Developing challenge models and vaccine efficacy tests against Streptococcus iniae for two ornamental cyprinid fish


Material Information

Developing challenge models and vaccine efficacy tests against Streptococcus iniae for two ornamental cyprinid fish the red-tail black shark (Epalzeorhynchos bicolor) and the rainbow shark (E. erythrurus)
Physical Description:
xiv, 236 leaves : ill. ; 29 cm.
Russo, Riccardo
Publication Date:


bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 2004.
Includes bibliographical references.
Statement of Responsibility:
by Riccardo Russo.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 003163591
oclc - 706801518
System ID:

Full Text
riy '^'

RED-TAIL BLACK SHARK (Epalzeorhynchos bicolor) AND THE RAINBOW
SHARK (E. erythrurus)






To my parents Francesca and Salvatore, my aunt Marilena, my uncle Ennio, and my wife
Juli-Anne. Thank you for your love and encouragement.


I want to thank my parents, Francesca and Salvatore; my aunt and uncle, Marilena

and Ennio; and my wife, Juli-Anne, for their help, support and love in these years. I want

also to thank Debra Anderson, Sonya Sampson, Marta Hartman, and Richard Miles for

their friendship, advice, and encouragement.

I thank the members of my supervisory committee (Drs. Roy Yanong, Jack Gaskin,

Richard Miles, Hugh Mitchell, John Tucker, Ruth Francis Floyd), Mr. Craig Watson

(Director of the Tropical Aquaculture Laboratory), and Dr.

Walter Parker (NOVARTIS-

Aqua Health LTD.) for their advice and help throughout my research. I thank Dr.


for teaching me about fish health and for the big revision of this dissertation, and Dr.

Gaskin for his great supervision and assistance during my stay in Gainesville.

I thank Dr. Debra Murie for helping me with the image analysis.

Special thanks go to Jason and Damon Diaz, Robin Sanderson, Scott Graves, and

Robert Leonard for assistance with fieldwork and in the laboratory. Thanks go to Carlos

Martinez, Sherry DeMayo, Tina Crosby, John Kao, and Bill Shields for their invaluable

assistance. I also want to thank all the University of Florida and University of South

Florida professors that have taught me so much in their courses.

This project was supported by a grant from USDA-CSREES.



. ... ..... ......... ..... .. ..... ... .. .......... .. .. .... ...... .. ......... ... ... ii

LIST OF TABLES .................................. .......................................... ..... .................. v

LIST OF FIGURES ........................... ................................................................................ x

A B ST R A C T .... ........... .. .. ........... .. ..... ....... .............. .... ......... ..... ...... .. .... .......... ...... ....... xiii



S *9..m**** C1(111)111 S1(1 C 111(1 111()1 CCC1111m****11$*.(11 *St m **C S m b e S S C e S* mS*SC 1m* *

RED-TAIL BLACK SHARK (Epalzeorhynchos bicolor) AND RAINBOW

SHARK (E. erythrurus) BIOLOGY



* CC *SC*lS Clm**SO CS *S C S C*S SC C C**S **Sl C.** CC S C... C S C*e me....C.C* C* SC
*.m ....... ....*em ....... m....S* mm *. *S.SC*...C S. CO *C1S ..m..S.C...* CCC1#.



Pathology ..........................................
Host-Pathogen Interactions ...............
Antibiotics Treatments and Vaccines

*..C1mC..e.SS1mm.SeeCSC1m#CS1eCS1e)eS1*S I

C.. CCC S eeseee S C*S*Ce .. S C.*m S 55CCCS S* C SC C CC5


Fish Immune System......... .............. ..
Ontogenesis of the Immune System.

Immune Organs and T-Cell Ontogenesis:

Histological Approaches..


1(111(11()11((1111111()1111)111111(11)11 5




Fish Vaccines...............
Adjuvants .....................
Injection Vaccination...
Immersion Vaccination
Oral Vaccination ..........
Biomolecular Vaccines

Genomics and Proteomics App
Herd Imm unity ........................

. ... ... .. ...... ................... ........................ ................ 55
S... .... ..*..... ... ... S ..*............... ..................... ........ ........ ..5

............ .......................... ................ .. ................. ...61
.. ...................................................... ..... .... ......... .... 6 2
.... .... ... .... .... .... .... ... .... .... .... 62..

lied to Vaccine Design .............. .................. 66
..... ..... .. ..... .... ........ .. ... .. .. ..... .....................7 4



Immunostimulants and Experiment Design

Vitamin C
Vitamin E
Vitamin A
Folic Acid
Iron ..........
Zinc .........

......... ............... ... ... ... ........ .....76

.. ... ...... ........... ..... ...... .... ........... ........... ......................... ............ .. 8

...... ........ ................. .......... ...... ..... .. .......................... ............. 8

as ....... ............. ................ .. ........ .. ............ ... ........... ......................... 8
ans........................................... ... .. .... ...........*-**--................,....91
ans. ...... ... ... ... ....... .... ................. ....................... ............ ..... ...9 1


Other Immunostimulants


........ ............ ............. ....... ...... ..... .. ............... .............. 9 4
.......................................................... .. .................9 6

Introduction ...............
Laboratory Design ...

Recirculating System s Design ...................................................................... 97
Sterilization and Start Up Protocols............................. ...................... ........ 97
Conclusion .................................. ......................... ..... ........................ ......... 98

RED-TAIL BLACK SHARK (Epalzeorhynchos bicolor) AND RAINBOW
SHARK (E. erythrurus), FAM. Cyprinidae

......... ................ ...................... .............. .............. 9 2


<*1C.. ,*..*.. . C*.. *n C- .* *in** **C **C in....'*4 *C *Ci '.*C***i.**C.* C'**'*C-*C *"


BLACK SHARK (Epalzeorhynchos bicolor), FAM. CYPRINIDAE

Materials and
Discussion ....

Methods ...........

......... .......................................... ................. 1
.......................................................... ........ 1
... ...... ................... ... .. ... .................. ....* .. 1
.*................ . . 9 .. S ... e.. t.... ..................1
---------- -- ............................ 1-

iniae IN JUVENILE RED-TAIL BLACK SHARK (Epalzeorhynchos bicolor),

Materials and
Discussion ....
Conclusion ...

M methods .... .. .. ... .... .... .... ... ... ... .. ..... ... .... ..... .. ... .
. .. ....eee..t. .. . C.* 5**tt eC**t* *9S .. .t. .et. C. .t**. **' .. .*S t St* ** .

* C t t. t e ..... *. .. *** q tt tC e .** S .* St. ..** .** t .* *"t.*t .* C t Ce* tt*9. .* *

................ 1
............. 1
...... ........ 1
............... 1
............... 1

erythrurus) AND RED-TAIL BLACK SHARK (E. bicolor), FAM.

Materials an(



..t...... .... .e.t... ..t....e........................ ............................... 1
.*e.e SC** 9. t .. .. *. t. t t t*.ee... ...**. 9 * .e.**.9**t"t*tt. I


Materials and
Discussion ....

. e. .i.O t.. .. .... ... i. ...............e.. .. *** ** ** "'"
M methods .......... ....... ....................... ......................................t

.* t ^ "^ * .*' .. .. .. .. .. .. .. .. .. ... .. .. .. .. .. .. *. .. .* *-- -- '. .. *- '- *' '* '- -




(Epalzeorhynchos bicolor, FAM. CYPRINIDAE) FED BETA-GLUCANS OR

Materials and
Discussion .....

M ethods . .. .... ....... .. ...... .. ............ .* ... .* .... 1
.....eth ds............ ................................................................... 1

. .. ..*.. ............ .. .... ... ... .. t. .. ....... .... ... .. .. .. ... ... 1
-- ..-. .... ............................I

BROODSTOCK RED-TAIL BLACK SHARK (Epalzeorhynchos bicolor, FAM.
CYPRINIDAE) AGAINST Streptococcus iniae

Introduction.............. .. ...... ... .................................. ......................... ............
M materials and M ethods ........ ...................... ................................................
Results ... .... ..... ....... ...S....t.. ..***..*. .*** *.***9. ........... ..... ...* ....... .....*. ...**
Discussion .................. ................... .................................... .. ......... .........
Conclusion ...................................................................................... .... .. ..........


Introduction. ............................ ............ ................... ......................................... 1
M materials and M methods .............................................................. .............. ................... .. 1
Results. ............... ........ .......... ................................. ............9.. 1
Discussion ..................................................... ................... .... ................... 1
Conclusion ................... ........... .................................................. .. ........... 1

17 CON CLUSION NI....................................................................................................197

LIST OF REFERENCES ......................................................... .....................................................200

BIOGRAPH ICA L SKETCH .......................................... ................... ..... ........... 236




First appearance of immune organs in different fish species. ..................................53

First appearance of lymphocytes in immune organs in different fish species .........53

Time of maturation of the immune organs .................... ...... ............. ......... .. ....54

Characteristic of S. iniae isolated from diseased rainbow shark................... .........11

Average cumulative mortality (%) 14 days after bacterial challenge for the

.............. 1

first experiment with RTB shark.

Average cumulative mortality (%) 14 days after bacterial challenge for the

two experiments with rainbow shark.

Average cumulative mortality (%) recorded in the first 12 hours after
injection of the vaccine formulations or of the bacterial challenge .......................124

Total average mortality (%) recorded for 12 days after challenge with S.


*.llll..........w. *. **g ...* m .*sla **# *S g. .S ,*. ** *S* *. q.* .


Average cumulative mortality (%) recorded in the first 12 hours after

injection of different volumes of the vaccine formulations

Total average mortality (%) in fish injected with different vaccines doses

recorded for 12

days after challenge with S. iniae .............

Total average mortality (%) recorded in RTB and rainbow sharks juveniles
and adults for 12 days after challenge with S. iniae ................... ............ .................. 153

Total average mortality (%) recorded in RTB,
12 days after challenge with S. iniae ...............

rainbow and blue sharks for

Total average mortality (%) recorded in RTB shark fed immunostimulants
and immunostimulants free-diets for 12 days after challenge with S. iniae. ......,..171

................ ..135


................... ..135


Total average mortality (%) recorded in RTB broodstock vaccinated 8
months before the experiment, for 12 days after challenge with S. iniae .............. 187

Range of koi and RTB shark serum dilutions tested in the indirect ELISA. .........196

Coating agents used in the indirect ELISA to detect koi or RTB or rainbow
shark Igs anti S. iniae, and coating agent dilutions that gave positive

reactions. ...................

