ported during the larval stage. Instead, high
the course of seed production in Japan. mortalities during the larval stage are usually associ-

58 NOAA Technical Report NAM 111

Table 1
Bacterial diseases in larval andjuvenile marine fishes.

Stage Name of disease Pathogen Host fish

Larval stage Abdominal swelling Vib7io alginolyticus Red seabrearn
Other pathogenic organisms Black seabream

Bacterial enteritis Vibrio sp. INFL Japanese flounder

juvenile stage Vibriosis Vibrio anguillarum Red seabream
Japanese flounder
Tiger puffer
Vib7io ordahi Rockfish,

Pasteurellosis Pasteurella piscicida Black seabream
Red grouper

Flexibacter infection Flexibacter ma7itimus Red seabream
Black seabream
Japanese flounder

'Sebastes schlegeli.

ated with intestinal infections. These intestinal infec- gram-negative, short rod, which is motile by its single
tions occur in the larvae of various marine fishes. polar flagellum. It grows well on ZoBell's 2216e agar
Abdominal swelling ("Fukubu-boman-sho" in Japa- and heart infusion or brain heart infusion agar with
nese) occurs in red and black seabreams (Iwata et al. 3% NaCl, forming small, circular, gray@colored colo-
1978; Kusuda et al. 1986; respectively). Vibrio nies. Based on biochemical characteristics and GC
alginolyticus has been reported to be the causative (guanine and cytosine) value (44.1 mol%) of DNA,
agent or to be associated with the disease in both fish the organism was placed in the genus Vibrio. How-
species. In an investigation of this disease in red ever, the characteristics of this pathogen differ from
seabream reared in extensive nursery ponds, various those of previously described fish-pathogenic Vib7io
species of the genus Vib7io, including V alginolyticus species. The pathogen was tentatively named Vibrio
and V vulnificus, were the predominant isolates from sp. INFL after the disease it causes: intestinal necrosis
different batches -of diseased larvae. The role of these of flounder larvae (Masumura et al. 1989b).
vibricis was not made clear, however, because signifi- Infection experiments using Vib7io sp. INFL were
cant mortality or abdominal swelling could not be carried out by oral administ 'ration via rotifers
produced by oral challenge in red seabream larvae (Brachionus plicatilis) and brine shrimp (Artemia
with these isolates (Yasunobu et al. 1988). A marine salina) nauplii, by addition to rearing tanks, and by
turbellarian (Allostoma sp.) was also reported to be intraperitoneal injection (Masumura et al. 1989b).
associated with abdominal swelling of black seabream The disease was reproduced in flounder larvae only
(Yamaguti 1987). At present, abdominal swelling of by oral challenge. On the contrary, V anguillarum,
seabreams can only be interpreted as being a syn- which was used for comparison, killed flounder juve-
drome caused by ingesting some pathogenic or niles by intraperitoneal injection, but did not kill
toxicogenic organisms. flounder larvae by oral challenge. The pathogenicity
A disease called "Chokan-hakudaku-sho" in Japa- of ViMo sp. INFL therefore seems to be quite differ-
nese, which means a disease condition characterized ent from that of V anguillarum.
by opaque intestine, occurs in larval Japanese Six different age groups of flounder (11, 16, 17,
flounder. The causative agent is thought to be a Vibrio 27, 41, and 60 day olds) were submitted to an oral
species isolated from an affected intestine (Murata challenge test with Vibrio sp. INFL (Muroga et al.
1987). According to Murata (1987), this disease oc- 1990). The bacteria were incorporated into brine
curs in 14 to 30- or 40-day-old larval flounder; shrimp nauplii or rotifers and were given for three
symptoms include darkening of body color and days. As a result, mortality of the test groups was sig-
opaqueness and shrinkage of the intestine. Mortali- nificantly higher than that of the respective controls
ties sometimes reach 90% or higher, especially when in the groups of 16, M and 27-day-old fish. The char-
it occurs in younger fish. The causative bacterium is a acteristic clinical sign of the disease, an opaque

Muroga: Bacterial and Viral Diseases of Marine Fish During Seed Production 59

intestine, occurred and subsequently Vibrio sp. INFL and Muroga 1989b), and recently, it was also re-
was isolated from the intestine in these test groups. ported that the bacterial contamination of rotifers
In the older fish groups (41- and 60-day olds), no and brine shrimp was significantly reduced by freez-
apparent changes were observed in either the test ing at -15' C for I month (Yamanoi and Katayama
groups or the controls. Based on these results and 1989).
the before-mentioned epidemiological data on natu-
ral outbreaks of the disease, it was concluded that
this bacterial enteritis is confined to the larval stage Viral Diseases
of flounder. Histopathological and electron micro-
scopic examinations revealed that pathogen Recently, several viral diseases have been reported in
multiplication and resultant pathological changes oc- the larvae and juveniles of marine fishes (Table 2).
curred only in the intestine (Miyazaki et al. 1990; These include viral ascites of yellowtail (Seriola
Muroga et al. 1990). Although pili-like structures quinqueradiata), viral epidermal hyperplasia of Japa-
were not observed on the cells of the pathogen, an nese flounder, viral epithelial necrosis of black
adhesive property was demonstrated on a chinook seabream, and viral nervous necrosis of Japanese
salmon embryo (CHSE-214) cell line. These findings parrotfish (OpL-gwathus fasciatus), red grouper and
led us to the conclusion that the necrotic enteritis striped jack (Pseudocaranx dentex). Except for the first
caused by Vihiio sp. INFL in flounder larvae is similar virus mentioned, the causative agents have not been
to enteric colibacillosis of young mammals (Muroga isolated in cultured cells.
et al. 1990). In 1983, an acute disease characterized by ascites
Investigations of the intestinal bacterial flora of lar- occurred among yellowtail juveniles cultured in a
val and juvenile marine fishes seem to be essential to hatchery, and an IPNV (infectious pancreatic necro-
elucidate the pathogenesis of the above-mentioned sis virus) -like virus, which was named YAV (yellowtail
intestinal infections in larval fishes. A method for iso- ascites virus), was isolated inseveral'cell lines includ-
lating and enumerating the aerobic intestinal ing RTG-2 and CHSE-214. It was confirmed by an
bacteria of larval and juvenile fish was devised by immersion challenge that the disease could be repro-
Muroga et al.:(1987a); the intestinal bacterial flora of duced in yellowtail juveniles and that the mortality
seabreams, flounder, and other fishes was subse- was higher in smaller fish (Sorimachi and Hara 1985;
quently investigated (Tanasomwang and Muroga Sorimachi and Egusa 1986). The same disease oc-
1988, 1989a). The compositions of the intestinal curred among. wild juveniles of yellowtail which were
flora of these marine fishes were characterized by two caught as seedlings for net cage culture (Isshiki et al.
predominating groups, Vibrio and Pseudomonas, which 1989). Histopathological observations of naturally
were thought to be derived from the diets of live roti- and experimentally infected fish suggest that acinous
fers and brine shrimp nauplii. The same genera, tissues of the pancreas and parenchymal tissue of the
Vib7io and Pseudomonas, were most frequently isolated liver are the primary tissues involved in YAV infection
from these live diets (Tanasomwang and Muroga (Egusa and Sorimachi 1986).
1990). Sodium nifurstyrenate bath proved effective in Since 1985, outbreaks of a disease resulting in mass
reducing bacterial contamination of rotifers (Hayashi mortalities have occurred in larvae and juveniles of
et al. 1976; Yamanoi and Sugiyama 1987; Tanasomwang the Japanese flounder cultured at several hatcheries.

Table 2
Viral diseases in larval and juvenile marine fishes.

Stage Name of disease Pathogen Host fish

Larval stage Viral epidermal hyperplasia Herpesvirus Japanese flounder
(Viral epidermal necrosis)
Viral epithelial necrosis Paramyxovirus-like virus Black seabrearn
Viral nervous necrosis Picornavirus-like virus Japanese parrotfish
Red grouper
Stripedjack
Seabass,
juvenile stage Viral ascites YAV (Yellowtail ascites Yellowtail
virus=IPW-like)

"Lates calcalif-

60 NOAA Technical Report NNM I I I

The disease occurred in 10- to 30-day-old fish reared system for marine fishes from the standpoints of
at 18-20' C, and mortality usually reached 80-90% in quarantine and hygiene, both of which have almost
.a few weeks. The affected fish had opaque fins and a been established for salmon hatcheries.
hyperplastic epidermis on the fins and skin. Electron
microscopy revealed hexagonal virus particles in the
nucleus and cytoplasm (diameter in enveloped state: Citations
200 nm) of the affected epidermal cells. Although Baxa, D.V., K. Kawai, and R. Kusuda.
isolation of the causative agent by the use of several 1986. Characteristics of gliding bacteria isolated from dis-
fish-cell cultures was not successful, the disease was eased cultured flounder, Paralichthys olivaceus. Fish Pathol.
transmitted to healthy larval flounder by exposing 21:251-258.
them to a 0.45 @Lni filtrate of diseased fish Egusa, S., and M. Sorimachi.
homogenate. Morphological and physiological char- 1986. A histopathological study of yellowtail ascites virus
acteristics of the virus indicate that the agent is a (YAV) infection of fingerling yellowtail, Seriola quin-
queradiata. Fish Pathol. 21:113-121. (in Japanese; English
herpesvirus (lida et al. 1989). When immersion chal- abstr.)
lenge tests of flounder larvae were made at three Fukuhara, 0.
different temperatures (15, 20, and 25' Q, the dis- 1987. Seed. production of red sea bream Pagrus major
ease progress was apparently delayed at the lowest (Sparidae) injapan. In Reproduction, maturation, and seed
temperature (15' Q, though the cumulative mortal- production of cultured species: proceedings of the twelfth
ity was the same for all three temperatures. Larvae U.S.-Japan meeting on aquaculture (CJ. Sindermann, ed.),
p. 13-20. U.S. Dep. Commer., NOAA Tech. Rep. NMFS 47.
younger than 20 days old (smaller than 9.5 min in Glazebrook, J.S., M.P. Heasman, and S.W. de Beer.
total length) were highly susceptible, but the suscep- 1990. Picorna-like viral particles associated with mass mor-
tibility was significantly decreased in fish 23 days old talities in larval barramundi, Lates calcarifer Bloch. J. Fish
or older (larger than 11.0 mm). Therefore, this dis- Dis. 13:245-249.
ease proved specific to the larval stage of flounder Hayashi, K., T. Kimura, and 1. Sugahara.
1976. Microbiological studies on the artificial seedling pro-
(Masumura et al. 1989a). A similar disease was re- duction of ayu fish, Plecoglossus altivelis (Temminck et
ported in Japanese flounder larvae and juveniles Schlegel)-rV. On the elimination of bacteria contaminated
under a name of viral epidermal necrosis (Miyazaki in Brachionusplicatilis and Moina macrocopa. Bull.Fac.Fish.
et al. 1989); however, a precise comparison of these Mie Univ. 3:87-99. (in Japanese; English abstr.)
lida, Y, K Masumura, T. Nakai, M. Sorimachi, and H. Matsuda.
herpesvirus infection's has not been made owing to 1989. A viral disease in larvae and juveniles of Japanese
the unsuccessful in-vitro culture of the agents. Aii- flounder Paralichthys olivaceus. J. Aquatic Animal Health
other epithelia] necrosis due to a paramyxovirus-like 1;7-12.
virus was reported from larval black seabream Isshiki, T., K. Kawai, and R. Kusuda.
(Miyazaki et al. 1989). 1.989. Incidence of yellowtail ascites virus (YAV) in wild yel-
lowtail fingerling. Nippon Suisan Gakkaishi 55:633-637.
Mass mortalities of hatchery-reared Japanese (InJapanese; English abstr.)
parrotfish larvae and juveniles have occurred in Iwata, K_ Y Yanohara, and 0. Ishibashi.
Nagasaki Prefecture, the westernmost part of Japan. 1978. Studies on factors related to mortality of young red
Light and electron microscopic examinations re- seabream (Pagrus major) in the artificial seed pro-
vealed an extensive necrosis of the nervous system in duction. Fish Pathol. 13:97-102. .(In Japanese; English
the spinal cord, spinal ganglia, and brain. Numerous abstr.)
Kusuda, R.,J. Yokoyarna, and K. Kawai.
nonenveloped virus particles, icosahedral in mor- 1986. Bacteriological study on cause of mass mortalities in
phology and measuring 34 nrn in diameter, were cultured black sea bream fry. Nippon Suisan Gakkaishi
found in the cytoplasm of the affected neurones and 52:1745-1751. (InJapanese; English abstr.)
glial cells. The disease was named viral nervous ne- Masumura, K., Y lida, T. Nakai, and T. Mekuchi.
crosis (Yoshikoshi and Inoue 1990). A similar disease 1989a. The effects of water temperature and fish age on a
herpesvirus infection of Japanese flounder larvae,
was found in red grouper larvae in 1989 and striped Paralichthys olivaceus. Fish Pathol. 24:111-114. (In Japa-
jack larvae in 1990 (Yoshikoshi 1990). Similar nese; English abstr.)
picorna-like viral particles were observed in degen- Masumura, K, H. Yasunobu, N. Okada, and K Muroga,
erative areas of the brain and retina in seabass (Lates 1989b. Isolation of a Vibrio sp., the causative bacterium of
calcafifff) larvae cultured in Australia (Glazebrook et intestinal necrosis of Japanese flounder larvae. Fish
Pathol. 24:135-141. (InJapanese; English abstr.)
al. 1990). A comparative study on these new disease Miyazaki, T., K. Fujiwara, J. Kobara, N. Matsumoto, M. Abe, and
agents is a matter of great interest because traffic of T. Nagano.
live fishes has been increasing among south and 1989. Histopathology associated with two viral diseases of lar-
southeast Asian countries. The frequent occurrences val and juvenile fishes: epidermal necrosis of the Japanese
of these new viral diseases remind us of the necessity flounder Paralichthys olivaceus and epithelial necrosis of
black sea bream Acanthopagrus schlegeli J. Aquatic Animal
of rapid improvement in the current seed production Health 1:85-93.

Muroga: Bacterial and Viral Diseases of Marine Fish During Seed Production 61

Miyazaki, T., N. Kajihara, K. Fujiwara, and S. Egusa. 1989b. Effects of sodium nifurstyrenate and tetracycline on
1990. A histopathological study on intestinal necrosis of lar- the bacterial flora of rotifer (Brachionus plicatilis). Fish
val Japanese flounder. Fish Pathol. 25:7-13. (In Japanese; Pathol. 24:29-35.
English abstr.) 1990. Intestinal microflora of marine fishes at their larval
Murata, 0. and juvenile stages. In Proceedings of the second Asian
1987. Infectious intestinal necrosis in flounder. Fish fisheries forum (R. Hirano, and I. Hanyu, eds.), p.647-
Pathol. 22:59-61. (Injapanese.) 650. Asian Fisheries Soc., Manila, Philippines.
Muroga, YL, and M. Tatani. Ueki, N., Y Kayano, and K- Muroga.
1982. Isolation of Vibrio.anguillarum from juvenile red 1990. Pasteurella piscicida infection in juvenile red grouper
seabrearn (Pagrus major). Fish Pathol. 16:211-214. (in (Epinephelus akaara). Fish Pathol. 25:43-44. (In Japanese.)
Japanese; English abstr.) Wakabayashi, H., M. Hikida, and K Masumura.
Muroga, IC, T. Sugiyama, and N. Ueki. 1984. Flexibacter infection in cultured marine fish in
1977. Pasteurellosis in cultured black seabream (Mylio Japan. Helgol. Meeresunters. 37:587-593.
macrocephalus). J. Fac. Fish. Anim. Husb. Hiroshima Univ. 1986. Flexibacter maritimus sp. nov., a pathogen of marine
16:17-21. fishes. Int.J. Syst. Bacteriol. 36:396-398.
Muroga, K, Y Jo, and K Masumura. Yamaguti, M.
1986. Vibrio ordahi isolated from diseased ayu (Plecoglossus 1987. Abdominal swelling in black seabream. Fish Pathol.
altivelis) and rockfish (Sebastes schlegeli). Fish Pathol. 22:58-59. (Injapanese.)
21:239-243. (InJapanese; English abstr.) Yamanoi, H., and I Sugiyama.
Muroga, K, M. Higashi, and H. Keitoku. 1987. Effects of sodium nifurstyrenate bath and ultraviolet
1987a. The isolation of intestinal microflora of farmed red- irradiation on the elimination of bacteria associated with
seabrearn (Pagrus major) and black seabream (Acanthopagrus rotifer. Suisanzoshoku (Aquiculture) 25:191-195. (InJapa-
schlegeli) at larval andjuvenile stages. Aquaculture, 65:79- nese.)
88. Yamanoi, H., and K Katayama.
Muroga, K., V. Tanasomwang, and K Momoyama. 1989. Effects of freezing on bacterial flora of rotifer and
1987b. Vib7io anguillarum infection in tiger puffer (Takifugu brine shrimp nauplii. Nippon Suisan Gakkaishi 55:2207.
rubripes) fingerlings. Fish Pathol. 22:29-30. (Injapanese.) (Injapanese.)
Muroga, K, H. Yasunobu, N. Okada, and K Masumura. Yamanoi, H., Y, Momoyama, H. Yasunobu, and K. Muroga.
1990. A bacterial enteritis of cultured flounder Paralichthys 1988. Vibrio anguillarum infection in flounder (Paralichthys
olivaceus larvae. Dis. Aquat. Org. 9:121-125. olivaceus) fingerlings. Fish Pathol. 23:69-70. (Injapanese.)
Sorimachi, M., and T. Hara. Yasunobu, H., K Muroga, and K Maruyarna.
1985. Characteristics and pathogenicity of a virus isolated 1988. A bacteriological investigation on the mass mortalities
from yellowtail fingerlings showing ascites. Fish.Pathol. of red seabream -Pagrus major larvae with intestinal
19:231-238. (InJapanese; English abstr.) swelling. Suisanzoshoku (Aquiculture), 36:11-20. (InJapa-
Sorimachi, M., and S. Egusa. nese; English abstr.)
1986. Experimental infection of yellowtail ascites virus (YAV) Yoshikoshi, K
to yellowtail fingerlings. Fish Pathol. 21:133-134. (In Japa- 1990. A virus found in larval marine fishes asssociated with
nese.) mass mortalities. In Abstracts of workshop of the Japanese
Tanasomwang, V., and K. Muroga. Society of Fish Pathology, p.5. Japanese Society of Fish Pa-
1988. Intestinal microflora of larval and juvenile stages in thology, Tokyo. (Injapanese.)
Japanese flounder (Paralichthys olivaceus). Fish Pathol. Yoshikoshi, K-, and Y- Inoue.
23:77-83. 1990. Viral nervous necrosis in hatchery-reared larvae and
1989a. Intestinal microflora of rockfish Sebastes schlegeli, tiger juveniles of Japanese parrotfish, Oplegnathus fasciatus
puffer Takifugu rub7ipes and red grouper Epinephelus akaara (Temminck & Schlegel). J. Fish Dis. 13:69-77.
at their larval and juvenile stages. Nippon Suisan
Gakkaishi 55:1371-1377.

An Ecological Study of the Parasitic Nematode
Hysterothylacium haze in the Japanese Common Goby
Acanthogobiusflavi.manus, in a Brackish Inlet

TOMOYOSHI YOSHINAGA

National Research Institute ofFisheries Science
Kachidoki, Chou-ku
Tokyo 104, Japan

ABSTRACT

The unique life cycle of the parasitic nematode. Hysterothylacium haze in the Japanese
common goby Acanthogobius flavimanus is presented. It is a direct cycle in which the fish
become infected by ingesting fully developed eggs or hatched larvae of the nematode or
by ingesting invertebrates that can act as transport hosts. Infections by the nematode
were investigated monthly in a wild population of common goby in.a brackish inlet over
a period of five years. Seasonally, the rate of infection in the body cavity peaked in
November in fish age 0+ and in May in fish'age I+. Annually, the infection rate declined
during the investigation period. The intensity of infection in individual fish tended to
fluctuate with the population's overall infection rate for both age 0+ and age I+ fish. The
influence of the infection on the host population is briefly discussed.

Introduction harmful to the host goby, heavy infections causing
the deaths of some hosts. There is no other known
Mass'mortalities of the Japanese common goby case in which an anisakid nematode infection is so
Acanthogobiusflavimanus (yellowfin goby) occurred in harmful as to kill the host fish. From the extensive
the inner part of Tokyo Bay every summer from 1973 damage observed in heavily infected fish, parasitism
to 1975. The body cavities of the dead fish were by the nematode may even reduce the host popula-
heavily infected with the eggs, larvae, and adults of a tion size.
nematode; visceral adhesions were often noted A previous study of the biology of H. haze has al-
(Takahashi et a]. 1976; Takahashi et al. 1977). ready elucidated its peculiar life cycle (Yoshinaga et
Takahashi et al. (1977) demonstrated that the heavily al. 1989). An ecological investigation was made for
infected fish were intolerant of low dissc4ved oxygen five years. In this paper, the seasonal and annual
concentrations and concluded that the mass mortali- changes in H. haze infections in goby inhabiting a
ties of the goby were caused by the nematode brackish inlet and the life cycle of this parasite are
infections in combination with the scarcity of dis- described. In addition, the influence of its infection
solved oxygen. The nematode was identified as a new on the population of the host is discussed briefly.
anisakid, Thynnasca7is haze, by Machida et al. (1978)
and later assigned to the genus Hysterothylacium as H.
haze, a new combination by Deardorff and Overstreet Life Cycle
(1981).
This nematode possesses unique features among In general, nematodes molt four times prior to be-
the anisakid nematodes. Adult worms and eggs of H. coming adults. The first-stage larvae are worms
haze were found in the body cavity of the host fish, before the first molting. Following the first, second,
whereas the adults of other anisakid nematodes har- and third moltings, the worms are called the second,
bor and deposit their eggs in the' digestive tract of third, and fourth-stage larvae, respectively, and after
their definitive hosts. It also appears that H. haze has the fourth molting (the fifth stage), they are adults.
a unique life cycle and that infection is apparently It was reported by Yoshinaga et al. (1988) that all

64 NOAA Technical Report NAM I I I

Ecological Study

An ecological study of H. haze was carried out in
EGG Invertebrates Lake Shonai, a brackish inlet of Lake Hamana, an
(transport host) estuary located on the Pacific coast of central Ja-
pan (Fig. 2). TheJapanese common goby has a life
span of one year and spawns in the spring
(Miyazaki 1940); however, some fish do not be-
come fully mature during the spawning season,
probably because of delayed growth. These fish sur-
ADULT -o- L4.*-L3 L34-L2 vive to become age 1+ in some areas. Thus, the
I +
EGG (Ll--o- L2) "P-12 _4*_j goby population in Lake Shonai consists of two
digestive year classes from late spring to autumn.
body cavity tract wall Hysterothylacium haze infections in the Japanese
common goby were investigated monthly from June
Japanese conunon goby 1981 to November 1982 and from December 1983 to
(definitive host) September 1986. Fish captured commercially by trap
net were sampled. They were examined for H. haze
Figure I infection in the body cavity or in the digestive tract
Life cycle of Hysterothylacium haze. L, L 2, L3, and L 4 repre- wall, or both, after age determination by scale read-
sent the first-, second-, third-, and fourth-stage larvae, ings. Infections in the body cavity were examined
respectively. The development from L 2 to L. in the body
cavity of the Japanese common goby (dotted arrow) is un-
certain. (From Yoshinaga et al. 1989.)

stages of H. haze are harbored in the body cavity of
heavily infected fish; this is an exceptional feature
among anisakid nematodes. Detailed observations of
H. haze infections in goby and experiments with in-
fected goby and invertebrates resulted in the R. liana
elucidation of the H. haze life cycle (Yoshinaga et al.
1989). It is summarized as follows (Fig. 1): when a
goby eats fully developed worm eggs, second-stage Th.e Paci-fiIC. ..2.m
larvae in the eggs hatch and penetrate the digestive
tract wall of the fish. The fish may also become in-
fected by eating invertebrate transport hosts
containing second-stage larvae in their body cavities; R. Isaji
various invertebrates (e.g., polychaetes, amphipods, .2
and mysids) can act as transport hosts. Second-stage
larvae develop to the third stage in the fissue of the
goby's digestive tract wall, then migrate into the body 2
cavity. In the body cavity, third-stage larvae develop
R. Shin.
through the fourth stage to become adults, that ma-
ture and deposit eggs in the body cavity of thg host.
These eggs are released into the water after the death 0 1 2 kn
of heavily infected goby. It should be noted that the
deaths of heavily infected goby are essential for egg
dispersion and subsequent invasion of new hosts. The
eggs hatch in the body cavity of the host. Although it
is uncertain whether the second-stage larvae can de- Figure 2
velop to third-stage larvae in the body cavity, it seems Map of Lake Hamana (A) and Lake Shonai (B).
that a large number of larvae are harmful to the host Lake Shonai is a brackish water inlet of Lake
and contribute to its death even when the host is Hamana, an estuary located on the Pacific Coast in
infected with only a small number of adults. Shizuoka Prefecture, Honshu.

Yoshinaga: Ecological Study of the para itic nematode Hysterolftylacium haze 65

100 1980 1983

0-
dP 100
1981 1984

U
OL *7 . . . . .......
H

100
1982 1985

Figure 3
OL Monthly changes in the level of
L-L 'L'-.L' L' _L'A' L L' I
M J A S 0 N D J F I . M J @ A S 0 N C1 1986 Hysterothylacium haze infection in the body
Month cavities of the Japanese common goby by year
[]ID 1: lightly infected class from 1980 to 1986. Asterisks indicate
IMID 2: moderately infected months when sample sizes were less than 30.
ID'= intensity index determined by macro-
ID 3: heavily infected scopical observation; ID 1 = lightly infe 'cted
M I J A S 0 N 0 J F M A M J J A S 0 N D with 1-9 worms: ID 2 = moderately infected
Month with 10-99 worms; ID 3 heavily infected
with more than 100 worms.

with a stereomicroscope during the entire investiga-
tion period. Infections in the digestive tract wall were 100 - 1984
examined microscopically by compressing the tract
between glass slides. 50-

& 0.
Seasonal Changes 100- 1985
Infections in the body cavities occurred in a clear ;0-
seasonal pattern, although there were slight differ-
ences in the pattern between. years (Fig. 3). 0.
Infections in age 0+ fish were first noticed in early H 100-
summer. The incidence of infected fish peaked in 1986
November and subsequently declined to a minimum BC
level. between January and March. A second peak oc- 50. IMBC+DTW
curred in May in age I+ fish that had survived after 0- EDTW
their spawning season, and then the peak declined.
When the prevalence of. the nematode was . high, MJ JASOND JFMAMJ JASON D
heavily infected fish containing worm eggs in the Month
body cavity were frequently found.
Although infection of the digestivetract walls was Figure 4
investigated for only two years, it was observed that Monthly changes in the incidence of infection with
the year when the infection level in the body cavity Hysterothylacium haze in the two infection sites in the
was lower and showed a less clear seasonal pattern, a 1984-86 year classes of the Japanese common goby.
seasonal pattern was still discernible in the incidence Blank, semidark, and dark areas represent fish in-
fected only in the body cavity (BC), both in the body
in the two sites of infection (Fig. 4). Characteristi- cavity and digestive tract wall (BC+DTW), and only
cally, infection in the digestive tract wall preceded in the digestive tract wall (DTW), respectively.

66 NOAA Te4knical Report NMFS I 11,

that in the body cavity. Infection in the digestive tract age 1+ to age 0+ fish occurred in summer, suggesting
wall began in June or July prior to or in the same that the survival of part of the goby population after
month as the infection in the body cavity. The inci- the spawning season is indispensable to the nema-
dence of nematode infection in the digestive tract tode life cycle.
walls peaked in August and declined to its lowest lev-
els in November and December (1985 year class)
when the incidence of infection in the body cavities AnntW Changes
began to peak. The incidence of nematodes in the
digestive tract walls continued to rise until April and During this 5-year investigation, the prevalence and
declined in May, when it reached a second peak in intensities of H. haze infection in goby body cavities
the body cavity. showed sharp annual declines when infection peaks
were observed (Fig. 6). The declines were observed
over the whole life span of the goby (Fig. 3). Also the
Age of beginning of the invasion of the body cavity was de-
the goby 0+ 1+ layed to July in 1985 and to August in 1986, whereas
it first occurred in June in the 1982 and 1984 year
Invasion to classes.
the digestive
tract vall

migration to
the body cavity 0100 2 100 2
0+ 1+
Depo ition of ONE ON11 OP
the :ggs

4J
infection level r4
in the body cavj H 50 so- 11
>
4J
. . . . . . . . . . .. ri
M J J 0, S 0 N D J F M A M J J
Figure 5 J J o J. Jo
A diagram of the seasonal pattern of Hysterothylacium '81182'83'841 85 '81'82'83'84'85
haze infection in the Japanese common goby in Lake
Shonai. Year class Year class

Figure 6
The seasonal pattern of H. haze infection is summa- Annual changes in the level of Hysterothylacium haze in the
rized from these results as follows (Fig. 5): body cavities of the Japanese common goby by year class
seIcond-stage larvae invade the digestive tract wall of from 1981 to 1985. The incidence and mean intensity in-
age 0+ goby in summer and develop to the third dex (total intensity indexes/number of fish examined) are
stage there. They migrate to the body cavity from shown for the first and second peaks in age 0+ and age 1+
summer to autumn, develop to the adult stage, ma- goby, respectively. See Figure 3 caption for definition of
ture, and then deposit their eggs there in late intensity index. Vertical bars are 95% confidence limits.
autumn through early winter when the infection level
in the body cavity in the body cavity first peaks. Be- If H. haze infection had regulated the host popula-
tween December and March, new invasions into the tion size, some relationship would have been found
digestive tract wall occur again and the infection between the levels of the host population and the
level in the body cavity reaches a second peak in May nematode population. Figure 7 shows that the goby
in age 1+ fish. catch in Lake Shonai from 1982 to 1986 (data from
Hysterothylacium haze has never been reported to oc- the Shirasu Branch, Hamana Fisheries Cooperation).
cur in any fish other than the Japanese common Though the total catch of goby showed considerable
goby; therefore it appears that the biology of the annual fluctuation, there was no apparent relation-
goby is closely related to the biology of the nema- ship between the infection levels and catch. It seems
tode. It is also apparent, from the seasonal pattern likely that the H. haze infection has little influence
above, that the transmission of the nematode from upon the population's size, despite the lethal damage

Yosbinaga: Ecological Study of the parasitic nematode Hysterothykxium haze 67

Another question arose from the annual decline of
2- the H. haze infection level. Why did the nematode
1982 population decrease during the investigation? Al-
though there were no data available to. interpret it,
two possible reasons may be assumed from the biol-
ogy of the nematode: one is that the populations of
0 n m n J
A " M J J A S the invertebrate transport hosts may have decreased,
and another is that the biological conditions of the
2- 1983 Japanese common goby in Lake Shonai (e.g., spawn-
ing season, survival after the spawning season, and
growth) may have changed. Nevertheless, some eco-
logical changes should have occurred in Lake
0 n n Q n Shonai, based on Price's (1980) contention that para-
0- ra -
A M J J A S sites generally exist in nonequilibrium` states, where
the departure of any one element in the patch may
render the population inviable.
2- 1984

Citations
U

Deardorff, T.L., and R.M. Overstreet.
0 S 1981. Reviews of Hyste,rothylacium and 1heringascafis (both pre-
viously = Thynnascafis) (Nematoda: Anisakidae) from the
x 2- northern Gulf of Mexico. Proc. Biol. Soc. Wash 93:1035-
1985 1079.
Machida, M., YL Takahashi, and S. Masuuchi.
U_ 1 1978. Thynnascaris haze n. sp. (Nematoda, Anisakidae) from
goby in the Bay of Tokyo. Bull. Natl. Sci. Mus., Ser. A
-AIMMMMAM (Zool.) 4:241-'244.
i J A S Miyazaki, 1.
1940. Studies on the Japanese common goby, Acanth6gobius
2- flavimanus (Temminck et Schlegel). Bull. Jpn. Soc. Sci.
1986 Fish. 9:159-180. (Injapanese; English abstr.)
Price, P.W.
1 - 1980. Evolutionary biology of parasites. Princeton Univ.
Press, Princeton, NJ.
Takahashi, K_ S. Masuuchi, T. Nakamura, and M. Saito.
0- J J A S 1976. On A nematode from Acanthogobius flavimanus. In
Annual report of ecological survey of fisheries and shell-
MONTH fishes in the inner area of Tokyo Bay (1974). Publication
of the Tokyo Metropolitan Fisheries Experimental Station
267:42-50. (Injapanese.)
Figure 7 Takahashi, K, S. Masuuchi, T. Nakamura, M. Ogura, and T. Arima.
Changes in the catch (metric tons) of the Japanese common 1977. On a nematode from Acanfhogobiusflavimanus (2). In
goby in Lake Shonai from 1982 to 1986 (based on data from Annual report of ecological survey of fisheries and shell-
Shirasu Branch, Hamana Fisheries Cooperation). Blank and fishes in the inner area of Tokyo Bay (1975). Publication
dark areas represent age 0+ and age I+ fish respectively. In of the Tokyo Metropolitan Fisheries Experimental Station
the semidark area, age was not determined. 276:89-93. (Injapanese.)
Tokyo Metropolitan Fisheries Experimental Station.
1981. Annual report of ecological survey of fisheries and
shell-fishes in the inner area of Tokyo Bay (1976-
to heavily infected hosts. This is probably due to the 1979). Publication of the Tokyo Metropolitan Fisheries Ex-
fact that the Japanese common goby has a life span of perimental Station 30, 73 p. (Injapanese.)
only I year and the influence of the infection does Yoshinaga, T., K Ogawa, and H. Wakabayashi.
not accumulate year by year, and because the goby 1988. Developmental morphology of Hysterothylacium haze
population is regulated more by environmental fac- (Nematoda: Anisakidae). Fish Pathol. 23:19-28.
tors than by the population size of the previous Yoshinaga, T., YL Ogawa, and H. Wakabayashi.
1989. Life cycle of Hysterothylacium haze (Nematoda:
generation (Tokyo Metropolitan Fisheries Experi- Anisakidae: Raphidascaridinae). J. Parasitol. 75:756-763.
mental Station 1981).
LAMI.