* ** *S ** *S SS I*Se* ** ** SS ............ .................. .. .. .. .. ... .. .. .. ... ..1 9



V iew of the biosecurity room .......................... .......................................... ...... 99

Close-up view of one of the recirculating systems ..............................................99

View of three of the four 400 L tanks used to store incoming water.....................99

View of one of the bead bioreactors ........... ........... ... ..... .. .................... 99

Average mortality rates (%) observed in Experiment 1 with RTB shark ............112

Average mortality rates (%) observed in Experiment

with RTB shark............l 12

Average mortality rates (%) observed in Experiment 3 with RTB shark............1 13

Average mortality rates (%) observed in Experiment 4 with rainbow shark....... 114

Average mortality rates (%) observed in Experiment

with rainbow shark.......1 14

Average mortality (%) recorded in the first 12 hours after injection of the
3 vaccine formulations or of the bacterial challenge ......... ....................... .........124

H&E staining of a posterior kidney section of a fish injected with the


H&E staining of a spleen section of a fish injected with the aluminum



Average mortality (%) recorded for 12 days after challenge with S. iniae.......... 126

Box plot of the mortality rate (%) in fish injected with different vaccines
doses recorded for 12 days after challenge with S. iniae ............ ......................... 136

H&E staining of a spleen section of a non-vaccinated fish ................................. 136


aluminum formulation ........................................

H&E staining of a posterior kidney section of a non-vaccinated fish .................137

H&E staining of a spleen section of a fish injected with the oil formulation......137

H&E staining of a posterior kidney section of a fish injected with the oil


H&E staining of a spleen section of a fish injected with the aluminum


H&E staining of a posterior kidney section of a fish injected with the

................... .. 138

*4S*S*5 *** S 55**. .139

Gram staining of a spleen section of a fish injected with the aluminum

formulation ......

H&E staining of a liver section of a fish injected with the aluminum


.......................... .................. 139

S......... S......... S *SS*SS*SSS *5955.1 0

Gram staining of a granuloma-like lesion in the spleen section of a fish

injected with the aluminum formulation...


Box plot of the mortality rate (%) recorded in fish vaccinated with the
aluminum formulation by bath or injection for 12 days after challenge


..................... 14 1

Rainbow and RTB juvenile sharks average mortality (%) recorded for 12

days after challenge with S.

nzae ...............................

............. 152

Rainbow and RTB adult sharks average mortality (%) recorded for 12

days after challenge with S. iniae ..........

Box plot of the rainbow and RTB sharks average mortality (%) recorded

for 12 days after challenge with S.

.... ..... ............... ........ 1 5 3

Box plot of average mortality (%) recorded in the first experiment with
RTB, rainbow and blue sharks for 12 days after challenge with S. iniae............ 161

Box plot of average mortality (%) recorded in the second experiment with
RTB, rainbow and blue sharks for 12 days after challenge with S. iniae............ 162

aluminum formulation ........

in iae .........,,,.. ...........

..,................ ..........138

with S. in iae ................... .

........._..,.., ..... 152

Box plot of average mortality (%) recorded in the first experiment with
RTB sharks fed immunostimulants and immunostimulants free-diets for
12 days after challenge with S. iniae ......... ... .... ..................... .... ...................172

Box plot of average mortality (%) recorded in the second experiment with
RTB shark fed immunostimulants and immunostimulants free-diets for 12

days after challenge with S. iniae.

Box plot of average mortality (%) recorded in RTB broodstock for 14 days
after challenge with S. in iae .......... ................... ......... ...... 18......6

Box plot of average mortality (%) recorded in RTB broodstock vaccinated
from 8 months, for 14 days after challenge with S. iniae................... ................. 18 7

............ 172

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

RED-TAIL BLACK SHARK (Epalzeorhynchos bicolor) AND THE RAINBOW
SHARK (E. erythrurus)


Riccardo Russo

December 2004

Chair: Roy P. E.


Cochair: Jack Gaskin
Major Department: Fisheries and Aquatic Sciences

The purpose of our study was to test three autogenous vaccine formulations (a)

basic formula; b) basic formula with aluminum salt adjuvant; c) basic formula with oil

adjuvant) developed by NOVARTIS-Aqua Health LTD against Streptococcus iniae, for

two ornamental fishes: Epalzeorhynchos bicolor (red-tail black shark) (RTB shark) and

E. erythrurus (rainbow shark). These two species of fish were chosen because they are

considered economically important for the Florida ornamental industry, and they are very

susceptible to streptococcal diseases.

We first developed a challenge model against

iniae for the RTB shark, and then

tested it on rainbow sharks. The protection afforded by the three formulations and

The three vaccine formulations were then tested on rainbow sharks, and

differences in vaccine efficacy among juvenile and adult groups of both RTB and

rainbow shark were investigated. Results of these experiments showed that the vaccine

conferred a higher protection level to RTB than to rainbow sharks. Farmers have reported

that the blue shark (a color variant of the rainbow shark) is more resistant to S. iniae. Our

study confirmed this.

In another set of experiments, the immunostimulatory capability of beta-glucans

and nucleotides were tested. Both compounds decreased the mortality of RTB sharks

challenged with

iniae compared to the controls.

The final step of our study was to

determine if the vaccines were effective in protecting RTB broodstock fish against S.

iniae infections resulting from the stress caused by artificial spawning. Results showed

that all three vaccines efficiently protected fish after artificial spawning.

Our study showed the efficacy of the two vaccine formulations with adjuvants

against S. iniae and the efficacy of two immunostimulant compounds, beta-glucans and

nucleotides for two species of ornamental fish. Furthermore, these challenge and vaccine

efficacy models can be used as templates in future studies with other species of

ornamental fish and other bacterial diseases.


The purpose of our study was to test three autogenous vaccine formulations (a

basic formula, and formulae with the addition of aluminum salt or oil as adjuvants)

against Streptococcus iniae, for two ornamental fish species, Epalzeorhynchos bicolor

(red-tail black shark) (RTB shark) and E. erythrurus (rainbow shark).

We chose these

two species of fish because they are considered economically important for the Florida

ornamental fish industry, and because they are very susceptible to streptococcal diseases.

We chose Streptococcus iniae infection because is an important disease in the

aquaculture industry, and has been reported worldwide causing up to 75% mortality.

The first step in our study was to build a biosecure laboratory (Chapter 8). A

challenge model against S. iniae was developed for both fish species (Chapter 9).

tested the protection offered by the three formulations and different routes of applications

on juveniles and adults of both species (Chapters 10-12).

We then examined differences

in vaccine efficacy between rainbow and blue sharks (a selected phenotypic variant of

rainbow shark) (Chapter 13).

Immunostimulatory capabilities of beta-glucans and

nucleotides were tested on RTB shark (Chapter 14).

Then, we examined whether the

vaccines were effective in protecting RTB broodstock against S. iniae infections resulting

from the stress caused by artificial soawnin2 (Chanter 1

). Agglutination and ELITSA



RED TAIL BLACK (Epalzeorhynchos bicolor) AND THE RAINBOW SHARK (E.


Red-tail black shark (Epalzeorhynchos bicolor) and rainbow shark (E. erythrurus)

are important economic species for the ornamental fish industry. The name "shark" may

lead to the misunderstanding that these fish belong to the Elasmobranches, real sharks

and rays. In reality these fish belong to the class Actinopterygii,

family Cyprinidae. This

family includes the common carp (Cyprinus carpio) and danio (Danio spp.).

These fish

are called "shark" only for marketing purposes and because the dorsal fin is vaguely

reminiscent of the fin of a real shark.

Epalzeorhynchos bicolor has a black body with a red tail, while E. erythrurus has

a greenish-brown body with a large black spot at the base of the caudal fin; and the

dorsal, anal, and pelvic fins are red. Both species attain a maximum total length (TL) of

cm (Axelrod and Burgess, 1997). These fish are benthopelagic, tropical freshwater

fish. Their natural environment is water with a pH range of 6.

70C, and a hardness of 1

(>0.04 ppm) of NO2

, a temperature of

Sharks are very sensitive to high concentrations

-. The diet of the shark includes algae, periphyton, plant matter,

worms, crustaceans, insects, and zooplankton (Yang and Winterbottom, 1998).


S---------------------------------------------at. -- -~ 2 C ~..14 irfl.
-S Sn '-a a

These species originate from Southeast Asia, particularly the Chao Phraya

(Thailand) and Mekong basins. It is not clear if E. bicolor still exists in the wild (Baillie

et al., 2004). Presently, this species has been placed in the IUCN Red List in the category

of extinct in the wild.

The reasons for its decline in the wild are not clear.

The aquarium

fish trade has been accused of driving the species to extinction by overfishing, but no

evidence is documented (Kottelat and Whitten, 1996). Furthermore, most of the sharks

sold in the ornamental industry are farm raised and not caught in the wild. Habitat

modification may be a more likely cause of their population decline. In the 1970s many

dams were constructed in Thailand, and these probably severely impacted the natural

ecosystems of several large basins in the region. Furthermore, large swamps have been

drained, decreasing the natural habitat of this species (Dai and Yang, 2003).

2.2. Aquaculture

Red-tail black and rainbow sharks can be raised and spawned commercially on

farms. Broodstock can be reared in ponds and in vats. The advantage of rearing

broodstock in vats is that they can be conditioned all year around. In ponds, fish are

conditioned naturally; and spawn in Florida only between May and September. The

indoor conditioning period lasts 6 weeks with fish kept at 280C and 16 h light.

Broodstock are then artificially spawned by induction with hormones (Rottmann et al.,

1991a,b). Each broodstock can be used for spawning up to 16 times in a period of 3 years

(personal communication by farmers).

In captivity, females become mature at 1.5 years

old, corresponding to a size of 10-13 cm. They can produce several thousand eggs.

In the third and fourth weeks, when fish are 4-8 mm long, they begin to be pigmented.

After 5-8 months, when the juveniles reach 2.5-5 cm, they are harvested and sold to the

pet market. Larger fish are also sold, but the market demand is less (personal

communication by farmers).