Epidemiology of Marine Fish Diseases
MP
in the Warm Waters Along the'Kuroshio Current

HIROKO ISHIOKA

Nansei National Fisheries Research Institute
2-17-5 Maruishi, Ohno-Cho, Saeki-Gun
Hiroshima, 739-04,japan

ABSTRACT

In Japan, epidemiological studies have been carried out on the diseases of marine fish
for many years. From 1981 to 1989, daily clinical record cards were used in every prefec-
ture to obtain information on the outbreak of diseases in cultured marine fish.
Observations were recorded on a standard form, translated into codes, and stored on a
data base. During the analysis of this data, fish diseases were classified into six groups
according to the causal agent: bacterial, viral, fungal, parasitic, nutritional, or unidenti-
fied disease. The most frequently reported group was bacterial disease, while the least
common one was fungal disease. This paper describes the species-specific features and
seasonal variations in outbreaks of these disease groups and discusses the relationship
between fish size and diseases. Also, the relationship between locality and diseases are
analyzed for yellowtail, Seriola quinqueradiata. From this epidemiological examination, it is
clear that disease outbreaks by bacteria that are normally nonvirulent, such as strepto-
cocci and gliding bacteria, suggest that fish cultivation itself disturbs the natural
relationship between fish and the environment. In other words, deteriorated environ-
mental quality and poor fish health led to a beneficial situation for the growth of
pathogens.

Introduction tured, 3) produce readily available seedlings, and 4)
must be commercially valuable as sashimi (raw fish).
Japan is surrounded by two major oceanic currents: Among the 30 species now being cultured in Japan
the Kuroshio Uapan) Current and the Oyashio are Seriola quinqueradiata (yellowtail), Pagrus major
(Okhotsk) Current. The Kuroshio Current is warmer (red sea bream), Paralichthys olivaceus (Japanese
and its surface temperature varies from about 10' C flounder), and Takifugu ru@rripes (tiger puffer). Yel-
in winter to about 30' C in summer (Fig. 1). Most lowtail. is the most popular and familiar species
marine fish cultures in Japan are carried out in the among these and accounts for about 70% of the total
coastal waters along this current. The marine envi- production of cultured marine fish.
ronment in these coastal waters is suitable for marine The production of marine fish by artificial cultures
fish cultivation because of its warmer temperature, and traditional methods is shown for the period
milder weather, abundant sunlight, and lower chlo- 1976-86 in Figure 2 (SID 1981-90). Recently, the an-
rinity. Nitrogenous and phosphorous nutrients nual catch by marine fisheries (excluding shellfish
supplied by many rivers make this area highly fertile. and seaweeds) has amounted to about 10 million
These factors support the high production of phyto- metric tons (t) yielding a gross income of about 210
plankton and zooplankton, which in turn assure food billion yen. The harvest from marine cultivation has
for fish and shellfish larvae. Also, the rias coastline is totaled about 0.2 million t valued at 20 billion yen.
well suited for the culturing of marine fish because Although the net production of fish both by tradi-
culture equipment can remain set up throughout the tional marine fisheries (A) and marine cultures (B)
year and the daily work routine of fish culturing can has increased slightly since the early 1980s, the ratio
be easily performed in the small gulfs and bays. between them (B/A) has remained fairly constant at
Fish species chosen for cultivation must be 1) about the 2% level. In contrast, the ratio of gross
adaptable to the culture environment, 2) easily cul- income (b/a, Fig. 2) has steadily increased and

70 NOAA Technical Report NPM I I I

and are fed nonliving raw food, such as minced sand
eels.
In the case of fish species such as red sea breams,
where seedling production is done mainly in land-
based ponds, fertilized eggs are collected by net and
41 then transferred into a rearing tank. After hatching,
4@ I fish are fed with live feed suitable for their size, such
as oyster eggs, rotifers, brine shrimp, and zooplank-
ton, which are collected from the sea. As they
W develop, they are eventually fed nonliving feed such
as minced raw fish, chopped shrimp, or shellfish until
G development to the young fish stage is almost com-
W
C W pleted.
I'-, Intensive rearing in the marine net pen is then
1 C carried out for about 1 or 2 years until the fish grow
j 2 41 PREFECTURES to a marketable size. During this rearing period, fish
W 'I
1. MIE are fed principally with sliced or cut, raw or frozen
4 1 2. WAKAYAMA fish meat.
3. HYOGO
4YAMAGUCHI Intensive culture techniques have an adverse effect
09 W
5KAGAWA
6TOKUSIMA not only on fish health, but also on the conditions of
7KOCHI
8. EHIME the surrounding environment. Food containing high
9.MIYAZAKI
10. KUMAMOTO levels of protein and fat, such as raw fish meat, is
11. KAGOS I MA supplied in large amounts in expectation of more
12. NAGASAKI
13, OITA rapid growth. The leftover food and fish excrement
14.SAGA
deposited on the sea bottom cause a deterioration of
CCold wafer water quality, which is believed to be a cause of red
WWarm water tides and diseases. In addition, excessive feeding and
restricted activity in the'limited confines of the rear-
ing pens also make fish fatty.
Figure 1
Currents around Japan.
Investigating Disease Outbreak
amounted to about 10% in 1986. This suggests that
the market not only demands cultured fish as a pro- At the request of an aquaculturist fish diseases are
tein source but also holds their specie s-spe cific diagnosed at the prefectural experimental stations,
quality and freshness in high regard. where daily clinical records are maintained. Diagnos-
Because of the importance of cultured marine tic data are recorded on a standard form (Table 1),
fishes to the Japanese economy, an epidemiological translated into codes, and stored in data bases of ma-
study of the fish diseases affecting the industry was rine fish diseases at the Nansei National Fisheries
undertaken. Data compiled by the various prefectural Research Institute. These data can then be entered
fisheries institutes from 1981 to 1989 were examined into a personal computer (NEC 9801) and be statisti-
to establish trends in disease outbreaks and relation- cally sorted and calculated using DATA BASE IV
ships between fish, their culture environment, and software (Nihon Integrated Software Co., Ltd.). The
disease organisms. program for compiling, sorting, calculating, and
printing disease case data was previously reported
(Umezawa and Ishioka 1988). The data base of the
Marine Fish Cultivation: present study represents diagnoses conducted by 20
Technical Features prefectures during the period from 1981 to 1989.
The standard form consists of five parts, as shown
The techniques used to culture marine fishes in Ja- in Table 1. The first part includes information essen-
pan vary little between species and localities. In the tial for registering one diagnostic case: fish species
popular case of yellowtail, fry or seed fish migrating name, region where the disease occurred, name of
with drifting algae in the surface of the Kuroshio the fish farm or fish culturist, diagnosis date, and the
Current are captured by a small purse seine. The fish disease name. Diagnosis relies upon physical observa-
are transferred to floating net pens located in a bay tions, parasitology, and bacteriology of the diseased

Ishioka: Epidemiology of Mar ine Fish Diseases 71

A: GROSS CATCH QUANTITY
OF SEA FISHERIES

110-

_100-
Z
0go- ------
.3
so.
B: PRODUCTION BY
>-' 70 RATIO B/A
MARINE FISH CULTURE
-2

L) 0
'd5 'd6
W

25- a:GROSS MARINE FISHERIES INCOME

2-
.0
20-

0 b:MARINE CULTURE INCOME
0
Z
RATIO b/a .10.
W
Cr 8
W
.6
LL
2- .4
0 2
a:
0 t2' "43' "61 __@85 0 Figure 2
YEARS Catch quantity and gross income by ma-
I rine fisheries.

fish. Because fish are cultured in a group, one disease years of fish cultivation experience accumulated by
is defined as one diagnostic case of one species in the aquaculturist.
one fish culture establishment in a region. The fifth part is a memorandum for recording
The second part of the form contains specific in- other information, such as the results of bacterial
formation on the diseased fish; date of observation of checks, symptoms of unidentified diseases, or infor-
unusual behavior or other abnormal symptoms, such mation on the occurrence of red tide.
as body color change or appetite loss; date of appear-
ance of mortality; number of dead fish per pen;
whether the progress towards death was sudden or Results
prolonged; fish size (body length, and weight); and
any treatment administrated to the diseased fish. Number of diseases reported
The third part details cultivation conditions such
as food type and species, region of food production, The rate of fish farms requesting disease diagnosis in
size of healthy fish in the pen, the basic information each prefecture was calculated by summing the total
necessary for calculating fish density, pen type, and number of establishments requesting diagnosis by using
seed information. the standard forms for that prefecture and dividing it
The fourth part contains environmental informa- by the total number of marine fish culture establish-
tion on the cultivation area, that is, water depth, ments quoted from governmental statistics (SID
weather, degree of rain, temperlature conditions, 1981-90). While both the number of disease cases re-
number of years the area was used as a fish farm, and ported and the. diagnostic rate differed from one

72 NOAA Ted@riiical Report NAM I I I

During the 9 years from 1981 to 1989, 56 species
Table I fish were reported to be diseased and over 20,000
Standard data form used to compile disease diagnosis disease cases were registered in 20 prefectures. About
information. three quarters of these cases involved yellowtail. Gov-
ernmental statistics (SID 1981-90) indicate an
Category Content increased number of fish species in marine cultiva-
Essential Species name tion. This increase in cultivation may contribute to
Information Region the increased number of fish with disease; an espe-
Name of fish farm cially high incidence of disease originated from the
Date of investigation recent increase in the culture ofjapanese flounder.
Name of disease
Information Date abnormality was detected
on the Date dead fish was observed
diseased fish Number of dead fish per pen Disease Diversity
Was death sudden or prolonged?
Size of dead fish (cm) Fish disease was classified into six groups according
Weight of dead fish (g) to the causal agent: bacteria, parasite, virus, nutri-
Treatment used
Information on Foodtype
cultivation Region of food production UNIDENTIFIED OTHERS
conditions Age of the fish NUTRITIONAL (8.4%) 10.6%)
Size of healthy fish (cm) (6.5%)
Weight of healthy fish (g)
Number of the fish per pen
Square measure of the pen(ml) VIRAL
(3.9%)
Length of the pen (m)
Pen type
Days after pen renewing
Place of seed production . . . . .... ..
BACTERIAL
Age of seed introduced (78.5%)
Date of the seed supplied F
PARASITIC
Information Water depth (m) (8.3%)
on the Weather
cultivation Degree of rain
environment Temperature conditions
Period the area was applied
for fish cultivation (years)
Length of experience of Figure 3
aquaculturist Disease groups observed in marine fish cultures located
Memorandum Other information such as along the Kuroshio Current.
bacterial checks, symptoms
of unidentified diseases,
occurrence of red tide, etc ... tion, fungus, and unidentified disease. When the dis-
eases occurred simultaneously, the causal agents were
classified into both partitions, so the total amount of
relative frequencies were over * 100%. Bacterial dis-
prefecture to another, a considerably high correlation eases were most frequently reported, accounting for
was calculated between the numbers of disease cases approximately 80% of all disease cases (Fig. 3). Para-
and numbers of marine fish culture establishments in sitic (8.3%), viral (3.9%), nutritional (6.5%), and
each prefecture. Spearman's rank correlation coeffi- unidentified (8.4%) diseases were less frequent. The
cient between number of establishments requesting partitioning of these values seems to be closely re-
diagnoses by their prefectural experimental station and lated to the fact that diagnoses were made during the
the total number of establishments in each prefecture advanced levels of disease progression. The number
was about 0.9 for yellowtail, and 0.7 for red sea brearn of dead fish recorded in diagnostic cards suggests
,in 1985. These results may suggest that collected data that the diagnoses had been performed after a con-
represent the trend of disease outbreaks in the fish of siderable number of dead fish were noticed.
the warm current on a smaller scale. Mean diagnostic
rates were about 20% for yellowtail, about 8% for red
sea bream, and about 15% for total marine fishes cul- Species-specificity of diseases
tured. Based on these values red sea breams appear to
be less sensitive to disease than yellowtail. The relative frequencies of the six disease groups var-

Ishioka: Epidemiology of Marine Fish Diseases 73

BACrERIAL VIRAL UNIDENTIFIED
PARASITIC El NUTRITIONALEM OTHER
N=2388 E M-illi 1111111111 Paralichthys olivaceus

. ..... ..
N=37 Sebastes schleseli
..... ........

N=901 Takifugu rubripes
MWW
E&TO

N=103 Seriola duserili

N=428 Seriola aureavittata

N=14689 Seriola quinqUeradiata

N = 171 Caranx delicatissimus

N=206 Trachuruef japmicus

S@.
N=1857 pagrus major

N =104 Ar-anthopagrus schlewli

N=108 Lateolabrax japonicus

N=59
oplegnathus puctatus

N =209
IFS W@N:: oplegnathus fasciatus
120 110100 90 80 70 60 5@O 4'0 A 2'0. 10 0'
Relative Frequency Figure 4
Composition of disease variet-
ies in each species.

ied with fish species (Fig. 4). Over 90% of the dis- pathogen -has not yet identified officially. The rate of
eased yellowtail experienced bacterial disease. In red unidentified disease was highest among the 13 fish
sea brearn, a high rate (22.1%) of parasitic disease species listed in Figure 4.
resulted from parasitism by Bivagina tai on the gill, by In Japanese flounder, bacterial disease accounted
Longicollum pagrosom in the rectum, and for about 67% of all the cases registered, while the
by other parasites. The high rate (33.4%) of nutri- rate of parasitic disease, mainly due to a! Tyichodina
tional disease involves yellow fat disease and peroxi- sp. and Cryptocaryon irritans, was 17%.
dative lipid intoxication caused by lipid metabolism The relative frequency of the six disease groups
disorders. was not related to any taxonomical order nor to the
In'tiger puffer, several serious parasitic diseases ecological habitat of the cultured fish species.
(Heterobothrium tetradonis, Trichodina sp., and The relative frequency of diseases caused by differ-
Cryptocaryon i mitans) break out frequently (39.2%). ent pathogenic bacteria, are shown in Figure 5 for 13
The specific viral disease (9.0%), named kuchi-jiro fish species. Yellowtail, purple yellowtail (Seriola
sho (white mouth symptom) is often reported; its aureovittata), and gold striped yellowtail (Seriola

74 NOAA Technical Report NMFS I I I

ORDER Percifornes Tetr tifo@s
SUBORDER Percoidei --I
dae Percichthyidae Pleuronectif6ree--
FAMILY
Oplegrathidae SPa id" Scarpe ifonles
SUBFAMILY I
Seriala Opiegmthus
GENUS

120-

110-
5100-

go-
r.
:4) 80-
70-

60-

50-
ca
40-

30-

20-

0
cc
C
A
1 z
to U
4d
A E
M 0

0
V 4J
0 U

Vibi.ig ftdtuherculasis StreptDcoccosis
Edwardsiellosis 0 Glidig bct.ri.1 Di..e Figure 5
Composition of bacterial
diseases in each species.

dumefili), showed higher rates of pseudotuberculosis bacteria and Streptococcus spp.
and streptococcosis diagnoses, whereas jack mackerel Tiger puffer suffered mainly from vibri.osis and
(Tachurusjaponicus), taxonomically closely related to gliding bacteria infections.
yellowtail, had high rates of vibriosis and With some exceptions, the relative frequency of
streptococcosis. bacterial disease seems to be related to the host by
Red sea bream were frequently infected with Vibrio the host's ecological niche and related physiological
spp., Edwardsiella tarda, and gliding bacteria (mainly characteristics. As shown in Figure 5, pelagic or mi-
Flexibacter sp.), but less so with Streptococcus spp. and gratory fish species such as ye'llowtail were more
Pasteurella piscicida. No edwardsiellosis was reported susceptible to Ptisteurella piscicida, while bottom fishes
from black sea bream (Acanthagrus schlegeli), which like tiger puffer, Japanese flounder, and Japanese
are related to red sea bream. rockfish (Sebastes schlegeli), were more susceptible to
Japanese flounder were seriously infected with gliding bacteria. Other species in Figure 5 fell be-
Edwardsiella tarda and suffered heavy damage; the tween these two extremes.
relative frequency of edwardsiellosis in Japanese A species' ecological habits are closely related to its
flounder amounted to 41.8% of all bacterial disease anatomical, physiological, and biochemical character-
cases. They were affected to a lesser extent by gliding istics. Pelagic and migratory fishes have a typical

Ishi oka: Epidemiology of Marine Fish Diseases 75

35-

30-

i@ 25-

20-

..............
15 ........... .. .........

04
10"

5
01 ............. ...........
JAN FEB MAR APR MAY JUN JILY AUG SEP OCT NOV DEC
MONTH Figure 6
Seasonal changes in
-YELLOWTAIL ---- RED SEA BREAM ..... JAPANESE -TIGER PUFFER number of diseases
FLOUNDER registered between
1981 and 1989.

spindle shape suitable for swimming, developed cer- the pathogen) are very important to understand the
ebellum and vascular system, abundant superficial phenomenon or mechanism of bacterial disease out-
dark muscle, and an abundance of myoglobin in break.
their ordinary muscle. On the other hand, the bot-
tom fishes of coastal areas, such asjapanese flounder,
rockfish, puffers, and sea breams have flat or Seasonal Variations of Disease Outbreaks
rounded shapes, less-developed cerebellums, and
characteristics that indicate that these fishes are gen- Monthly variation in the relative frequency of disease
erally less active. Their white ordinary muscle has less cases reported is shown in Figure 6, for four fish spe-
myoglobin and hemoglobin, and their superficial cies: over 900 cases for each species were reported
dark muscle is not as developed. during the 9-year period 1981-89.
Because Pasteurella piscicida, a causative bacterium for The highest evidence of disease occurred during
pseuclotuberculosis, migrates in the bloodstream of in- July for yellowtail (30%) and tiger puffer (17%), dur-
fected fish, active fish, with well-developed vascular ing August for Japanese flounder (15.5%), and
systems, may be more likely to reach an exhausted state. during September for red sea bream (13%). Reports
Infections by gliding bacteria tended to break out of disease for the latter three species peaked in the
in benthic fish or fish living near the beach. Because summer fforn June through September when the wa-
these fish are less active than the pelagic species and ter temperature is higher than 20' C. These peaks
may have a susceptible body surface, gliding bacteria, were not as steep as that of yellowtail, yet the summer
which move by means of surface mucus, may have an peaks suggest that fish under cultivation are exposed
advantage in expanding their living area. Also, be- to physiological and environmental conditions favor-
cause these fish are more likely to come into contact able to infection by pathogens, in spite of possible
with other individuals via their body surface, new in- efforts to promote the defensive activities of fish,
fections and propagation of gliding bacteria are such as immunological resistance.
prevalent. In yellowtail, viral and bacterial diseases, especially
It is believed that the accumulation and integra- pseudotuberculosis, show significantly sharp seasonal
tion of information on fish species and pathogens changes (Fig. 7). Since the official identification
(such as cell or tissue susceptibility to pathogens; ana- (Sorimachi and Hara 1985) of the viral disease viral
tomical, histological, and physiological aspects of ascites (VA), the rate of this viral disease has in-
fish; environmental conditions; and characteristics of creased. VA disease breaks out in late May, reaches a

76 NOAA Technical Report NMFS 111

30-
12-
25- -VIRAL 11-
NUTRITIONAL lo-
20 . . ...... 9- %%
Z
8-
15- 0 7-
W
10 6-
w 5-
5 4 ............
3- ..... ...............%
01 2-
L) I I I I I I I
Z 17
1.2-
0
.I- VIRAL JAN FES MAP APR MAY JUN JLY AUG SEP OCT NOV DEC
0.9. ...... PARASITIC MONTH
LU 0.8-
> -NUTRITIONAL -VIRAL PARASITIC - - - NUTRITIONAL -UNIDENTIFIED BACTERIAL
0.7-
UNIDENTIFIED
0.6-
UJ 0.5- ........
0.4-
Figure 9
0.3- ........... Seasonal changes in diseases ofjapanese flounder.
0.2-
0.1- . ....... .

JAN FEB MAR APR MAY JUN JLY AUG SEP OCT NOV DEC
MONTH

Figure 7 October. Parasitic, nutritional,. and unidentified dis-
Seasonal changes in yellowtail diseases. eases showed gentle seasonal changes in their
occurrence. Parasitic and unidentified diseases broke
out during summer, and the level of nutritional dis-
eases peaked in autumn.
6- In red sea bream, viral diseases represented only
3.6% of the total disease cases reported (Fig. 8), and
-Z@ 5- for the most part were observed as lymphocystis dis-
ease every month of the year. Parasitic diseases were
.1 4-
reported least in February and March and most fre-
3- quently between May and December. Nutritional
> . ........
disease outbreaks appeared between May and Sep-
2-
tember, indicating that they are chronic in nature.
1 Half of the bacterial disease cases reported were
0 vibriosis (53.2%), which occurs all year round with a
JAN FEB MAR AM MAY JUN JLY d* 41' 0& NOrV OTC peak at winter (see Fig. 14). Gliding bacterial disease
MONTH also occurred all year round but the higher rates
VIRAL ..... PARASITIC --- NUTRITIONAL - UNIDENTIFIED -BACTERIAL were found in January and from May through July
(see Fig. 15). Edwardsiellosis accounted for about
Figure 8 10% of all disease cases in red sea bream and was
Seasonal changes in diseases of red sea bream. prevalent between July and November (see Fig. 13).
Since the successful development of seedling pro-
duction methods, production of the Japanese
flounder has gradually increased. The main viral dis-
peak in June, and ends in mid-July. The rate of VA ease observed in cultural flounder was epidermal
disease diagnoses was very low and account for on ly necrosis, which occurred in spring during the fry
1% of yellowtail disease reported in June; its occur- stage and accounted for only 0.,8% of all disease cases
rence fluctuates yearly. The rate at which the major (Fig. 9). Japanese flounder suffered from parasitism
bacterial diseases were reported showed similar sea- by protozoa such as Cryptocaryon spp., Trichodina spp.,
sonal variations. Vibriosis accounted for about 10% and other Ciliata. Because there are no effective
of all the bacterial diseases reported and the number treatments for these parasites, infected fish are left
V' A

A
L
@
S'TIC
AR '110-Al
NIT @nM
IN

...........
. ..........

of cases peaked in June. Cases of pseudotuberculosis untreated, resulting in a great deal of infection and
peaked sharply in July (see.Fig. 11), and strepto- death. A trial is now being conducted to see if condi-
coccosis epidemics seem to occur between July and tions leading to infection can be effectively removed

Ishioka: Epidemiology of Marine Fish Diseases 77

by increasing the water supply to rearing areas. It is
unknown why nutritional diseases are rare in Japa- 25 -
nese flounder. Bacterial diseases are serious for this 20 -
fish accounting for about 67% of all disease cases
reported. The most virulent, edwardsiellosis, 15-
amounts to about 28% of all disease cases. >_ 10-
0
Z
Edwardsiellosis is most prevalent during the summer, W
from June through October, with the number of re- 5-
ports peaking in August. Streptococcosis is most
UJ 0.1 -----------------
prevalent from July to October and vibriosis and glid- ?!
0.5-
ing bacterial disease are both most prevalent between
April and August. Unidentified disease accounts for cc 0.4-
16% of all disease cases reported.
0.3-

0.2

6-
0.1 A
5-
.. ......... DI V N ................
Z
w: 4- JAN FEB MAR APR MAY JUN JLY AUGG SEP OCT NOV DEC
MONTH
3- ................ YELLOWTAIL RED SEA BREAM

..... JAPANESE TIGER PUFFER
2- FLOUNDER

Figure I I
JAN FEB MAR APR MAY JUN JLY AUG ShP OL N6V DEIC Seasonal changes in pseudotuberculosis for each species.

MONTH
VIRAL .... PARASITIC--- NUTRITIONAL - UNIDENTIFIED

Figure 10 10
9-
Seasonal changes in diseases of tiger puffer. K
8-

Z
7-
W
36-
In tiger puffer (Fig. 10), kuchi-
jiro sho was regis ff 5-
tered often throughout the year. Parasitic disease 4-
amounted to about 39% of all disease cases (Fig. 10). 3-
2-
Fatal parasites included a Dichodina spp., a protozoa
which was prevalent between April and October, and 0-
Heterobothrium tetrodonis, a monogenea prevalent JAN FEB MAR A6 MAY JUN JLY AUG SEP OCT NOV DEC
mainly from April through August. Nutritional dis- MONTH
ease was also prevalent from spring through autumn. -YELLOWTAIL ----- RED SEA BREAM --JAPANESE -TIGER PUFFER
FLOUNDER
Higher levels of bacterial diseases were diagnosed
from June to September: most were cases of vibriosis
and gliding bacterial disease. Figure 12
Pasteurella piscicida was a virulent pathogen for Seasonal changes in streptococco-sis for each species.
pseudotuberculosis in yellowtail and to a less extent
in other cultured fish species (Fig. 11). This disease
broke out seasonally, appearing during warmer
months in all fish species. June through October; the level. of reported cases
Streptococcus sp. was most common in yellowtail and peaked during August in Japanese flounder and
Japanese flounder and occurred less frequently in showed a plateau from July to November in red sea
red sea bream and tiger puffer, showing similar sea- brearn (Fig. 13). The optimal water temperature
sonal variation patterns that peaked in September ranges from 15 to 24' C for Japanese flounder and is
(Fig. 12). Edwardsiella tarda was virulent for Japanese slightly broader from 13 to 28' C for red sea brearn.
flounder and red sea brearn and most prevalent from Cultured Japanese flounder suffer from exposure to

78 NOAA Tectinical Report NMFS I I I

high temperatures above 25' C in summer and become
7- more susceptible to bacterial infection. On the other
6- hand, red sea brearn grow wdll during summer and suf-
fer from nutritional disorders in autumn. Because
5-
EdwardsieHa tarda can multiply more actively during the
0 4-
summer season, Japanese flounder, which are less toler-
3- ant of higher temperatures, are more susceptible to this
pathogen. Conversely, red sea bream, which are more
2-
W ----------
----------
tolerant of higher temperatures, are less susceptible to
this pathogen thanjapanese flounder.
0. .............. . Seasonal variations in the relative frequency of
JAN FEB MAR APR MAY JUN III AUG SEP OCT NOV DEC
MONTH vibriosis are complicated and differ according to fish
---- @REO SEA BREAM JAPANESE FLOUNDER species (Fig. 14). This complexity may result not only
from the difference in susceptibility of each fish spe-
cies but also from the presence of different varieties of
Figure 13 Vibfio. Vibrio anguillarum is known to have varied sero-
Seasonal changes in edwardsiello-sis for each species. types. Susceptibility of fish to this pathogen would vary
with the serotype and fish species involved.
. The term "gliding bacteria" implies several bacte-
rial species. The gliding bacteria resulting in marine
5 fish diseases are mostly Flexibacter, Cytophaga, and
Sporocytophaga spp. The unique characteristics of each
4- gliding bacterium may contribute to the complicated
0
Z patterns of seasonal variations in reported infection
3-
among the different fish species (Fig. 15).

w 2-

Fish size and disease
............V

0
JAN FEB MAR APFI MAY JUN JLY AUG SEP OCT NOV DEC The variation in relative frequency of disease cases
MONTH reported in yellowtail is illustrated according to fish
-YELLOWTAIL ----- RED SEA BREAM JAPANESE -TIGER PUFFER size (body weight) in Figure 16. For this analysis, rela-
FLOUNDER
tive frequency was calculated by dividing the total
number of reports for each disease in the defined
Figure 14 size class by the total number of disease cases with
Seasonal changes in vibriosis for each species. body weight (bw) data for each species. Viral disease
was more frequent in younger fish (<50g bw), while
nutritional diseases were more prevalent in older fish
(>250g bw) and season also an influencing factor. Oc-
6- currence of bacterial disease appears closely related
5- to body weight. Yellowtail seemed to be successively
attacked by VA virus first, then Vibrio and Pasteurella
4-
piscicida, and finally Streptococcus. This sequence of in-
a fection suggests that susceptibility of yellowtail to
3-
different pathogens varies with body weight or
2
growth. This may imply not only the development of
defense mechanisms to the pathogens but may also
01 ---------- -- represent the changes in the overall physiology of the
JAN FEB MAR APR MAY JUN JLY A6G SEP OCT NOV DE'C fish under cultivation.
MONTH In red sea bream, parasitic diseases occurred more
-YELLOWTAIL ---- RED SEA BREAM JAPANESE TIGER PUFFER frequently in fish between 50 and 250 g bw while
FLOUNDER reports of nutritional disease increased in fish over
250 g bw (Fig. 17). The rate of gliding bacteria diag-
Figure 15 noses was highest in young fish below 10 g bw and
Seasonal changes in gliding bacterial diseases for each that of vibriosis was highest in fish between 50 and
species. 100 g bw. The diagnostic rate of edwardsiellosis was

Ishioka: Epidemiology of Marine Fish Diseases 79

2.4- VIRAL
2.2-
2- PARASITIC

NUTRITIONAL
IA-
1.4- UNIDENTIFIED

W
0.2
Uj
cr
0 AGE
LLJ 20-
VIBRIOSIS
F-
PSEUDOTUBERCULOSIS
W Is-
14- STREPTOCOCCOSIS

12 -

=@Ill =@Ioo =<511 =@Illl =<2110 =<4110 =<1000 =<1010
=<51 =<250 =<750 =<1511 =<3111 =<5110 =<7000 Figure I .6
BODY W .EIGHT (9) Relationship between fish size and disease
I outbreaks in yellowtail.

13-
12-
11- 13 VIRAL

10 PARASITIC

0 NUTRITIONAL
6- 0 UNIDENTIFIED
5-
4-

Z
w 2-

W
LL
LU AGE
>

LU
X 5- 0 VIBRIOSIS
0 PSEUDOTUSERCULOSIS
4-
0 STREPTOCOCCOSIS
0 EDWARDSIELLOSIS

GLIDING BACTERIA

[email protected]
=<580 loll =@2000
=-751 =<1590 =@31109
BODY WEIGHT (9) Figure 17
Relationship between fish size and disease
outbreaks in red sea bream.

80 NOAA Tectinical Report NNM I I I

VIRAL

I- PARASITIC

11-
NUTRITIONAL

4-
UNIDENTIFIED

0
LU 0 AGE
LL 21@
El VIBRIOSIS
Uj

< 14- PSEUDOTUBERCULOSIS

LU 12-
STREPTOCOCCOSIS
Is-

11-
0 EDWARDSIELLOSIS

4 GLIDING BACTERIAL

=@111 -251 511 =@110 =@1111 =@1610 =<2110
BODY WEIGHT(g) Figure 18
Relationship between fish size and disease
I outbreaks in Japanese flounder.

4 0-

VIRAL

PARASITIC

NUTRITIONAL

UNIDENTIFIED
L) 16-
Z

U. 0-
0

W 4-
a:
El VIBRIOSIS

3- PSEUDOTU13ERCULOSIS

STREPTOCOCCOSIS
2-
E3 GLIDING BACTERIA

Figure 19
BODY WEIGHT (9) Relationship between fish size and disease
outbreaks in tiger puffer.

Ishioka: Epidemiology of Marine Fish Diseases 81

higher in larger fish, especially those over 250 g bw. streptococcosis were synchronized with nutritional
All ages of Japanese flounder experienced out- disorders.
breaks of parasitic diseases, while viral diseases were
found only in fish below 10 g bw (usually herpes vi-
rus) (Fig. 18). Vibriosis and gliding bacterial disease Locality and Disease
outbreaks were more common in younger fish
(<50g). Although streptococcosis and edwardsiellosis As shown in Figure 20, for yellowtail in 1986, some
were both more prevalent in larger fish (>250 g), characteristics are common to all the prefectures
edwardsiellosis was more common than streptococcosis listed. Although the relative frequency of bacterial
in the fish of 50-100 g bw. disease was about 80% in all prefectures, the compo-
In tiger puffer, a unique feature was found: fish sitions of different bacterial diseases were to some
250 to 500 g bw appear to be most susceptible to extent different by locality. Mie Prefecture is the
pathogens (Fig. 19). This may be dependent on the most northern of all the prefectures listed in Figure
method used to culture this species. Though this 20, while Kagoshima Prefecture @is located in most
species is in great demand by the market, its cultiva- southern part of Japan. Interprefectural analysis sug-
tion from the egg through the spawning stage is gests that the composition of different bacterial
very difficult. The eggs for seedling production are diseases seems not to be related to geographical lo-
generally raised by the public sector and are in most cality. A possible similarity exists between the
cases supplied from the natural resource by composition of bacterial infections observed and the
fisheries, which also provides some large sized seed- type of cultivation management used in some prefec-
lings. Because the aquaculturist cultures mainly tures. There are two types of management:
large-size seedlings, diagnoses may concentrate on cultivation of large seedlings and production of fish
the larger sized fish. of marketable size. Streptococcosis tended to be re-
In all species, nutritional disease tended to occur ported at a higher rate in the prefectures managing
in larger fish, probably as a result of prolonged nutri- the larger fish (2-year-olds) of marketable size.
tional treatment. Outbreaks of edwardsiellosis and No characteristic difference in @seasonal outbreak

783 129 288 100 111 164 327 4 9 139 154 45 72 19 23 (N)
too- S-d
0 C
)0 UNIDENTIFIED
0 C
N:
)0
...... OTHERS
XX

...:.:.X*. NUTRITIONAL
. ........
-X -:::j
.... .......
....... ...
........ ......
VIRAL
F0
nn]

so
..N PARASITIC
Z
OTHER
BACTERIAL DISEASE
... .......

W
LL NOCARDIOSIS
40
Uj
.. ... . . . ......
PSEUDOTUBERCULOSIS

.... VIBRIOSIS
LU
20- COMB NED DISEASES
WITH STREPTOCOC-
costs
STREPTOCOCCOSIS

LU 0 0 W < < <
0 < < U
T < 7 <
LU 0 ro 0 A0 &0
@4 < < < <
0 D < <
<
Z 0 <

PREFECTURE
X. -X

Figure 20
Composition of diseases of yellowtail by prefecture during 1986,

82 NOAA Technical Report NNn I I I

501 M I E 40' KOCHI
40- N=327 30- N=288
30- 20-

20- 10

10
SAGA
0
20 WAKAYAMA 20. N= 45
1.0 to-
N= 19
0 1 . .
30- HYOGO 59 N AGASAK I
20- 40
N=129 N=139
10- 0
Z
W In 20
D 0
a 20,
W YAMAGUCHI 1
CC 10,
LL N=23 01. . .
W 0 01@113 30 KUMAMOTO
> 40- TOKUSIMA - 20
30- N=81
N= 72 1
W
cc 20- r,3
10, 39 OITA
2
0. . . . .
40 -KAGAWA 10 N=154
30- N=100
20- 48, MIYAZAKI
10. 30- N= 49
0' 0 20-
40 EHIME 10.
30 N= 783 4
20 20 KAGOSIMA
10 10
0 ri 0 N=164
> 0 Z co cr cc -11 CL 6 @i io d: C*C > a 0. Figure 21
0 W < W a. D :) W 0 0 W < W 5 M :3 LU
Z 0 n LL M 1) -3< 9) 0 Z a , LL < rn 0 Patterns of the outbreak of
pseudo tuberculosis in yellow-tail
in each prefecture in 1986.

of virulent disease was found according to geographi- gests that all fish culture fields along the warm cur-
cal location. In the example of 1986 shown in Figure rent are in an area favorable to the multiplication
21, pseudotuberculosis broke out earlier in warmer and propagation of pathogens.
areas. This disease was reported to occur at a higher
rate in June and July in Kagoshima, Miyazaki, and
Kochi Prefectures, while the rates were higher in July Discussions
and August in other prefectures. Because the optimal
range (20 to 25' Q for multiplication of Pasteurella From this epidemiological examination, it is clear
piscicida is strict, the location of outbreaks of this dis- that the natural and delicate relationship between
ease could be determined by calendar and locality or fish and their environment is disturbed on a local
0

0
0
"1 0
01"
LN

latitude. basis by fish cultivation in coastal areas and induces
As a whole, no characteristic local feature was ob- the propagation of pathogens and the outbreak of
served that could explain particular disease disease in the cultured animals. As shown in Figure
outbreaks in the area surrounded by warm oceanic 22, the means and methods of fish culture affect not
current, except perhaps slight differences. This sug- only fish health but also water quality and pathogen

Ishioka: Epidemiology of Marine Fish Diseases 83

tal factor. Higher temperature is beneficial in activat-
ing various physiological functions, including
0 LI T BREAK immunological processes in healthy fish. But this fac-
0 F tor can also be a harmful to unhealthy fish under
D ISE A SES poor environmental conditions.
The results of this study help us to direct future
study and research in this field. Unidentified diseases
PATHOGEN INFECT ION F I S H must be clarified as soon as possible, especially in
new fish species introduced for cultivation. In this
type of research, it is most important to have a thor-
ough knowledge of the anatomy, physiology, ecology,
ENVIRONMENTAL pathology, and immunology of the fish and patho-
CONDITION gens. This may lead to the development of improved
cultivation techniques.
Effective methods to diagnose fish health should
CULTIVATION TECHNIQUE
be developed, not only to prevent mass mortality by
pathogens that are generally nonvirulent, but also to
Figure 22 regulate cultivation techniques. A large amount of
Disease outbreak flowchart for marine fish cultured in the research is needed in this area, for example: physi-
Kuroshio Current. ological studies on poikilothermal fish, pathological
studies that clarify the mode of action of pathogen
distribution; major factors include fish density in a within their hosts during infection, and immunologi-
pen, location of the pen, pen density, handling, ex- cal .studies to explain fish defense mechanisms
cess food supply, food quality, sanitary treatment of against pathogens.
nets, type of food and other materials, and treatment Epidemiological research in a small area such as
of dead fish. Supplying unsuitable or unsanitary food one bay unit or a fisheries cooperative association
induces the degeneration of fat in fish, whereas sup- unit should be carried out in detail to clarify the di-
rect relationships among disease outbreaks,
plying excessive food makes them generate fat. environmental factors, and cultivation techniques.
Unsuitable handling techniques injure the body sur- Such research could provide knowledge for a more
face, especially in younger fish, and allows pathogens successful cultivation of healthy fish.
to invade. Fish are also susceptible to pathogens un-
der some nutritional conditions and during certain
developmental stages. Culturing fish under high den- Citations
sity adds additional stress.
A high density of pens interrupts the water current Sorimachi, M., and T. Hara.
and causes water quality to deteriorate. As excess 1985. Characteristics and pathogenecity of a virus isolated
food and fish excrement are deposited on the bot- from yellowtail fingerlings showing ascites. Fish Pathol.
tom and dissolve into seawater, the quality of water in 19:231-238. (Injapanese.)
the cultivation area degrades. These changes in envi- SID (Statistics and Information Department)..
1981-90. Annual statistical report on fisheries and culture
ronmental conditions are beneficial for pathogen production of Japan. SID, Jpn. Min. Agri. Forest. Fish.,
multiplication, which, in turn, lead to infection of Tokyo. (InJapanese)
the fish and to outbreak of disease. Umezawa, S., and H. Ishioka.
As previously described, the seasonal changes of 1988. Program for compiling disease cases diagnosed in pre-
fectures. In Personal computer programs for fish popula-
disease cases are very remarkable, and it is thought tion dynamics, vol. 1, p. 314-339. Tokai Regional Fisheries
that temperature is the most influential environmen- Research Laboratory, Tokyo. (Injapanese.)

Characterization of Hematic Neoplasia in the
Softshell Clain Mya arenaria

PAUL W. RENO*, ANDREA ILLINGWORTH** ' and MICHAEL DORITY***

*Coastal Oregon Marine Experiment Station/Microbiology
Mark 0. Ha0eld Marine Science Center
2030 S. Marine Science Dr
Newport, Oregon 97365-5296

"Institutefor Cellular Research
345 State St.
Bangor, ME 04401

***Department of Microbiology
181 Hitchner Hall
University of Maine
Orono, ME 04469LO131

ABSTRACT

A leukemia-like disease, variously termed hemic, hematopoietic, or hematic neoplasia,
(HN) has been detected in bivalves from diverse geographic locales. The wide geo-
graphic range of the disease, coupled with mortalities due putatively to HN, makes it
important to determine the nature of the abnormal cells characteristic of the disease.
The studies reported here determined certain biochemical and genomic characteristics
of HN in the softshell clam M@a arenaria. No significant difference in the total amount of
lipid of normal and abnormal hemocytes was noted. Fatty acids of the omega 3 and
omega 6 families were elevated in clams with the highest HN intensities. Similarly, in
vitro incorporation of acetate into phospholipids and neutral lipids by normal lympho-
cytes was significantly different from HN hemocytes but unlike the pattern seen in
mammalian leukemias. Examination of the DNA content of normal and HN cells by flow
cytometry indicated a high degree of aneuploidy in the HN cells. This consisted of clams
with hyperdiploid DNA content of 1.6 to 2.0 times normal, and a hypodiploid population
-With DNA 0.85 times normal. Hyperdiploid cells had significantly higher DNA synthesis
than diploid cells. In this respect, HN cells are similar to true leukemias of mammals, but
the combination of hypodiploid and hyperdiploid cells in one individual is uncharacter-
istic of mammalian neoplasia. In sunimary, hemocytes of M. arenaria affected with HN
have some features coincident with mammalian leukemias but also several unique char-
acteristics.

Introduction diseases of cultured and feral bivalves in order to pre-
dict sustainable harvests of this significant marine
The harvest of bivalves from commercial and cul- resource accurately.
tured sources is of considerable economic im- Among the most widespread diseases of bivalves
portance on both worldwide and national scales. Dis- are the putative neoplastic diseases which have been
ease affects natural as well as artificially propagated detected in a wide variety of commercially important
populations. Unfortunately, our knowledge of dis- species and from a geographically disparate range
eases in bivalves is quite restricted in comparison to
the information base available for finfish. It is impor-
tant that we establish a more detailed database for *Send correspondence to this author.

86 NOAA Technical Report NMEFS I I I

(Couch 1969; Farley 1969; Brown et al. 1977). The Materials and Methods
primary tissue distribution of these diseases ap-
pears to be gonadal ("germinomas") (Yevich and
Barszcz 1976, 1977) and hernatic (variously termed Sample Collection, Evaluation of HN, and
hernatopoietic neoplasms, hernatic neoplasms, or Animal Maintenence
hernic neoplasms). Hernatic neoplasia (HN) has
been associated with elevated levels of mortality in Softshelled clams ranging in size from 45 to 120-mm
field and laboratory studies with the softshell clam valve length were collected from Long Cove,
Mya arenaria (Appeldoorn and Oprandy 1980; Coo- Searsport, Maine, during the summer, placed on ice,
per et al. 1982) and the edible oyster Ostrea edulis and returned to the laboratory for evaluation of HN
(Alderman et al. 1977). The prevalence rates for levels. When necessary, animals were held in the wet
HN in M. arena?ia range from 0.02 to 84% , and laboratory at. the University of Maine in a re-
intensities of HN in individuals span the range circulating artificial seawater system at 14' C until
from less than 0.01% to greater than 99% of hemo- needed. Animals were generally evaluated within I
cytes affected (Couch 1969; Mix 1986; Reno, week of collection. Evaluation of HN was performed
unpubl. observations). The pervasive nature of this by a modification of the method of Cooper (1982).
disease and its potential for killing feral and cul- Approximately 100 @LL of hemolymph was removed
tured bivalves makes it essential to understand the from the cardiac or anterior adductor sinus, 50 IiL
nature of the disease and its etiology. The results was placed in one well of a 96-well, flat-bottomed
reported here reflect an initial attempt to define microtitration plate, and hemocytes were allowed to
mo re clearly the nature of the disease in M. adhere for 20 to 30 minutes. Samples were then
arena,ria. evaluated for the levels of HN by microscopic exami-
Hernatic neoplasia is characterized by its prolif- nation under phase optics at 20OX 'with an inverted
erative nature and the resulting cytological mor- cell culture microscope. The number of nonadherant
phological abnormalities assessed by microscopic or loosely adherant, rounded cells in the hemolymph
examination of either unstained, or fixed and (HN cells) was compared to the number of fully ad-
stained cells (Couch 1969; Farley 1969; Cooper et herent spreading normal hemocytes by counting 500
al. 1982). These criteria, however, are not adequate to 1000 cells and converting the proportion to
for accurately defining HN in bivalves as neoplastic percentages.
because these invertebrates are evolutionarily dis-
tant from the mammals, the phylum on which Lipid Analysis
virtually all definitive work has been done on the
characteristics of neoplasia. This lack of definition is The major lipids that are altered in neoplasia are the
noted in two prominent reviews of presumed neo- fatty acids, the neutral lipids, and the phospholipids.
plasms of bivalves: "Some of the neoplasms may be Several methods were used to evaluate lipid changes
of questionable validity ... and little is known about in HN cells, including gravimetric total lipid content,
the exact nature of the cells involved but, in gen- fatty acid methyl ester (FAME) profiles, and in vitro
eral, they are considered neoplastic" (Mix 1986); incorporation of radiolabelled acetate into neutral
and "In spite of intensive study, the true nature of lipids and phospholipids.
these neoplastic cells is not yet quite clear." Gravimetric analysis was carried out on washed he-
(Lauckner 1983). There are scores of structural, mocytes from normal and abnormal clams using the
biochemical, immunological, and genomic alter- method of Sasaki and Capuzzo (1984). This involved
ations in mammalian cells associated with the a modified Folch procedure employing several ex-
transformed or neoplastic state (reviewed in Wood traction cycles with chloroform: methanol. Total lipid
1972; Marchesi 1976; Vasiliev and Gelfand 1981; by weight was obtained using an ultramicrobalance
Heim and Mitelman 1987; Iversen 1988). Among (Cahn model 25 automatic electrobalance). Fatty ac-
the most well characterized biochemical alterations ids were esterified to fatty acid methyl esters (FAMEs)
that occur in transformed cells are those occurring to stabilize them prior to analysis, which was carried
in two fundamental classes of molecules: the nucleic out on hemocytes from clams with various levels of
acids and the lipids. The present work was designed HN and clams with no evidence of HN. Fatty acid
to monitor potential differences between normal separation was carried out using the modified Bligh
hemocytes'and HN cells from M. arenaria compared and Dyer (1959) method of Jerkofsky and DeSiervo
by flow cytometric analysis and in vivo and in vitro (1986). Washed, pelleted hemocytes were extracted
lipid ccinstitution. with chloroform:methanol followed by a saline phase

Reno et al.: Hematic Neophisia in the Softshell Clam Mya amuwia 87

separation, and final suspension in chloroform. A Cells were washed in buffered saline with 0.1% bo-
known quantity of a 19-carbon fatty acid was added to vine serum albumin, counted, and suspended in a
the cell pellet prior to extraction to serve as an inter- solution of I jig/mL RNAse and 0.01% NP-40 deter-
nal standard for the extraction process. Samples were gent to increase membrane permeability. This was
transesterified with a methanolic base reagent fol- followed by the addition of propidium iodide
lowed by methylene chloride and final suspension in fluorochrome at the ratio of logg/106 cells in a solu-
hexane. Esterified samples were then applied to a gas tion of NP-40 and 0.1% polyethylene glycol 8000 at
chromatograph (Hewlett Packard model HP5890A) pH 5.0. Examination of the cells under a ultraviolet
equipped with a 30 m DB carbowax capillary column. light microscope confirmed that nuclear morphology
Peak areas were calculated with a disc integrator, and was normal and that no artifacts were present. The
retention times were compared with known standard cells were examined in a flow cytometer (Becton-
FAMEs. Dickinson, EPICS model CS) and the relative
In vitro assimilation of radiolabelled acetate into fluorescence (DNA content) and cell cycle analysis
neutral and polar lipids was carried out using hemo- was performed using a Modfit analysis program (Ver-
cytes that were removed from the clams within, 8 ity So ftware Corp.), which integrated the area under
hours of capture. To each of four replicate tubes con- the various peaks and estimated the percentage of
taining the hemocytes, an equal volume of 0.45 total area for each peak, as well as their coefficients
@Lrn-filtered sterile seawater containing 10 RCi of 14C_ of variation.
acetate was added. The cell-acetate mixture was
incubated at 15' C for 24 hours and extracted as de-
scribed for FAMEs, except that, in order to improve Results
the precipitation of labelled lipid, non-labelled car-
rier lipid was added. The carrier lipid was prepared
from two shucked clams by removing the digestive Lipid Analysis
diverticula and skin of the siphons, blending them in
a blender, and extracting as previously described. Comparison of hemocytes from HN-affected and nor-
The lipid extract was weighed and resuspended in mal clams by gravimetric analysis indicated that the 're
chloroform to a concentration of 300 mg/mL; 30 mg was no significant difference between the total ex-
of carrier lipid was added to each sample. A final tractable lipid in the cells (normal cells = 15.97�8.51
suspension in solvent was divided into 2 aliquots pg lipid/cell; HN cells = 23.21�11.88 pg lipid/cell,
which were then evaluated for either neutral lipids by P = 0.223). In order to more fully define any differ-
one-dimensional silicic acid paper chromatography ences that might exist between the two cell types,
(Wuthier 1966) or for polar lipids by two-dimen- both the basic fatty acid components, as well as the
sional silicic acid paper chromatography, followed by complex lipid moieties derived from them were ana-
autoradiography Uerkofsky and DeSiervo 1986). lyzed.
Gas chromatographic analysis of FAMEs from he-
mocytes taken from HN-affected clams was compared
Flow Cytometric Analysis with that of normal clams as well as with the pattern
found in normal and SV40 virus-transformed WI-38
In order to determine the configuration of the DNA cells (human diploid fibroblasts) (Table 1). When
in the genome of HN cells versus normal hemocytes, compared to normal clams, only clams with rates of
flow cytometric analysis of the total DNA content of HN in excess of 90% abnormal cells exhibited signifi-
the cells as well as the cell cycle status of the cells was cant alterations in FAME composition. The saturated
carried out. Preliminary experiments were per- class of FAMEs remained stable in affected clams,
formed to determine whether live or fixed hemocytes with the exception of a decrease in palmitic acid and
were optimal for flow cytometric analysis. Fixation of in stearic acid. Abnormal cells had significantly lower
HN hemocytes in either 2% formalin-seawater or 2% levels of monounsaturated fatty acids, as well as sig-
glutaraldehyde-seawater led to reduced resolution of nificantly higher levels of the longer chain fatty acids
DNA peaks compared to unfixed materials sus- (20 and 22 carbon polyunsaturated groups). The
pended in the anticoagulant 0.01 M cysteine in changes found in the FAME profiles of HN cells were
seawater. Consequently, fresh, unfixed, washed hemo- opposite in direction from those found in SV40-trans-
cytes were used for the flow cytometric analysis in all formed human cells.
further studies. The processing of hemocytes for flow Evaluation of the in vitro incorporation of lipid
cytometric analysis of DNA followed standard meth- precursors (acetate) into the more complex lipids
ods used in human clinical oncology (Raber 1988). also indicated significant changes in hemocytes from

88 NOAA Technical Report NKB 111

Table I
A summary comparison of lipid composition of normal and SV40@virus transformed mammalian cells and normal
and HN cells from Mya arenaria. Levels of HN ranged from 5 to >95%.

SV-40
infected Normal HN
Lipid class WI-38a WI-38a hemocytes hemocytes

Total lipid/cell (pg) 65.0 59.0 16.0 23.2 (P--0.223)
Neutral lipids (% of total cpm) 29.0 41.0* 5.0 8.0 (P<0.05)
Triacylglycerides (% of neutral) 27.0 25.0* 40.0 62.0 (P<0.005)
Free fatty acids (% of neutral) 21.0 26.0* 36.0 22.0 (P<0.01)
Cholesterol (% of neutral) 35.0 35.0 6.2 4.2 (P<0.05)
Cholesterol esters (% of neutral) 5.0 4.0 18.0 8.0 (P<0.05)
Phospholipids (% of total cpm) 71.0 59.0* 95.0 92.0 (P<0.001)
Cardiolipin (% of phospholipids) 6.1 2.7* 6.0 2.0 (P<0.001)
Phosphotidylethanolamine 12.0 13.0 75.0 71.0 (P<0.001)
(% of phospholipids)
Phosphotidylinositol 13.0 10.0* 1.0 2.0 (P<0.001)
(% of phospholipids)
Phosphotidylcholine 57.0 57.0 2.0 8.0 (P<0.005)
(% of phospholipids)
Fatty acid methyl esters
Palmitic acid (16:0) (% of FAME) 44.0 50.0* 70.0 59.0 (P<0.01)
Stearic acid (18:0 ) (% of FAME) 37.0 30.0* 20.0 25.0 (P<0.05)
18:1 Fatty acids (% of FAME) 63.0 78.0* 25.6 22.5 (P<0.05)
20:4 Fatty acids (% of FAME) 37.0 17.5* 1.2 4.0 (P<0.01)

a Data from Howard et al. 1977.
*Significant difference between WI-38 and SV40-transformed WI-38.

diseased clams relative to normal ones. The percent- and the phosphotidylcholines(P<0.001 and P<0.05 re-
age of labelled acetate incorporated into the total spectively). Many of the changes occurring in the
neutral lipid fraction of the HN cells was significantly lipids of HN hemocytes differ with those documented
higher, especially the triacylglycerols (P<0.005), the in mammalian neoplastic cells (Howard et al. 1973).
primary storage form of lipids (Table 1). In fact,
while the level of the other classes of neutral lipids
evaluated, including free fatty acids, cholesterol, and Flow Cytometric Analysis
cholesterol esters, decreased significantly, the eleva-'
tion in triacylglycerides masked the depressed Preliminary experiments indicated that using forma-
synthesis of these compounds in the total neutral lin or glutaraldehyde fixatives for flow cytometric
lipid fraction. Oil red 0 staining, which preferen- analysis of hemocytes altered the chromosomal mate-
tially stains triacylglycerols (Luna 1986), specifically rial andjesulted in poor separation of the normal
stained the vacuoles or "droplets" seen in HN cells by diploid and aneuploid peaks in HN-positive clams.
phase contrast microscopy; these droplets are not ap- Therefore, for all of the experiments reported here,
parent in normal hemocytes. unfixed hemocytes were rapidly (within 5 minutes of
As with the neutral lipids, there were also signifi- removal from the clam) processed for flow
cant changes in the incorporation of acetate into the cytometric analysis. This resulted in a clear separa-
polar lipids, which consist almost entirely of the fion of DNA peaks in the cells.
membrane-associated phospholipids. The overall Flow cytometric analysis indicated that the DNA
level of incorporation into the total phospholipid content of the abnormal hemocytes was distinctly
class decreased significantly (P<0.001). Likewise, de- aneupoid. As shown in Figure 1, there was a strong
creases occurred in the amount of labelled positive linear correlation (r = 0.95) between the ab-
phosphatidylglycerol and cardiolipin, which is de- normal microscopic appearance of the nonadherent
rived from phosphatidylglycerol (P<0.001) (Table 1). cells and abnormal DNA content as detected by flow
By contrast, there were significant increases in the cytometry. Two forms of aneuploid cells were de-
levels of incorporation into the phosphoddylinositols tected: those with a hypodiploid DNA content and

Reno et al.: Heinatic Neoplasia in the Softshell Clam Mya arenatia 89

those with a hyperdiploid DNA content. Examples of higher (mean = 16.4%, P = 0.0001) proportion of
the DNA contents of normal and abnormal hemo- cells which were either undergoing DNA synthesis, in
cytes are shown in Figure 2. The amount of DNA in preparation for mitosis, or actually in the process of
diploid clam hemocytes was approximately 3.2 pg, mitosis (Fig. 3). The proportion of FIN cells in some
compared with 6.8 pg in a human lymphocyte stan- phase of replicative activity ranged from 10 to 25%;
dard. In contrast to diploid clam hemocytes, DNA most of this activity was in the hyperdiploid popula-
profiles of cells from abnormal clams invariably had a tion, but the hypodiploid and even the diploid
population of cells containing hyperdiploid DNA populations also were undergoing more extensive
comparable in number to the percentage of abnor- multiplication than normal.
mal cells as assessed by microscopic observation. The
DNA index (DI)-the quantity of aneuploid DNA
relative to the DNA content of the normal diploid Discussion
hemocyte-of clams showing hyperdiploidy varied
from 1.6 to 2.0 in 12 HN-positive individuals. In most, The present study was designed to help characterize
but not all clams that were HN-positive, a hypodip- the abnormal hemocytes in M. arenayia with hematic
loid cell population also occurred which comprised neoplasia. This disease, and similar diseases in a vari-
between 0.5 and 10% of the cells (Fig. 2). The DI of ety of other bivalves, have attracted much attention
the hypodiploid cell population, ranging from 0.82 in the scientific community for two reasons. In the
to 0.86, was more constant than the DI of the marine community, they can cause significant mor-
hyperdiploid population. This hypodiploid peak was talities in the affected animals (Cooper et al. 1982;
not noted in the hemocytes of normal clams. Brousseau and Baglivo 1991; Reno, Leavitt, and
One of the hallmarks of neoplastic transformation Capuzzo, unpubl. observations). In addition, these
is the proliferative nature of the cells. Cell cycle diseases have become the focus of epidemiological
analysis of hemocytes from HN-positive clams indi- investigations assessing the- potential causal interac-
cated that this was indeed the case for this particular tion between anthrop ogenic pollutants in marine
malady. Normal cells showed some indication of pro- environments and the generation of neoplasia, and
liferative activity; 2 to 5% (mean = 4.4%) were in the the potential for using FIN Idisease in bivalves as
S (DNA synthesis) phase and an approximately equal sentinals in appraising the potential carcinogenic ac-
percentage were in the G2/M (premitotic and mi- tivity of these pollutants in humans. While the
totic) phases of the cell cycle. Hemocytes from clams rationale of the first reason still holds'true irrespec-
that were aneuploid, however, had a significantly tive of the true nature of the disease, the assumption

y 1.1029 + 0.91207x RA2 0.907

1W_

so-

60,

40,

0
20 40 60 so 100 Figure I
% HN (microscopy) Relationship between the percentage of HN
by microscopic evaluation and the percent-
age of DNA aneuploidy by flow cytometry.

90 NOAA Technical Report NMFS I 11

1000-
A a

800- HN-

600-

0
400-

200.

0
20 40 60 80 100 120 140 160 180
IRFL (DNA) Figure 2
Examples of DNA distrib ution
1000- and cell-cycle analysis of normal
B a and abnormal hemocytes. (A)
Normal clam; (B) clam with ap-
800- 78% HN+ proximately 78% abnormal
cells. IRFL = integrated red
fluorescence (DNA content).
60o- Key to letters on graphs:
a: resting stage of cell cycle (G,
and G, phases) for diploid
400- cells
b: DNA synthesis phase of cell
a cycle (S phase)
200- c: Premitotic and mitotic phases
of cell. cycle (G, and M
A phases) for diploid cells.
0 - d: Hypodiploid DNA peak
20 40 60 80 100 120 140 160 180 e: Hyperdiploid DNA peak
IRFL (DNA) f: S phase for hyperdiploid cells
g: G,/M phases for hyper-
diploid cells
on which the second is based is subject to invalida- 1979; Lauckner 1983), prudently note that the crite-
tion if the disease is not a tr 'ue neoplasm, but is ria presently used to determine the nature of the
rather a hyperplastic or anaplastic disease-or even disease are inadequate when compared with the
due to an exogenous source such as a protozooan manifold criteria that are demonstrable in mammals,
parasite. and that it may be premature or unwarranted to de-
Information is readily available in the literature scribe the leukernia-like disease in bivalves as a true
with respect to the characteristics of neoplasia and neoplasm without further investigations into the na-
transformed cells in mammals and birds. Relatively ture of the disease.
little is present that deals stringently with the disease Mammalian and avian neoplastic cells derived from
and nature of the altered cell in poikilothermic verte- tumor tissue or transformed by biological, physical,
brates; virtually none is available with respect to the or chemical agents have several dozen characteristics
"diseased cell" in invertebrates. Mix's comprehensive in common that differ significantly from normal
review (1986) of the nature of the interaction be- cells. These characteristics are most readily evaluated
tween anthropogenic pollutants and "neoplastic" when the cells are separated from their normal resi-
L78*/* HN+

diseases in aquatic vertebrates and invertebrates rep- dence in vivo and are cultured in vitro, if only for
resents the most critical analysis of the extant data. short periods. This process removes the major con-
The author, as well as others (Harshbarger et al. founding factor of the large number and wide variety

Reno et al.: Heinatic Neoplasia in the Softshell Clam Mya arenaria 91

of normal cells that are present in tumor tissue in
E %S vivo. Affected cells of the circulatory system are most
M % G2-M readily divested of these contaminants by simply
0 % GO-GI drawing off hematic fluids which contain the abnor-
mal cells and placing them in culture for further
A 100- analysis. This is the approach taken in the current
investigations.
080- Lipids are the major constituents of cellular mem-
branes, as well as a source of stored energy, and the
Q60-
M specific lipid components that are present have a
.2 40- marked effect on the properties of the membrane
20- (Wood 1972; Shinitzky 1984). Because a wide variety
A of the alterations occurring in neoplastically trans-
formed cells are associated with the plasma
0-
0 62 88 92 94 95 membrane surface, it is not surprising that many al-
terations in the lipid composition of neoplastic cells
% HN have been documented in mammalian systems. Wood
B (1972) and others have published a variety of papers
100- dealing with the alterations in lipid metabolism and
biochemistry in mammalian neoplasia (e.g., Awad
So- and Spector 1976; Yau et al. 1976; Montaudon et al.
- 1981; Schroeder and Gardiner 1984; Kier et al. 1988;
60- Calorini et al. 1989). The SV40-induced transforma-
tion of human WI-38 lung fibroblast cells (Howard et

0
40- al. 1973; Perkins and Scott 1978) is an in vitro system
CL resembling HN of M. arenafia because transforma-
20- tion by viruses produces a homogeneous population
of altered cells, as is found with HN hemocytes, and
0 because it has been indicated that FIN in M. arenafia
0 62 88 92 94 95 is also viral-induced (Oprandy et al. 1981).
There,are significant alterations in more than one
% HN half of the lipid moieties examined in the SV40-WI-38
C cells in culture. With the essential building blocks of
100- lipids, the fatty acids, there are increases in the levels
of the saturated fatty acid palmitic acid (16:0), the
80-
monounsaturated fatty acid oleic acid (18:IA9), and
V
'3 60- the diunsaturated fatty acid linoleic acid (18:2M). In
CL contrast, in M. arenaria FIN cells show a decrease in
40- palmitic acid but a similar increase in the two unsat-
urated fatty acids. In the virally transformed WI-38
CL
20-
cells, a significant decrease occurs in the levels of
ge 0- stearic (18:0), palmitoleic (16:lA7), linolenic
0 62 88 92 94 95 (18:3M), and arachidonic (20:4A6) acids. In M.
arenaria HN, however, there is a marked increase in
% HN the levels of stearic acid in conjunction with de-
creases in the levels of 16:IA7, 18:3A3, and 20AA6
fatty acids. The changes seen in mammalian and bi-
Figure 3 valve cells are at variance in their levels of two major
Percent of diploid and aneuploid cells in various phases of essential fatty acids derived from de novo sources:
the cell cycle. (A) Aiploid cells; (B) hypodiploid cells; palmitic and stearic acids. These variations may re-
(C) hyperdiploid cells. % S = cells in DNA synthesis phase flect a difference in the process of synthesis of the
of cell cycle; G2/M = cells in premitotic and mitotic phases fatty acids in invertebrates compared with mammals
of the cell cycle; GO/G, = cells in the resting stages of the
cell cycle. or may reflect a basic difference in the nature of the
abnormal HN cell and the transformed WI-38 cell.
With respect to the more complex lipids, a significant

92 NOAA Technical Report NWS I I I

increase is seen in the levels of cholesterol in both The available data on the etiology of HN in
virally transformed WI-38 and in murine hepatoma bivalves is scant. There is evidence that the abnormal
cells (Steele and Jenkin 1977). Cholesterol is one of hemocytes of HN clams possess unique surface anti-
the most important lipids in the cellular membrane gens as detected with both polyclonal and
and contributes much to the attributes of the cell monoclonal antibodies (Reinish et al. 1983; Miosky et
surface. In studies with transformed mammalian cells al. 1989). In Smolowitz et al. (1989), this same group
there is a significant increase in the levels of choles- has also produced a monoclonal antibody that reacts
terol in the cytoplasmic membrane (Howard et al. both with a subpopulation of cells in the gut of nor-
1973). In our study, the incorporation of labelled ac- mal clams and with HN cells, indicating an internal
etate into cholesterol, and its derivatives, the source for HN cells. However, with the well-known
cholesterol esters, of HN cells was significantly lower possibility of epitopic crossreactivity, or even with ac-
than the levels of incorporation seen in normal he- tive mimicry of epitopes by parasites, the reaction of
mocytes. This is in contrast to uptake observed in a single monoclonal antibody with a subpopulation
mammalian neoplastic cells. Likewise, the changes in of M. arenafia hemocytes is insufficient to confirm
another class of neutral lipids, the free fatty acids, the endogenous source of the HN cells. As men-
remained unchanged while in mammalian cells there tioned above, several authors have noted that the
is an increase in free fatty acids. Taken together, available information on the origin of the HN cell in
these major differences between the lipid biochemis- clams does not preclude the possibility that the cells
try of HN cells and that of neoplastically transformed are parasitic rather than neoplastic in nature
mammalian cells indicate that it would be (Lauckner 1983; Mix 1986). Unfortunately, the data
presumptious to declare HN cells neoplastic 6n the presented here does not alleviate this problem be-
basis of their lipid biochemistry. cause, for example, the heteroploid cells may be
A fundamental change that occurs in neoplastically parasitic protozooans or even algae, which would
transformed cells is the alteration of the genome of have a considerably different chromosomal makeup
the cell, which is heritable and leads to the initiation than clam hemocytes. Further work on genetic analy@
and progression of tumors. These alterations may be sis of these cells using cytogenetic techniques, as well
"considerable, such as heteroploidy, which can be eas- as molecular biological techniques, must be carried
ily detected by flow cytometric analysis, or more out before the resolution of this problem can be
subtle, such as chromosomal rearrangements or alter- achieved.
ations in specific gene sequences which must be
detected by karyotyping or molecular DNA analysis.
In this study, flow cytometric analysis was used -to de-
termine the potential lesions in the DNA content of Citations
HN cells and also to assess the proliferative potential
of the disease. It was found that HN cells were dis- Alderman, D.G., P. Vin Banning, and A. Perez-Colomer.
tinctly aneuploid - in nature and that the most 1977. Two European oyster (Ostrea edulis) mortalities associ-
consistent change in the DNA content in HN cells ated with abnormal hemocytic condition. Aquaculture.
10:35-340.
was to a hyperdiploid state. The level of this Appledoorn, R.S., andj.J. Oprandy.
hyperdiploid population correlated well with the pro- 1980. Tumors in soft-shell clams and the role played by a
portion of cells which appeared abnormal by virus. Maritimes. August, 1980:4-6.
microscopic examination. There was significant varia- Awad, A.B., and A.A. Spector.
tion, however, in the amount of aberrant DNA in the 1976. Modification of the fatty acid composition of Erhlich
cells, which ranged from 1.6 to 2.0 times normal. ascites tumor cell plasma membranes. Biochim. Biophys.
Acta. 426:723-731.
This would be consistent with the clonal develop- Bligh, E.G., and W.J. Dyer.
ment of the transformed cell (Iversen 1988). 1959. A rapid method of total lipid extraction and
However, it was also found that most of the clams that *purification. Can. J. Biochem. Physiol. 37:911-917.
were HN positive had a second, markedly stable hy- Brousseau, D.J., andj.A. Baglivo.
1991. Field and laboratory comparisons of mortality in nor-
podiploid population of cells, a finding that is mal and neoplastic Mya arenmia. J. Invertebra. Pathol.
inconsistent with the clonal development of trans- 57:59-65.
formed cells. This finding is problematic if one is Brown, R.S., R.E. Wolke, S.B.N. Saila, an&C.W. Brown.
making the assumption that HN in M. arenari .a is a 1977. Prevalence of neoplasia in 10 New England popula-
true neoplastic disease because it is unlikely that a tions of the soft shell clam (Mya arenaiia). Ann. N.Y Acad.
Sci. 198:522-532.
stable hypodiploid population would coexist with a Calorini, L., A. Fallani, D.Tombaccini, E. Barletta, G.Magnai, M. Di
more variable hyperdiploid population in the same Renzo, P. Comoglio, and S. Ruggieri.
animal if this were true. 1989. Lipid characteristics of RSV-Lransformed Balb/C 3T3