Streptococci are gram-positive, not motile, and catalase-negative bacteria. In

actively spreading lesions within host tissues, diplococcal and individual coccal forms are

common; in artificial media and purulent exudate, chains are more common. The length

of the chain tends to be inversely related to the adequacy of the culture medium. Factors

that tend to promote exaggerated chaining include conditions that impair growth (such as

unfavorable media, cold temperature, or antimicrobial agents); and also the presence of

antibody that reacts with cell wall antigens (Woo and Bruno, 1998). All streptococci are

lactic-acid bacteria, and derive their energy primarily from fermenting sugars, regardless

if they are grown aerobically or anaerobically. The accumulation of lactic acid in media

with high glucose content can limit bacterial growth if the pH is not buffered or


Several kinds of streptococci can be isolated from humans and other animals. The

first proposed classification of streptococci was based on the description of the type of

hemolytic reactions observed on blood agar plates. In this classification the bacteria are

divided into three groups: alpha, beta and gamma. A narrow zone of hemolysis surrounds

alpha-hemolytic streptococcal colonies, with unhemolyzed RBCs in an inner zone, and

complete hemolysis in an outer zone surrounding the colonies.

The mechanism


medium, and on the incubation time. Because of this feature, this streptococcal group is

also called the viridans group. Most of the streptococcal strains that cause human disease

belong to Group A.

Beta-hemolytic streptococci have an outer capsule, and produce a

wide clear zone of complete hemolysis in which no red cells are visible upon microscopic

examination. These bacteria can release two types of beta-hemolysin enzyme:

streptolysin O and streptolysin S.

Streptolysin O is destroyed by atmospheric oxygen,

and is therefore demonstrable only in deep colonies. Streptolysin S is stable in contact

with atmospheric oxygen, and is responsible for surface colony hemolysis. Beta

hemolytic streptococci are differentiated into a number of immunologic groups

designated by the letters A through O; in each group, different streptococcal types are

present. The groups are characterized by different external carbohydrates that act as

antigens, while specific antigen proteins characterize the types. Precipitin tests

(Lancefield) and aggregation reactions (Griffith) showed this difference in antigens.

Streptococcus iniae is included in the beta-hemolytic streptococcal group.

Gamma-streptococci produce no hemolysis either on the surface or within the agar.

classification based on the hemolysis pattern is not very satisfactory (Woo and Bruno,

1998) for the following reasons:

* Many species classified as beta-hemolytic are in fact not hemolytic (certain

members of the antigenic groups B, C, D, H, K, O and all group N).

* Most streptococci found in the gastrointestinal tract are nonhemolytic, but some

strains produce beta-hemolysis (group D).


4.1 Pathology

Streptococcal infections in fish result in septicemic diseases that have been

reported worldwide. They cause an estimated $150 million in annual losses of fish

production (Shoemaker and Klesius, 1997). Streptococcosis can cause up to


mortality in closed culture systems (Perera et al. 1994) and up to 50% in ponds (Eldar et

al. 1997). It is one of the most important diseases in South Africa, and Japan (Ferguson et

al. 1994), in yellowtail farming (Bang and Sung-Hee, 1993); and is among the most

important summer diseases in Italian and Israeli trout farming (Bercovier et al. 1997).

Streptococcosis can affect wild and farmed fish in both fresh and saltwater. Case reports

in saltwater fish include yellowtail, Seriola quinqueradiata (Aoki et al. 1990; Ikeda and

Minami, 1982; Park et al. 1987

Sako, 1998), turbot, Scophthalmus maximus (Romalde et

al. 2000), Japanese flounder, Paralichthys olivaceus (Nakatsugawa, 1983), mullet, Mugil

cephalus (Bunch and Bejerano, 1997), red drums, Sciaenops ocellatus (Eldar et al. 1999),

striped piggy, Pomadasys stridens, and variegated lizardfish, Synodus variegatus

(Colorni et al. 2002). Case reports in freshwater fish include tilapia, Oreochromis spp.

(Shoemaker et al. 2001), Mozambique tilapia, Tilapia mossambica (Ming-Chen et al.

1985), Nile tilapia, Tilapia nilotica (Kitao et al. 1981), hybrid tilapia, Tilapia aurea x T

al. 1995a; Kitao et al. 1981), eel, Anguillajaponica (Kusuda et al. 1978), ayu,

Plecoglossus altivelis (Kitao et al. 1981

Ugajin, 1981), snakeheads, Ophicephalus

striatus (Lio-Po et al. 1998), walking catfish, Clarias batrachus (Lio-Po et al. 1998),

zebra danio, Danio rerio, and pearl danio, D. albolineatus (Ferguson et al. 1994), clown

coaches, Botia macracanthus, rainbow shark, E. erythrurus, red-tailed black sharks, E.

bicolor, rosy barbs, Barbus conchonius (Russo data not published), tetras,

Hyphessobrycon sp., and African cichlids of the genera Nimbochromis and

Pelvicachromis (Yanong and Floyd, 2002). Cases of streptococcosis in wild fish are

reported in several publications (Baya et al. 1990; Colorni et al. 2002; Zlotkin et al.


The possibility of streptococcal outbreaks is increased if fish are stressed, as

happens with non-optimal water temperature, low dissolved oxygen, high nitrite levels,

and high culture densities (Bunch and Bejerano, 1997

Perera et al. 1997).

Several species of streptococci have been isolated from fish. Alpha-hemolytic

streptococci generally cause granulomatous inflammation; whereas beta and gamma-

hemolytic streptococci cause systemic infection localized in various organs (mainly

spleen, kidney, brain, liver, and eye), with the presence of granulomas in some


(Ferguson et al. 1994; Chang and Plumb, 1996). An alpha-hemolytic streptococcus was

isolated from hybrid tilapia (Al-Harbi, 1994) showing erratic and whirling swimming,

exophthalmia, hemorrhagic areas on the body surface, and the presence of fluid in the

abdominal cavity as the main clinical signs. Another alpha-hemolytic streptococcus, with

characteristics similar to S. faecalis and

faecium, was isolated from Siganus



affected the central nervous system. No external ulcerations and hemorrhages or other

signs were observed. Streptococcus difficile and S. shiloi were first isolated in Israel from

tilapia and trout with meningoencephalitis (Eldar, et al. 1994; 1995b). Two efficacious

vaccines against S. difficile, a formalin-killed and a bacterial extract based, have been

developed (Eldar et al. 1995b). Other streptococci isolated from fish are S. agalactiae

isolated from seabream and wild mullet (Evans et al. 2002), and S. parauberis isolated

from turbot (Domenech et al. 1996; Romalde et al. 2000).

Streptococcus iniae is the most commonly isolated streptococcus from fish

(Bromage et al. 1999; Eldar et al. 1999; Fiske and Johnson, 1998; Perera et al. 1994;


Sako, 1998). Streptococcus iniae causes a systemic disease characterized by

histological and external pathology and by characteristic fish behavioral changes. The

disease is associated with a strong inflammatory response by the host where acute

necrotizing reactions and systemic neutrophil and macrophage exudation can be observed

in several organs and tissues in conjunction with the presence of high numbers of

gram-positive cocci (Ferguson et al. 1994; Neely et al. 2002).

Externally, streptococcosis-induced lesions include pronounced congestion and

hemorrhages, especially around the base of the pectoral fins and over the heart.

Exophthalmia, corneal opacity, and intra-ocular and periorbital hemorrhages can also be

observed (Ferguson et al. 1994; Liu et al. 1990; Perera et al. 1998).

The main organs affected by S. iniae are the brain, spleen, and posterior kidney.

Pathological changes associated with the brain and nervous system are typical of


granulomas and massive cellular infiltration can be observed (Eldar et al. 1995a; Perera et

al. 1998). Spleen and posterior kidney are common targets because of their high

capability to trap bacteria. In the later stages of infection, there is an increase in size,

expressed as percent body weight, of the spleen and kidney (Ikeda and Minami, 1982).

Analysis of the spleen can reveal congestion and fibrinoid necrosis of the ellipsoidal

sheaths with loss of reticular macrophages. Bacteria can be widely scattered throughout

the parenchyma of the spleen, but can also be observed within macrophages (Ferguson et

al. 1994; Perera et al. 1998). Microscopic examination of the posterior kidney can reveal

severe congestion of the renal parenchyma, and pronounced degeneration and necrosis of

many tubules (Ferguson et al. 1994; Perera et al.


A large number of bacteria can be observed in the pericardial area, especially

within the hemorrhagic musculature associated with the pectoral girdle (Ferguson et al.

1994; Liu et al. 1990; Perera et al. 1998). Other organs that can be affected are the liver,

where sub-capsular petechial hemorrhages and granulomas have been observed; and the

intestinal tract, where bacteria have been observed in the lamina propria; but not in the

mucosa (Ferguson et al. 1994; Perera et al. 1998). An elevated number of bacteria can be

observed within blood vessels, especially in the capillary network of the gills, cranial

meninges, retina, and choroidal rete. The presence of bacteria in the capillaries caused

occasional thrombosis (Ferguson et al. 1994; Perera et al. 1998). Streptococcus iniae can

cause typical behavioral changes such as a swimming spinning pattern (Eldar et al.

1995a; Perera et al. 1998).

However, differently from the studies of Ferguson et al. (1994) and Perera et al. (1998),

during histological examination no granulomas were observed.

One concern with Streptococcus iniae is the possible transmission of the bacteria

from fish to humans. Few cases have been reported. Apparently the probability of such

transmission is low. It is a concern mainly with immunocompromised individuals (Goh et

al. 1998; Lau et al. 2003; Weinstein et al.

1996; 1997). However, recent reports in China

suggest a possible higher risk of S. iniae infections in Asian populations due to their habit

of cooking and eating fresh fish (Lau et al. 2003). Pulse-field gel electrophoresis (PFGE)

has showed that S. inae strains that cause disease in fish are genetically diverse from

commensal strains isolated from nondiseased fish (Fuller et al. 2001; Weinstein et al.


4.2. Host-Pathogen Interactions

Streptococcus iniae, like streptococci belonging to the Group B (GBS), shows a

tropism for the brain. This tropism is one of the main causes of death. The invasion rate

of S. iniae in the different organs has been experimentally observed in yellowtail (Kusuda

and Kawai, 1982

Kusuda and Kimura, 1978; Sako, 1998). In the first

days after

challenge, there is an increase in the bacterial count of blood, spleen, and kidney;

followed by a decrease or a constant count value. In contrast,

days after challenge

there is a constant increase of bacterial count in the brain until the onset of mortality. The

exact mechanism used by S. iniae to cross the blood-brain barrier (BBB) is still not well

understood. It involves production of the cytolisin streptolysin

S (SLS) required for local


could elicit its own endocytosic uptake from the microvascular endothelial cells and so

pass the BBB (Nizet and Rubens, 2000). Otherwise, as observed with S. suis, S. iniae

could adhere and damage the microvascular endothelial cells through the action of SLS,

and then proceed from the circulation to the central nervous system (Charland et al.