Reno et al.: Hematic Neoplasia in the Softshell Chun Mya arenaria 99

cell lines with different spontaneous metastatic Montaudon, D.J.C. Louis, andj. Rober.
potentials. Lipids. 24:685-690. 1981. Phospholipid acyl group composition in normal and
Cooper, K-R., R.S.Brown, and P.W. Chang. tumoral nerve cells in culture. Lipids. 16:293-297.
1982. Accuracy of blood cytological screening techniques for Oprandy, jj., P.W. Chang, A.D. Pronovost, K.R. Cooper, R.S.
the diagnosis of a possible hematopoietic neoplasm in the Brown, and V.J. Yates.
bivalve mollusc, Mya arena7ia. J. Invertebra. Pathol. 1981. Isolation of a viral agent causing hernatopoietic
39:281-289. neoplasia in the soft-shell clam, Mya arenaiia. J. Invertebra.
Couch, J.A. Pathol. 38:45-51.
1969. Sarcoma-like disease in a single specimen of the Perkins, R.G., and R.E Scott.
American oyster. Comparative Leukemia Research Bibliog- 1978. Differences in the phospholipid, cholesterol, and fatty
raphy: Haernatology. No. 36. acyl composition of 3T3 and SV3T3 plasma
Farley, C.A. membranes. Lipids. 13:653-657.
1969. Probable neoplastic disease of the hematopoietic Sys- Raber, M.
tem in oysters, Crassostrea virginica and Crassostrea 1988. Clinical applications of flow cytometry. Oncology.
gigas. Nad, Cancer Inst. Monogr. 31:541-555. 2:35-43.
"arshbarger,J.C., S.V. Otto, and S.C. Chang. Reinisch, C.L., A.M. Charles, andj. Troutner.
1979. Proliferative disorders in Crassostrea virginica and Mya 1983. Unique antigens on neoplastic cells of the soft shell
arenafia from the Chesapeake bay and intranuclear virus- clam, Mya arenaria. Dev. Comp. Immunol. 7:33-39.
like inclusions in Mya arenaria with germinomas from a Sasaki, G.C., andj.M. Capuzzo.
Maine oil-spill site. Haliotis. 8:243-248. 1984. Degradation of Artemia lipids; under storage. Comp.
Heim, S., and F. Mitelman. Biochem. Physiol. B. Comp. Biochem. 78:525-531.
1987. Cancer Cytogenetics. Alan R. Liss, Inc., New York, Schroeder, F., andj.M. Gardiner.
NY 1984. Membrane lipids and enzymes of cultured high and
Howard, B.V.J.D. Butler, andj.M. Bailey. low-metastatic B16 melanoma variants. Cancer Res.
1973. Lipid metabolism in normal and tumor cells in cul- 44:3262-3269.
ture. In Tumor lipids: biochemistry and metabolism (R. Shinitzky, M.
Wood, ed.), p. 200-214. American Oil Chemists' Society 1984. Membrane fluidity in malignancy, adversative and
Press, IL. recuperative. Biochim. Biophys. Acta. 738:251-261.
Iversen, O.H. Smolowitz, R.M., D. Miosky, and C.L. Reinisch.
1988. Theories of carcinogenesis: facts, fashion or 1989. Ontogeny of leukemic cells of the soft shell clam. J.
fiction? Hemisphere Publishing Co., Washington, DC, Invertebra. Pathol. 53:41-51.
327 p. Steele, W., and H.M.Jenkin.
jerkofsky, M., and Aj. DeSiervo. 1977. Lipids and lipid metabolism of Novikoff rat hepatoma
1986. Differentiation of strains of varicella-zoster virus by cells. In Tumor lipids: biochemistry and metabolism (R.
changes of neutral lipid metabolism of infected cells. J. Wood, ed.), p. 215-225. American Oil Chemists' -Society
Virol. 57:809-815. Press, IL.
Kier, A.B., M.T. Parker, and F. Schroeder. VaseleivJ.M., and I.M. Gelfand.
1988. Local and metastatic tumor growth and membrane 1981. Neoplastic and, Normal Cells in Culture. Cambridge
properties of LM fibroblasts in athymic (nude) Univ. Press, Cambridge, UX
mice. Biochim. Biophys. Acta. 938:434-446. Wood, R., ed.
Lauckner, G. 1972. Tumor lipids: biochemistry and metabolism. American
1983. Diseases of Molluska:Bivalvia. In Diseases of marine Oil Chemists' Society Press, Champaign, IL.
animals. (0. Kinne, ed.), p. 477-962. Biol. Anst. Wuthier, R.E..
Helgoland, Hamburg, FRG. Vol 11. 1966. Two-dimensional chromatography on silica gel-loaded
Luna, L.G. paper for the microanalysis of polar lipids. J. Lipid Res.
1968. Manual of histological staining methods of the Armed 7:544-550.
Forces Institute of Pathology, 3rd ed. McGraw-Hill Inc., Yau, T.M., T. Buckman, A.H. Hale, and M.J. Weber.
New York, NY. 1976. Alterations in lipid acyl group composition and mem-
Marchesi, V.T. (ed.) brane structure in cells transformed by Rous sarcoma
1976, Membranes and neoplasia: new approaches and strate- -virus. Biochemistry. 15:3212-3219.
gies. Alan R. Liss, Inc., New York, NY Yevich, P., and C. Barszcz.
Miosky, D.L., R.M. Smolowitz, and C.L. Reinisch. 1976. Gonadal and hematopoietic neoplasms in Mya
1989. Leukemia cell-specific protein of the bivalve mollusc, arenatia. U.S. Natl. Mar. Fish. Serv. Mar. Fish. Rev. 38:
Mya arenafia. J. Invertebra. Pathol. 53:32-40. 42-43.
Mix, M.C. 1977. Neoplasia in soft-shell clams, Mya arenairia, collected
1986. Cancerous diseases in aquatic animals and their asso- from oil-impacted sites. Ann. N.Y. Acad. Sci. 298:409-426.
ciation with environmental pollutarfts: a critical literature
review. Mar. Environ. Res. 20:1-92.

Kinetics of Bovine Serum Albumin Administered by the Immersion
Method in Fishes Acclimatized to Seawater and to Fresh Water

MITSURU OTOTAKE and TERUYUKI NAKANISHI

Inland Station of National Research Institute ofAquaculture
Tamaki, Mie 519-04, Japan

ABSTRACT

To clarify plasma BSA kinetics in.marine fish after hyperosmotic infiltration (HI) treat-
ment, chum salmon (Oncorhynchus keta) and blue tilapia (Tilapia aurea) were acclimatized
to seawater or to fresh water, and were administered bovine serum albumin (BSA) by HI
treatment or intravenous injection. First, by using a least squares program, regression
curves were computed from the plasma BSA leve Is of fish injected with BSA to obtain the
clearance rate of BSA from plasma. Then, by using a deconvolution method, the BSA
release rate from the organ of BSA entry to plasma after HI treatment was calculated from
plasma BSA levels after HI treatment and the clearance rate..The results showed that the
clearance rates of both fishes acclimatized to seawater were higher than those of the fish
acclimatized to fresh water. On the other hand, the release rates of both fishes acclimatized
to seawater were much.lower than those of the respective fish acclimatized to fresh water.
Thus, it is quite plausible that the high clearance rates and low release rates of fish in
seawater work synergistically, resulting in lower plasma BSA levels of fish in seawater com-
pared with fish in fresh water after HI treatment.

Introduction plasma BSA levels of these animals, the rate of BSA
release into plasma and BSA clearance from the
The immersion method, first reported by Amend and plasma were calculated.
Fender (1976), is a useful method for mass immuni-
zation in aquaculture. Many papers have shown the
effectiveness of this method on freshwater fish (Ellis Materials and Methods
1988). On the other hand, there is a little knowledge
as to the immqrsion method's effectiveness on ma-
rine fish. In a preliminary study (unpubl. data), the Experimental Animals
authors used hyperosmotic infiltration (HI) treat-
ment to administer bovine serum albumin (BSA) to Six-month-old chum salmon, 16 � 1 g (mean � stan-
yellowtail Seriola quinqueradiata, which is the most im- dard deviation) in body weight, and 1-year-old blue
portant aquaculture species in Japan, and rainbow tilapia, 157 � 20 g in body weight, were used in this
trout Oncorhynchus mykiss, and then measured the experiment. Tilapia were fed commercial floating dry
BSA levels in their plasma. We found that plasma BSA pellets in a I-ton tank supplied with running fresh
levels were much lower in yellowtail than in rainbow water at 25' C, whereas chum salmon were kept in a
trout after HI treatment. This difference may have I-ton tank supplied with running fresh water at
been caused by two factors: species specificity and the 15' C, and fed commercial dry pellets for rainbow
type of environmental water-fresh water or seawater. trout. These tanks were kept indoors.
To analyze this difference it is necessary to compare
the kinetics of BSA in the plasma of several species of
fish that can reside in both seawater and fresh water. Acclimatization to Seawater
Thus, in the present study, chum salmon (0. keta)
and blue tilapia (Tilapia aurea) were acclimatized- to Before the administration of BSA, one half of the
seawater or to fresh water, and then administered tilapia and chum salmon were acclimatized to seawa-
BSA by HI treatment or injection. By using the ter, and the other half were kept in running fresh

96 NOAA Technical Report NWS I I I

water throughout the experiment. Fish were acclima- 1978). The gel plate was run at I V/mm and 15' C
tized to seawater (salinity: 34 ppt) by increasing the for 3 hours. The exact amounts of BSA in normal fish
proportion of seawater day by day to 50% on the first plasma were run in parallel with all tests for standard
day, 75% on the second day, 87.5% on the third day, reference.
and 100% on the fourth day. Afterwards the fish were
kept in the tank supplied with running seawater for 3
weeks before the experiment. During the acclimatiz- Calculations
ing period, no fish died or showed abnormal
behavior. Experiments were carried out with four Plasma BSA Clearance By means of an unweighted
groups of fishes: chum salmon in fresh water (fresh- least squares program, the plasma BSA levels Cpi"i(t)
water salmon); chum salmon acclimatized to seawater at each time t after injection were fitted to following
(seawater salmon); tilapia in fresh water (freshwater exponential equations:
tilapia); and tilapia acclimatized to seawater (seawa-
ter tilapia). Water temperature was maintained at Cp ini(t) = a, - exp (- bt); or
15' C for freshwater or seawater salmon and at 25' C
for freshwater or seawater tilapia. Cp'-i(t) = a, . exp(-b,.t) + a 2 *exp (-b 2* 0.

For the index of BSA clearance from plasma, the
Administration of BSA by HI Treatment mean residence time (MRT) of each experimental
group was given by
Fifty fish of each group were administered BSA (frac-
tion V, Sigma Co.) by the two-step hyperosmotic MRT = l1b,; or
infiltration technique according to the method of
Amend and Fender (1976) with slight modification. MRT = l1bi + l1b 2- (a, + a,)I(a, - b 2+ a 2 -b).
Distilled water was used to prepare the two bath solu-
tions; 5.3% NaCl solution and 2% BSA solution. Fish The BSA distribution volume V of each experimental
were immersed in 5.3% NaCl solution for 3 minutes group was calculated by
and then placed directly into 2% BSA solution for 3
minutes. After the bath treatment, the treated fish V = D/ Cp i-j(O),
were placed in tanks supplied with either running where D is the amount of BSA administered by injec-
seawater or fresh water for 3 minutes. Afterwards
they were returned to their original tanks. Blood tion (I mg/kg body weight).
samples were taken 0.5, 1, 2, 4, and 8 hours after the BSA Release into Plasma-The BSA release rate into
bath treatment. At each sampling time six fish were the plasma U(t) at each time t was given by'
sacrificed and the blood was withdrawn from the cau-
dal blood vessels using heparinized syringes. CpHl(t) = f.U(t) - Cpili(t-r) dT,

where Cp(t) is the plasma BSA level after the HI
Administration of BSA by Injection treatment and r is the time constant. The gross BSA
release into the plasma after the HI treatment was
Fifty fish of each group were administered BSA by calculated from U(t) by integration.
intravenous injection. Bovine serum albumin was dis-
solved with sterilized 0.85% NaCl solution to 2 mg/
mL and then injected into fish via the caudal blood Statistics
vessels at the dose of 1 mg/kg body weight. Blood
samples were taken 0.5, 1, 2, 4, 8, and 24 hours after Data were analyzed using the F test, and then by the
the administration by the same method as mentioned Student's t test to detect significant differences.
above.

Results
Immunoelectrophoresis
Plasma BSA Levels After the HI Treatment
Bovine serum albumin content in the plasma was
quantitatively assayed by rocket immunoelectro- Concentration of plasma BSA in the freshwater
phoresis (Laurell 1966; Wallenborg and Andersson salmon increased until 4 hours after the HI treatment

Ototake and Nakanishi: Kinetics of Bovine Serum in Fish 97

A B

120- 20

-100-
E15
80
----------
M
C< 60 <io. Figure I
n
co 40 1110 Changes in plasma concentrations of
bovine serum albumin (BSA) after
20 5 hyperosmotic infiltration treatment in
II freshwater-acclimatized fish (A) and
041 0 il Ll@* - seawater-acclimatized fish (B). Key:
0 2 4 6 8 0 2 4 6 8 (o) = tilapia; (*) = chum salmon. Data
HOURS AFTER BATH HOURS AFTER BATH are presented as means � standard
errors.

and reached a maximum of 90.6 � 6.5 @Lg/mL (mean was adopted for express plasma BSA concentration
standard error), and then decreased gradually (Fig. for freshwater salmon after injection. On the other
IA). On the other hand, plasma BSA levels of the sea- hand,
water salmon (Fig. 1B) were always significantly lower
(NO.01) than those of freshwater salmon.'A maximum Cpi@i(t) = a, - exp(-b,-t)
plasma BSA level of 18.8 � 2.6 @Lg/mL was observed in
the seawater salmon 2 hours after the bath; from then was adopted as the regression equation for seawater
on, plasma BSA stayed at a rather constant level of I I- salmon, freshwater tilapia, and seawater tilapia
19 I.Lg/mL. The plasma BSA level of freshwater tilapia (Fig. 2). The MRT and the distribution volume of
increased until 4 hours following' the bath when it BSA after the injection are shown in Table 1. Both in
reached a maximum of 119.0 � 8.1 @tg/mL and subse-
quently decreased to 64.1 � 2.8, @Lg/mL 8 hours after
the treatment (Fig. IA). In contrast, plasma BSA levels
of seawater tilapia (Fig. IB) were always significantly 60-
lower (P<0.001) than those of freshwater tilapia. A low A
level of BSA (0.09-1 `1 I.Lg/mL) was detected in the 50
plasma of the seawater tilapi.a for the initial 2 hours, 40@
but no BSA was detected in the plasma after 4 hours 30 p'=31 OexP(- I -20-0+26-9-exp(-0.027-t)
(minimum detectable level, 0.01 pg/mL). 20
E
10
OL Cp"--29-3-exp(-0.088-
Plasma BSA Levels After the Injection <
co 40-
C0=26.9-exV-0.964-0 B
Following BSA injection, the plasma BSA levels were 30.
always lower in seawater salmon than in freshwater 20
salmon, and the difference was significant (NO.01) 10- "=36-6-exW0.145-0
0.5, 1, 2, 8, and 24 hours after the injection (Fig. 2A). 0
The difference between seawater and freshwater ac- 0 4 8 12 16 20 24
climatized fish was more distinct in tilapia than in HOURS AFTER INJECTION
salmon, and the plasma BSA levels of seawater tilapia
were always significantly lower (P<0.001) than those Figure 2
of freshwater tilapia (Fig. 2B). Changes in plasma concentrations of bovine serum
albumin (BSA) after injection of BSA into chum
Plasma BSA Clearance salmon (A) and tilapia (B)."Open circles indicate the
fish in fresh water and closed circles indicate the fish
The regression equation acclimatized to seawater. Data are presente d as
C

means standard errors. Regression equations were
Cp a, exp (-b I - t) + a2 - exp (-b.2- t) calculated with a least squares program:

98 NOAA Technical Report NWS I I I

Table I
Kinetics of plasma BSA administered by hyperosmoticinfiltration treatment.

Experimental Maximum BSA Mean residence Gross release Distribution
group level in plasma time (MRT) into plasma volume
(Rg/mL) (hours) (pg/mL) (mL/100gb.w. a

Freshwater salmon 84 36 167 1.7
(37) b (95) b (3.7) b
Seawater salmon 19 11 24 3.4
Freshwater tilapia 119 6 146 2.7
Seawater tilapia 1 1 14 3.7

'b.w. = body weight.
bCalculated by the modified regression equation.

chum salmon and in tilapia, the plasma BSA clear-
ance, as represented by the MRT, was more rapid in 60- A
seawater acclimatized fish than in respective freshwa-
3 40-
ter acclimatized fish. The MRT of seawater salmon
and seawater. tilapia were 1/3 and 1/6 of that of 20.
respective freshwater acclimatized fish. The distribu-
tion volume of freshwater salmon was smaller than a 0
UJ 100
those of other experimental groups. tR B
cc 80
ul
co
< 60
BSA Release into the Plasma
LU
Ir 40.
The BSA release rate into the plasma, calculated <
co 20.
from the clearance rate and the plasma BSA levels,
is shown in Figure 3. The BSA release rate was al- 00
ways lower in seawater acclimatized fish than in 0 1 2 3 4 5 6
freshwater acclimatized fish. In all experimental HOURS AFTER BATH
groups, the BSA release rate was high during the Figure 3
early period, and the peak of the rate was observed Changes in the BSA release rate into plasma after
within one hour after the bath treatment. The rate hyperosmotic infiltration treatment in chum
rapidly decreased after one hour and became low salmon (A) and tilapia (B). Open circles indicate
six hours after the bath. The gross BSA release into the fish in fresh water and closed circles the fish
plasma, which is given by the area under the curve acclimatized to seawater. Data are presented as
in Figure 3, is shown in Table 1. Both in chum mean � standard errors.
salmon and tilapia, the gross BSA release of freshwa-
ter acclimatized fish was larger than that of seawater
acclimatized fish.
(1982) reported that plasma BSA administered by the
immersion method was trapped in the spleen and the
Discussion kidney of rainbow trout. Similarly, the trapping of
plasma BSA by the spleen and the kidney has been
In the present study, BSA clearance from the plasma observed in plaice administered BSA by intravenous
after the intravenous injection differed greatly within injection (Ellis 1980). The kidney is one of the most
the same species; in each species, the BSA clearance important organs for osmoregulation in fish, and its
was much more rapid in seawater-acclimatized fish function changes following seawater acclimatization
than in freshwater-acclimatized fish. These results (Hickman and Trump 1969). Thus it is possible that
strongly suggest that the environmental water affects the seawater acclimatization mechanism influenced
fish physiology and consequently, by some mechan- the BSA trapping in the kidney and affected the
ism(s), changes the plasma BSA clearance. Smith plasma BSA clearance. The plasma BSA clearance was

Ototake and Nakanishi: Kinietics of Bovine Serum in Fish 99

also different between species. Tilapia showed a rapid rates in freshwater salmon were examined by modify-
BSA clearance, while the chum salmon showed a slow ing the regression equation to ignore the first term,
BSA clearance. This difference may not only be at- which is the main component for the two hours after
tributed to species specificity, but also to the the injection. The distribution volume, the MRT, and
temperature of the rearing water. the gross BSA release were recalculated by using this
The estimated gross BSA release into the plasma of modified regression equation and the deconvolution
seawater salmon and seawater tilapia was much method. Recalculation with the modified regression
smaller than the amount released into respective equation also showed more rapid BSA clearance and
freshwater acclimatized fish. This indicates that the less gross BSA release in seawater salmon than in
physiological changes that resulted from seawater freshwater salmon (Table 1).
residence reduced the gross BSA release after the HI It has been reported that the plasma kinetics of
treatment. Amend and Fender (1976) suggested that several ions, such as Na', Ca2', Mg2+, and Cl-, are
hyperosmotic pretreatment had a hydrating effect on different in seawater- and freshwater-acclimatized fish
membranes of fish tissues, and thus made BSA infil- (Hirano and Uchida 1972; Bentley 1971). There have
trate into the fish more easily. However, fish residing also been reports comparing the kinetics of medi-
in seawater are adapted to a hyperosmotic environ- cines in plasma of seawater- and freshwater-
ment, and their membranes are more resistant to acclimatized fish, but the results are contradictory:
hydration than those of fish in fresh water. There- among the chemicals. Endo and Onozawa (1987)
fore, it is reasonable to consider that hyperosmotic and Ishida (1990) reported that when oxolinic acid
pretreatment stimulates BSA uptake more strongly in was administered to ayu, Plecoglossus altivelis, and
fish in fresh water than in fish in seawater. In any coho salmon, 0. kisutch, by bath and oral methods
case, it is quite plausible that some factors in seawater the seawater-acclimatized fish of both species showed
influenced the fish physiology and induced the low lower oxolinic acid concentration and shorter reten-
BSA release rate and high BSA clearance rate. These tion time than freshwater-acclimatized fish. On the
two rates of seawater-acclimatized fish worked syner- other hand, BergsJo and Bergsjo (1978) reported
gistically to make their plasma BSA levels after the HI that when sulfanilamide or sulfadimidine was admin-
treatment much lower than those of -freshwater- istered by bath treatment to rainbow trout, the
acclimatized fish. plasma concentration of seawater-acclimatized fish
The BSA distribution volume after the intravenous was higher than that of freshwater-acclimatized fish.
injection is expected to agree with the whole plasma It still remains unknown why there are such differ-
volume estimated by tracer-dilution methods using ences between medicines. In any case, from the view
Evans blue dye, serum albumin, or red blood cells, all of vaccine application to marine fishes, it is quite im-
of which have been widely used to determine whole portant to clarify whether the phenomenon observed
plasma (or blood) volumes. Using these methods, in the present study for BSA (seawater-acclimatized
whole plasma volumes of fish have been determined fish have a lower release rate and higher clearance
to be 2.5-4.7 mL/100 g body weight in salmonids rate) is common to protective antigens such as li-
(Conte et al. 1963; Smith 1966; Huggel et al. 1969, popolysaccharides and proteases.
Nikinmaa et al. 1981; Gingerich et al. 1987) and 3.5
mL/100 g body weight in yellowtail (Itazawa et al.
1983). The distribution volumes of the present study Acknowledgments
for seawater salmon, freshwater tilapia, and seawater
tilapia agree well with those reported values. In fresh- Financial assistance for this study was provided in part
water salmon, however, the distribution volumes in by a grant-in-aid (Bio-Media Program) from the Minis-
the present study were smaller than the whole plasma try of Agriculture, Forestry, and Fisheries (BMP-91-
volumes reported so far in the literature. Smith IV-1-9). We would like to thank Yasuo Inui 'for his criti-
(1966) suggested that blood sampling periods of less cal review of this manuscript. We also acknowledge the
staff of the Hokkaido Salmon Hatchery for supplying us
than two hours after tracer injection tended to pro- with the chum salmon eggs used in this study.
duce low plasma volume and high clearance rate
estimates in salmonids because these periods did not
allow for adequate circulation of tracers in the vascu- Citations
lar system. Therefore, in the case of freshwater
salmon in the present study, there is a possibility that
Amend, DY, and D.C Fender.
the plasma BSA clearance rate was overestimated. In 1976. Uptake of bovine serum albumin by rainbow trout
such a case, the gross BSA release is likely to be over- from hyperosmotic infiltration: a model for vaccinating
estimated. Therefore, the BSA release and clearance fish. Science 192:793-794.

100 NOAA Technical Report NAM I I I

Bentley, P.J. par la method de la dilution isoto-pique de 1'1311 et du
1971. Endocrines and osmoregulation. Springer-Verlag, Ber- 5ICr. J. Physiol. (Paris), 61:145-154. (In French.)
lin/Heidelberg/New York, 300 p. Ishida, N.
Bergsjo, T., and T.H Bergsjo. 1990. Comparison of tissue level of oxohnic acid in fresh
1978. Absorption from water as an alternative method for and seawater fishes after the oral administration. Bull. jap.
the administration of sulphonamides to rainbow trout, Soc. Sci. Fish. 56(2):281-286. (Injapanese; English abstr.)
Salmo gairdne7i. Acta. Vet. Scand. 19:102-109. Itazawa, Y, T. Takeda, I- Yamamoto, and T. Azuma.
Conte, F.P., H.H. Wagner, and T.O. Haris. 1983. Determination of circulating blood volume in three
1963. Measurement of blood volume in the fish (Salmo teleosts, carp, yellowtail and porgy. jpn. J. Ichthyol.
gairdnerigairdne7i). Am.J.Physiol.205:533-540. 30(l):94-101. (Injapanese.)
Ellis, A.E. Laurell, C.B.
1980. Antigen trapping in the spleen and kidney of the pla- 1966. Quantitative estimation of proteins by electrophoresis
ice Pleuranedes platessa L. J. Fish Dis. 3:413- 426. in agarose gel containing antibodies. Anal. Biochem.
1988. Current aspects of fish vaccination. Dis. Aquat. Org. 15:45-52.
4(2)-.159-164. Nikinmaa, M., A. Soivio, and E. Aiolo.
Endo, T., and M. Onozawa. 1981. Blood volume of Salmo gairdneri: influence of ambient
1987. Effects of bath salinity and number of fish on the up- temperature. Comp. Biochem. Physiol. A Comp. Physiol.
take of oxolinic acid by ayu. Bull. jap. Soc. Sci. Fish 69A(4):767-769.
53(4):557-562. (Injapanese; English abstr.) Smith, L.S.
Gingerich, W.H., R.A. Pityer, andjj. Rach. 1966. Blood volumes of three salmonids. J. Fish. Res. Board
1987. Estimates of plasma, packed cell and total blood vol- Can. 23:1439-1446.
ume in tissues of the rainbow trout. Comp. Biochem. Smith, P.D.
Physiol. A Comp. Physiol., 87A(2):251-256. 1982. Analysis of the hyperosmotic and bath methods for
Hickman, C.P.Jr. and B.F. Trump. fish vaccination: Comparison of uptake of particulate and
1969. The kidney. In Fish physiology I (W.S. Hoar and Dj. nonparticulate antigens. Dev. Comp. Immunol., Suppl.
Randall, eds.) p. 91-241. Acad. Press, New York/ L6ndon. 2:81-186.
Hirano, T., and S.'Uchida. Wallenborg, B., and U.B. Andersson.
1972. Water and ion movements in the intestine and the uri- 1978. Immunoelectrophore tic techniques with the LKB 2117
nary bladder of the teleosts and their hormonal multiphor. LKB Application Note, 249:1-12. (Pharmacia
control. Medical Science 23(2):56-68. (Injapanese.) LKB Biotechnology Inc., 800 Centennial Ave., Piscataway, NJ
Huggel, Hj., H.C. Lane, and C.G. Ducret. 08854.)
1969. Determination de la courve D'homogeneisation et du
volume sanguin circulant de la truite Salmo gairdneri Rich.

The Epidemiological Study of Furunculosis
m Salmon Propagation

TETSUICHI NOMURA*, MAMORU YOSHIMIZU**, and TAKAHISA KIMURA**

*Hokkaido Salmon Hatchery, Fisheries Agency
Nakanoshima 2-2, Toyohiraku
Sapporo, Japan.

"Faculty, ofFisheries, Hokkaido University
Minatomachi 3-1
Hakodate, Japan

ABSTRACT

The authors attempted to determine the distribution and prevalence of Aeromonas
salmdnicida in mature chum (Oncorhynchus keta), pink (0. gorbuscha), and masu
salmon (0. masou) in Hokkaido that showed no apparent clinical signs of furunculo-
sis. From September 1979 to November 1989, a total of 12,891 chum, pink, and masu
salmon were collected from 30 rivers. The changing pattern of the annual prevalence
of A. salmonicida in salmon was closely related to changes in fish density in the
holding ponds: the prevalence of A. salmonicida increased in proportion to the in-
crease in the number of fish in the ponds. We concluded from the results of
histological and bacteriological examinations that fish with A. salmonicida in the kid-
ney were not diseased but were carriers of A. salmonicida. The agent could not be
isolated from the immature fish examined. A. salmonicida was also isolated from the
ovarian fluid of fish showing no apparent clinical sign of furunctilosis. Few A.
salmonicida were found on the surface of,the eggs one hour ' after fertilization. A
survey of agglutination titers against A. salmonicida in sera of chum, pink, and masu
salmon showed great variability within the species. The isolated strains were identi-
fled as A. salmonicida subsp. salmonicida and were pathogenic to salmonids. We
concluded that the A. salmonicida carrier state in fish poses a serious problem in the
prevention of furunculosis and its reduction plays a key role in salmon propagation.
Both maturation of fish under conditions of low density in ponds, and disinfection of
their eggs, are necessary to prevent fish furunculosis during artificial propagation of
salmon.

Introduction documented in juvenile amago salmon (0. rho-
durus), and masu salmon (0. masou) in accordance
Furunculosis of salmonid fishes, caused by Aeromonas with increased production of these fish. In
salmonicida, was first reported in 1890s by Emmerich Hokkaido, outbreaks of furunculosis have been re-
and Weibel (1890, a and b). Since these first reports, ported to occur in chum salmon (0. keta) by
furunculosis has been reported in virtually all parts Nishino (1967), and in masu and pink salmon (0.
of the world where wild or cultured salmonids occur gorbuscha) by Kimura (1970) during the maturation
(Smith 1960; Herman 1968; Snieszko 1972; Austin of these species in holding ponds. Nomura and
and Austin 1987). Kimura (1981), Nomura (1983), and Nomura et al.
Furunculosis is not a serious problem in rainbow (1983, 1991, a and b) reported isolating A.
trout (Oncorhynchus mykiss) culture in Japan be- salmonicida from the kidneys of mature chum, pink,
cause this species is resistant to the causative agent and masu salmon that showed no apparent clinical
of the disease. However, serious mortality has been signs of furunculosis.

101

102 NOAA Technical Report NNM I 11,

10 11 Sea of Okhotsk
9 12
13 14 is
Sea of Japan 16 17 19
7 8 18 20 21 Nernuro Strait
22 23 24
6 25 Figure 1
4 5 Rivers where chum, pink, and masu salmon were col-
27 26 lected (after Nomura et al. 1991a).
30 29 1 Moheji 2 Hekirichi 3 Shiriuchi 4 Shubuto
28 5 Shiribetsu 6 Ishikari 7 Shokanbetsu 8 Nobusha
9 Teshio 10 Tonbetsu 11 Hrobetsu 12 Tokushibetsu
2 1 Cape Erimo Pacific Ocean 13 Horonai 14 Shokotsu 15 Yuubetsu 16 Tokoro
3 17 Abashiri 18 Shari 19 lwaobeLsu 20 Ichani
Tsupru Strait 21 Shibetsu 22 Tohoro 23 Nishibetsu 24 Furen
25 Bettouga 26 Kushiro 27 Tokachi 28 Shizunai
29 Niikappu 30 Yurappu

Recently, no systematic epidemiological studies We isolated A. salmonicida from -chum salmon in 11 of
have been done to establish control measures for fu- the 22 rivers examined; the percent occurrence of the
runculosis in salmonids from which quarantine and bacterium in this species of fish ranged from 0.6 to
disease control policies could be based. 49.2%. Populations Of pink salmon, from 13 rivers were
In this paper, we report the recent epidemiological tested and A. salmonicida was isolated from 6 of these
study of A. salmonicida which was carried out for the rivers with percent occurrence ranging from 0.2 to
purpose of establishing control measures for the dis- 13.3%. In masu salmon the bacterium was isolated from
ease. 5 of 10 rivers examined and the percent occurrence
ranged from 1.0 to 5.6%. Hence, A. salmonicida was de-
termined to be distributed widely in the salmonid
Distribution of A. salmonicida populations of Hokkaido, except those of rivers located
in Salmonids in Hokkaido between Tsugaru Strait and Cape Erimo (Fig. 2).
In the Ishikari, Shari, Iwaobetsu, Shibetsu, and
We attempted to determine the distribution and Tokachi rivers, the prevalence of A. salmonicida was
prevalence of A. salmonicida in mature chum, pink, found to vary yearly. In the Ishikari river, the preva'-
and masu salmon populations in Hokkaido that lence of A. salmonicida in chum salmon was high from
showed no apparent clinical signs of furunculosis 1979 to 1984 but has gradually been decreasing since
(Nomura et al. 1991 a). 1985 (Fig. 3). In the chum salmon of the Tokachi
From September 1979 to November 1989, a total of River and in all three species in the Shibetsu River,
12,891 chum, pink, and masu salmon were collected the prevalence of the bacterium remained high
from 30 rivers (Fig. 1). At each sampling, a total of 60 throughout the examination period. In the Shari
fish of each species were randomly selected from the river from 1979 to 1988, A. salmonicida was not iso-
rivers' salmonid populations in accordance with Amos lated from any of the fish examined; however, it was
(1985). The fish were separated by species and river isolated from 4 of 60 fish examined in 1989.
and held in individual ponds at each river for about I From 1979 to 1984, changes in the monthly preva-
month until maturity.- After spawning, they Were pro- lence of the agent could be observed in fishes in the
cessed for examination. Kidney materials were Ishikari river. The incidence of A. salmonicida in-
streaked onto nutrient agar plates (Eiken Co., Tokyo, creased until the middle of October and then
Japan) and cultured at 20'C for 7 days. No clinical decreased thereafter. The pattern of change was
signs of furunculosis were observed in the examined closely related with changes in fish density in the
fish. Bacterial colonies that produced a soluble brown holding pond; the prevalence of the bacterium ap-
pigment and showed the following characteristics were peared to increase proportionately as density of fish
classified as A. salmonicida: Gram-negative staining, in the pond increased (Fig. 4).
lack of motility, failure to grow at 37'C, tested positive The number of A. salmonicida bacteria found in
for cytochrome oxidase, and had the ability to ferment kidney tissues ranged from 10' to 105 colony forming
on oxidative fermentative basal medium. units per gram (cfu/g) (Nomura et al. 1991 a).