The pathogenesis mechanisms of GBS have been studied in detail and resemble

those of systemic infection with S. iniae. For this reason, the GBS pathogenesis

mechanisms may be very useful in understanding the possible mechanisms used by S.

iniae (Miller and Neely, 2004). Invasion of the brain microvascular endothelial cells by

GBS causes an evident cellular injury and increased cell permeability because of the

cytotoxicity activity of the beta-hemolysin/cytolisin (b-h/c) that causes formation of pores

on the epithelial tissue (Nizet et al. 1997

Ring et al. 2002a). Streptococcal mutants with a

hyperhemolytic phenotype give evidence of the involvement of b-h/c on host tissue

damage. Histological examination of animals infected with these mutants shows higher

degree of necrotic tissue with these strains of bacteria (Ring et al. 2002a). The b-h/c is

also important for penetrating the BBB. In animals infected with mutants lacking b-h/c,

fewer bacteria penetrate the BBB, and the mortality rate and bacterial numbers in brain

increase with increased levels of b-h/c activity (Nizet et al. 1996). In addition, b-h/c has

proinflammatory, proapoptotic, and proinvasive properties that could contribute to

disease pathogenesis. Mice b-h/c production is associated with increased local and

systemic levels of proinflammatory cytokines, such as IL-6 and IL-1 released by injured


migration, and stimulate the respiratory burst, degranulation, and adherence of leukocytes

and neutrophils. The development of inflammatory exudates then promotes further

damage of the BBB and stimulation of apoptosis of neurons and other cells (Doran et al.

2003; Nizet et al. 1997). Leukocytic infiltration into the cerebrospinal fluid is a major

clinical diagnostic marker for acute bacterial meningitis.

Group B Streptococci may also induce epithelial damage indirectly by altering

host cell processes, such as the cell peroxidase-H202 system that can damage adjacent

cells by producing oxygen radicals. The observed hypotension in septic shock could also

be caused by an excess production of nitric oxide (NO) stimulated by the GBS (Ring et

al. 2000; 2002b). The GBS can also modify the arachinoic acid metabolism favoring the

production of prostaglandin E2 (Sbarra et al. 1987). Bacteria replication within CNS may

then provoke an overwhelming host inflammatory response that, in combination with the

disruption of the BBB, could be fatal (Doran et al. 2003). Streptococcus iniae, as with

GBS, could also secrete other factors that break down the extracellular matrix helping to

spread the bacteria, or that inactivate the inflammatory mediators, such as C5a, thus

decreasing recruitment, and the opsonophagocytic capacity of neutrophils (Cheng et al.

2002; Liu and Nizet, 2004).

The polysaccharide capsule of Streptococcus protects the bacteria by decreasing

adhesion of immune cell receptors to the bacteria, (Tamura et al. 1994). Sialic acid

residues on the capsule can inhibit activation of the complement alternative pathway,

thereby interfering with the opsonophagocytic killing mechanism of the neutrophils, and

macrophages. The ability of S. iniae to survive in macrophages may be useful in

establishing a systemic bacteremia and for crossing the BBB within macrophages

migrating in the CNS. The ability to overcome the immune response of macrophages is

one characteristic that allows S. iniae to establish an infection. Another defense

mechanism used by S. iniae is inhibiting apoptosis and increasing necrosis of natural

cytotoxic cells in fish (NCC) (Doran et al. 2003; Fuller et al. 2001; Martin et al. 1992;

Taylor et al. 2001

Zlotkin et al. 2003).

Data collected in recent research shows that recognition of GBS by immune cells

and their later activation involves Toll-like receptors (TLRs). The same receptors may

also be involved in recognition of S. iniae (Henneke et al. 2001; 2002).

Both the innate and acquired parts of the immune system are involved during

Streptococcus infections. Nonspecific cytotoxic cells (NCC) are part of the nonspecific

immune system, and play an important role in fish during bacterial infections (Evans et

al. 1998; 2000a; Taylor et al. 2001; Weinstein et al. 1997). Nonspecific cytotoxic cells

have cytotoxic action. They release cytokines that increase phagocytosis by macrophages,

activate antigen-processing cells, and augment the inflammatory response. Nonspecific

cytotoxic cells become biochemically activated; this increases their cytotoxicity after the

first encounter with a stressor agent.

These responses are different from those occurring

during normal homeostatic regulation, where the encounter of NCC with a target cell

represented by a parasite or a tumor cell, causes apoptosis of the NCC. Prokaryotic DNA

can activate the cytotoxicity mechanism in NCC, and it is involved in the interactions of

cytotoxicity. Also treatment of bacterial DNA with DNAase or methylation of all

cytosine bases, abrogates the cytotoxic enhancement (Oumona et al. 2002).


observations indicate that DNA binding receptors on NCC and other immune cells have a

mechanism for preferring certain nucleotides. This mechanism should be similar to MHC

recognition and presentation of peptides, where certain amino acids serve to anchor the

peptide to the MHC antigen. Studies using synthetic oligonucleotides shown that

cytotoxic activation is enhanced by bacterial DNA fragments containing -GACGTT- and

-GTCGTT- sequences. These results also suggest that binding and internalization of

bacterial DNA is required for initiation signaling and cytokine release. Recognition of

bacterial DNA by NCC appears to be determined by a combination of sequence and

conformational structure.

This mechanism may be similar to the recognition of antigenic

determinants by T and B-cell receptors. Apparently, different subsets of NCC could exist

with different specificities for bacterial DNA receptors, or individual NCC could contain

receptors with different specificities (Babiuk et al. 2000; Krieg, 2000; Scheule, 2000).

These results are important to better design and understand the mechanism of action of

DNA vaccines.

Streptococcus iniae inhibits apoptosis and increases necrosis of NCC. The

advantage of a decreased number of apoptotic cells is that apoptotic cells are

phagocytosed and do not initiate an inflammatory response. In contrast, cells undergoing

necrosis are eventually lysed, and release several compounds that will activate the

inflammatory response. Another effect is increased antigen processing which increases


some stains of S. iniae where phosphatidylserine expression is inhibited in the NCC outer

membrane. Recognition of phosphatidylserine by macrophages is needed for initiating

the phagocytosis process (Taylor et al. 2001).

The possibility that protection against bacteria in fish mainly involves the cellular

immune system of the non-specific defense system could explain the results of several

vaccine studies (Eldar et al. 1995b; Evans et al. 2000b; Lillehaug et al. 1993; Taylor et al.

2001). They observed no correlation between antibody titers and protection. In other

studies the protection given by vaccination lasted only for a few months (Eldar et al.

1995b; Hortvitz et al. 1997

Iida et al. 1981).

Nevertheless, the humoral part of the immune system is important in the host

defense mechanism. In yellowtail (Kusuda and Tanaka, 1988) specific antibodies

increased the phagocytic activity and index of macrophages. This effect could not be

duplicated if complement alone was added to cell culture. However with both antibody

and complement, the phagocytic activity and index were significant higher. These results

suggest that during bacterial infections in fish, activation of the classical pathway of the

complement by antigen-antibody reaction is an important mechanism of defense.

Antibody production appears to be stimulated more by toxins released by

Streptococcus than by the bacteria itself (Kusuda and Takagi, 1983).

These results

highlight the possibility that toxoid vaccines may work better than bacterial vaccines.

4.3. Antibiotics Treatments and Vaccines

The antibiotic used most to treat Streptococcus spp. infections is erythromycin

50S subunit of the bacterial ribosome eukaryotess do not have the 50S ribosomal subunit)

(Treves-Brown, 2000). Some erythromycin resistant strains of bacteria present mutations

in components of the 50S ribosomal subunit, and for this reason the antibiotic fails to

bind the ribosome. The non ionized from of the drug is considerably more permeable to

cells, and this probably explains the increased antimicrobial activity that is observed in

alkaline pH. The oral recommended dose for fish is 25-50 mg erythromycin/kg fish/day

for 4-7 days. The injectable dose is 10 to 20 mg/kg fish for 14 days.

Oxytetracycline was tested as a possible treatment for S. iniae, but efficacy varied

among experiments. Oxytetracycline inhibits the bacterial protein synthesis by blocking

the attachment of the complex tRNA-amino acid to the ribosome. Oxytetracycline can

less likely inhibits the protein synthesis in the host, because eukaryotic cells do not have a

tetracycline uptake mechanism (Treves-Brown, 2000). In blue tilapia infected with S.

iniae, administration for 14 days of medicated feed containing oxytetracycline at levels

between 50 and 100 mg OTC/kg fish/d fed at 2% body weight/day increased the survival

rate between 45 and 98% (Darwish et al. 2002). The minimum inhibitory concentration

(MIC) of oxytetracycline against multiple S. iniae isolates indicated a general sensitivity

at concentrations of 0.25-0.50 jig/mL. Completely different results were observed in

hybrid striped bass infected with S. iniae (Stoffregen et al. 1996). In this case, treatment

with Terramycin for Fish (Pfizer, Inc., Connecticut, USA) at doses of 82.7 mg drug/kg

fish per day for 10 days was unsuccessful. The different results could be explained partly

by the higher resistance of this bacterial strain to oxytetracycline with a MIC of


Amoxicillin has been used to treat infections from S. iniae with a range of dose of

30-120 mg/ kg fish body weight administered for 8 consecutive days (Darwish and

Ismaeil, 2003). Amoxicillin, as other penicillins, interferes with the bacteria cell wall

synthesis preventing the development of the peptidoglycan layer (Treves-Brown, 2000).

Another antibiotic tested as a possible treatment for streptococcosis in one study

is josamycin (Takemaru and Kusuda, 1988). Josamycin inhibits the protein synthesis by

binding the 23S rRNA or 50S bacterial ribosomal subunit (Treves-Brown, 2000). In the

study of Takemaru and Kusuda (1988), the MICs ofjosamycin, erythromycin, and

spiramycin were compared. Observed josamycin MIC values ranged between 0.10 and

0.39 pg/mL. They were higher than the values recorded for spiramycin, but lower than

the ones for erythromycin. The development of resistance of the bacteria to the antibiotics

after 9 successive subcultivations was 4-fold higher in MIC value for josamycin and 8-

fold higher for the other two antibiotics. These results support josamycin as a possible

treatment against streptococcosis.

Florfenicol is one of the newest antibiotics on the market, and it has been used

against several bacteria such as Aeromonas salmonicida, Yersinia ruckeri,

Flavobacterium psychrophilum, Vibrio salmonicida (Michel et al. 2003; Nordmo et al.

1998), and Edwardsiella ictaluri (Gaikowski et al. 2003). This antibiotic has also been

tested in ornamental fish (Yanong et al. 2002). Florfenicol binds reversibly the large

ribosomal subunit of bacteria and eukaryotes, and it also binds the peptidyl transferase

enzyme inhibiting the transfer of the growing polypeptide to the next amino acid


Streptococcus infections, as is reported by the label of Aquaflor (active ingredient

Florfenicol 500 mg/g, Schering Plough, Union, New Jersey) and from the results of some

investigations (Yanong et al. 2004).

Several vaccines have been developed against streptococci, mainly for tilapia and

rainbow trout. Protection levels given by the vaccines varied among experiments and

vaccine formulations (Akhlaghi et al. 1996; Bachrach et al. 2001; Eldar et al. 1995b;

Eldar et al. 1997

Klesius et al. 2000; Klesius et al. 2001; Lim and Klesius 2001a;

Palacios et al. 1993; Sakai et al. 1987

Shelby et al. 2002a). An effective combined

vaccine against Streptococcus sp. and Vibrio anguillarum has also been developed (Sakai

et al. 1995).


Fish Immune System

Fish are a heterogeneous group of animals that are physiologically and

anatomically very different. They are divided in the classes Agnatha, Chondrichthyes,

and Osteichthyes. Agnatha (Cyclostomes: hagfish and lamprey) lack a well-structured

mouth with bones such as maxillary and palatine, and has a cartilaginous skeleton.

Chondrichthyes (Elasmobranches: shark, rays) do not have a swim bladder and ossified

bones, but only a cartilaginous structure. Osteichthyes (salmon, tilapia) have ossified

bones. Teleost are included in the class Osteichthyes. Complexity of the immune system

differs substantially among Cyclostomes, Elasmobranches, and Osteichthyes. It is

possible to observe a gradual increase in complexity of the immune system (immune

organs, cells, soluble factors, and receptors) among these classes.

In mammals, the main lymphoid organs are thymus, spleen, bone marrow, lymph

nodes, and gut-associated tissues. Fish have lymph vessels, but lymph nodes have not

been observed, at least not with the structure present in mammals.

Bone marrow and

germinal centers also do not exist in fish, so other organs and tissues perform their

functions (Zapata et al. 1992


The main lymphoid organs in teleost fish are

thymus, kidney, spleen, and gut-associated lymphoid tissue. In Elasmobranches the main


(Culbreath et al. 1991; Rumfelt et al. 2002). Lampreys do not have true lymphoid organs,

but lymphohemopoietic tissues are present in various locations such as within the nephros

and intestinal lamina propria. Neither lampreys nor hagfish have a thymus, but lymphoid

accumulations that could be homologous to thymus, are present in the branchial area

(Mattison and Fange, 1977

Rowley et al. 1988). In fish, as in mammals, there is an

involution of the thymus with age; and in fish this process can also be related to season or

hormonal changes associated with sexual maturation (Honma and Tamura, 1984). In

some fish such as rays, flatfish and cod, the thymus continues to grow after sexual

maturation, whereas in eels it has already involuted before sexual maturity. In carp and

primitive sharks, the thymus does not appear to involute. In Zoarces viviparous (Bly,

1985) the thymus reaches is maximum size at 1 month preparturition when it occupies

0.83% of the total body volume. Thereafter, it decreased in size and by 3 months

postparturition occupies only 0.048% of the total body volume. Thymus involution is

characterized histologically by an increase in connective tissue, a decrease in lymphocyte

density, infiltration of adipocytes, the presence of reticular cells and collagenous fibers,

and the formation of epithelial cysts. However, the relationship of such changes to the

immune capabilities in older fish is not clear.

In mammals, the outer layer of the skin (stratum lucidum) is keratinized and

confers a physical barrier to the penetration of pathogens. In fish this layer is not

keratinized but is covered with mucus. The mucus protects the skin, decreasing the ability

of pathogens to attach to the skin, and it also contains several antibacterial compounds,


The fish immune system includes cells and soluble mediators that are involved in

the innate and acquired immunity. Fish present both B and T-cells that are involved in

acquired immunity. Although T-cells have not been formally identified in fish because

they lack.specific molecules such as CD4, CD8 or CD3, the genes encoding polypeptides

homologous to the TCR molecules have been reported in teleosts and elasmobranches

(Partula et al. 1995). Although in mammals the T-cells develop only in the thymus, it

appears that in fish an extrathymic population of T-cells is present. The results of some

studies have suggested the possibility that the lymphoid cell populations of the thymus

and kidney are acquired independently, as it has been observed in early thymectomy fish

that still have a normal lymphocyte population in the kidney (Grace and Manning, 1980).

B-cell lymphocytes can be recognized in fish by the presence of immunoglobulins on the

cell membrane.

Macrophages, polymorphonuclear cells and non-specific cytotoxic cells are

involved in innate immunity in fish, as in mammals. Non-specific cytotoxic cells (NCC)

in fish are considered to be equivalent functionally to mammalian natural killer cells

(NK). NCC can lyse a wide spectrum of mammalian tumor cells and protozoan parasites

(Evans and Jaso-Friedmann, 1992) through the mediation of function-associated

molecules on the cell membrane (Harris et al. 1992). NCC cells are the smallest

leukocytes possessing a very pleomorphic, clefted nucleus and relatively little cytoplasm

with no cytoplasmic granules.

Villi are present on their surface, and they attach to their

target using long, membranous filaments, allowing many binding sites (Secombes, 1996).


and Jaso-Friedmann, 1992). In hagfish only a single population of granulocytes has been

reported (Mattison and Fange, 1977), but different subpopulations are distinguishable in

lampreys (Rowley et al. 1988). In Elasmobranches several subpopulations of

granulocytes have been identified, but there is an enormous heterogeneity among species

(Rowley et al. 1988). In teleost fish, different types of polymorphonuclear cells have been

detected, but there is an enormous variation among species. Furthermore, eosinophilic

and basophilic cells, when present, are not equivalent in function to those corresponding

to mammalians (Rowley et al. 1988; Zapata et al. 1996). In fish, as in mammals,

macrophages act as antigen presenting cells through MHC II, they activate lymphocytes

releasing IL-1, and they release reactive oxygen and nitrogen molecules for killing

pathogens (Ellsaesser and Clem, 1994). Fc receptors have also been identified on fish

macrophages (Haynes et al., 1988).

In Elasmobranches and Osteichthyes, as in mammals, antibodies represent the

humoral section of the adaptive branch of the immune system, but differences are

present. In Cyclostomes imunoglobulins have been observed in hagfish (Kronenberg et

al. 1994), but not in lamprey (Varner et al. 1991). IgM is present in fish, and some

species also produce IgG or unique Igs, but IgA, IgD, IgE have not been found. In

cartilaginous fish, at least sharks, skates and rays, three different types of imunoglobulins

have been found: IgM, IgNAR and IgW (Rumfelt et al. 2002; Voss and Sigel, 1972). Igl

exists in approximately equal amounts as a monomer (7S) and a pentamer (19S). In

teleost Igs appears to be a tetramer composed of four monomeric subunits, each


been observed, but this could be caused from a lack of detection (Lobb and Olson, 1988).

In fish the assembly of antibody molecules via disulfide crosslinking appears to be a

means by which structural diversity is generated.

This is distinctly different from what is

observed in mammalian species where intersubunit disulfide cross-linking appears to be

employed in a more uniform manner (Sanchez and Dominguez, 1991). In mammals,

somatic mutations cause a production of antibody with an increased affinity for the

antigen. Generally the process of somatic mutation has not been detected in fish, with the

exception of some species of elasmobranches (Hinds-Frey et al. 1993). In mammals, a

second exposure to an antigen stimulates a greater immune response with a faster and

greater production of imunoglobulins (mainly IgG). in contrast, in fish there is a lower

production of imunoglobulins after a second exposure to a pathogen compared to the

logarithmic increase in monomeric IgG observed in mammals (Arkoosh and Kaattari,

1991). Studies showed that during a secondary response, the increase of specific

antibody-producing cells is caused by a clonal proliferation of native and primed

lymphocytes, indicating that in fish immune memory is not well developed (Brooks and

Feldbush, 1981). Another important difference among the immune system of fish and

mammals is the absence of IgA in fish, even if a homologous function of IgA could be

present. Some studies have detecteded antigen specific antibody induced in the mucus of

the skin in orally vaccinated fish, while little serum antibody was produced (Rombout et

al. 1993). This is similar to what happens in humans, where the local exposure to a

pathogen on a mucosa stimulates IgA production in all the body mucosa.


identified in fish (Culbreath et al. 1991; Yano, 1996). In general, the complement system

in fish is effective against non-virulent gram-negative bacteria, but not against virulent

gram-negative and gram-positive bacteria. The alternative complement pathway can be

inhibited by the large amount of sialic acid present on virulent strains of gram-negative

bacteria (lida and Wakabayashi, 1993). In Cyclostomes, only components of the

alternative complement pathway have been detected, but the complement system lacks a

cytolytic activity (Ishiguro et al. 1992; Nonaka, 1994). Elasmobranches possess the

classical and alternative complement pathways that stimulate the formation of the

membrane attack complexes on target cells (Culbreath et al. 1991). The complement

system pathways in bony fish are very similar to those of mammals (Yano, 1996).

Lysozyme is important for fighting all types of pathogens. Lysozyme is

distributed mainly in tissues rich in leucocytes, such as the head kidney, skin, gills, and

the alimentary tract. Fish lysozyme, in contrast to mammalian, has an antibacterial

activity against not only against gram-positive bacteria but also gram-negative bacteria in

the absence of complement (Grinde, 1989). Interferon (IFN) has been detected in bony

fish, but not in cyclostomes or cartilaginous fish (Yano, 1996). Transferrin and C-reactive

proteins have been found in fish with a similar activity as in mammals (Nakanishi et al.


Yano, 1996).