Nomura et al.: Furuculosis in Salmon Propagation 103

10 c Sea of Okhotsk
9 c 12 CPM
13 CP
16 CP
Sea of Japan 17 P
8 M 18 20 cp Nemuro Strait
21 CPM
23 C
6 C Figure 2
Rivers where Aeromonas salmonicida was isolated from
M 26 C chum, pink, and iasu salmon from 1979 to 1989. C:
27U isolated from chum salmon; P: isolated from pink
salmon; M: isolated from masu salmon (after'
Pacific Ocean Nomura et al. 1991a).
Cape Frimo 5 Shiribetsu 6 Ishikari 8 Nobusha
Teshio 10 Tonbetsu 12 Tokushibetsu
Tsugarn Strait 13 Horonai 16 Tokoro 17 Abashiri
18 Shari 20 Ichani 21 Shibetsu
23 Nishibetsu 26 Kushiro 27 Tokachi

Yew Year
1979 n=263 1984 n=240 The kidney materials of chum salmon in which A.
salmonicida was isolated were fixed with Bouin's solu-
50 - 0 - tion for histopathological examination. The kidney
0 organs were dehydrated and embedded in Paraplast,
0 0 and sections of the samples were made and stained
a , i I I- % with HE and Gimsa stain.
1980 n7--360 1985 n=274 Histopathological examination of the infected fish
did not, however, reveal colonies of A. salmonicida
50- - typically observed in fishes with furunculosis. Also,
000 no outbreaks of furunculosis were recorded in the
0 examined populations during the research period
10 R I I ocb I I (Nomura et al. 1991a).
1981 0 n=274 1986 n--240 There are few reports examining the prevalence
50 - 0. - of A. salmonicida in the organs of apparently nor-
0 mal mature fish. In fact, as far as we know, there is
0
0 only one report by Daly and Stevenson (1985).
_L 011 1 1 10 Ino They reported that A. salmonicida was detected in
19112 n=300 1987 n=300 31 of 286 brown trout (Salmo trutta) sampled from
spawning runs in the Ganaraska River, Ontario,
50 - Canada, over a period of two years. Our results
0 showed that the incidence of this agent in appar-
_L 00, -dcbO b- 1 ently normal chum salmon was higher compared to
1983 0n=469 1988 n=541 that of Daly and Stevenson's (1985) estimated for
brown trout, and that A. salmonicida is distributed
50 - 0 - widely in the river populations of salmonids in
Hokkaido.
000 Morikawa. et al. (1981) reported that the number
001 1 of A. salmonicida in the kidneys of moribund amago
Sep. Oct. Nov. Sep. Oct Nov. salmon was 10' to 101 cfu/g. The reason why diseased
Month fish were not found in the population we examined,
even though they had A. salmonicida in their kidneys,
Figure 3 was that the degree of infection was not high
Changes in the monthly incidence of Aeromonas salmonicida enough. From the results of histological and bacte-
riological examinations, we'conclude that fish with A.
'o C ea 0'0
@12
M
cp
9 C
@Japa '6 CP17,PS
18
8 M 20 C
2,

6 C

Par
@iflc
'0

pe Fri..

in chum salmon collected from the Ishikari river, and held
for maturation periods during the period September to salm .onicida in the kidney are not diseased fish but
November, 1979-88 (Modified from Nomura et al. 1991a). carriers of A. salmonicida.

104 NOAA Technical Report NMB I I I

in ovarian fluids ranged from 101 to 101 cfu/mL in
the populations from the Shibetsu and Teshio rivers.
Ovarian fluid containing A. salmonicida flows out of
01981 the fish at the time eggs are stripped or during the
40 process of maturation in the pond. Consequently, in-
fected water and infected ovarian fluid are expelled
into the river because the sewage from the egg strip-
0 1980 ping areas and the holding ponds is not disinfected
0 im in Hokkaido. We suspect from these results that the
20 0 im agent drained from the fish may be a source of infec-
tion for other anadromous salmon that ascend the
01982
0 1984 river for spawning.
01M
0190 Horne and Maj (1928), McCraw (1952), and
19W , -0 IM Hastein and Lindstad (1991) stated that the most im-
0 portant source of A. salmonicida in the spread of
0 2 4 6
furunculosis is the existence of fish carrying this
Fish stock density (flsh/m@) agent. Fish carrying the bacterium pose a serious
problem to the prevention of furunculosis, and their
Figure 4 reduction in fish plays a key role in salmon propaga-
The relationship between average density of the fish in tion.
hol 'ding ponds and the incidence (% occurre@ce) of
Aeromonas salmonicida in fish take n from the Ishikari River,
Numbers in the figure indicate the year of examination A. salmonicida on Egg Surfaces
(unpubl. data).
The existence of the bacterium in the ovarian fluid
suggests that the surface of eggs taken from the fish
A. salmonicida in Immature Fish will also be contaminated. Contaminated eggs may
spread the agent to areas where the eggs will be
We attempted to isolate A. salmonicida from immature transplanted. We studied the existence of A.
chum and masu salmon (Nomura et al. 1991b). A salmonicida on the surface of eggs by artificially con-
total of 680 fish were collected in four coastal set nets taminating chum salmon eggs with A. salmonicida.
off Hokkaido, and a total of 1,200 juvenile masu The A. salmonicida 20-1, strain, which was isolated
salmon and chum salmon fry were collected from I I from the kidneys of chum salmon in the Tokachi
hatcheries of the Hokkaido Salmon Hatchery system. River, was used as innoculum. The strain was cultured
The bacterium was not isolated from any of the and harvested, then was suspended in phosphate
examined fish. buffer saline (PBS). The chum salmon eggs were
bathed in PBS containing the agent for an hour. The
eggs were incubated in well water at 8'C in the labo-
A. salmonicida in Ovarian Fluid ratory. At one hour and at 24 hours after fertilization,
we took 20 eggs from the incubator and put them
In 1989 i *n Hokkaido, we attempted to isolate A. into sterilized water. The flask was shaken strongly for
salmonicida from the ovarian fluids of mature chum, 5 minutes, we then measured the viable number of A.
pink, and masu salmon. salmonicida in the water according to the method of
Ovarian fluids were collected according to the Nomura et al. (1991b).
method of Yoshimizu et al. (1985). The ovarian fluids Egg surfaces were initially bathed with 1.1 x 101 to
were streaked onto nutrient agar plates (Eiken Co., 4.3 x 101 cfu/egg of A. salmonicida. The number of A.
Tokyo, Japan) and the plates were cultured at 20'C salmonicida present on the egg surfaces decreased
for 7 days. Number of viable counts of A. salmonicida from 68 to 4.6 cfu/egg an hour after fertilization and
in ovarian fluid and kidney were measured in accor- A. salmonicida could not be isolated from the eggs
dance with the method of Nomura et al. (1991a). cultured on plates 24 hours after the initial bath
A. salmonicida was i solated from the ovarian fluid of treatment.
fish showing no apparent clinical signs of furunculo- We also attempted to isolate A. salmonicida from
sis. For example, A. salmonicida was isolated from the eggs in the incubation boxes at the Satsunai,
ovarian fluid of 22 of 120 fish examined from the Nakagawa, Nemuro, and Tokachi hatcheries. These
Tokachi River. The number of A. salmonicida bacteria eggs were taken from brood fish in which the preva-

Nomura et at.: Furunculosis in Salmon Propagation 105

lence of A. salmonicida was high (Nomura et al. It is believed that A. salmonicida is not able to exist
1991a). Fortunately, A. salmonicida was not isolated for long time in water without fish, but McCarthy
from any of the 15 hatcheries' eggs (Nomura et al. (1980) studied the survival of the agent in water us-
1991b). ing an antibiotic-resistant strain of A. salmonicida and
In Hokkaido, fertilized eggs are transported to a found the agent could survive for 8 days in water.
hatchery from the egg collection location I hour af- The results of McCarthy (1980) and this study indi-
ter fertilization. From the results of our experiment, cate that A. salmonicida survives long enough to infect
it appears that A. w1monicida is able to exist on the other fish in the water.
surface of an egg. This makes us concerned that we
may be transporting the bacteria to the hatchery with
the fertilized egg. We believe that it is necessary to Variation of ALmdutination Titer
prevent the transfer of A. salmonicida via eggs in or- Against A. salmonicida in the Serum
der to control furunculosis.
A serological survey of adult salmon was -made from
blood samples collected in 1988 in Hokkaido, from
Survival of A. salmonicida in Water mature chum, pink, and masu salmon. Blood was
aseptically extracted from the donsal artery with 10
By definition, A. salmonicida is considered to be an mL of Vacteinor (Terumo Co., Tokyo, Japan). The
obliga te pathogen (Popoff 1984) and is never found resulting serum was separated from the blood-cell
in surface water. Its ability to survive and remain in- clot by centrifugation and was stored at -90'C until
fectious in the external environment may be a major assayed. The serum was tested for agglutinating anti-
determinant in the spread of furunculosis. We stud- body titers individually, by test-tube methods with A.
ied the viability of A. salmonicida in nonsterile, sterile salmonicida ATCC 14174.
filtered, and autoclaved fresh water and in salt water. Agglutinin titers against A. salmonicida in the se-
A. salmonicida strain 20-1 isolated from chum rum of mature chum, pink, and masu salmon in
salmon in the Tokchi River was used in this experi- Hokkaido in 1988 are shown in Table 1.
ment. The strain was cultured at 20'C and harvested, Of a total of 75 serum samples taken from mature
then suspended in fresh water or in sea water. The chum salmon, 73.3% did not have 'the agglutinin,
suspended cells were inoculated into 200 mL of and the range of titers was 8 to 32. In pink salmon,
nonsterile, sterile filtered, and autoclaved fresh water 10% of the sample did not have the agglutinin, and
and saltwater and were incubated at 10'C. the range was 4 to 32. In masu salmon, 16.6% of the
In sterilized fresh water, A. salmonicida survived for examined serum did not have the agglutinin, and the
60 days and in nonsterile water, only 4 days. The sur- range was 4 to 128.
vival of A. salmonicida in sterile salt water was 8 days; The diversity in the incidence of agglutination titer
this was a shorter survival period than that in sterile within each of the three species indicates a continu-
fresh water. ous, widespread interaction between individuals of

Table 1
Prevalence and ranges of agglutinin titers against A. salmonicida in the serum of mature chum, pink, and masu salmon
ascending various rivers in Hokkaido.

Species River No. of fish examined Agglutinin titers
Negative Modes Range

Chum salmon lchani 15 80.0 32 16- 32
Tokachi 30 76.6 8 8- 32
Shizunai 30 66.6 16 8- 32

Pink salmon Shokotsu 30 16.6 16 4- 32
lwaobetsu 30 6.7 16 4- 32
Nishibetsu 30 6.7 16 4- 32

Masu salmon Shiribetsu 39 15.4 8 4-128
Shari 18 0 16 4-128
Nishibetsu 27 29.6 8 4- 32

106 NOAA Technical Report NNM I I I

the host populations and A. salmonicida. The differ7 suspected that A. salmonicida can survive for a long
ence in the amount of agglutination titer is time in the river water in Hokkaido after leaving the
proportionate to the period of A. salmonicida infec- fish and that its existence may be a source of infec-
tion. tion to salmonid fish. The results suggest that such
In general, the percentage of serologically reactive bacteriophage could be very useful for studying the
salmon increased as their length of freshwater resi- existence of the agent in water.
dency increased. Weber and Zwicker (1979) reported
that of a total of 43 serum sampled from Atlantic
salmon (Salmo salar) in the Miramichi or Margareen Table 2
rivers in Canada, none had A. salmonicida aggluti- Isolation of Aeromonas salmonicida bacteriophage
nin, but of 27 Restigouche River salmon, four had a from the samples of river water and hatchery water.
titer of 10, five had a titer of 20, and one had a titer
of 640. They confirmed that Atlantic salmon have Sample Samples
previously contacted A. salmonicida in, the Numbers examined Isolated
Restigouche River. containing phage
In our study, agglutinate titers in the serums were River water 4 2
low. It was suggested that the fish were infected with Hatchery water 9 2
A. salmonicida shortly before their eggs were stripped. Sewage of hatchery 13 3
Kimura (1970) reported that the immunological
method of preventing furunculosis was useful in
adult masu salmon during the holding period be-
cause these salmon stay in fresh water for a long Pathogenicity of the
enough period to allow them to produce antibodies Isolated A. salmonicida
after antigen inoculation. In chum salmon, however,
the freshwater residency period is short, so this The isolated strains were identified as A. salmonicida,
method of prevention would not be practical. subspecies salmonicida, by their biological, biochemi-
cal, and immunological characteristics. All of the
isolated A. salmonicida strains showed auto-agglutina-
Isolation of the Bacteriophage tion and produced protease in the medium, so we
of A. salmonicida from Water also expected them to be pathogenic.
In order to examine the pathogenicity of the iso-
There is no sensitive medium for selecting A. lated strain, we injected it into chum and masu
salmonicida. This means that when the number of A. salmon fry and adult chum salmon.
salmonicida in water is low, the isolation of A. A. salmonicida 20-1 was cultured for 48 hours at
salmonicida from the water will be difficult. This is 20'C. The cells were washed three time in PBS and
because A. salmonicida cannot grow on the culture were suspended in PBS. The strain was injected into
medium under competitive conditions with other chum salmon fry, yearling masu salmon, and chum
natural bacteria populations. We attempted to isolate salmon brood fish at concentrations of 1.7 X 102, 1.8
the bacteriophage of A. salmonicida to ascertain the X 101 and 6.0 X 101 cfu/fish, respectively.
existence of A. salmonicida in the water. All of the examined fish showed typical signs of
Water samples, from 11 hatcheries and 4 rivers furunculosis 3 to 4 days after injection. The number
were examined. Nutrient agar (Eiken Co., Tokyo, Ja- of A. salmonicida in the kidneys of moribund fish was
pan) was employed for the routine culture, dilution, around 108 cfu/g kidney tisstie, the same number re-
and enumeration of A. salmonicida and its phage ported by Morikawa. et al. (1981) in the kidneys of
strain. One hundred mL of sample was added to 500 moribund amago salmon. On the basis of these re-
mL'of cultured A. salmonicida Ar-32, Ar-43, Ar-71, and sults, we suspect the isolate is a pathogenic strain.
H-70 strains in the logarithmic phase. Detection and
enumeration of phage were achieved using the me-
dium and double agar layer technique (Paterson et Control of A. salmonicida
al. 1969). The results are shown in Table 2. on the Surface of Egg
McCraw (1952) stated that when the bacteriophage
of A. salmonicida exists, its presence may indicate the To establish a method of controlling A. salmonicida
existence of the bacterium. The bacteriophage was on the eggs, the bactericidal effect of popidon-iodine
isolated from two samples of river water and five (Isodine), and the toxicity of this agent to the chum
samples from hatchery water. From this result, it was salmon egg were studied.

Nomura et al.: Furunctdosis in Salmon Propagation 107

The bactericidal effects of popidon-iodine to A. Chum salmon in the Ishikari River were randomly
salmonicida were determined in accordance with the assigned to experimental holding ponds.and held un-
method of Amend and Fryer (1972). der low (4.9 fish/M2 ) and high density (14.7 fish/M2)
A. salmonicida was completely killed by treatment conditions until maturation. The kidney tissues of all
with 25 ppm isodine for five minutes and this solu- the fish used in experiment were cultured on nutri-
tion was not toxic to the chum salmon eggs for ent agar in accordance with the method of Nomura
treatments lasting up to one hour. Thus, the authors et al. (1991).
confirmed that isodine solution has a sanitizing ef- As we expected, we found that 12.4% of the fish
fect on the agent, and thatit.does not have adverse examined harbored A. salmonicida when they were
effects on chum salmon eggs. stocked at a high density, but no examined fish con-
tained the agent when they we're stocked at a low
density (Fig. 5A). The incidence of the bacterium in
Method for Decreasing the Prevalence fish that were held under low dissolved. oxygen condi-
of A. salmonicida in Chum Salmon tions was higher than that of fish held under high
dissolved oxygen levels (Fig. 51)). These results
From the results of our epidemiological study, we sus- clearly indicate that high stocking densities and low
pected that the incidence of A. salmonicida was dissolved oxygen levels in holding ponds have a
affected by the density of fish during their maturation marked effect on the prevalence of the agent in the
period in the holding ponds; as the average density of fish. We concluded that fish maturation in the pond
brood fish stocked in ponds decreased, the incidence under low density conditions and disinfection of the
of A. salmonicida in examined fish also decreased (Fig. eggs, are necessary to prevent fish furunculosis in the
4). Therefore, we examined the relationship between artificial propagation of salmon.
the stocking density of fish in the pond and the preva-
lence of A. salmonicida in,the fish.
Citation

FiSh/M2 A FUh stock density Amend, D.F., andj.L. Fryer.
15 1972. Virucidal activity of two iodo.-phors to salmonid
virus. J. Fish. Res. Board Can. 29:61-65.
10 Amos, ILH., ed.
1985. Fish health blue book-procedures for the detection of
5 certain fish pathogens. Fish health section, Am. Fish. Soc.,
114 p.
0 Austin, B., and D.A. Austin.
PPM 11' Dissolved oxygen 1987. Bacterial fish pathogens: disease in farmed and wild
10 - 0 fish. Ellis Horwood Ltd., Chichester, West Sussex, En
gland, 195 p.
5 - Daly, J.G., and R.M.W. Stevenson.
1985. Importance of culturing several organs to detect
0 Aeromonas salmonicida in salmonid fish. Trans. Am. Fish.
% C Survival rate Soc. 114:909-910.
100 - Emmerich, R., and C. Weibel.
1890a. Ober eine durch Bacterien verursachte Infection-
50 - skrankheit der Forellen. Allg. Fish. Zgt. 15:73-77.
1 1 1890b. Ober eine durch Bacterien verursachte Infection-
0 D Incidence skrankheit der Forellen. Allg. Fish. Zgt. 15:85-92.
% Hastein, T., and T. Lindstad.
2 1991. Disease in wild and cultured salmon: possible
0 interaction. Aquaculture. 98:277-288.
Herman, R.L.
0 1
0 1968. Fish furunculosis 1952-1966. Trans. Am. Fish. Soc.
Lot A Lot B 97:221-230.
Hornej.H., and I.M.S. Maj.
Figure 5 1928. Furunculosis in trout and the importance of carriers
The relationship between the density in the spread of the disease. J. Hyg. 28:67-78.
of fish stock in holding ponds, the Kimura, T.
concentration of dissolved oxygen in 1970. Studies on a bacterial disease occurred in the adult
the water, survival rate, and the inci- "Sakuramasu" (Oncorhynchus masou) and pink salmon (0.
gorbuscha) rearing for maturity. Sci. Rep. Hokkaido Salmon
dence (% occurrence) of Aeromonas Hatchery. 24:9-100. Onjapanese; English abstr.)
salmonicida (unpubl. data).

108 NOAA Technical Report NNM I I I

McCraw, B.M. Nomura, T., M. Yoshimizu and T. Kimura.
1952. Furunculosis of fish. U.S. Fish and Wildl. Serv. Spec. 1991b. Prevalence of Aeromonas salmonicida in the kidney of
Sci. Rep. 84:87 p. chum salmon (Oncorhynchus kela) and masu salmon (0.
McCarthy, D.H. masou) at various life stages. Fish Pathol., 26:149-
1980. Some ecological aspects of the bacterial fish patho- 153. (InJapanese; English abstr).
gen-Aeromonas salmonicida. In Aquatic microbiology: sym- Paterson, W.D., RJ. Douglas, I. Grinyer, and L.A. McDermott.
posium of the Society of Applied Bacteriology 6:299-324. .1969. Isolation and preliminary characterization of some
Morikawa, S., S. Miki, and F. Tashiro. Aeromonas salmonicida bacteriophages. J. Fish. Res. Board
1981. Changes in hemato logical properties and viable cell Can. 26:629-632.
number of bacteria in amago salmon artificially infected Popoff, M.
with Aeromonas salmonicida. Fish Pathol. 16:43-49. (In 1984. Family 11. Vibrionaceae. In Bergey's manual of system-
Japanese; English abstj%) atic bacteriology, Vol. I (Noel R. Krieg and John G. Holt
Nishino, K. eds.), p. 545-550. Williams and Wilkins, Baltimore.
1967. On a bacterial disease of adult "sake" Oncorhynchus Smith, I.W.
kela, in captivity. Fish Pathol. 2(l):73-74. (InJapanese.) 1960. Furunculosis in salmon kelts. Nature 186:733-734.
Nomura, T.- Snieszko, S.F.
1983. Incidence of Aeromonas salmonicida among anadro- 1972. Panel review on furunculosis of salmonidae. In Report
mous salmonids, 1980-1982. Sci. Rep. Hokkai- do Salmon of the symposium on the major communicabable fish dis-
Hatchery. 37:63-65. (InJapanese; English abstr.) eases in Europe and their control; 20-22 April 1972,
Nomura, T., T. Kimura. Amsterdam (William A. Dill, ed.): p 157-163. European In-
1981. Incidence of Aeromonas salmonicida among anadro- land Fisheries Advisory Commitee EIFA Technical paper 17
moussalmonids. Fish Pathol. 16:69-74. (Injapanese;En- (suppl. 2).
glish abstr.) Yoshimizu, M., T Kimura, andJ.R. Winton.
Nomura, T., I Kimura, I. Shimizu, and K. Nara. 1985. An improved technique for collecting reproductive
1983. Incidence of Aeromonas salmonicida among chum fluid samples from salmonid fish. Prog. Fish-Cult. 47:199-
salmon, Oncorhynchus keta, in Chitose River. Sci. Rep. 200.
Hokkaido Salmon Hatchery. 37:53-61. (In Japanese; En- Weber, J.M. and B.M. Zwicker.
glish abstr.) 1979. Aeromonas salmonicida in Atlantic salmon (Salmo
Nomura@ T., M. Yoshimizu, and T. Kimura. salar). J. Fish. Res. Board Can. 36:1102-4107.
1991a. Prevalence of Aeromonas salmonicida in the chum
salmon (Oncorhynchus keta), pink salmon (0. garbuscha) and
masu salmon (0. masou) returning to rivers in
Hokkaido. Fish Pathol. 28, 139-147. (In Japanese; En-
glish abstr.)

Functions of Hemocytes During the Wound Healing Process
in the Pearl Oyster

TOHRU SUZUKI

National Research Institute of Aquaculture
Nansei, Mie 516-01, Japan

ABSTRACT

Morphological evidence from previous studies suggests that bivalve hemocytes function
in hemostasis and extracellular matrix production during the wound healing process.
The present paper describes the use of an in vitro cell culture system to show that the
agranular hemocytes of the. pearl oyster Pinctada jucata have the ability to perform these
healing functions. A possible wound healing system is also presented.

Introduction muscle with a plastic syringe attached to a 24 gauge
needle (Fig. 1). Approximately 2 mL of blood was
Defense reactions against infection and the removal of collected from each animal. Blood pooled from five
tissue debris are the responses of the immune system animals was centrifuged at 100 X g for 8 minutes and
that occur after wounding. On the other hand, tissue a pellet of hemocytes was obtained. The pellet was
regeneration at the wound site is an organogenetic washed twice with a balanced salt solution for marine
event in which cell-cell and extracellular matrix (ECM)- molluscs (MMBSS) prepared according to Machii
cell interactions occur. Thus, the wound healing and Wada (1989). The pellet was suspended on 2 mL
process can be a useful model to study the functions of of culture medium Pf35 (Machii and Wada 1989),
proteins and other cell components in both the im- developed specifically. for pearl oyster tissue. A 0.5-
mune and organogenesis systems. Because bivalves have mL sample of the suspension was pipetted into a
.no humoral clotting factors, platelets, or capillary ves- plastic culture flask (50 mL: Nunc). After an hour, to
sels, it is assumed that their wound healing. system allow the hemocytes to adhere to the plastic surface,
differs from that of vertebrates and is probably less com- 2.0 mL of P05 was added to the flask. The cells were
plex. The wound healing process has been described then incubated at 25'C. The culture medium was re-
for several bivalves at the morphological level. It is newed on the fourth day of culture.
known that the hemocytes prevent blood loss at the
wound site by forming a cellular sheath and producing Microscopy
an ECM (DesVoigne and Sparks 1968; Pauley and
Heaton 1969; Ruddell 1971). The healing process in The cultured cells were fixed in Karnovsky's fixative
the pearl oyster Pinctadafucata and the associated func- containing 8% sucrose at VC for I hour. After wash-
tions of hemocytes have been demonstrated (Suzuki et ing in 0. 1 M phosphate buffer (PB; pH 7.2), they were
al. 1991). In this paper, the author describes a study post-fixed in 1% osmic acid in PB for I hour, washed
using cultured hemocytes from the pearl oyster in tv@ice in PB, dehydrated through graded alcohols, and
which agranular hemocytes show the ability to form a embedded in Taab 812 resin. Ultrathin sections were
cellular sheath and to produce an ECM. A possible stained with uranyl acetate and lead acetate. All trans-
wound healing system is also presented. mission electron microscopy was performed with a
JEOLJEM- 1200EX electron microscope.
Methods Results and Discussion
Hemocyte Culture Aggregate Formation
While the valves of R fucata were held slightly open The hemocytes formed cellular aggregates during
with a wedge, blood was drawn from the adductor blood collection and rinsing. Three hours after the

109

Ito NOAA Technical Report NNUS I I I

viewed under phase-contrast microscopy (Fig. 5).
Blood collection The matrix continued to develop until about the sev-
enth day of culture, after which time the number of
living hemocytes rapidly decreased, Agranular hemo-
cytes of 1-3 layers thickness were attached to the
matrix, which was composed of fine fibrils and
flocculent substances (Figs. 6 and 7). The fibrils were
Washing 20 nm in diameter and showed a faint banding pat-
tern indicating that they are collagenous fibrils. The
flocculent substances are probably proteoglycans. It
has also been ascertained at the biochemical level
Suspension that the matrix includes collagen (Suzuki et al.
1991).

Standing for 1h Wound Healing System
In the pearl oyster, four noticeable cellular reactions
occur during the healing process (Suzuki et al.
Culture at 25*C 1991): 1) removal of tissue debris, 2) cellular sheath
formation, 3) ECM deposition, and 4) epithelial re-
generation. The first reaction is attributed to the
Figure I phagocytic ability of -agranular hemocytes. In addi-
Procedure used for hemocyte culture. tion, these cells also perform the second and third of
these healing reactions, cellular sheath formation
and ECM deposition, as suggested by the results of
beginning of culture, the aggregates were 80-200 Rm this paper.
in diameter and adhered well to the plastic surface . of Based on these results, a possible wound healing
the culture flask (Fig. 2). In bivalves, two categories system in the pearl oyster is presented in Figure 8.
of hemocytes occur-agranular. and granular hemo- When the pearl oyster.is wounded, agranular hemo-
cytes (Feng et al. 1971). It has been demonstrated cytes infiltrate the injured area to remove tissue
that the former are macrophage-like cells with active debris (and possibly foreign particles) and then form
phagocytic capabilities (Reade and Reade 1972; a cellular sheath to prevent blood loss. After sheath
Moore and Lowe 1977). One hemocyte of the pearl formation, they begin to produce an ECM, which
oyster is able to take in 1-9 vertebrate erythrocytes functions as a template for the regenerating epithe-
(Suzuki and Mori 1990). As shown by transmission lium. It is possible that the wound site is repaired
electron microscopy, the aggregate is formed only by through this sequence of processes. Assuming this ex-
agranular hemocytes (Fig. 3). They adhere to one planation is accurate, the wound healing system of
another at an adhesion plaque from which microfila- this bivalve is simple compared with that of verte-
ments can be seen running into the cytoplasm (Fig. brates.
4). Thus, agranular hemocytes can form a cellular It is well known that in vertebrates growth factors
aggregate in a relatively short time (perhaps several secreted from the platelets, macrophages, and
minutes if stimulation is provided), although it is not fibroblasts are the humoral factors that control cell
understood what factors stimulate the reaction. It is growth and ECM production at the wound site. In
possible that this aggregate formation is a homolo- bivalves, however, neither a growth factor nor its
gous reaction to the cellular sheath formation at the production cell has been identified. The agranular
wound site. hemocyte is the most likely candidate to be a growth
factor production cell, because, like macrophages and
fibroblasts in vertebrates, it is the primary cell type
ECM Production in bivalves that appears at the wound site. In addi-
tion, platelets are absent from bivalve hemocytes.
During the first 2-3 days of culture, deposition of an Future studies with in vitro cultures will improve our
ECM started in the agranular hemocyte aggregate. understanding of both the wound healing system
The matrix was observed as a transparent gel when and organogenesis in bivalves.

Suzuki: Function of Hemocytes During Wound Healing in the Pearl Oyster III

V
<L

'_kw4

0
'N*' JA
@PIN J,

!!k, 3;

:f'4

J,
6

0
9;0 11M 'K M
45

Figure 2 Hemocyte aggregate after 3 hours of culture. X 135.
Figure 3 Electron microscopy of aggregate after I day of culture. X 2,000.
Figure 4 Adhesion plaque formed at contacting site of agranular hemocytes. X 6,700.
Figure 5 Matrix (arrowhead) formed in a hemocyte aggregate after 7 days of culture. X 135.
Figure 6 Electron microscopy of matrix and agranular-hemocyte layer. X 2500.
Figure 7 Fibrils in the matrix at high magnification. X 31,000.

Citation

DesVoigne, D.M., and A.K. Sparks. Moore, M.N., and D.M. Lowe.
1968. The process of wound healing in the pacific oyster, 1977. The cytology and cytochemistry of the hemocytes of
Crassostrea gigas. J. Invertebr. Pathol. 12:53-65. Mytilus edulis and their responses to experimentally injected
Feng, S.Y.JS. Feng., C.N. Burke, and L.H. Khairallah. carbon particles. J. Invertebr. Pathol. 29:18-30.
1971. Light and electron microscopy of the leucocytes of Pauley, G.B., and L.H. Heaton.
Crassostrea virginica (mollusca: pelecypoda). Z. Zellforsch. 1969. Experimental wound repair in the freshwater muisel
120:222-245. Anotionta oregonensis. Invertebr. Pathol. 13:241-249.
Machii, A., and K.T. Wada. Reade, P., and E. Reade.
1989. Some marine invertebrates tissue culture. In Inverte- 1972. Phagocytosis in invertebrates 11-the clearance of car-
brate cell system application, Vol. 11 U. Mitsuhashi, ed.) p. bon particles by the clam, Didacna maxima.
225-223. CRC Press, Boca Raton, Florida. Reticuloendothel. Soc. 12:349-360.

112 NOAA Technical Report NMFS I I I

9000 0
'.. O/V 9SE
0

Infiltration of agranular hemocytes Cellular sheath formation
Removal of tissue debris by them by agranular hemocytes

><-

Regeneration of epithelium Secretion of ECIVI
along the new ECM by agranular hemocytes Figure 8
Growth factors? A possible wound healing system in the pearl oys-
ter: ECM=extracellular matrix.

Ruddell, C.L. Suzuki, T., R. Yoshinaka, S. Mizuta, S. Funakoshi, and F, Wada.
1971. The fine structure of oyster agranular amebocytes 1991. Extracellular matrix formation by amebocytes during
from regenerating mantle wounds in the paciU oyster, epithelial regeneration in the pearl oyster Pinctada
Crassostrea gigas. J.. Invertebr. Pathol. 18:260-268. fucata. Cell Tissue Res. 266:75-82.
Suzuki, T., and Y, Mori.
1990. Hemolymph lectin of the pearl oyster, Pincladafucata
martensii: a possible non-self recognition system. Dev.
Comp. Immunol. 14:161-173.
0;1

Skeletal Abnormalities of Fishes Caused by
Parasitism of Myxosporea

YUKIO MAENO and MINORU SORIMACHI
National Research Institute of Aquaculture
Fisheries Agency
Nansei-cho, Mie 516-01, Japan

ABSTRACT

The relationship between skeletal abnormalities and parasitic infection was studied for
deformed cultured yellowtail, Seriola quinqueradiata, Japanese bluefish, Scombrops boops,
and mullet, Mugil cephalus. Soft radiographic observations indicated that the deformities
were due to skeletal abnormalities: deformed yellowtail, Japanese bluefish, and mullet
were characterized by scoliosis, lordosis, and lordo-scoliosis, respectively. Myx6sporean
cysts were found in various parts of the brain of deformed fish such as the fourth ven-
tricle' the cavity of the optic tectum, the surface of the olfactory lobe and bulb, and the
optic lobe. The cysts were observed in the fourth ventricle in all deformed fish, but not in
normal fish. From morphological characteristics, myxosporeans from the deformed yel-
lowtail and Japanese bluefish were identified as Myxobolus buri, and that from deformed
mullet was identified as M. spinacurvatura. These results suggest that Myxobolus species are
responsible for the skeletal abnormalities observed in these deformed fish.