Cytokines, TCR components, major histocompatibility complex I and II (MHC I

and MHC II) receptors have been identified in Elasmobranches and Osteichthyes, but the

genetic polymorphism of MHC genes is considerably lower than in mammals (Knight et


One of the major differences between the immune system of fish and mammals is

the effect that temperature has on the system. In fish, the rate of the development of the

immune system is influenced more by the external temperature and by the size of the fish

than by age (Tatner, 1996). The higher the temperature, the faster the fish grows and the

earlier the immune system develops. Rainbow trout start to produce antibodies when they

reach a weight of approximately 0.24 g without any dependence on their age (Tatner and

Manning, 1983; Tatner, 1986).

Seasonal changes in water temperature influence the functions of the fish immune

system. For example, the number of peripheral blood lymphocytes is minimum in winter

and maximum in the summer, although there are exceptions as in the brown trout where

the levels of lymphocytes are low in winter and summer and high in spring and autumn.

In winter, it is possible to observe regression in lymphoid tissues, particularly the thymus.

In spring and summer there are waves of thymocyte production, while degenerated

lymphoid cells can be found from the beginning of autumn and throughout winter. Higher

temperatures cause faster and higher antibody production (Rijkers et al. 1980).

Temperature has a different influence on the various parts and cells of the immune

system. For example, virgin T-cells rather than memory T-cells, B-cells or accessory

cells are particularly susceptible to the inhibitory influences of low temperature (Bly and

Clem, 1992; Miller and Clem, 1984). Low temperatures have a higher inhibitory effect on

T-cells than on B-cells. A possible explanation for this phenomenon is that B-cells have a

higher capability to modify the fluidity of their plasma membrane. B-cells can rapidly


phospholipids causing an increase of membrane viscosity. This may explain the

suppressed function at low temperatures. In fact, a lower membrane viscosity would

affect receptors and signal pathways that involve membrane-associated proteins (Bly et

al. 1990). Even if low temperatures have an inhibitory effect, the fish immune system is

able, at least in part, to counterbalance this effect. For example, in channel catfish serum

complement lytic activity is low in winter, while the antibody affinity increases. In this

way antibody can act as an opsin and activate the complement cascade (Clem and Small,

1970; Hayman et al. 1992). A similar mechanism involves C-reactive proteins in channel

catfish. CRP levels are low in winter, but at the same time their ligand avidity increases.

Temperature also influences the mammalian immune system, but in a minor way. This

difference is caused mainly by the ability of mammals to maintain a constant body


The immune system of sharks (Epalzeorhynchos sp.) has not been studied in any

research. However, because sharks belong to the family Cyprinidae, it is very plausible

that there are similarities between the shark, the common carp, Cyprinus carpio, and the

danio, Danio rerio, immune systems and ontogenesis of the immune systems. Because

sharks are warm water fish, as danio, it is plausible the timelines of the ontogenesis of the

immune system is more similar between these two species than between shark and carp.

5.2. Ontogenesis of the Immune System

The ontogenesis of the fish immune system has been studied only in a few species

(Ellis, 1988; Honma and Tamura, 1984; Tatner, 1996), such as nurse shark,


Zapata, 1998; Trede and Zon, 1998; Trede et al. 2001; Willett et al. 1997ab; Willett et al.

1999), common carp, Cyprinus carpio (Botham and Manning, 1981), rainbow trout,

Salmo gairdneri (Grace and Manning, 1980; Tatner and Manning, 1983; Tatner, 1986),

viviparous blenny, Zoarces viviparous (Bly, 1984; 1985), European flounder, Platichthys

flesus (Pulsford et al. 1994), channel catfish, Ictalurus punctatus (Petrie-Hanson and

Ainsworth, 2001), and angler fish, Lophius piscatorius (Fange and Pulsford, 1985).

Histological, immunological and biomolecular techniques have been used in these

studies. In histological analyses, tissues are first stained and then observed

microscopically. Immunological techniques utilize rabbit or mouse monoclonal

antibodies against the putative B and T-cells that can be detected by ELISA or

fluorescent staining techniques. Biomolecular techniques utilize the expression of several

genes involved in the ontogenesis process. One of the most delicate steps in this type of

analysis is the design of specific mRNA primers. Usually mRNA is isolated, amplified by

PCR and identified by Northern Blot analysis. In situ hybridization techniques can be

used for the detection of the time and of the organs where specific genes are expressed.

5.3. Immune Organs and T-Cell Ontogenesis: Histological Approaches

The thymus is the first immune organ to develop in fish and the first to become

lymphoid (Bly, 1985; Botham and Manning, 1981

Grace and Manning, 1980; Tatner and

Manning, 1983; Tatner, 1986; Trede et al. 2001; Willett et al. 1999). The second organ to

develop is the kidney followed by the spleen. The gut-associated lymphoid tissue

becomes lymphoid later than the kidney and the spleen. In general, the spleen remains


organs in rainbow trout from 1 to 15 months of age. Thymus, kidney and spleen grew

steadily over this period, but all attained their maximum relative weights expressed as a

percentage of body weight, at 2-3 months of age, indicating a time of intense activity and

importance in the maturation of the lymphoid system. The total number of leukocytes in

the organs increased with age, but when the size of the organ and the size of the fish were

taken into account, the actual leukocyte numbers decreased with age. The thymus showed

intense mitotic activity in the first few months, then slowed down. There were no signs of

massive cell death, and it was assumed that emigration of cells to the peripheral lymphoid

organs must have been occurring. Histological changes indicative of involution were

present in the thymus from 9 months onward. The spleen and the kidney showed an

increase in melanin deposition with age, with the kidney retaining a more varied range of

leukocyte types than the spleen. The number of circulating lymphocytes in the blood

remained remarkably constant throughout the first year of life.

The progressive development of the immune organs was also observed in

common carp (van Loon et al. 1982). The thymus contained the largest

pool, 70% or 3

lymphocyte cell pool, 5

ymphoid cell

x 105 cells at day 28, in the first 4 weeks. At two months of age the total

x 106 cells, was distributed for 38% in the mesonephros, 32% in

the thymus, 15% in the pronephros, 12% in the peripheral blood and 3% in the spleen.

In Zoarces viviparous, a viviparous marine fish, thymus has a paired structure

localized behind the last gill arch (Bly, 1985). The thymus lies in the angle between the

operculum and the body wall, and is covered by a single layer of epithelium. It is evident


densely stained, and is comprised of mainly lymphocytes and few reticular cells. The

other inner zone is lightly stained and contains fewer lymphocytes. However, there is no

distinct boundary between the two zones. The thymus is covered by a thin connective

tissue capsule and by collagen fibers. Macrophages can also be found. The head kidney is

seen at 3 months preparturition, and it contains mainly a few kidney tubules surrounded

by small erythrocytes. At 1 month preparturition the kidney also contains lymphoid

tissue, where granulocytes and macrophages can be found. The spleen is first observed at

two months preparturition. The organ is in close contact with the pancreas.

Differentiation in the spleen is evident at two months postparturition when two zones are

apparent: lightly stained lymphoid areas aggregated around the trabeculae, and dark

staining areas composed mainly of erythrocytes. These two zones may be analogous to

the white and red pulp of mammalian spleen. In the spleen, granulocytes and

macrophages are also present.

In rainbow trout (Grace and Manning, 1980; Razquin et al. 1990; Tatner and

Manning, 1983; Tatner, 1986), thymus is first observed at 4 days before hatching. At this

stage it consists of a thickened area of the pharyngeal epithelium containing a few

lymphoid cells interspersed among the epithelial cells, and of a thin connective tissue. At

4 days after hatching, thymic stellate cells, numerous thymocytes, and some macrophages

are present. The first connective tissue trabeculae in the thymic parenchyma are observed

about 14-21 days after hatching, and incipient subcapsular and outer thymic zones appear

at 1 month age. Thymic cysts develop from 1-2 months after hatching. Renal


intertubular loose connective tissue of the developing kidney. At hatching, those foci tend

to fuse and contain myelocytic cells interspersed among the cell processes of reticular

cells. Scattered large lymphoid cells can be clearly observed at 4 days after hatching, and

from this stage the amount of lympho-haemopoietic tissue increases in the kidney,

retaining a similar structure in both pro- and mesonephros, except for the absence of renal

tubules in the former. The spleen develops slowly, containing a poorly organized

haemopoietic red pulp at 12 days after hatching, formed by loose cell cords, mainly

occupied by erythroid cells and enlarged blood sinusoids. Some scattered lymphocyte,

macrophages and melanomacrophages occurred in the spleen about 14 days after

hatching. Small amounts of splenic white pulp can be observed in

month old trout,

when ellipsoids also develop.

The thymus of the channel catfish, Ictalurus punctatus (Petrie-Hanson and

Ainsworth, 2001), is first colonized by monocytes, suggesting that these cells may help

thymocyte proliferation and differentiation. The first immature lymphocytes can be seen

in the thymus after 5 days. Mature lymphocytes are detectable at 10 days post-hatch. The

intermediate stage of T lymphocytes is concentrated in the peripheral lighter staining area

(cortical zone); meanwhile mature lymphocytes are concentrated in the central darker

staining area (medulla region). Until approximately 3 months post hatch, the total number

of lymphocytes increases, then reaches a plateau.

In zebrafish, Danio rerio, a thymic primordium consisting of two layers of

ePithelial cells appears at 54 h after fertilization (Willett et al. 1997a). By 65 h


morphological mature. At 72 h post-fertilization kidney glomeruli can be seen. At 96 h

post-fertilization the pronephros begins to contain hematopoietic cells, mainly developing

erythrocytes and granulocytes. Fibroplastic reticular cells, that will constitute the

supporting apparatus for the developing hematopoietic cells, appear at 7 days

post-fertilization. Lymphocytes are detected after 3 weeks.

The first appearance of lymphocytes within an organ, rather than the first

appearance of the organ itself, is considered to be a better index of the maturation of the

lymphoid system (Tatner, 1996). The reproductive strategy among fish (such as the

length of egg incubation) and the environmental temperature are the main factors that

influence the time of maturation of the immune organs. The warmer the water

temperature, the faster development occurs. In general, the longer the hatching time, the

slower the development. For example trout eggs are incubated at 140C and hatch around

18 days; however, when they are incubated at lower temperatures, they develop slower.

Carp eggs are reared at 220C and hatch after 36 h. The gestation period in

viviparous is

4-5 months. Table 5-1 summarizes the first appearance of the immune system organs;


the first appearance of lymphocytes in those organs, and Table 5-3 when the

thymus can be considered immunologically mature.

Development of cell-mediated protection can be better analyzed using allograft

rejection responses. Allograft rejection in carp can occur as early as day 16 post-hatch

(Secombes and van Groningen, 1983), at day 26 post-hatch in rainbow trout, and at

months in Tilavia mossambica. These results suggest that cytotoxic T-cells mature earlier


It still is not well understood if during ontogenesis lymphocytes migrate from the

thymus to the kidney. The results of several studies have outlined the possibility that the

lymphoid cell populations of the thymus and kidney are acquired independently. In fact,

early thymectomized fish have a normal lymphocyte population in the kidney (Grace and

Manning, 1980).

5.4. Immune Organs and T-Cell Ontogenesis: Biomolecular Approaches

Zebrafish have been used as a model in biomolecular studies for immune system

ontogenesis (Trede and Zon, 1998; Trede et al. 2001). One main reason is that the

genome of the zebrafish has been almost completely cloned. Moreover, because of the

small size (5 cm) of this fish, they are reared easily, they can be stocked at high density

(up to 70 adult per 19 L tank), their generation time is relatively short (3 months), and

females produce a large number of progeny (200-300 every week).

outside the mother's

The embryos develop

body and are easily accessible for physical manipulation, visual

inspection under dissecting microscope, whole mount in situ hybridization, and gene

transfer experiments. Thanks to the relatively short generation time, zebrafish lineages

may be created for specific mutant genes, and may be used for positional cloning of such

mutant genes. Zebrafish mutants with severe hematopoietic defects can still live beyond

the onset of thymopoiesis, this permits the analysis of lymphocyte development in

mutants that would be destined in the mouse to suffer early embryonic death from anemia

or other developmental defects (Trede et al. 2001). The juveniles of this fish are almost

transparent allowing a direct visualization of the temporal and spatial development of


The utilization of green fluorescent protein genes, the expression of which is

under the control of the promoter of the studied genes (such as Ick, for example), helps in

the direct visualization of the ontogenesis of the immune system. Additionally, proto-

oncogenes can be expressed under the control of the studied genes so as to cause tumors.

The animals with these malignant tumors can be utilized in several ways. For example,

monoclonal antibodies could be produced against zebrafish homologs of T-cell markers

such as CD3, CD4 and CD8. These transgenic experiments would also provide an animal

model of leukemia in lower vertebrates (Trede and Zon, 1998).

New techniques to study gene expression will apply retroviral insertions for the

generation of zebrafish mutants. A major advantage of this approach is the ease with

which genes responsible for the mutant phenotypes can be cloned (Trede et al. 2001).

These characteristics make the zebrafish an ideal model for studying the ontogenesis of

the vertebrate immune system.

As soon as the first lymphocytes are visible they are still not functionally active or

mature. For this reason some studies (Hansen, 1997

Hansen and Zapata, 1998; Trede and

Zon, 1998; Trede et al. 2001; Willett et al. 1997ab) have analyzed the gene expression of

the enzymes including RAG], RAG2, TdT,

Ick, GA TA, ikaros, or TdT, as markers of the

maturation of the immune system. These enzymes are involved in the maturation process

of the T and B-cells acting on the TCR and Igs

genes rearrangement. These studies have

confirmed the results obtained by histological investigations (Willett et al. 1997a). Brain

and thvmus express ikaros, indicating where early lymphoid development occurs.


Zon, 1998) expression of RAGI and RAG2 were first observed in the thymus at 92 h

post-fertilization marking, perhaps, the first rearrangement of TCR genes. At this time the

thymus is heavily populated with lymphocytes. The thymus enlarges for at least 1 month

continuing to express the RAG genes and changing shape from spherical to crescent. No

distinction of medulla and cortex is present at this stage. The expression of the gene Ick is

temporally and spatially parallel to the expression of the two RAG genes. Ikaros's

is found for the first time in the Intermediate Cell Mass (an intraembryonic region

between the somites and the yolk sac) at 24 h after fertilization. At day


Ikaros's mRNA

is expressed in the dorsal aorta; and at day 3, it is expressed in the thymus. Pronephros

and kidney, which are respectively the larval and adult sites of hematopoiesis, do not

express RAG1 until day 21 indicating the later lymphoid function development of these

organs (Trede et al. 2001). The DNA polymerase Terminal Deoxynucleotidyl Transferase

(TdT) is one of the enzymes involved in the generation of functional diversity during

immunoglobulin and T-cell rearrangement. Northern blot analysis shows that TdT mRNA

is highly expressed within the thymus and to a lesser extent in the pronephros (Hansen,

1997). This research has also showed that RAG1 and TdT genes are expressed also in the

intestine, where few lymphocytes are present. This result suggests that the intestine of

fish, as in mammals, may serve as an alternative site for lymphocyte development. That

could indicate that the GALT is a location for gamma-delta T-cell homologues.

5.5. B-Cell Ontogenesis

The kidney seems to be the first lymphoid organ where IgM bearing cells can be


determine the age of maturation of B-cells. In common carp imunoglobulins were

detected on the surface of cells in thymus and pronephros from 14 days after fertilization

at 210C, and in mesonephros and spleen, from days 28 (Koumas-van Diepen et al. 1994).

From the same research, it was observed that the age of fish and the percentage of surface

immunoglobulin (sIg) cells in the lymphoid organs were positively correlated, with the

lowest percentage in 2

week old fish and the highest in the oldest fish. The percentage of

sig cells increased gradually from 3 to 16 months of age, with values being respectively

15.8-48.1% for blood, 9.7-21.6% for pronephros, 7.2-16.8% for mesonephros, 15.9-

21.6% for spleen and 1.5-3.7% for thymus. The percentages of cytoplasmic Ig cells were

very low or absent in spleen, thymus and blood. Plasma cells were not found in the

kidney until the fish were 1 month old (0.17%), and increased to 1% of the total cell

population by 8 months. This was supported by a steady increase in serum Igs from 3

weeks onward. In S. salar membrane IgM was detected at 41 days post-hatch, this

coincided with the onset of first feeding (Ellis, 1977). From this stage, the percentage of

positive lymphocytes increased steadily reaching 80% by day 48 post-hatch.

In rainbow trout cytoplasmic IgM was first observed in embryos at 12 days before

hatch with eggs reared at 140C (Castillo et al. 1993; Razquin et al. 1990). At this stage no

cells with sIgM were present. Lymphocytes with sIgM were first observed at 8 days

before hatch. Unfertilized eggs contained detectable amounts of IgM indicating that

transfer of IgM from mother to embryo is possible in salmonids. Cells with sIgM first

annear in the kidney 4-5 days after hatching. By 1 month after hatching, IgM positive


activities in the non-lymphoid cells indicated that a certain degree of maturation of the

cellular stroma of the developing lymphoid organs of trout was reached before or at the

time when IgM-expressing cells could be observed. This could suggest the possibility

that both macrophages and some stromal cells are involved in the maturation of B-cells.

However, reticular cells with alkaline phosphatase are not observed in the spleen red pulp

of trout. This could explain the difference in haemopoietic and lymphoid functions

between kidney and spleen. In splenic parenchyma, reticular cells with acid phosphatase

activity are present before the appearance of the first IgM positive cells. That cell type

appears to form the extravascular microenvironment of the red pulp, which is an

important erythropoietic site (Zapata, 1983). In contrast, the development of alkaline

phosphatase reticular cells in the white pulp of the spleen is not related to the occurrence

of the first IgM positive cells in this organ. The development of IgM positive cells in the

thymus might be related to the development of the connective tissue trabeculae, rather

than with the differentiation of alkaline phosphatase and acid phosphatase-positive cells.

In fact, cells with IgM are not present in the thymus until the development of thymic

trabeculae, and IgM positive cells are concentrated at the border between the thymus

inner and outer zones where trabeculae ramify into several branches.

In mammals, macrophages play an important role in the B-cell maturation

process. It appears that in fish these cells have the same function. Melanomacrophage

centers may be functionally analogous to the germinal centers of higher vertebrates where

the localization of antigen on the surface of dendric cells has been implicated in the


al. 1990). Furthermore, kidney and spleen of trout fry are functional phagocytic organs

prior the acquisition of lymphoid cells (Tatner and Manning, 1983). Development of

splenic ellipsoids, which are considered involved in antigen trapping in fish, appear to be

related in time to the differentiation of the white pulp rather than with the early

differentiation of splenic IgM positive cells (Lamers and De Hass, 1985).

Similar results have been observed in channel catfish (Petrie-Hanson and

Ainsworth, 2001). In renal interstitial tissue, macrophages are present before B-cells

suggesting a possible role in B-cell differentiation. In channel catfish larvae mature

B-cells appear in the anterior kidney before their detection in any other organ, confirming

the role of the kidney in early lymphopoiesis. B-cells are detected in the kidney at 7 days

post-hatch, in the thymus at 10 days, and in the spleen at 14 days after hatching.

Interestingly, in research conducted by Petrie-Hanson and Ainsworth (2001) Igs positive

cells were absent in the thymus from day 14 to 19. On day 21 putative B-cells were

observed again in the thymus. This suggests a wave-like, B-lymphocyte maturation

process in the thymus. At hatch, the spleen of catfish contains only erythrocytes and blast

cells. Monocytes are present by day 3. By day 5 neutrophil and immature B-lymphocytes

are present. Mature B-cells are observed from day 14 after hatching. At this time there is

still not any recognizable organization of the splenic lymphoid compartment. T-cells can

be found by day 28. There is an increase in the T-cell population until

months post-

hatch and at this stage T-cells appear to be more numerous than B-cells. T and

B-lymphocyte populations are localized adjacent to the arterioles and along some of the


The response to thymus-dependent and thymus-independent antigens in fish

involves different immune cells, and these responses develop at different ages. B-cells

respond to thymus-independent antigens, while helper T-cells and accessory cells

respond to thymus-dependent antigens. Antibody production has been observed in 23

days old rainbow trout, however, juveniles respond earlier to thymus-independent than to

thymus-dependent antigens (Paterson and Fryer, 1974). Carp fry, 4- weeks old, were

unable to mount a plaque-forming cell response to sheep red blood cells, that are

thymus-dependent antigens (van Loon et al. 1982). In contrast, 8-week-old carp fry were

able to mount a primary humoral response against human gamma globulin, a

thymusdependent antigen, and against formalin-killed Aeromonas salmonicida, which is

a thymus-independent antigen. Rainbow trout also mount a response to thymus-

independent antigens earlier (2 months old age) than to thymus-dependent ones

(3 months old age) (Tatner, 1986). If juveniles are exposed too early to thymus-

dependent antigen, they can develop a tolerance state. These results are important for

vaccination strategies. Fry can be successfully vaccinated only when they can respond to

antigen, and vaccines based on thymus-independent antigens can be used earlier.