Introduction tu re from 1987 to 1989.Four deformed Japanese
bluefish and four deformed mullet were captured by
Skeletal abnormalities of fish have been reported to a set net in Mie Prefecture in 1989.
be caused for various reasons, such as dietary defi- After external observation, radiography of all fish
ciencies (Halver and Shanks 1960), pesticide was performed with Fuji Lx film at 40 milliamps and
exposure (Couch et al. 1977), heavy metal exposure 50kV for 20 seconds and subsequently examined for
(Holcombe et al. 1976), and bacterial (Kaige et al. parasitism in viscera by necropsy. From these animals,
1984) and parasitic infections (Halliday 1976). Many three deformed and three normal yellowtail, three
studies report sporulation of myxosporeans in fish Japanese bluefish, and three mullet were used for
brains. However, only a few studies have described a subsequent histological examination.
possible correlation between skeletal abnormalities The brain, liver, kidney, and spleen were excised
and myxosporean parasitism in the brain. While from each fish and fixed in 10% formalin, embedded
studying skeletal abnormalities in three species of in paraffin, cut serially at 4 gm, stained with Giemsa
fish, cultured yellowtail (Se7iola quinqueradiata), Japa- or hematoxylin and eosin, and then observed for the
nese bluefish (Scombrops boops), and mullet (Mugil presence and location of cysts.
cephalus), we found spores of a myxosporean in the For the study of spore morphology, some myxospor-
brains of these fish. The present paper elaborates on ean cysts were removed from the brain and crushed in
the possible relationship between skeletal abnormali- phosphate buffer saline to get a suspension of fresh
ties and parasitism by myxosporeans with special spores. Light microscopic observations were made both
reference to parasitized regions of the brain. on fresh spores and those stained with Giemsa.
For the scanning electron microscopy, spore-sus-
pensions were fixed in Karnovsky's fixative for 2
Materials and Methods hours at room temperature. They were then dehy-
drated in a graded ethanol series, treated with
Seventeen yellowtail (age I+), including 7 deformed isoamyl acetate three times, each for 15 minutes, pro-
ones, were collected from a fish farm in Mie Prefec- cessed in a critical point dryer, sputter-coated with

113

114 NOAA Technical Report NMFS I I I

Pw""

IIN#

tow Figure I
Scoliosis found in yellowtail
W Se7iola quinqueradiata (dor-
sal view).
4'
Figure 2
A cyst of Myxobolus bu7i found
n the brain of yellowtail
(Seriold quinqueradiata) Gi-
entsa stain (X300).

gold, and examined with the JEOL Uapan Electron- curvatures consisting of a smooth S-like bend in the
ics Optical Limited) T220A scanning electron lateral plane, scoliosis (Fig. 1).
microscope. In the seven deformed yellowtail, myxosporean
cysts were found in one or more of the following
regions: beneath the meninges of the olfactory bulb
Results and lobe, surface of optic lobe, cerebellum, medulla
oblongata, inferior lobe, the cavity of optic tecturn,
Cultured Yellowtail and the fourth venticle. In particular, the cysts were
found in the fourth ventricle of the brain in all de-
The deformities were externally apparent as curva- formed fish examined (Table 1). In the normal fish,
tures from the trunk to caudal region. Radiography on the other hand, no cysts were found in the fourth
of the deformed specimens revealed extreme skeletal ventricle of three fish, even though seven fish had

Maeno and Sorimachi: Skeletal Abi@ormalities of Fishes Caused by Myxosporea 115

V11

III
IV(

Table I
Distribution of cysts in the brains of deformed (n--7) and apparently normal (n=10) yellowtail Seriola quinqueradiata.
(I=olfactory bulb; Il=olfactory lobe; III=optic lobe; IV=the cavity of optic tectum; V=cerebellum; VI=f6urth ventricle;
VII=inferior lobe.)

I II III W V VI VII

Deformed fish 5/7 5/7 2/7 3/7 1/7 7/7 2/7
Normal fish
Infected 2/5 3/5 2/5 1/5 0/5 0/5 1/5
Non-infected 0/5 0/5 0/5 0/5 0/5 0/5 0/5

cysts in some other region of the brain. No cysts were In deformed Japanese bluefish, cysts were always
observed in regions other than the brain. found in the fourth ventricle and less frequently
The cysts range from 0.5 to 3 mm in diameter and found in the cavity of the optic tectum. In some cases
were enveloped by a collagenous capsule with cysts were found attached to the surface of the olfac-
fibroblasts (Fig. 2). Most cysts were filled with nearly tory bulb and olfactory lobe and in the cavity of the
mature or mature myxosporean spores. Morphologi- optic tecturn. No cysts were found in tissue other
cal observations showed spores to be broadly than the brain. Both light and scanning electron mi-
ellipsoidal and symmetrical in shape and to possess a croscopy revealed that the spores from these fish
prominent sutural ridge in the front view. In the side I were morphologically similar to those obtained from
view they were lenticular and symmetrical and con- deformed yellowtail. The spores were broadly ellip-
sisted of two shell valves. The polar capsules were soidal in frontal view and broadly lenticular in side
pyriform and nearly equal in size, and a small view; the shell valve of the spore had a sutural ridge
intercapsular appendix was clearly seen. The length, with fold; each spore had two polar capsules nearly
width, and thickness of spores were 10.5, 9.0, and 6.3 equal in shape and size; and an intercapsular appen-
gm, respectively. From these morphological charac- dix was present (Figs. 4 and 5). The average length of
teristics they were identified as Myxobolus buyi Egusa, fresh spores was 10.4 gm, width 9.1 9m, and their
1985. thickness 6.2 gm.

Japanese Bluefish Mullet

Radiographic observations demonstrated that the The four deformed mullet showed spinal curvature
abnormality in the four Japanese bluefish was and lordo-scoliosis in radiography (Figs. 6 and 7).
dorso-ventral deformity of the vertebral column Histological examination revealed many
and flexure of the neural spine and hernal spine myxosporean cysts (in the surface of the gut and in
(Fig. 3). various visceral organs such as the liver, kidney,

1.16 NOAA Technical Report NNM I I I

V-
W."

1. VW
X-
Q':

Njl@'

Figure 3
V Soft radiograph of lordosis in a Japanese blue-
fish (Scombrops boops) (side view).

Fig ure 4
Fresh spore of Myxobolus sp. obtained from the
brain of Japanese bluefish. (a) frontal view (ar-
row indicates intercapsular appendix); (b) side
view (bar=10 lim). Phase contrast microscopy.

Figure 5
Scanning electron micrographs of spores of
Myxobolus sp.obtained from the brain of Japa-
nese bluefish. (a) frontal view; (b) side view.

spleen, pancreas, and brain of the deformed mullet. tural plane, and lack of an intercapsular appendix
As in yellowtail and Japanese bluefish, cysts we@e (Fig. 9), these myxosporean spores were considered
found in various regions of the brain, such as the to belong to the genus Myxobolus (Lom and Noble
fourth ventricle, the cavity of the optic tecturn, the 1984).
surface of the olfactory lobe and bulb, and the optic
lobe. Cysts were observed in the fourth ventricle of
all four fish examined (Fig. 8).The average length of Discussion
fresh spores was 11.5 gm, their width was 9.8 gm, and
thickness 6.7 gm. From the morphological character- The morphological characteristics of the spores
;eool@@
,e "@7@

istics of the spores, such as their rounded frontal clearly indicate that the myxosporean found in de-
view, lenticular side view, two smooth shell valves, two formed yellowtail of the present study is M. bwri. The
teardrop-shaped polar capsules situated on the su- light and scanning electron microscopic examina-

Maeno and Sorimachi: Skeletal Abnormalities of Fishes Caused by Myxosporea 117

Figure 6
Soft radiograph of lordosis in a mullet (Mugil
cephalus) (side view).

Figure 7
Soft radiograph of scoliosis in a mullet (Mugil
cephalus) (dorsal view).

Figure 8
Diagrammatic view of the distributions of the
T
& r
4'r cysts in the brain of mullet Mugil cephalus
(*=loci of the cyst).

Figure 9
L T-WW'L

Fresh spore of Myxobolus spinacurvatura ob-
tained from the brain of mullet. (a) frontal
view; (b) side view (bar=10 gm).

118 NOAA Technical Report NNOS 111

tions also demonstrated a complete similarity in the These studies, as well as the present study, indicate
morphology between the spores from the deformed some correlation between skeletal abnormalities and
Japanese bluefish and those from deformed yel- myxosporean parasitism in the brains of fish. In addi-
lowtail. Thus, the spores parasitic in the brain of the tion, the present study strongly suggests that skeletal
deformed Japanese bluefish are considered to be abnormalities occur when cysts infect particular re-
those of M. bu7i. gions of the brain such as the fourth ventricle. Thus,
To date, more than 400 species of Myxobolus have it is quite possible that myxosporean infection in this
been described. The morphological characteristics of region of the brain mechanically affects central ner-
Myxobolus from the deformed mullet differ from vous system function to produce skeletal
those of most species described. Only three Myxobolus abnormalities in fish.
species, M. achmerovi Shulman, 1966, M. bu?i, and M.
spinacurvatura Maeno et al., 1990, have morphologi-
cal similarities to those found in deformed mullet. Citations
However, M. achmerovi, which has been found from
the gill, fin, and mesentery of common carp Couch, J.A., J.T. Winstead, L.R. Goodman.
(Cyprinus carpio) and mullet (Mugil cephalus) 1977. Kepone-induced scoliosis and its histological
(Shulman 1966) is different from the present cosequences in fish. Science 197:585-w587.
HalverjE., and W.E. Shanks.
Mymbolus in that the former spore has an ellipsoidal 1960. Nutrition of salmonid fishes VIII. Indispensable
frontal view, and a distinct intercapsular appendix. aminoacids for sockeye salmon. J. Nutrition 72:340-346.
M. buri also differs from the present Myxobolus in Halliday, M.M.
mullet because the spore of M. buyi has a broadly 1976. The biology of Myxosoma cerebralis: the causative organ-
ellipsoidal frontal view, distinct folds around the ism of whirling disease of salmonids. J. Fish Biol. 9:339-
357.
edge, an intercapsular appendix, and a distinct polar Holcombe, G.W., D.A. Benoit, E.N. Leonard, andj.M. McKim.
filament inside the polar capsule. On the other hand, 1976. Long-term effects of lead exposure on three genera-
the spores from the deformed mullet were quite simi- tions of brook trout (Salvelinus fontinalis). J. Fish. Res.
lar to those of M. spinacurvatura in both Board Can. 33:1731-1741.
morphological characteristics and in spore dimen- Hoshina, T.
1952. Notes on some myxosporidian parasites on fish of
sions. Thus, the Myxobolus found in deformed mullet Japan. J. Tokyo Univ. Fish. 39:69-89.
in the present study is considered to be M. Kaige, N., T. Miyazaki, and S. Kubota.
spinacurvatura. 1984. The pathogen and the histopathology of vertebral de-
In the present study the myxosporean cysts were formity in cultured yellowtail. Fish Pathol. 19:173-179. (In
invariably observed in the fourth ventricle of the Japanese; English abstr.)
.Langdon,J.S.
brain in all deformed fishes examined. Myxosporean 1987. Spinal curvatures and encephalotropic myxosporean,
parasitism in the central nervous system of fish have Triangula percae sp. nov. (Myxozoa: Ortholineidae), enzootic
been studied in the bullbead Cottus gobio (Lom et al. in redfin perch, Percafluviatilis L., in Australia. J. Fish Dis.
1989), the redfin perch Perca fluviatilis (Langdon 10:425-434.
1987), the yellowtail Sefiola quinqueradiata (Sakaguchi Lom, J., and E.R. Noble.
1984. Revised classification of the class Myxosporea Butschi,
et al. 1987), the fathead minnow Pintephales promelas 1881. Folia Parasitologia (Praha). al:193-205.
(Mitchell et al. 1985), and the Japanese river goby Lomj, S.W. Feist, 1. Dykova, and T. Kerp.
Acanthogobius flavimanus (Hoshina 1952). However, 1989. Brain myxoboliasis of bullhead, Coitus gobio L., due to
even in a heavily myxosporean-infected bullhead, mo- Myxobolus firoved sp. nov.: light and electron microscope
tor or sensory disturbances were not observed. Also, observation. J. Fish Dis. 12:15-27.
Mitchell, L.G., C.L. Seymour, andj.M. Gamble.
in myxosporean-infected fathead minnow and Japa- 1985. Light and electron microscopy of Myxobolus
nese river goby, no clear external signs of deformity hendiicksoni sp. nov. (Myxozoa: Myxobolidae) infecting the
were found. On the other hand, skeletal abnormali- brain of the fathead minnow, Pimephales promelas
ties were observed in the myxosporean-infected Rafinesque. J. Fish Dis. 8:75-89.
redfin perch and yellowtail. In the redfin perch, cysts Sakaguchi, S., T. Hara, T. Matsusato, I Shibahara, Y Yamagata, H.
Kawai, Y. Maeno.
were located in the regions of the mesencephalon, 1987. Scoliosis of cultured yellowtail caused by parasitic
diencephalon, third ventricle, and medulla oblon- Myxobolus bu7i. Bull. Natf. Res. Inst. Aquaculture. 12:79-86.
gata, while in the yellowtail, cysts were found in the (Injapanese; English abstr.)
cavity of the optic tectum, the surface of the optic Shulman, S.S.
and olfactory lobes, and in the fourth ventricle. 1966 Myxosporidian fauna of the USSR. Nauka, Moscow and
Leningrad, 251 p.

Presence of Oncogenes in Fish Tissues and in Fish CeR Unes

E. READ-CONNOLE, C. A. SMITH, and F. A HETRICK*
Department of Microbiology
University of Maryland
College Park, MO 20742

ABSTRACr

Work during the past decade has shown that avian and mammalian proto-oncogenes
are centrally involved in cell transformation in vitro and in the formation of tumors in
vivo. The fish systems in which oncogenes have been described are reviewed as are the
general methodologies used to detect oncogenes and their gene products. Also discussed
is preliminary work on the development of a test system that measures oncogene activa-
tion in fish cell lines in order to evaluate carcinogenic chemicals in the environment.
Sequences related to the rask, raP, v-raf, v-erb-B, c-src, c-myc, c-abl, and c-fos proto-
oncogenes, and to the p53 suppressor gene, were detected by Southern, Northern, and
Western blots. Taken together, the above findings indicate that proto-oncogenes are well
conserved evolutionarily in vertebrates and may be responsible for development of the
transformed phenotype in fish.

Introduction 1990). However, wild type ras protein can be trans-
formed in a variety of carcinogen-induced anim"al
Evidence has been accumulating in recent years that tumor model systems. For example, in rats almost
in mammals, particularly humans, tumor develop- 90% of mammary carcinomas induced by a single
me nt and progression are correlated with changes in dose of nitrosomethylurea possess a c-H-ras-I
the structure or expression of cellular genes. Yet in a oncogene, which is the activated form of the *c-H-ras
number of higher vertebrates, oncogenic retroviruses gene having a G to A transition at codon 12
have long been known to be causally involved in the (Barbacid 1986, 1987). In humans, highly amplified
initial appearance and subsequent growth of a variety N-ras DNA sequences were found in small cell lung
of naturally arising tumors (Bishop 1985; Klein and cancer, and this amplification seems to correlate with
Klein 1985). These two statements are not contradic- tumor progression and prognosis Uolinson et al.
tory. The retroviral oncogenes (v-oncs) responsible 1987). Changes in the structure (mutation) of one or
for the transformation event are now recognized to more oncogenes, their amplification, and the level of
have arisen from the capture and processing of whole expression of their mRNA, are all potential indicators
or portions of essential cellular genes termed proto- of tumorigenic initiation and progression.
oncogenes or c-oncs (Bishop 1983; Stehelin et al. In view of the widespread occurrence of neoplasia
1976). It is now considered important to understand in freshwater and saltwater fish species, and because
the extent of evolutionary conservation of these cel- fish are becoming recognized as convenient model
lular genes in order to fully characterize their normal systems for the study of natural and environmentally
function and delineate their involvement in the induced diseases (Powers 1989), the search for proto-
tumorigenic conversion of normal cells either di- oncogene-like sequences in teleosts is warranted.
rectly or through retrovirus-induced transformation. Experimental evidence indicates that some fish
The,family of ras proto-oncogenes, rash (Harvey), species, when compared with rodents, are less sensi-
ras' (Kirsten), and N-ras are highly conserved from tive to the toxic effects of chemical agents and more
mammals to yeast. The 21 kilo dalton ras protein is susceptible to their carcinogenic effects. Because of
expressed in a wide variety of nontransformed cells, their aquatic habitat, fish are fully exposed to chemi-
suggesting a normal cellular function. Xenopus laevis cals in the water at the gill, eye, skin, and gut levels,
oocytes espress the ras protein throughout oogenesis
and embryonic development (Baum and Beberriitz *Send correspondence to this author.

120 NOAA Technical Report N?M I I I

and therefore may be good indicators of carcinogens are detected. Comparisons of the number and size of
in the environment. Thus, it is worth investigating oncogene homologous fragments from the Southern
oncogenes, and their expression in tumorigenesis in blots with the more quantifiable data from the dot
fish, especially when they are induced by chemicals. blot experiments should reveal whether any of the
Such studies have now begun. Here we describe the fish genes detected have been rearranged or ampli-
general methodology used in detecting oncogenes fied in the tumor tissues under study.
and their expression and review those systems where
oncogenes have been found in fish tissues and in
continuously cultivated fish cell lines. Detection of Oncogene Expression

For any gene to exert an effect on cells and tissues,
Methodology mRNA must be transcribed and eventually translated
into protein. One can therefore screen RNA isolated
from the various sources for the presence of tran-
Oncogene Detection scription products homologous to those of
mammalian oncogenes. Enhanced expression of vari-
For the detection of oncogenes at the DNA level, it is ous oncogenes following the amplification of DNA
important to isolate high molecular weight genomic may contribute to malignant progression as is sug-
DNA with a minimal size of 100-200 kb. Standardized gested by the human c-N-myc gene involvement in the
procedures are available for the rapid isolation of progression of human neuroblastomas (Schwab et al.
good quality DNA from either cell cultures or directly 1984), or it may involve the loss of a gene and its
from tissues excised from organisms. One procedure expression as has been observed in the case of the
(Smith et al. 1988) employs a gentle sodium dodecyl human retinoblastoma susceptibility gene (Friend et
sulfate (SDS) and proteinase-K lysis step followed by al. 1986; Lee et al. 1987).
phenol extraction, ethanol precipitation, and spool- Described briefly, cells and tissues being examined
ing of DNA (contaminating RNA can be removed by for oncogene transcription products are first pro-
RNAse treatment), Tissue samples are processed by cessed by the guanidinium thiocyanate/hot phenol
this method by including liquid nitrogen freezing procedure to extract the total cellular RNA. The total
and tissue crushing steps prior to the lysis. Methods cellular RNA obtained is subjected to oligo-dT cellu-
for isolating genomic DNA from fish cell lines are lose affinity chromatography to purify or enrich the
described in Smith et al. (1988) and Read-Connole et poly A+ RNA fraction (putative mRNA). The poly A+
al. (1990). RNA species are then separated by size by denaturing
Samples of DNA from the various sources are then (formaldehyde) agarose gel electrophoresis. After
digested to completion with one or more various re- transfer of the RNA to nitrocellulose or ilylon mem-
striction endonucleases (e.g., EcoRl, Hind III and Bam branes, mature mRNA species with homology to
HI), and the resulting DNA fragments are resolved oncogene probes can be detected by northern blot-
by size by agarose gel electrophoresis. After transfer ting procedures. In addition, RNA dot blot
of the DNA fragments to nitrocellulose or nylon experiments can also be performed to quantify the
membrane supports, the DNA samples can then be levels of mature mRNA hybridizing with the
screened by Southern blotting for the presence of oncogene probes (Smith et al. 1987; Louis et al.
sequences homologous to nick-translated labelled 1988). The mRNA studies give valuable information
DNA probes specific for mammalian and viral on whether any of the specific oncogenes detected in
oncogenes. In addition, serial dilutions of samples of the various tumor tissue .preparations are of different
restriction-endonuclease-treated DNA can be directly sizes (suggesting either truncation of the gene or re-
spotted onto nitrocellulose or nylon membranes. arrangements) and whether they are overexpressed
These dot blots can then be probed in parallel with or aberrantly expressed when compared with the nor-
the Southern blots to aid in the quantification of mal counterparts.
hybridizable DNA in each sample, and to estimate
any gene amplification that may occur in the tumor
tissues. Detection of Oncogene Products
Comparisons between normal and tumor tissues
can be made from the banding patterns obtained in The third approach to monitor oncogenes in cells is
Southern blots probed by the different oncogenes. In to assay for their gene products (e.g., tyrosine kinase
this way one can obtain some idea on the similarity of for c-src expression, epidermal growth factor receptor
the gross structure of the various fish oncogenes that for c-erb-B expression). Monoclonal antibodies

Read-Connole et al.: Oncogenes in Fish Tissues and Cell Lines 121

(Mabs) have been raised against synthetic peptides gene. It was also reported (van Beneden et al. 1988)
spanning the active site regions of the src and ras that the blots, when rehybridized to other viral
gene families and are available commercially. oncogenes, gave no indication of gross amplification
Nonidet P-40 (non-ionic detergent) lysates of cells or rearrangement. Clearly, cloning and sequencing
are cleared by high speed centrifugation and the pro- of the entire gene is necessary to show putative dif-
teins imm,unoprecitated with Mab mixtures. ferences.
Following electrophoresis on 10% SDS-polyacrylam- Similarly c-ras-related sequences were cloned from
ide gels, the bands are transferred to nitrocellulose the genomic libraries of the goldfish (Cerassius
membranes, and immunoblotted with an appropriate auratus) by Nemoto et al. (1986). Comparison of the
antibody. These western hybridizations, in which nucleotide sequences of one of these clones with
Mabs are used against oncogene products, can reveal those in mammalian c-ras genes showed extensive ho-
the size and level of expression of these proteins. mology to the- gene coding for the mammalian p2l.
protein (96% homology to the c-ras' protein). Subse-
quent work (Nemoto et al. 1987) involved
Oncogenes Detected in Fish Tissues sequencing and comparison of the first exon and its
flanking regions in the c-ras-related genes from nor-
A certain population of platyfish (Xiphophorus mal goldfish liver tissues with those from goldfish
maculatus) carry special genes (Tu or tumor genes) erythrophoroma cells cultured in vitro. No. differ-
for macrome Ian oph ores. When they are inter- ences were apparent in the first 245 nucleotides
specifically hybridized with swordtails (X. hellmi), the which covered the first exonic region and whose
F, offspring carrying these genes develop a length was identical to the first exonic region in
preneoplastic state. When the F, offspring are back- mammalian c-ras genes. Enhanced expression of c-ras
crossed with swordtails, a certain percentage of the gene was seen in the erythrophoroma cells.
backcrossed offspring develop a. heritable form of The high incidence of hepatomas present in winter
melanoma with a characteristic inheritance pattern flounder Pseudopleuronectes ame7icanus in Boston har-
(Ozato and Wakamatsu 1983; Anders et al. 1984). Ex- bor has been associated with the high levels of
pansion of these classic and elegant genetic studies polycyclic aromatic hydrocarbons (PAHs) found in
involves the use of modern biotechnology techniques the harbor's sediment. Genomic DNA from the liver
to define the molecular basis of the melanoma for- tumors produced foci in the NIH-3T3 transfection
mation. For example, elevated levels of cellular src assay and a c-rask oncogene was identified in a
mRNA and phosphoprotein (pp 60) kinase activity transformant derived from one of the tumors
are detected in melanomas or tumors induced by car- (McMahon et al. 1988). Sequencing of c-ras@ se-
cinogens in Xiphophorus (Schard et al. 1985) and quences from the tumor and from the transformants
more recent research indicates that the signal to start indicated the oncogene to be of fl6under origin. Se-
the events leading to melanoma development may quencing of the tumor cell DNA amplified by
come from the platyfish erbB region (Zechel et al. polymerase chain reaction (McMahon et al. 1987)
1988). They have named this gene Xmrk. An addi- showed GC to AT or GC to TA conversions in ,the
tional copy of the Xmrk gene is found linked to either 12th codon of this gene (McMahon et al. 1990). Such
sex chromosome of the platyfish when hybridized to mutations in the 12th codon can activate c-ras proto-
a probe specific for this region (Schartl 1990). The oncogenes to oncogenic forms (Barbacid 1987).
cellular myc (c-myc) gene has been cloned from rain- DNA samples from livers taken from fish at a less
bow trout Oncorhynchus mykiss and sequenced (van polluted site did not transform NIH-3T3 cells, and
Beneden et al. 1986). The results of this work were only wild-type sequences (GGT) were seen at the 12
used for evolutionary comparison with other verte- codon of c-rask.
brate c-myc genes. Using the rainbow trout c-?nyc A cellular 53 kilo dalton nuclear phosphoprotein,
clone as a -probe, DNA was examined from various denoted p53, was discovered more than a decade ago
fish tumors, including hepatocellular carcinomas in- in SV40 virus transformed mouse cell lines (Chang et
duced in the medaka Oryzias latipes by di- al. 1979). Because p53 was found complexed to the
methylnitrosamine. Compared with DNA isolated transformation antigen (T antigen) of SV-40 virus,
from normal tissues, there was no apparent increase and because the p53 gene was thought by many inves-
in the intensity of hybridizing bands or differences in tigators to function as a dominantly acting oncogene,
restriction patterns noted in Southern blots (van it has been intensively investigated during the past
Beneden et al. 1988). However such experiments decade. Current work on mutations in the p53 gene
would not detect mutations, which could be detected indicates that the wild-type gene product actually
only by cloning and sequencing the entire c-myc functions like a tumor suppressor gene (Finlay et al.

122 NOAA Technical Report NNEFS I I I

Table I
Oncogenes Detected in Fish Tissues.

Transforms
Fish Species Tissue Tested Oncogene NIH-3T3 Cells Reference

Platyfish - swordtail Hereditary melanoma Xmrk (tu) Schard et al. 1985
hybrid (Xiphophorus sp.) Wittbrodt et al. 1989

erb-B Zechel et al. 1988

Rainbow Trout Normal liver c MYC van Beneden et al. 1986
(Oncorhyncus mykiss)

Goldfish Normal liver ras Nemoto et al. 1986
(Carassius auratus)

Winter flounder Liver tumor ras' + McMahon et al. 1988, 1990
(Pseudopleuronectes
americanus)

Tomcod Liver tumor raSk + Wirgin et al. 1989
(Microgadus tomcod)

Northern pike External ? + van Beneden et al. 1990
(Esox lucius) lymphomas

1989). Mutations in the p53 gene are frequently de- Southern blots with v-erb-b, v-src, and v-ras' revealed a
tected in diverse human tumor types (Nigro et al. striking similarity in the banding patterns of homolo-
1989), and currently it is believed that p53 gene mu- gous sequences between the rainbow trout 0. mykiss
tations play an important role in the development of and the chinook salmon 0. tshawytscha cell line. This
many common human cell malignancies. The hy- suggests that, as expected, these genes in the rainbow
pothesis is that the mutated p53, which develops trout and chinook salmon are more closely linked in
during the process of tumorigenesis, binds to the an evolutionary sense than they are to the homolo-
wild-type p53 gene product, creating an inactive com- gous genes in the other fish species studied. A very
plex. Further loss'of growth control can occur when high degree of homology has been shown in the pro-
the wild-type allele is deleted. Thus, p53, like the tamine genes of chum salmon (0. keta) and rainbow
retinoblastoma gene, is under intensive study by trout (Moir and Dixon 1988).
many investigators as a potential tumor suppressor Many of the monoclonal antibodies raised against
gene (Sager 1989). The tumor suppressor genes are mammalian viral or cellular oncogene protein prod-
wild-type alleles of genes that play regulatory roles in ucts recognized fish proteins in Western blots.
cell proliferation and differentiation, and it is their Immunoprecipitation followed by Western blotting is
loss or inactivation that is oncogenic. The p53 gene is considered a highly sensitive and discriminatory pro-
highly conserved in vertebrates including fish (Smith cedure, especially when monoclonal antibodies are
et al. 1988). used that recognize discrete and specific epitopes on
the proteins of interest. The apparent recognition of
fish proteins by this technique shows that these re-
-Oncogenes Retected- iry Fish-Cell-Lines- gions of the oncoproteins may be well -conserved
from viruses to mammals and fish.
A number of oncogenes have been described in some Both the c-ras' and c-rask gene products were de-
of the more commonly used fish cell lines (Read- tected in all the fish cells, showing that these proteins
Connole et al. 1990). These have been detected by are conserved in vertebrates and may be essential. It
Southern, Northern, and Western blots (Table 2). is possible, however, that the c-ras' and c-ras' antibod-
The presence of many bands in Southern blots sug- ies recognized the same fish proteins, but the
gests that, as in mammals, some of the fish monoclonal antibodies used were raised against dif-
proto-oncogenes may exist as families of homologous ferent pepticles representing regions of the two viral
sequences that share common functions. Probing proteins and did not cross-react.

Read-Connole et al.: Oncogenes in Fish Tissues and Cell Lines 123

Table 2
Oncogenes Detected in Fish Cell Lines.

Detection Method Oncogene(s) Detected Cell line (s)

Southern blots rash, raf, BB, RTG-2,
Using mammalian erb-B, src EPC, CHSE-214
viral onc probes p53 EPC

Western blots BB, EPC,
Using Mabs against c-myc, c-ab4 c-fos, CHSE-214
human and viral v-ras, p53 (murine), v-ras,
oncogene products v-raf v-sre
Northern blots rash High expression in BB and EPC
Using mammalian Moderate expression in RTG
viral onc probes Low in CHSE-214

abl High expression in CHSE-214
Moderate in RTG

src, abl, myc, rask BB, EPC,
RTG-2, CHSE-214

p53 EPC, CHSE-214

*BB = Brown bullhead catfish line (Wolf and Quimby 1969); RTG-2 rainbow trout gonad (Wolf and Quimby 1962); EPC
epithelioma papulosurn cyprini (Fijan et al. 1983); and CHSE-214 chinook salmon embryo (Fryer et al. 1965).

Significance ing to establish tissue culture cell lines from these
tumors with the intent of comparing the genomic
The mechanisms by which normal cells acquire the DNA of the tumor cells to that of the normal skin
malignant phenotype form the central focus of can- tissue and the brown bullhead (BB) cell line to deter-
cer research. Transforming genes have been detected mine if any of the potential fish oncogenes are
in a variety of tumors, including human tumors, and altered or expressed differently in the tumor cells. If
in cell lines derived from tumors by their ability to mutations identified in specific fish oncogenes can be
tranform NIH-3T3 mouse cells in culture. Genetic al- shown to result from chemically-induced DNA dam-
ternations also accompany tumorigenesis in feral age, then such mutations could serve as indicators
populations of aquatic organisms exposed to environ- for environmental disease.
mental carcinogens. Any unambigious signal
regarding oncogene or tumor-repressor-gene expres-
sion that correlates with tumor formation could serve Citations
as the basis for a test system for evaluating carcino-
genic chemicals in the environment. Changes in the Anders, F., M. Schard, A. Barkenow, and A. Anders.
structure (mutation) of one or more oncogenes, 1984. Xiphophorus as an in vivo model for studies on normal
their amplification, or the level of expression of their and defective control of oncogenes. Adv. Cancer Res. 42:
191-275.
mRNA are all potential indicators of tumorigenic ini- -Barbacid, M.
tiation and progression. 1986. Involvement of ras oncogenes in the initiation of car-
The number of genes homologous to either the cinogen-induced tumors. Int. Symp. Princess Takamatsu
viral oncogenes or the mammalian cell counterparts Cancer Research Fund. 17:43-53.
found in fish genomes suggests that these genes are 1987. Ras genes. Annu. Rev. Biochem. 56:779-827.
Baum, EX, and G.A. Berbernitz.
highly conserved evolutionarily, and therefore their 1990. K-ras oncogene expression in Xenopus laevis. Oncogene
role in both normal cells and those cells in various 5:763-767.
stages of tumorigenic transformation may be com- Bishop,J.M.
mon to all vertebrates. The brown bullhead Ictalurus 1983. Cellular oncogenes and retroviruses. Annu. Rev.
nebulosus is of particular interest to us because of the Biochem. 52:301-354.
1985. Viral oncogenes. Cell 42:23-38.
high prevalenceof tumors that occur in them during Chang, C., D.T. Simmons, M.A. Martin, and P.T. Mora.
the late spring and summer in certain areas of the 1979. Identification and characterization of new antigens
Chesapeake Bay (unpubl. data). We are now attempt- from SV,-transformed mouse cells. J. Virol. 31:463-471.

124 NOAA Technical Report NWS I I I

Fijan, N., D. Sulimanovic, M. Bearzotti, D. Muzinic, L. Zwillenberg, Read-Connole, E., C.D. Smith, and F.M. Hetrick.
S. Chilmonezykj Vautherot, and P. de Kinkelin. 1990. Nucleotide sequence homologous to mammalian
1983. Some properties of the epithelioma papulosum cyprini proto-oncogenes and their expression in fish cell lines. J.
(EPC) cell line from carp. Annales Virologie (Institut Aquat. Animal Health 2:77-84.
Pasteur) 134:207-220. Sager, R.
Finlay, C.A., P.W. Hinds, and Aj. Levine. 1989. Tumor suppressor genes: the puzzle and the
1989. The p53 Proto-oncogene can act as a suppressor of promise. Science 246:1406-1412.
transformation. Cell 57:1083-1090. Schard, M.
Friend, S.H., R. Bernards, S. Roge1j, R.E. Weinberg, J.M. Rapaport, 1990. Homology of melanoma-inducing loci in Xiphophorus.
D.M. Albert, and T.P. Drya. Genetics 126:1083-1091.
1986. A human DNA segment with properties of the gene Schard, M., C.R. Schmidt, A. Anders, and A. Barkenow.
that predisposes to retinoblastoma and osteosar- 1985. Elevated expression of the cellular sre gene in tumors
coma. Nature 323:643-646. of different etiology in Mphophorus. Int. J. Cancer 36:199-
Fryer, J.L., A. Yusha, and K-S. Pilcher. 207.
1965. The in vitro cultivation of tissue and cells of Pacific salmon Schwab, M., J. Ellison, M. Busch, W. Rosenau, H. Varmus, and
and steelhead trout. Ann. NX Acad. Sci. 126:566-586. J.M. Bishop.
Johnson, B.E., D.C. lhde, RX Makuch, A.F. Gazdar, andj. Minna. 1984. Enchanced expression of the human gene N-myc con-
1987. Myc family oncogene amplification in tumor cell lines sequent to amplification of DNA may contribute to malig-
established from small cell lung cancer patients and its rela- nant progression of neuroblastorna. Proc. Nad. Acad. Sci.
tionship to clinical status and course. J. Clin. Invest. 81:4940-4944.
79:1629-1634. Schwab, M., R. OehImann, S. Bruderlein, and Y. Wakamatsu.
Klein, G. and E. Klein. 1989. Amplified DNA in cells of genetic melanoma of
1985. Evolution of tumors and the impact of molecular Xiphophorus. Oncogene 4:139-144.
biology. Nature 315:190-195. Smith, C.A., M.J. Louis, and F.M. Hetrick.
Lee, W.H., R. Bookstein, F. Hong, LJ. Young, JX Shew, LY.M.P. 1988. A sequence homologous to the mammalian p53
Lee. oncogene in fish cell lines. J. Fish. Dis. 11:525 -530.
1987. Human retinoblastorna susceptibility gene: cloning, Smith, C.A.D., Dj. Winterbourne, VW. McFarland, and P.T. Mora.
identification, and sequence. Science 235:1394-1399. 1987. Changes in Heparan Sulfate pattern but not in
Louis, Mj., VW. McFarland, P. May, and P.T. Mora. oncogene expression correlate with tumor growth in spon-
1988. The phosphoprotein p53 is down-regulated post- taneous transformation of cells. Oncogene Research
transcriptionally during ernybrogenesis in verte- 1:325-341.
brates. Biochimica. Biophysica Acta. 956:395-402. Stehelin, D., H.E. Varmus,J.M. Bishop, and P. Vogut.
McMahon, G., E. Davis, and G.N. Wogan. 1976. DNA related to the transforming genes of avian sar-
1987. Characterization of c-Ki-ras oncogene alleles by direct coma virus is present in normal avian DNA. Nature
sequencing of enzymatically amplified DNA from carcino- 260:170-173.
gen-induced tumors. Proc. Natl. Acad. Sci. (USA), van Beneden, R.J., D. R. Watson, T.T. Chen, J.A. Lautenberger, and
84:4974-4978. T.S. Papas.
McMahon, G., Lj. Huber, JJ. Stegeman, and G.N. Wogan. 1986. Cellular myc (c-?nyc) in fish (rainbow trout): its rela-
1988. Identification of a c-Ki-ras oncogene in a neoplasm iso- tionship to other vertebrate myc genes and to the transform-
lated from winter flounder. Marine Environ. Res. 24:345- ing genes of the MC29 family of viruses. Proc. Natl. Acad.
350. Sci. (USA) 83:3698-3702.
McMahon, G., L.J. Huber, Mj. Moore, Jj. Stegeman, and 1988. Teleost oncogenes: evolutionary comparison to
G.N. Wogan. other vertebrate oncogenes and possible roles in teleost
,1990. Mutations in c-Ki-ras oncogenes in diseased livers of neoplasms. Mar. Environ. Res. 24:339-345.
winter flounder from Boston Harbor. Proc. Natl. Acad. van Beneden, Rj., K.W. Henderson, D.G. Blair, T.S. Papas, and
Sci. (USA) 87:841-845. H.S. Gardner.
Moir, R.D., and G.H. Dixon. 1990. -Oncogenes in hematopoietic and hepatic fish
1988. Characterization of a protamine gene from chum neoplasms. Cancer Res. 50:5671s-5674s.
salmon Oncorhynchus keta. J. Mol. Evol. 27:8-16. Wirgin, L, D. Currie, and Sj. Garte.
Nemoto, N., K. Kodama, A. Tazawa, Prince Masahito, and 1989. Activation of the K-ras oncogene in liver tumors ' of
T. Ishikawa. Hudson River tomcod. Carcinogenesis 10:2311-2315.
1986. Extensive sequence homology of the goldfish ras gene Wittbrodt, J., D. Adam, B. Malitschek, W. Maueler, F. Raulf,
to mammalian genes. Differentiation 32:17-23. A. Telling, S.M. Robertson, and M. Schard.
Nemoto, N., K. Kodama, A. Tazawa, J. Matsumoto, Prince 1989. Novel putative receptor tyrosine kinase encoded by the
Masahito, and T. Ishikawa. melanoma-inducing Tu locus in Xiphophorus. Nature
.1987. Nucleotide sequence comparison of the predicted first 341:415-421.
exonic region of goldfish ras gene between normal and neo- Wolf, IL, and M. Quimby.
plastic tissues. J. Cancer Res. Clin. Oncol. 113:56-60. 1962. Established eurythermic line of fish cells in vitro. Sci-
Nigro, J.M., S. Baker, A. Preisinger, et al. ence (Washington, D.C.) 135:1065-1066.
1989. Mutations in the p53 gene occur in diverse human 1969. Fish cell and tissue culture. In Fish physiology, vol. 3
tumour types. Nature 342:705-708. (W.S. Horr and D.J. Randall, eds.), p. 253-305. Acad.
Ozato, K_ and Y. Wakamatsu. Press, N.Y.
1983. Multi-step genetic regulation of oncogene expression Zechel, C., U. Schleenbecker, A. Anders, and F. Anders.
in fish hereditary melanoma. Differentiation 24:181-190. 1988. v-erb-B sequences in Xiphophorus that map to melanoma,
Powers, D.A. determining Mendelian loci and overexpress in a mela-
1989. Fish as model systems. Science 246:352-358. noma cell line. Oncogene 3:605-617.

Streptococcal Infection in Cultured Yellowtail

HIROSHISAKO

Nansei National Fisheries Research Institute
Ohno, Saeki
Hiroshima, 739@04, Japan

ABSTRACT

Streptococcal infection is widespread in yellowtail (Seriola quinqueradiata) cultured in
southwestern Japan and results in extensive losses. It is urgent that control measures be
established against this disease. This paper presents the etiology, prophylaxis, and treat-
ment of streptococcal infections in cultured yellowtail.

Introduction clearly distinguishable by their biochemical and
physiological characteristics (Table 1; Sako, unpubl.
Yellowtail (Seriola quinqueradiata) culture developed data). The species reported by Kusuda et al. is gener-
rapidly in the 1960s, mainly in southwestern Japan. ally called alpha-hemolytic Streptococcus sp. Almost all
By 1988, the farming of yellowtail was being practiced of the streptococci isolated from yellowtail were con-
in 27 prefectures (Fig. 1), the total production was firmed to be the alpha-hemolytic Streptococcus sp.
about 166,000 metric tons, or 68% of the total pro- (Kitao 1982). Recently, this species was transferred to
duction of marine fish cultured in Japan (Fig. 2). genus Enterococcus (Riichi Kusuda, Kochi Univ., pers.
The incidence of diseases in cultured yellowtail is commun., April 1990). Another species isolated from
gradually increasing as the culture of this fish be- a yellowtail juvenile in 1976 by Minami et al. (1979)
comes more popular. About 30 diseases have been displayed beta-hemolysis on sheep blood agar, so it
identified so far. Figure 3 shows the loss of cultured has been called beta-hemolytic Streptococcus sp. This
yellowtail due to disease in 1988; the cost to fish species is classified with pyogenic streptococci based
farmers was about 11,300 million yen. Bacterial dis- on its biochemical characteristics. Its biochemical
eases are responsible for the majority of the losses, characteristics also agreed well with those of S. iniae
whereas those due to viruses and parasites are minor. ATCC 29178 isolated from the Amazon freshwater
acterial diseases, streptococcosis was re- dolphfin, Inia geof
Among the b 'frensis (Pier and Madin 1976). The
sponsible for the highest losses (64%), pasteurellosis last species is a nonhemolytic Streptococcus reported
ranking second'(21%). Thus, with yellowtail culture, by Iida et al. (1986), which has biochemical charac-
it is very important to control the incidence of strep- teristics similar to those of S.agalactiae and has the
tococcal infection. Lancefield grouping antigen B, type lb.
The author describes here some aspects of strepto- A species that shows the same characteristics as al-
coccal infection in cultured yellowtail with special pha-hemolytic Streptococcus sp. was also isolated from
emphasis on the etiology of the causative bacteria, cultured yellowtail in the Republic of Korea (Park et
prophylaxis, and methods used to treat the disease. al. 1987). The nonhemolytic, group B, type Ib Strepto-
coccus was also isolated from several marine fishes in
the United States (Plumb et al. 1974).
Causative Bacteria

Streptococcal infection in cultured yellowtail was first Host Range
identified by Kusuda et al. in 1974 (Kusuda et al.
1976). Since then, three species of fish pathogenic Large numbers of Enterococcus sp. and S. iniae have
Streptococcus have been reported. These bacteria are been isolated from wild and cultured marine fishes as

125

126 NOAA Technical Report NPM I I I

their minced meats, are a major feed for yellowtail, it
is possible that they are the origin of streptococcal
Total Woduction infection in cultured fish.
166,000 niet& tons Pathogenicity

Eac h of the Streptococcus spp. and their strains ex-
hibit different degrees of virulence. Enterococcus sp.
produces both intra- and extracellular toxins
(Kimura and Kusuda 1979). In artificially induced in-
fections, even small amounts of Enterococcus sp. cells
10 killed yellowtail (Taniguchi 1982a). However, S. iniae
killed yellowtail only occasionally although it fre-
quently caused vertebral deformity (Sako, unpubl.
data).
0 10,000 - 40,000 Streptococci is an indigenous bacteria in warm-
1,000 - 9,999 blooded animals and grows rapidly at 30-35' C. S.
- 999 iniae isolated from yellowtail have been confirmed to
mtric tons be pathogenic to mice by intraperitoneal injection.
However, Enterococcus sp. and nonhemolytic group B,
type lb Streptococcus isolated from yellowtail were
Figure I nonpathogenic to mice (Sako, unpubl. data).
Yellowtail production in Japan by prefecture, 1988. Source:
Statistics and Information Department, Ministry of Agricul-
ture I Forestry, and Fisheries, Japan. Diagnosis

well as in small quantities from cultured freshwater Major clinical signs caused by Enterococcus sp. are
fishes in Japan (Table 2; Kusuda et al. 1976; Kusuda exophthalmos, petechiae, and ulceration on the in-
et al. 1978; Minami 1979; Minami et al. 1979; Kitao et side of the opercula, and congestion of the pectoral
al. 1981; Ohnishi and Jo 1981; Kusuda and Kawai and caudal fins. Signs such as congestion and
1982; Kusuda et al. 1982; Ogawa et al. 1982; Yasunaga haernorrhagia are also found in the intestine, liver,
1982; Hatai et al. 1983; Nakatsugawa 1983; Kaige et spleen, and kidney (Kusuda et al. 1976). Epicarditis is
al. 1984; lida et al. 1986; Sakai et al. 1986; Atsuta et also seen often. Pathogenic Streptococcus is frequently
al. 1990). Because sardines and anchovies, as well as isolated from the brains of diseased yellowtail

JAPANESE FLOUNDER

JAPANESE HORSE MACKEREL
@@the@rs

COHO SALMON
YELLOWTAILS
242,000
metric tons

SEA BREAMS Figure 2
Marine fish culture produc-
tion for 1988. Source: Sta-
166,000 metric tons
tistics and Information
Department, Ministry of Ag-
riculture, Forestry, and Fish-
eries,japan.

Sako: Streptococcal Infection in Cultured YeHowtail 127

Nocardlosis others
Vibriosis

COMO' ion*

119300 Streptococcosis
Figure 3 F_ million yen
Loss of cultured yellowtail to Pasteurellosis
diseases in 1988. *This classifi-
cation contains cases that the 7,270 million yen
cause of a disease are two or
more: e.g. mixed infection.
Source: Research Division, Fish-
eries Agency ofjapan.

(Shiomitsu 1982; Kaige et al. 1984; lida et al. 1986). Prevention
Immunological identification using fluorescent an-
tibody techniques is also useful for the rapid It was reported that fish meats used as feed for yel-
identification of yellowtail pathogenic streptococci lowtail have fish pathogenic Streptococcus (Minami
(Kusuda and Kawahara 1987). 1979; Yasunaga 1982). The organisms grow quickly in
A rapid identification system is also commercially the minced meat of fish,within several hours under
available for the identification of fish pathogenic optimal temperature (Fig. 4; Sako, unpubl. data). In
streptococci. Three species of streptococci from yel- laboratory experiments, it was demonstrated that yel-
lowtail can be easily distinguished using API 20 Strep lowtail can be infected orally by Streptococcus through
System (BIO MERIEUX S.A.) within 4 hours (Table feed such as minced meat (Taniguchi 1982a). In or-
3; Sako, unpubl. data). der to prevent contamination by the pathogen and to

Table 1
Biochemical characteristics of ot, P and nonhemolytic Streptococcus sp. isolated from yellowtail.

ot-hemolytic P-hernolytic nonhernolytic
Streptococcus sp. Streptococcus sp. Strepwoccus sp.
Characteristics (Enterococcus sp.) (S. iniae) (S. agalactiae)

Hemolysis U, -
Voges-Proskauer reaction + +
Esculin hydrolysis + +
Starch hydrolysis +
Hippurate hydrolysis +
Pyrrolidonylarylarnidase + +
P-galactosidase - +
Alkaline phosphatase - + +
Sensitivity to bacitracin - + +
cAMP test - +
Lancefield group antigen not A, B, C, D, F, G not A, B, C, D, F, G, BIbb

'No reaction at least against the antisera of A-G.
'Blb denotes type of group antigen.

128 NOAA Technical Report NMFS I I I

Table 2
Fish -species in Japan from which a pathogenic streptococci was isolated (Kusuda et al. 1976; Kusuda et al. 1978;
Minami 1979; Minami et al. 1979; Ohnishi and Jo 1981; Kitao et al. 1981; Yasunaga 1982; Kusuda and Kawai 1982;
Kusuda et al. 1982; Ogawa et al. 1982; Nakatsugawa 1983; Hatai et al. 1983; Kaige et al. 1984; Iida et al. 1986;
Sakai et al. 1986; Atsuta et al. 1990).

Marine fish Freshwater
Isolates Cultured Wild cultured fish

Enterococcus sp. Yellowtail (Seriola quinqueradiata) Sardine (Sardinops melanosticta) Eel (Anguillajaponica)
Japanese horse mackerel (Trachurusjaponicus) Anchovy (Engraulis japonica)
Purplish amberjack (Seriola dumerils) Round herring (Etrumeus microps)
Goldstriped amberjack (Seriola lalandi) Sand lance (Ammodytes personatus)
Largescale blackfish (Girella punctata) Chub mackerel (Scomber japonicus)
Black scraper (Thamnaconus modestus) Moon dragonets (Repomucenus lunatus)
Black scraper (Thamnaconus modestus)
Multicolorfin (Halichoeres poecilopterus)
Silver jewfish (Pennahia argentata)
Red sea bream (Pagrus major)

Streptococcus Yellowtail (Seriola quinqueradiata) Yellowtail (Seriola quinqueradiata) Ayu (Plecoglossus altivelis)
iniae Japanese horse mackerel (Trachurusjaponicus) Japanese horse mackerel (Trachurus japonicus) Rainbow trout
Japanese flounder (Paralichthys olivaceus) Chub mackerel (Scomber japonicus) (Oncorhynchus mykiss)
Knifejaw (Oplegnathus fasciatus) Sardine (Sardinops melanosticta) Amago salmon
Rabbit fish (Siganus fuscescens) Red sea bream (Pagrus major) (Oncorhynchus rhodurus)
Largescale blackfish (Girella punctata) Tilapia (Tilapia nilotica)
Triggerfish (Aluteres monoceros)
Jacopever (Sebastes schlegeli)
Coho salmon (Oncorhynchus kisutch)

Streptococus Yellowtail (Seriola quinqueradiata)
agalactiae

slow its growth, the following techniques are recom- Therapy
mended: 1) washing feed fish in uninfected water, 2)
feeding fish without defrosting, and 3) feeding pro- It is not easy to cure yellowtail that are in the ad-
cessed pellets (Taniguchi 1982b). vanced stages of streptococcal infection. Minimal
Yellowtail can acquire immunity after streptococcal inhibitory concentrations (MIC) of antimicrobial
infection (Kusuda and Takagi 1983). Indeed, many agents against Streptococcus from yellowtail are shown
attempts have been made to prevent streptococcal in- in Figure 5 (Sako, unpubl. data). Sensitivities to sev-
fection using bacterin. lida et al. (1982) reported on eral drugs are different between Enterococcus sp. and
the efficacy of vaccination for control of streptococ- S. iniae, but both species have high sensitivities to
cal infection. penicillins, macrolides, tetracyclines, and lincomycin.

Table 3
Differential characteristics among Enterococcus sp., Streptococcus iniae, and Streptococcus agalactiae isolated from
yellowtail by API 20 Strep System incubated at 35' C for four hours (Sako, unpubl. data). VP = Voges-Proskauer reaction;
HIP = Hippurate hydrolysis; ESC = Esculin hydrolysis; PYRA = Pyrrolidonylaryl amidase; PAL = Alkaline phosphatase.

Test (characteristics)
Species VP HIP ESC PYRA PAL

Enterococcus sp. + + +
Streptococcus iniae + + +
Streptococcus agalactiae + + +

Sako: Streptococcal Infection in Cultured Yellowtail 129

Table 4
7 301C Drugs used to cure yellowtail streptococcal infection
a 6 in Japan. Source: Fisheries Agency ofJapan.
5 Drugs Daily dose (mg/kg body weight)
201C Alkyl trimethyl ammonium 50
4 calcium oxytetracycline
Doxycycline hydrochloride 20-50
3 10 V Spiramycin embonate 25-40
Kitasamycin 80
2 Erythromycin 25-50
josamycin 30-50
0 3 6 9 is Oleandomycin polystyrene 25
hr sutfonic acid
Lincomycin hydrochloride 20-40
Florfenicol 40
Figure 4
Growth of Streptococcus iniae in minced meat of sand
lance (Sako, unpubl. data).

At present, two kinds of tetracyclines, five kinds of Citations
macrolides, lincomycin, and florfenicol are permit-
ted as the chemotherapy of streptococcal infection in Atsuta, S.j Yoshimoto, M. Sakai, and M. Kobayashi.
1990. Streptococcosis occurred in the pen-cultured coho
yellowtail in Japan (Table 4). Recently, streptococci salmon, Oncorhynchus kisutch. Suisanzoshoku 38:215-219.
that have aquired resistance to macrolides and tetra- (Injapanese.)
cyclines were found in many yellowtail farms (M. Hatai, K., S. Yasumoto, J. Tsukahara, E. Hirakawa, N. Yasunaga,
Fukudome, Kagoshima Pref. Fish. Exp. Sta., pers. and T. Ichiki.
1983. Drug sensitivity of various fish-pathogenic bacteria iso-
commun., July 1990). Therefore, chemotherapy lated from cultured fish in Nagasaki Prefecture,
treatments for yellowtail infected with drug-resistarit 1982. Bull, Nagasaki Pref. Inst. Fish, 9:13-22. (In Japa-
streptococci are likely to be ineffective. nese; English abstr.)

Vx
:0.00625: 0.0125: 0.025 0.05 0.10 0.20 0.39 0.78 1.56 3.13 6.25 12.5 25 so 100 a200

Peni c i I I i n G 0.
Ampi C i I I i n 0
0
Erythromycin 0:
a 0
Oleandomycin
Kita,samycin 0 0
Spiramycin 0 0 0
Jor.amycin 0 49
Chloramphenicol 0 0

Oxytetracycline
0 0

Doxycycline 0 0
Streptomycin i 0 0: 0

Lincomycin
0 0::

Colistin 00
Sulfadimethoxine so,
Sodium nif :.0 0
ar-
styrenate
Trimethoprim 0 0 0

Oxolinic acid
:49 0

Figure 5
z

Sensitivities of Enterococcus sp. and Streptococcus iniae isolated from yellowtail against various drugs: 0 Enterococcus sp.; 0
Streptococcus iniae (Sako, unpubl.data).

130 NOAA Technical Report NAM I I I

lida, T., H. Wakabayashi, and S. Egusa. Minami, T.
1982. Vaccination for control of streptococcal disease in cul- 1979. Streptococcus sp., pathogenic to culture7d yellowtail, iso-
tured yellowtail. Fish Pathol. 16:201-206. (In Japanese; En- lated from fishes for diets. Fish Pathol. 14:15-19. (In Japa-
glish abstn) nese; English abstr.)
lida, T., K_ Furukawa, M. Sakai, and H. Wakabayashi. Minami, T., M. Nakamura, Y Ikeda, and H. Ozaki.
1986. Non-hemolytic Streptococcus isolated from the brain of 1979. A beta-hemolytic Streptococcus isolated from cultured
the vertebral deformed yellowtail. Fish Pathol. 21:33-38. yellowtail. Fish Pathol. 14:33-38. (In Japanese; English
(Injapanese; English abstr.) abstr.)
Kaige, N., M. Miyazaki, and S. Kubota. Nakatsugawa, T.
1984. The pathogen and the histopathology of vertebral de- 1983. A Streptococcal disease of cultured flounder. Fish
formity in cultured yellowtail. Fish Pathol. 19:173-179. (In Pathol. 17:281-285. (Injapanese; English abstr.)
Japanese; English abstr.) Ogawa, S., N. Yasunaga, EL Hatai, and E. Hirakawa.
Kimura, H., and R. Kusuda. 1982. A beta-haemolytic Streptococcus isolated from epizootics
1979. Studies on the pathogenesis of streptococcal infection of cultured seawater fishes. Bull. Nagasaki Pref. Inst. Fish
in cultured yellowtail Se?iola spp.: effect of the cell-free cul- 8:81-90. (Injapanese; English abstr.)
ture on experimental streptococcal infection. J. Fish Dis. Ohnishi, K., and Y. Jo.
2:501-510. 1981. Studies on streptococcal infection in pond-cultured
Kitao, T. fishes-1. Characteristics of a beta-hemolytic streptococcus
1982. The methods for detection of Streptococcus sp., caus- isolated from cultured ayu and amago in 1977-1978. Fish
ative bacteria of streptococcal disease of cultured yellowtail Pathol. 16:63-67. (In Japanese; English abstr.)
(Seriola quinq Iueradiata), especially their cultural, biochemi- Park, S.W., D.S. Sim, and Y.C. Ybo.
cal and serological properties. Fish Pathol. 17:17-26. (In 1987. Pathogenic Streptococcus sp. isolated from an epizootic
Japanese; English abstr.) of cultured yellowtail. Seriola quinqueradiata. Bull. Nat.
Kitao, T., T. Aoki, and R. Sakoh. Fish. Res. Dev. Agency 40:75-79. (In Korean; English abstr.)
1981. Epizootic caused by 0-haemolytic Streptococcus species Pier, G.B., and S.H. Madin.
in cultured freshwater fish. Fish Pathol. 15:301-307. (In 1976. Streptococcus iniae sp. nov., a beta-hemolytic streptococ-
Japanese; English abstr.) cus isolated from an Amazon freshwater dolphin, Inia
Kusuda, R., and K. Kawai. geofftensis. Int. J. Syst. Bacteriol. 26:545-553.
1982. Characteristics of Streptococcus sp. pathogenic to Plumb,J.A.,J.H. Schachte, and].L. Gaines.
yellowtail. Fish Pathol. 17:11-16. (In Japanese; English 1974. Streptococcus sp. from marine fishes along the Alabama
abstr.) and Northwest Florida Coast of the Gulf of Mexico. Trans.
Kusuda, R., and S. Takagi. Am. Fish. Soc. 103:358-361.
1983. Antibody production against Streplococcus sp. in natu- Sakai, M., S. Atsuta, and M. Kobayashi.
rally infected yellowuail Se?iola quinqueradiala. Rep. USA. 1986. A streptococcal disease of cultured jacopever Sebastes
Mar. Biol. Inst. Kochi Univ. 5:21-28. (In Japanese; English schlegeli. Suisanzoshoku 34:171-177. (In Japanese; English
abstr.) abstr.)
Kusuda, R., and E. Kawahara. Shiomitsu, K.
1987. Direct and indirect fluorescent antibody identification 1982. Isolation of Streptococcus sp. from the brain of cultured
of yellowtail pathogens. Bull. Jpn. Soc. Sci. Fish. 53:389- yellowtail. Fish Tathol. 17:27-31. (In Japanese; English
394. (Injapanese; English abstr.) abstr.)
Kusuda, R., K Kawai, T. Toyoshima, and I. Komatsu. Taniguchi, M.
1976. A new pathogenic bacterium belonging to the genus 1982a. Experiment on peroral inoculation via food to in-
Streptococcus, isolated from an epizootic of cultured yel- duce yellowtail streptococcicosis. Bull. Jpn. Soc. Sci. Fish.
lowtail. Bull. Jpn. Soc. Sci. Fish. 42:1345-1352. (In Japanese; 48:1717-1720. (Injapanese; English abstr.)
English abstr.) 1982b. Studies on the streptococcicosis of yellowtail and its
Kusuda, R., 1. Komatsu, and K. Kawai. preventions in Kochi Prefecture. Fish Pathol. 17:55-59.
1978. Streptococcus sp. isolated from an epizootic of cultured (Injapanese; English abstr.)
eels. Buli.jpn. Soc. Sci. Fish. 44:295. Yasunaga, N.
Kusuda, R., K. Kawai, and T. Shirakawa. 1982. Occurrence of Streptococcus sp., a pathogen of cultured
1982. Serological study of Streptococcus sp. pathogenic to cul- yellowtail, in muscle of sardine for diets. Fish Pathol.
tured yellowtail. Bull. Jpn. Soc. Sci. Fish. 48:1731-1738. 17:195-198. (Injapanese; English abstr.)
(Injapanese; English abstr.)

Stress Induced Pathologies in Fish: The Cost of Stress

GARY P. MOBERG

Department ofAnimal Science
and
Fisheries Program
University of California, Davis
Davis, CA 95616

ABSTRACT

Pathology is the consequence of prolonged stress. Maintaining fish under aquaculture
conditions intensifies the problems of stress by adding the impact of stressoIrs that are
unique to culture conditions. Unlike terrestrial domestic animals, fish have not
benefitted from the genetic pressures of generations of domestication that favor those
individuals most suitable for culture. The physiological mechanisms by which stress in-
duces the development of pathology remain unknown, although it is recognized that one
of the major factors affecting health during stress is the response of the neuroendocrine
system, a stress responsive system that directly regulated growth, reproduction, and the
immune system. During stress, the neuroendocrine system shifts biological resources
from pre-stress activities to new functions at a biological cost to the fish. As the biologi-
cal cost of shifting these resources rises during stress, the fish is placed into a
prepathological state, rendering it vulnerable to the development of pathology. It is by
focusing on this biological cost of stress that it is possible to develop strategies to reduce
stress in aquaculture and to understand the biological basis for the development of
disease.

Introduction Practical considerations have dictated most culture
conditions with emphasis on such factors as the ease
Pathology is the consequence of prolonged stress. of cleaning, maximizing the number of fish, or using
While we can not identify the exact mechanisms re- any holding facility 'that is readily available. Little
sponsible for the development of pathologies during concern is given to the fish, at least until stress be-
stress, there is indisputable evidence that stress can comes so severe that pathology occurs. Certainly this
result in the development of a diverse number of pa- is an area of aquaculture that needs to be addressed,
thologies that include not only disease, but the loss especially as increasingly intensified culture condi-
of reproduction, the failure to grow normally, and tions are adopted.
even the development of abnormal, deleterious be-
haviors (Moberg 1985).
Fish in culture are especially at risk to the adverse Problem of Defining Stress
effects of stress. Unlike domesticated mammals, fish
have not benefitted from generations of genetic se- Although there is considerable evidence demonstrat-
lection for traits that would assist them in adapting to ing the disruptive, effects of stress on the well-being
the restraints of confinement. In essence, fish are still of fish (Schreck 1982; Pickering and Pottinger 1987;
wild animals, and when culture conditions do not du- Maule et al. 1989; Vijayan and Leatherland 1990), it
plicate their natural habitat, they are forced to make is surprising that there is not a better understanding
biological adjustments to survive. Exacerbating this of how stress actually leads to the development of
problem is the failure of aquaculturists to make any pathologies. One reason for the difficulty in discover-
serious effort to systematically define those 'culture ing the mechanisms involved is a lack of consistent
conditions that the fish would find least stressful. responses to stress. Most fish do not develop a pathol-

132 NOAA Technical Report NMFS I I I

ogy during stress. In fact, it is not possible to predict work see Pickering 1981 and Adams 1990). Because
which fish will become vulnerable to a pathology due these responses are so varied and complex, I will use
to stress. This is because stress, per se, is not necessar- a model of animal stress (Fig. 1) to organize current
ily harmful to an individual. Animals have developed concepts of stress biology and examine how the bio-
defenses for coping with stress. Stress is a part of life, logical cost of stress results in the development of
and for a species to be successful, it must have pathology. The development of this model has been
evolved biological defenses to stress. Therefore, stress discussed elsewhere (Moberg 1985, 1987), but, in
responses are not bad. In fact, they are desirable and brief, the model divides the biological response to
necessary. Pathology occurs only when the animal is stress into three major components: recognition of a
confronted with stress of such a magnitude that the threat to homeostasis, the stress response, and the
very biological responses that evolved for defense re- consequences of stress. It is the central nervous sys-
sult in a biological cost to the individual and renders tem that perceives a threat to homeostasis and
it vulnerable to pathology. To understand how the organizes the fish's biological defense. If a threat
biological cost of stress leads to the development of (also referred to as a stressor) is perceived, three gen-
pathology, it is necessary to first examine how fish eral types of biological responses are available to the
respond to stress. fish: behavioral, autonomic nervous system, and neu-
roendocrine system responses. In the wild, behavior
can be the most biologically efficient way for a fish
Model of the Stress Response to cope by simply allowing the animal to remove
itself frorn'the stressor. However, under the confine-
Numerous studies have identified the various biologi- ment of culture conditions, this option is severely
cal responses that fish use in responding t6 and limited. If the fish is unable to avoid the stressor,
coping with stress (for monographs summarizing this then the roles of the autonomic nervous and neu-
roendocrine systems become critical. While
activation of the autonomic nervous system repre-
sents an important way for a fish to avoid or cope
with a stressor, its effects are rapid in onset, of short
STIMULU duration, and relatively specific. For this reason, the
importance of the autonomic nervous system in in-
RECOGNITION OF A ducing stress-related pathologies is questionable. In
STASIS CDM"4"A@@
THREAT TO HOMEO
t, contrast to the autonomic nervous system, the by-
PERCEPTIION pophyseal hormones of the neuroendocrine system
OF sTRESSOR
undoubtedly play an important role in stress-in-
duced pathology (Moberg 1985). These hormones
have a widespread action on the animal and long-
ORGANIZATION OF
BIOLOGICAL DEFENSE lasting effects on such diverse biological functions
as reproduction, growth, metabolism, resistance to
diseases, and behavior. Each of these functions are
BIOLOGICAL RESPONSE vital to the fish's well-being.
STRESS RESPONSIE (Behavimal. Aoonoffk, When the central nervous system organizes the
NemwWocrlne)
biological defense to a stressor, it is a combination of
these three general biological systems that alters
CHANGEIN function. In mammals, we have found that a number
BIOLOGICAL FUNCTION of factors influence the pattern of this biological re-
I sponse, and we have found it impossible to predict
CONSEQUENCES GICA how the individual systems will respond (Moberg
OF STRESS PREPATHOLO 1985). While the data for fish is limited, there is evi-
STATE dence indicating that in fish there are also a number
4 of factors that can alter the stress response. For ex-
LEE I
I

M'4ELOPMENT ample, water quality and temperature (Pickering and
OF PATHOLOGY Pottinger 1987, 1989), genetics (Refstie 1982), and
social interactions (Peters et al. 1988) have all been
Figure I found to influence the stress response. Thus, mea-
Model for the response of animals to a stressful event surement of the three general biological systems have
(from Moberg 1985). not proved to be reliable as an indicator of stress

Moberg- Suvss Induced Pathologies in Fish: The Cost of Su-ess 133

(Moberg 1987). However, what is important to the The concept that there is a biological cost associ-
fish is not the pattern of the biological response but ated with coping with stress also explains how the
the resulting change in biological function-the ulti- effects of subclinical stressors can accumulate, result-
mate consequence of stress. ing in a significant stress that results in pathology
Regardless of which of the general biological sys- (Moberg 1985, 1992). Separately, none of the
tems the fish chooses in response to the stressor, the subclinical stressors would result in a significant ex-
result is a change in biological function. Depending penditure of the fish's resources, but combined, the
upon the appropriateness of the response, this subclinical stressors could cost the animal sufficient
change in biological function may alleviate, or even resources to induce a prepathological state and lead
eliminate the stressor. But the change in biological to the development of a pathology. For example,
function can also lead to the development of a Jarvi (1989) found a greater mortality rate in Atlantic
prepathological state which can eventually result in salmon smolts. (Salmo salar) that were exposed simul-
the development of a pathology (Fig. 1). For ex- taneously to both osmotic stress and the presence of
ample, if the change in biological function during predators than if the smolts experienced only one of
stress results in the suppression of the immune sys- the stressors. Because of the accumulation of biologi-
tem (a prepathological state), then the animal is cal costs, a series of apparently innocuous events can
vulnerable to pathogens, and the opportunity for dis- lead to' disease, loss in reproduction, diminished
ease (a pathological state) exists. The longer the fish growth, or mortality.
is stressed, the longer the immune system is sup-
pressed and the greater the opportunity for disease
to occur. However, disease is only one possible patho- Conclusion
logical state. Other examples would be the inability
to reproduce, abnormal behavior, or the failure to The biological cost of stress is the key to under-
grow normally. standing why fish expo@ed to stress develop
pathologies. For a fish to cope with a stressor, it-
must expend biological resources. The more pro-
Biological Cost of Stress longed the stressor or the greater the effort needed
to cope with a severe stressor, the greater the diver-
The change in biological function that occurs as part sion of biological resources, and thus the greater
of the stress response results in a biological cost to the biological cost of the stress. This diversion of
the fish. The fish's resources are diverted from such resources occurs at the expense of other biological
pre-stress activities as growth to new activities. For ex- functions, leading to a prepathological state where
ample, the glucocorticosteroid hormone cortisol the fish is at risk to the development of pathology.
secreted during stress will induce gluconeogenesis, For some fish, the biological cost of coping with a
diverting metabolic resources supporting such func- stressor is greater than for other fish and these are
tions as growth to the production of glucose. Barton the first to succumb to the stressor. As we manage
and Schreck (1987) found in juvenile steelhead fish in culture, every effort must be made to reduce
(Oncorhynchus mykiss) that acute stress reduced by stress and, as a result, lower the biological cost for
about 25% the amount of energy available for other fish living in culture.
activities. Likewise, the biological cost of responding
to a stressor may suppress the' ability of the immune
system to respond to pathogens, rendering the fish Citations
vulnerable to disease (Maule et al. 1989). It does not
matter which pattern of responses the animal Adams, S.M.
chooses; a change in biological function occurs that 1990. Biological indicators of stress in fish. Am. Fish. Soc.
imposes, a cost, whether or not the change in func- Symposium 8, Bethesda, Maryland, 191 p.
Barton, B.A. and C.B. Schreck.
tion is effective in helping the animal to cope with 1987. Metabolic cost of acute physical stress in juvenile
the stressor. steelhead. Trans. Am. Fish. Soc. 116:257-263.
Fortunately, most stressors last for only a brief time Jarvi, T.
and the biological cost of coping with them is rela- 1989. Synergistic effect on mortality in Atlantic salmon,
tively small. However, if a stressor is severe, if it Salmo salar smolt caused by osmotic stress and presence of
predators. Environ. Biol. Fishes 26:149-152.
persists, or if the fish experiences a series of stressors, Maule, A.G., R.A. Tripp, S.L. Kaattari, and C.B. Schreck.
the resulting biological cost may be sufficient to in- 1989. Stress alters immune function and disease resistance in
duce a prepathological state which in turn can lead chinook salmon (Oncorhynchus tshauytscha). J. Endocrinol.
to the development of pathology. 120:135-142.

134 NOAA Technical Report NXB I 11

Moberg, G.P. Pickering, A.D. and T.G. Pottinger.
1985. Biological response to stress: key to assessment of ani- 1989. Stress responses and disease resistance in salmonid
mal well-being? In Animal stress (G.P. Moberg, ed.), p 24- fish: effects of chronic elevation of plasma cortisol. Fish
29. American Physiological Society, Bethesda, MD. Physiol. and Biochem. 7:253-258.
1987. Problems in defining stress and distress in animals. J. Refstie, T.
Am. Vet. Med. Assoc. 19 1: 1207-1211. 1982. Preliminary results: differences between rainbow trout
1992. Stress: diagnosis, cost and management. In The well- families in resistance against vibriosis and stress. Dev.
being of agricultural animals in biomedical and agricultural Comp. Immunol. Supplement 2: 205-209.
research U.A. Mench, Sj. Mayer and L. Krulisch, eds.), p Schreck, C.B.
M-61. Scientists Center for Animal Welfare, Bethesda, 1982. Stress and rearing of salmonids. Aquaculture 28:241-
MD. . 249.
Peters, G., M. Faisal, T. Ung, and I. Ahmed. Vijayan, M.M., andj.F. Leatherland.
1988. Stress caused by social interaction and its effect on 1990. High stocking density affects cortisol secretion and tis-
susceptibility to Aeromonas hydrophila infection in rainbow sue distribution in brook charr, Salvelinus fontinalis. J,
trout Salmo gairdneri. Dis. Aquat. Org. 4:83-89. Endocrinol. 124:311-318.
Pickering A.6.
1981. Stress and fish. Acad. Press. New York, NY, 354 p.
Pickering, A.D., and T.G. Pottinger.
1987. Poor water quality suppresses the cortisol response of
salmonid fish to handling and confinement. Fish Biol.
30:363-374.

Control of Fish Disease mijapan

AKIRA MURATA

Fisheries Agency, Research Division
Kasumigaseki 1-2-1, Tiyoda-ku
Tokyo, Japan

ABSTRACT

The damage to Japanese aquaculture by fish, disease in 1988 was estimated at 20,000
metric tons, worth 144 million dollars, and amounted to 6% of the total aquaculture
production. The Fisheries Agency of the Government of Japan financially supports fish
disease control projects carried out by prefectural governments and fisheries cooperative
associations. It also entrusts the Japan Fisheries Resource Conservation Association
UFRCA) with the following tasks; 1) voluntary pathogen inspection of imported larval
fish and fish eggs; 2) training of advisory personnel on fish disease control; 3) improve-
ment of diagnostic methods; and 4) improvement of disease prevention methods and the
publication of their findings. The main body of the Agency is augmented by the
Shimonoseki University of Fisheries and nine national research institutes. One of these is
the National Research Institute of Aquaculture, which has a Fish Pathology Division
totalling 11 researchers. Each of the 47 prefectural governments in Japan has at least one
Fishery Experimental Station. There are a total of 96 stations. Prefectural governments,
with their institutions, undertake disease diagnosis and drug 1@esidue testing and provide
guidance on fish disease control and drug use. Fisheries co-operative associations play
important roles in fish disease prevention by giving guidance to fishermen and control-
ling local fishery practices. There are 17 universities that have a fishery faculty and 16
universities with a veterinary faculty. All these universities have curriculums for fish dis-
ease and conduct research on it.

Aquaculture and Fish Disease There are two types of aquaculture; feeding aquac-
ulture (i.e., fish culture) and aquaculture without
feeding (i.e., shellfish culture). Production by feed-
Fishery Products and Aquaculture Production ing aquaculture has grown markedly in recent years
and has reached 343,000 t or $2.5 billion in 1988-
In 1988, fishery production in, Japan totaled about 7 and 19 times the same values for those in
12,785,000 metric tons (t) valued at $19.4 billion (1 1965, respectively.
dollar = 140 yen) (MAFF 1988). Changes in amount The production of marine fish by aquaculture in
and value of aquaculture products over the last 23 1988 was valued at $1,811 million and amounted to
years are shown in Figure 1. Aquaculture production 245,000 t. Yellowtail Seriola quinqueradiata production
in 1988 reached 1,426,000 t or I I % of the total fish- was greatest at $949 million (165,900 t), followed by
ery output. These products were valued at $4.9 red sea bream, Pagrus major, and coho salmon,
billion or 25% of the total fishery value. Oncorhynchus kisutch. (Table 1). Coho salmon, which
Aquaculture has been rapidly developing and has is cultured in. fresh water until the smolt stage
become more important because 1) consumers are and then transferred to seawater, is one of the
demanding higher quality and more variety in fishery species the production of which has increased rapidly
products, 2) aquaculture techniques have improved, during recent years. Likewise the production of
and 3) the total amount of aquacultural grounds has horse mackerel, Trachurus japonicus, striped jack,
increased owing to the development of new aquacul- Caranx delicatissimus, and tiger puffer, Takifugu
tural facilities, and the expansion of older ones. rubripes, have increased markedly. Increased variety

135

136 NOAA Technical Report NMOFS 111

Total fishery

10-

E
C
0
5-

Aquaculture

0 Feeding aquaculture
1965 1970 1975 1980 1985
Year

20, Total fishery

IUI

Aquaculture
Figure I
I Feeding aquaculture Cultured fish production (feeding aquacul-
0 ture and aquaculture without feeding) and
1965 1970 1975 Year 1980 1985 1990 the total fishery production since 1965
(MAFF 1988) in metric tons and dollars. (1$=
1 140 yen.)

of cultured species is a current trend in Japanese decline in the health of the cultured fish owing to
aquaculture. environmental deterioration and overpopulation,
Production of freshwater fish by aquaculture was and partly by the recent entry of foreign pathogens.
$732 million (97,800 t) in 1988. Japanese eel, For example, infectious pancreatic necrosis (IPN),
Anguilla japonica, production was largest at 39,600 t infectious hematopoietic necrosis (IFIN), and bacte-.
($417 million), followed by Ayu, Plecoglossus altivelis, rial kidney disease (BKD) are diseases considered to
salmonids, and common carp, Cyprinus carpio. A di- have been introduced from abroad. Both IHN and
versity of species is also seen in freshwater BKD are thought to have entered through eggs im-
aquaculture. Among these are Japanese native salmo- ported for propagation, while IPN probably entered
nids such as landlocked salmon, 0. masou, dwarf rill through eggs imported for culture. These diseases
trout, 0. rhodurus, and mountain trout, Salvelinus have become prevalent in salmonid farms and are
pluvius. causing severe damage to salmonid production.
Based on a questionnaire sent to all aquaculture
farmers through the prefectural governments every
Fish Disease and Its Damage year, the Fisheries Agency has estimated the annual
damage by fish diseases to feeding aquaculture. The
The rapid development of aquaculture has also results are shown in Figure 2.
brought about a problem, namely with the occur- The damage to aquaculture by fish disease in 1988
rence of diseases. These are caused partly by the was estimated at 20,000 t or $144 million which is

Murata: Control of Fish Disease in Japan 137

Table I
Cultured fish production by species and damage due to disease in 1988 (MAFF 1988).

Production Damage Ratio (Damage/Production)

Amounta Valueb Amount Value Amount Value
Fish (1,000 t) (million dollars) (1,000 t) (million dollars) M M

Marine fish 245.0 1811 17.3 119 7 7
yellowtail 165.9 949 14.4 81 9 9
red sea bream 45.5 431 0.5 5 1 1
coho salmon 16.5 109 1.0 7 6 6
flounder 3.1 64 0.4 6 13 10
horse mackerel 6.5 37 0.4 2 6 6
tiger puffer 1.2 37 0.3 7 27 19
kuruma prawn 3.0 140 0.2 10 7 7
others 3.3 44 0.1 1 3 4

Freshwater fish 97.8 732 2.8 25 3 3
eel 39.6 417 1.1 14 3 3
ayu 13.6 120 0.3 3 2 3
salmonids 19.1 73 1.0 5 5 7
common carp 18.1 57 0.2 1 1 1
others 7.4 65 0.2 2 3 4

Total 342.8 2,543 20.1 144 6 6

1 in metric tons, t,
b 1$ = 140 yen.

about 6% of the total value of all aquaculture prod- culture generally showed low damage rates, while
ucts. In 1979 estimated damage was $137 million and species with a short history of culture, such as tiger
increased year by year to reach a peak of $190 million puffer and japanese flounder, showed high damage
in 1982. It then decreased and has been maintained rates. This difference may be caused by the lack of
at around $140 million since 1985. During the 1980s, experience and less-developed management tech-
the total value of aquaculture products increased rap- niques for culturing new species.
idly while the rate of damage by diseases either In recent years, fish disease problems have become
decreased or remained nearly constant. It is thought more diversified and complicated. In yellowtail cul-
that two factors have worked effectively in reducing ture, streptococcosis and pseudotuberculosis are the
damage by diseases'in recent years. One is the prac- two major diseases. When an episode of mass mortal-
tice of control measures, such as disinfection, ity occurs, both of these agents are often found in a
disposal of dead fish, and suitable layout of ponds. single fish. Nocardiosis, viral ascite disease (diseases
The other is the improvement of husbandry tech- for which treatments are not yet established), and a
niques, such as improvement of food and the control disease causing peduncle curvature have recently be-
of population levels and pen location with regard to come an additional problem.
the environmental capacity.
The rate of disease damage (estimated damage
value/total product value) for each species is shown Fish Disease Control
in Table 1. Disease reduced yellowtail production by
9% in 1988. The damage rates of tiger puffer, Japa- System-Because of the complexity and variety of dis-
nese flounder (Paralichthys olivaceus) and kuruma eases *occurring in Japanese aquaculture, the
prawn (Penaeus japonicus) were 19, 10, and 7%, re- Fisheries Agency is carrying out systematic fish dis-
spectively, while only 1% of red sea brearn ease control in cooperation with prefectural
production was affected. As for freshwater finfish, governments, fisheries cooperative associations, and
rainbow trout (0. mykiss) showed a, high damage rate some universities in order both to decrease disease
of 7%, while that of common carp was only 1%. In damage and to ensure the safety of cultured fish for
marine aquaculture, species with a long history of human consumption. It is important for aquaculture

138 NOAA Technical Report NNM 111

tural governments and the fisheries cooperative asso-
VALUE OF PRODUCTION ciations. It also entrusts the Japan Fisheries Resource
Conservation Association UFRCA) with the following
2WO responsibilities: 1) voluntary pathogen inspection of
imported live fish and eggs; 2) training of advisory
0 personnel on fish disease control; 3) improvement of
diagnosis methods; and 4) improvement of disease
prevention methods and the publication of their
2250-
-0 findings (Fig. 4).
The Fisheries Agency oversees nine national re-
search institutes, the Shimonoseki University of
Fisheries, and the Hokkaido Salmon Hatchery. In
2000 these facilities either research or education in
1979 1981 1983 1985 1987 fisheries is carried out, but among these national re-
search facilities, only the National Research Institute
VALUE OF DAMAGE of Aquaculture, which was established in 1979 to
carry out basic and leading research for aquaculture,
has a fish pathology division. This division has 11 re-
searchers and consists of 4 research sections;
175 Pathogens, Pathophysiology, Pharmacology, and Im-
munology (Fig. 5).
150-
C
a Prefectural Government-Control measures on
the aquaculture grounds are very important for dis-
125 ease prevention. Prefectural governments play im-
portant roles in guiding aquaculture farmers either
100, directly or indirectly through the fisheries coopera-
1979 1981 1983 1985 1987 tive associations.
Prefectural governments and their research institu-
DAMAGE RATE tions undertake disease diagnosis and drug residue
testing, and provide guidance on fish disease control
8- and drug use (Fig. 6). , i
There are 96 prefectural fisheries experimental sta-
tions in Japan: at least one station in each prefecture.
Investigation and research for aquaculture and fish
6. disease is carried out in their propagation and aqua-
culture divisions. In most of these prefectures, @sh
disease control centers (20 in total) have been estab-
4- lished, either as a part of the fishery experimental
1979 1981 1 @83 485 1987 station or as an independent establishment. Prefec-
YEAR tural governments have also established fisheries
extension offices to give technical guidance to aqua-
Figure 2 culture farmers. These offices and fishery
Changes in cultured fish production (feeding aquaculture) experimental stations undertake fish disease preven-
and damage by fish disease. (I$= 140 yen.) tion measures in a cooperative manner.
The number of staff engaged in disease control
measures in prefectural governments was 470 in
farmers to practice proper feeding and fish disease 1986. In addi 'tion, 30 staff members -from various cit-
control. Thus, prefectural governments and fisheries ies, towns, and villages were also engaged in these
cooperative associations have been requested to es- measures.
tablish a system to assist the aquaculture farmer's
practice of disease control (Fig. 3). Fisheries Cooperative Association-Fisheries coop-
erative associations play a very important role in fish
Fisheries Agency@This agency financially supports disease control by giving guidance to aquaculture
fish disease control projects carried out by prefec- farmers and controlling local fishery grounds. How-

Murata: Control of Fish Disease in Japan 139

- Funding Japan Fisheries
@islheries --------- Resource
Agency -*- Conservation
Report Association

Information
Guidance Rq-Z
Financial support I I - Guidance, Assistance
Prefectural
government

Information

Guidance
Financial support Rep., Farmer

Fisheries
Co-operative Guidance, Assistance
Association
Infognation Figure 3
Fish disease control system in Japan.

Fisheries @g7ency7] Fisheries Agency

National Research institute of Aquaculture
Funding Report Fish Pathology Division

-Japan Fisheries Resource National Fisheries Research Institute (8)
Conservation Association
Shimonoseki University of Fisheries

� Voluntary pathogen examination of imported live fish and eggs Figure 5
� Training on fish disease ctwacteristics Fish disease research system of the Fisheries Agency.

Improvement of diagnosis and disease prevention methods
Extension (Films, Video tape recordings, Textbooks) Universities-Fish disease education is carried out
. I by 17 fishery faculties in universities (including the
Figure 4 Shimonoseki University of Fisheries, Fisheries
Role of the Japan Fisheries Resource Conservation Asso- Agency) and 16 veterinary faculties. All -*these facul-
ciation in the control of disease in aquaculture. ties also have fish disease research facilities.

Prevention of the Introduction and Outbreak of
ever, many of these associations are small in scale, Pathogens-In Japan, neither laws nor regulations
and the level of guidance varies widely. Only a few of have been established concerning fish quarantine.
them carry out drug residue testing, diagnosis of fish However, in order to avoid the introduction of fish
disease, and observation of environmental conditions pathogens from foreign countries, the Fisheries
in their aquaculture grounds (Fig. 7). The number of Agency requests importers (through prefectural gov-
staff engaging in disease control from those associa- ernors) to take the following steps. Importers are
tions was only 80 in 1986. requested 1) to import living eggs or fish in conjunc-

140 NOAA Technical Report NNIFS I I I

tion with health certificates issued by the government
authority of the.exporting country, 2) to receive an Pref ural Governments (47)
examination for pathogens in the fish or eggs at the
time of import, and 3) to disinfect eggs immediately
after importation. Fisheries Experimental Stations (96)
Examination for pathogens at the time of import is
carried out by the JFRCA as entrusted by the Fisheries
Agency. Examination is done on viral diseases such as
VHS (viral hemorrhagic septicemia), BKD and whirl- Fish Disease Control Centers (20)
ing disease of salmonid fish (Mxyosoma cerebralis), and
on viruses, bacteria, and parasites for eel. Examina-
tions are also conducted for other species. Fisheries Extension Offices
To prevent the spread of fish diseases, prefectural
fish disease control centers and fisheries experimen-
tal stations examine live fish and eggs for pathogens Disease diagnosis
upon the request of the aquaculture. farmer. The Drug residue testing
Providing guidance on fish disease control and drug use
Fisheries Agency encourages farmers to transport live To survey and observe enviromnental conditions in aquaculture grounds
fish and eggs with records of species, place of produc- Research
tion, name and address of producer, dates of Figure 6
fertilization or hatching, history of fish disease, and Role of the prefectural governments in the fish disease con-
medication history. trol system.

Guidance, Assistance
Prefectural
Guidance Government infortnation
Financial support I I -Report
Fisheries Guidance. Assistancet
Co-operative
Association Inforniation

Drug residue testing
Diagnosis of fish disease Figure 7
Observation of enviromental conditions in aquaculture ground. Role of fisheries coopefartive associations in the con-
trol of fish disease.

Training-Fish disease specialists are indispensable There are various training courses offered by the
for carrying out adequate control measures such as JFRCA for fish disease technical workers. A basic class
providing advice to the aquaculture farmers and di- consists of 20 day-lectures and practical exercises
agnosing and disinfecting facilities and equipment. each year for 3 years. In 1989, there were 30 to 40
Since 1973, the Fisheries Agency has been training trainees for each year course; altogether, 100 people
the staffs of prefectural governments-in 1984 staffs have participated in the training. Other training on
of fisheries cooperative associations were included in fish disease is also carried out. Since 1974, thejapan
the training program. This training was aimed at e.du- Veterinary Medical Association has been providing a
cating new fish disease specialists and at improving fish disease course for veterinarians every year as
the knowledge of the current specialists. In recent postgraduate education. In Japan, training for fish
years, the Fisheries Agency has entrusted this train- disease technical workers is carried out separately in
ing to JFRCA. the fishery and veterinary fields. On the other hand,

Murata: Control of Fish Disease in Japan 141

aquacultural guidance and leadership are mainly car-
ried out in the fishery field. Fisheries Agency
Investigation and Extension-Under the present cir- Japan Fisheries Resource
cumstances where fish disease can cause large-scale Conservation Association
damage and is becoming more complicated to pre- textbooks
vent, diagnose, and treat, it is becoming more films
important to promote research on fish disease and to video tapes
give aquaculture farmers information and techniques
to control this threat. Prefectural government
Studies on fish disease by the Fisheries Agency are
mainly conducted at the National Research Institute guidance
of Aquaculture in cooperation with six of the nine textbooks and pamphlets
regional national fisheries research institutes and the
Farmer]

Figure 9
Fisheries Agency The Fisheries Agency's extension system for
National Fisheries Research Institutes (8) providing fish disease informat ion in Japan.
National Research institute of Aquaculture (1)

Universities Fishery faculties (17) Drugs-Appropriate aquaculture management is im-
Veterinary faculties (16) portant to prevent or'to reduce damage by fish
disease. Drugs are also important to reduce damage
when disease breaks out.
Prefectural government At present, there are 26 antibacterial medicines
Fisheries Experimental Stations available for fish in Japan. In addition, there are vac-
cines for vibriosis, insecticides, anesthetics, and
Figure 8 nutritive drugs. In 1988, all the fishery drugs sold in
Major organizations involved in fish disease research Japan amounted to 1,540 t, which was worth $55 mil-
in Japan. lion.
All drugs are strictly evaluated for safety, effi-
cacy, residue accumulation in fish tissues, and
Hokkaido Salmon Hatchery. The Agency entrusts other properties by the Central Pharmaceutical Af-
some universities with fundamental research, such as fairs Council prior to receiving manufacture
exploring the mechanisms of outbreak and infection approval from the Ministry of Agriculture, Forestry,
of various fish diseases. It also entrusts prefectural and Fisheries, based on the Pharmaceutical Affairs
fisheries experimental stations with applied research Law. Use of antibacterial drugs for fish as well as
,for fish disease control techniques, such as disinfec- livestock is regulated by "The Standard to be Ob-
don and vaccines (Fig. 8). served by User," which is based on the same law
Through its extension program, the Agency dis- (Table 2).
tributes textbooks about disease diagnosis and Staffs of prefectural governments and fisheries co-
techniques of disease prevention and lends films operative association assist all aquaculture farmers to
and video tape recordings through the JFRCA to the use drugs properly by visiting, distributing pam-
prefectural staffs who are in charge of guiding . phlets, or carrying out guidance courses at regular
aquaculture farmers. Prefectural governments also intervals. The public health divisions of prefectural
offer guidance courses and distribute textbooks or governments carry out drug residue testing by sam-
pamphlets on fish disease to the aquaculture farmer. pling cultured fish from shops. The fisheries divisions
These measures are used to impart knowledge on of prefectural governments and fisheries cooperative
fish disease characteristics, disease prevention, associations also examine aquaculture products sub-
aquaculture management, and use of drugs, and mitted on a voluntarily basis for drug residues before
how to put all of this knowledge into practice appro- harvest, in order to ensure that foods are safe for
priately (Fig. 9). human consumption.

142 NOAA Technical Report NMOFS I I I

Table 2
The standard to be observed by user of drug (MAFF 1980).

Drug Sub ect animal Administration and dosage Withdrawal perioda

Feed additive Yellowtail Administer orally, mixing not more than 20 days
containing 50 g (potency) in I t of feed'
alkyltrimethyl-
ammoniumcalcium-
oxytetracycline
Feed additive Yellowtail Administer orally, mixing not more than 5 days
containing 20 g (potency) in I t of feed
ampicillin
Feed additive Yellowtail Administer orally, mixing not more than 10 days
containing 50 g (potency) in I t of feed
chlortetracycline
hydrochloride Eel Administer orally, mixing not more than 15 days
50 g (potency) in I t of feed
Feed additive Yellowtail Administer orally, mixing not more than 30 days
containing 50 g (potency) in I t of feed
erythromycin
Feed additive Yellowtail Administer orally, mixing not more than 5 days
containing 10 g in 1 t of feed
florphenicol

Feed additive Yellowtail Administer orally, mixing not more than 20 days
containing 80 g (potency) in I t of feed
kitasamaycin

Feed additive Yellowtail Administer orally, mixing not more than 20 days
containing 50 g (potency) in I t of feed
oxytmaccycline
hydrochloride
Red sea brearn Administer orally, mixing not more than 30 days
50 g (potency) in I t of feed
Coho salmon Administer orally, mixing not more than 30 days
50 g (potency) in I t of feed

Eel Administer orally, mixing not more than 30 days
50 g (potency) in I t of feed
Rainbow trout Administer orally, mixing not more than 30 days
50 g (potency) in I t of feed
Feed additive Yellowtail Administer orally, mixing not more than 16 days
containing 30 g in I t of feed
oxolinic acid
(except liquid) Coho salmon Administer orally, mixing not more than 21 days
20 g in I t of feed

Eel Administer orally, mixing not more than 20 days
20 g in 1 t of feed
Rainbow trout Administer orally, mixing not more than 21 days
20 g in 1 t of feed

Ayu Administer orally, mixing not more than 14 days
20 g in I t of feed

Carp Administer orally, mixing not more than 28 days
10 g in I t of feed
Feed additive Yellowtail Administer orally, mixing not more than 16 days
containing 20 g in I t of feed
oxolinic acid
(liquid)

Murata: Control of Fish Disease in Japan 143

Table 2 (Continued)

Drug Subject animal Administration and dosage Withdrawal period a

Bath agent Eel Give bath, dissolving not more than 25 days
containing 5 g in 1 t of water
oxolinic acid
Ayu Give bath, dissolving not more than 14 days
10 g in I t of water

Feed additive Yellowtail Administer orally, mixing not more than 30 days
containing 40 g (potency) in 1 t of feed
spiramaycin
embonate

Feed additive Rainbow trout Administer orally, mixing not more than 30 days
containing 100 g in I t of feed
sulfadimethoxine
or its sodium salt

Feed additive Yellowtail Administer orally, mixing not more than 15 days
containing 200 g in I t of feed
sulfamonomethoxine
or its sodium salt Eel Administer orally, mixing not more than 30 days
200 g in I t of feed
Coho salmon Administer orally, mixing not more than 30 days
100 g in I t of feed
Ayu Administer orally, mixing not more than 15 days
100 g in I t of feed
Rainbow trout Admini@ter orally, mixing not more than 30 days
150 g in 1 t of feed

Bath agent Rainbow trout Give bath, dissolving not more than 15 days
containing 10 g in I liter of salt water (less
sulfamonomethoxine than 1% concentration)
or its sodium salt

Feed additive. Ayu Administer orally, mixing not more than 15 days
containing 15 g of sulfamonomethoxine and
sulfamonomethoxine 5 g of ormetoprim in I t of feed
and ormetoprim
(compound drug)

Feed additive Yellowtail Administer orally, mixing not more than 15 days
containing 50 g in I t of feed
thiamphenicol

Withdrawal period defined as the required waiting period following drug administration prior to use of fish for human consumption.
b t = metric ton.

Citations

MAFF (Ministry of Agriculture, Forestry, and Fisheries)
1980. Ministerial ordinance regarding the control and use of
drugs for animals. MAFF Ordinance No. 42, Sept. 30, 1980,
MAFF, Government ofJapan, Tokyo, Japan. (In English.)
1988. Annual statistics of fisheries and aquaculture produc-
tion, MAFF, Government of Japan. (In English.)

-U.S. GOVERNMENT PRINTING OFFICE 1992-0-790-258/80000

Guide for Authors
NOAA Technical Report NMFS

PREPARA TION footnoted with italic lower case letters (a, b, c) for general com-
ment, with asterisks for probability in statistical data. Because
Title of the paper should be as brief as possible (between 6 and tables are typeset, they need only be submitted typed and
12 words) and should not include the Latin binomial except when formatted.
it is needed for clarification, i.e., to distinguish between species
sharing the same common name or between species known in- Figures include line illustrations and photographs. Unless photo-
ternationally-by different common names. graphs are submitted on glossy paper with good contrast, we can-
not guarantee a good final printed copy. Figures must be labeled
Abstract should be one paragraph of about 200 words or less. with author's name and number of figure. Lettering within the
It should state the main scope of the paper but emphasize its con- figure should not be so heavy and large as to upstage the impact
clusions and relevant findings. Because abstracts are circulated of the entire figure. Avoid "outlier" words in a figure which take
by abstracting agencies, it is important that they represent the up needless space. Each figure must include a legend that explains
research clearly and concisely. all symbols and abbreviations. These figure legends should
be typewritten on a separate piece of paper at the end of the
Text must be typed double-spaced throughout and should be manuscript.
divided into the following sections: Introduction, Materials and
Methods, Results, Discussion (or Conclusions), and Acknowl- Copies of Published Reports are available free of charge to the
edgments. Headings within each section must be short, reflect senior author (50 copies) and to his or her laboratory (50). If the
a logical sequence, and follow the rules of multiple subdivision article is part of a proceedings or collection of articles, the senior
(i.e., there can be no subdivision without at least two items of author receives 50 free copies of the individual article and one
subdivision). The entire text should be intelligible to interdisci- copy of the collective work. Additional copies (reprints or of1prints)
plinary readers; therefore, all acronyms, abbreviations, and of an article may be purchased in lots of 100 at a cos( of $1.40
technical terms should be defined the first time they are used. per page at the time of first printing. Additional copies of a
The scientific names of species must be written in full the first collective work must be sought from the editor assigned to coor-
time they are mentioned and abbreviated in subsequent refer- dinate the research papers for submission as a NOAA Technical
ences. Footnotes should be avoided as much as possible. We follow Report.
the U. S. Government Printing Ofi5ce, Style Manual (1984 ed.) and the
CBE Style Manual (5th ed.) for general format and style; the
American Fisheries Society's most recent edition of Common and SUBMISSION
Scientific Names of Fishes ftom the United States and Canada for fish
nomenclature. Dates should be written as I I November 1991. Send manuscript (original and two copies) to the Scientific Editor:
Measurements should be expressed in metric units, e.g., tons as Dr. Linda Jones, Scientific Editor
metric tons (t), but if the work is in British long tonnes, please National Marine Mammal Laboratory, F/AKC3
make this fact explicit to the reader. The numeralcone (1) should National Marine Fisheries Service, NOAA
be typed as a one, not a lower-case el (1). 7600 Sand Point Way NE
Citations comprise both unpublished and published works. Seattle, WA 98115-0070.
Authors are advised to avoid references to nonstandard material Once the manuscript has been accepted for publication, you will
such as internal and project reports wherever possible. for these be asked to submit a software copy of your manuscript to the
works, include whether they are available from NTIS (National Scientift: Editor. The software copy should be submitted in ASCII
Technical Information Service) or from some other public format (i.e., in a MS-DOS "print" or "nondocument" file) and
depository. Personal communications and unpublished data must should be placed on a 5.25-inch (preferably) or 3.5-inch disk that
be cited in parentheses in the text with full address of com- is double-sided, double or high density, and that is compatible
municator or author. Follow the name-and-year system for cita- with either IBM or Apple Macintosh systems.
tion format. In the text, cite Smith and Jones (1977) or (Smith
and Jones 19 7 7). If there is a sequence of citations, list chron- Once the manuscript is being typeset and prepared for publica-
ologically: Smith 1932; Green 1947; Smith and Jones 1985. tion, all inquiries should be made to the Managing Editor:
Abbreviations of serials should conform to abbreviations given
in Serial Sourcesfor Biosis Data Base". AUTHORS ARE RESPONSIBLE Sharyn Matriotti, Managing Editor
FOR THE ACCURACY AND COMPLETENESS OF ALL CITATIONS. NOAA Technical Reports NMFS
National Marine Fisheries Service
Tables should not be excessive in size and must be cited in Scientific Publications Office, F/NWRl
numerical order in the text. All unusual symbols must be ex- 7600 Sand Point Way NE
plained in the table heading. Other incidental comments can be Seattle, WA 98115-0070.

UNITED STATES
DEPARTMENT OF COMMERCE
NATIONAL OCEANIC AND ATMOSPHERIC BULK RATE
ADMINISTRATION POSTAGE & FEES PAID
NATIONAL MARINE FISHERIES SERVICE
SCIENTIFIC PUBLICATIONS OFFICE U.S. Department of Commerce
BIN C15700 Permit No. G-19
SEATTLE, WA 98115
OFFICIAL BUSINESS
Penalty for Private Use, $300

NOAA SCIENTIFIC AND TECHNICAL PUBLICATIONS

The National Oceanic and Atmospheric Administration was established as part of the Department of Commerce on
October 13, 1970. The mission responsibilities of NOAA are to assess the socioeconomic impact of natural and
technological changes in the environment and to monitor and predict the state of the solid Earth, the oceans and
their living resources, the atmosphere, and the space environment of the Earth.

The major components of NOAA regularly produce various types of scientific and technical information in
the following kinds of publications:

PROFESSIONAL PAPERS- Important definitive TECHNICAL SERVICE PUBLICATIONS-Reports
research results, major techniques, and special containing data, observations, instructions, etc. A par-
investigations. tial listing includes data serials; predictions and outlook
periodicals; technical manuals, training papers, plann-
CONTRACT AND GRANT REPORTS-Reports ing reports, and information serials; and miscellaneous
prepared by contractors or grantees under NOAA technical publications.
sponsorship.
TECHNICAL REPORTS-Journal quality with exten-
ATLAS-Presentation of analyzed data generally in the sive details, mathematical developments, or data listings.
form of maps showing distribution of rainfall, chemical
and physical conditions of oceans and atmosphere, TECHNICAL MEMORANDUMS-Reports of
distribution of fishes and marine mammals, ionospheric preliminary, partial, or negative research or technology
conditions, etc. results, interim instructions, and the like.

Information on availability ofNOAA publications can be obtainedfrom:

U.S. Department of Commerce
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161

3 6668 00003 7327