In Elasmobranches, Leydig's

organ, the intestinal spiral valve, spleen and the

epigonal organs are important in B-cell maturation (Rumfelt et al. 2002). In

Elasmobranchs the epigonal and the Leydig organs are homologous to the mammalian

bone marrow. The Leydig organ is attached to the esophagus and consists of a reticulum

packed with large number of leukocytes with few arteries and capillaries present. The


development. Putative lymphocytes of various sizes are abundant in this organ and they

form loose follicle-like aggregates with scattered plasma cells. The epigonal organ is

similar in structure and organization, and it is physically attached to the gonads. It is a

primary lymphoid tissue for the entire life of the animal. At least one of these organs is

present in all cartilaginous fish; for example in nurse shark only the epigonal organ is

present. As in other vertebrates, the spleen is highly differentiated, has a vascular

structure, and is the major site for antigen presentation for antibody production.

In the cartilaginous sharks, skates, and rays, three different types of

imunoglobulins have been found: IgM, IgNAR and IgW (Rumfelt et al. 2002).

IgM exists

in approximately equal amounts as a monomer (7S) and a pentamer (19S). In neonatal

nurse sharks there are very low levels of IgM; at 1 month of age, the level of IgM is

similar to the adult one. However, some IgM may be passed from the mother by the egg

yolk. In neonatal sharks, high levels of a different IgM are present in the blood. Cells of

the spleen and epigonal organs secrete other forms of IgM, and a germline-joined

non-diverse VDJ gene encodes it. This type of IgM is secreted throughout the adult life

only in the epigonal organ. IgM is first expressed in the liver, than in the kidney, spleen,

and Leydig/epigonal organ when the juveniles are close to hatch. IgNAR is expressed

mainly in adults, while IgW has been identified only in the supernatant of in vitro

cultures of splenocytes. It is also interesting to note that in adult sharks MHC class II

mRNA is highly expressed in the spleen, thymus, gill, and intestine (Ohta and Flajnik,

unpublished results). In the spleen, MHC II is expressed on slgM+

cells (B-cells) and on

the surface, (2) large white pulp areas presenting cells without IgM and with few

scattered IgM and IgNAR secretary B-cells, (3) red pulp with IgM putative secretary

cells localized principally close to venous sinuses. The Igs" and MHC II1

second zone are associated with MHC II'

cells in the

cell (dendric cells) and they resemble

lymphocytes in size and appearance. For these reasons they are considered putative T-

cells. In newborn sharks, the spleen is populated mainly by sIgM

cells (B-cells), and

these cells have low expression of MHC II. They are also present in lower numbers than

in adults. T-cell zones are not yet present. In newborn shark the epigonal organ contains

developing granulocytes, blast cells, lymphocytes and plasma cells. The genes RAG 1 and

TdT are expressed in these organs during the entire life of the animal, and the

transmembrane form of IgM mRNA is predominantly found. These data and the fact that

the present sIgM cells are not secretary, suggest that in reality these cells are more likely

developing naive B-cells. The expression of RAG1 gene is 10 times higher in the thymus

than in the epigonal organ, and this gene is not expressed in spleen or muscle. This

confirms that the thymus is the site for T-cell maturation in shark (Rumfelt et al. 2002).

Data collected in several studies (Irwin and Kaatari, 1986; Kaatari and Irwin,

1985; Sanchez et al. 1995) have shown that there is an organ dependent heterogeneity in

B-cell populations. These differences are also related with the specific function of the

immune organs. The anterior kidney of coho salmon possesses lymphocytes with a more

restricted antigen recognition profile of antibody specificities than do lymphocytes from

the posterior kidney or spleen. The heterogeneity profiles of the posterior kidney


by variation in antibody affinity or to the actual absence of select antigenic specificities in

different B-cell populations. Differences in the response to mitogens can be also observed

among B-cell populations of different organs. Anterior kidney cells can respond only to

LPS, while splenic cells can respond to LPS and PPD. These results must be related to

the function of these organs: the anterior kidney is more a hemapoietic center and can be

considered the homologue of mammalian bone marrow.

Differences in organ function can be also observed in the antigen trapping

process. In the spleen antigens are probably held as immune complexes, and are located

extracellularly, associated with the metalophil reticular fibers present mainly in the

ellipsoid walls, scattered throughout the peri-ellipsoid areas, and the melanomacrophage

centers. In the kidney, tissue antigens are present mainly intracellularly within the

reticulo-endothelial system and free macrophages. The thymic and the

reticulo-endothelial system of the heart atrium play no part in the uptake of antigen. The

antigen trapping process could be involved in the induction of B memory cells and in the

negative feedback control of antibody production (Ellis, 1988).

5.6. Nonspecific Immunity

Non-specific immunity involving both cellular and soluble factors plays an

important role in the defense mechanisms of fish both in the adult and during earlier

stages of development. Fish are most likely entirely dependent on non-specific immunity

during the period of free-living existence prior to immune system maturation. Injection of

colloidal carbon particles has been used to study the phagocytic activity in different


present in connective tissues, under the skin, in the gut, and in the gills. Some

macrophages are also observed in the kidney. By day 18 the kidney has become efficient

at trapping carbon. At 8 months the main sites of localization of phagocytosed material

are.the spleen, kidney, heart, and macrophages in the blood. In 2 week old carp, the

phagocytic pattern is similar to that in the adult, and the kidney and spleen trap the largest

amount of carbon.

In addition to the nonspecific lectins and hemagglutinins found in fish eggs and

newly hatched fry, fry possess an efficient phagocytic system (Tatner and Manning,

1983; Tatner and Home, 1984). Rainbow trout fry 4 days post-hatch have an efficient

phagocytic system, and administered carbon is engulfed by free-wandering macrophages

that accumulate under the skin and in the connective tissue, gut, and gills. By day 14

post-hatch the pattern of antigen-trapping was similar to that found in adults, with the

fixed macrophages of the spleen and kidney sequestering the carbon particles from the

blood stream.

Passive immunity from mother to fry has been demonstrated in a few species, and

C-reactive protein and lectin-like agglutins, that should be maternally derived, have been

found in ova of several fish (Tatner, 1996). In rainbow trout the levels of IgM in the

whole fish increased gradually to reach a peak, in terms of milligrams per gram tissue, at

time of hatch, and then slowly declined to initial values by 2 months post hatch (Castillo

et al. 1993). Differences in the feto-maternal and neonate-maternal relationships were

observed among oviparous plaice, Pleuronectes platessa, the viviparous blenny with


into plaice eggs, the young blenny, and swordtail. The earlier lymphoid development in

the blenny compared to the oviparous species could result in the embryo being

immunologically competent while still within the ovary, and capable of mounting an

immune response against both maternal and sibling antigens. The finding that

immunoglobulin can be detected in the eggs and larvae of fish raised the possibility that

hen (mother) fish could be vaccinated to provide their offspring with protection against

various pathogens. This would be particularly beneficial in the case of pathogens that are

transmitted vertically and to which the young are particularly susceptible, such as IPN


5.7. Interactions between the Immune and Neuroendocrine Systems

Neuroimmunoendocrinology is a new interdisciplinary field of research that has

recently emerged, and it involves the study of the bidirectional communication between

the immune and the neuroendocrine systems. These systems are closely and finely

interconnected through innervations and a bidirectional communication that uses the

same set of signal molecules, such as hormones and lymphokines. Most of the studies in

the neuroimmunoendocrinology field have been conducted on mammals, but fishes have

also been target of study, as several recent reviews have shown (Harris and Bird, 2000;

Weyts et al. 1999; Yada and Nakanishi, 2002). There are several lines of evidence that

the sympathetic nervous system (SNS) plays a role in immune regulation: (1) primary

and secondary lymphoid organs are innervated by sympathetic noradrenergic nerves;

(2) synapse-like contacts among sympathetic nerve terminals and T and B-cells in the


Quinn et al. 1997); (4) norepinephrine (NE) is released in lymphoid organs following

immunization (Kohm et al. 2000); (5) NE can alter lymphoid cell functions (Moynihan et

al. 2004); (6) the nervous system acts as an endocrine system when activated neurones

produce an electric stimulation of the terminal part of axon, thus releasing

neuromediators, such as norepinephrine, that will influence the functions of the target

cells. In addition, the histological structure of some endocrine glands resembles the

structure of some parts of the nervous system, as can be observed in the neurohypophysis,

composed mainly of axons originating in the hypothalamus. Also in fish, the sympathetic

nerves have a regulatory function on the immune system (Flory, 1989).

External and internal stimuli are registered by the nervous system and passed to

the hypothalamus. The hypothalamus, through nerve axons or hormones, inhibits or

stimulates the pituitary to release a second group of hormones that will target specific

organs, tissues, and glands regulating their physiology. During these processes several

negative feedback mechanisms are present to finely control the release of hormones and

their effects. For example, under specific stimuli, the nervous system stimulates the

hypothalamus to release the thyrotropin-releasing hormone (TRH) that will stimulate the

hypophysis to secrete thyroid-stimulating hormone (TSH). TSH will stimulate the thyroid

to synthesize and release the thyroid hormones triiodothyronine (T3) and thyroxine (T4).

Both these hormones will target specific tissues and organs to stimulate precise

physiological processes, but they will also produce a negative feedback on both

hvyothalamus and hypophysis inhibiting the release of TRH and TSH. In this way an


As mentioned previously, hormones can influence and be released by the immune

system. It is well known that hormones released under stress conditions can affect the

immune system (Blalock, 1989; Sanders et al. 1997

Schreck et al. 2001), but they are not

the only ones. For example, TRH stimulates both the adenohypophysis and T-cells to

release TSH (Kruder and Blalock, 1986). TSH, produced by the hypophysis or T-cells,

stimulates the thyroid to synthesize and release the thyroid hormones triiodothyronine

(T3) and thyroxine (T4). Chorionic gonadrotopin (CG) can be also produced by T-cells

under gonadotropin releasing hormone (GnRH) stimulation by the hypothalamus.

Somatostatin can block CG production both from the adenohypophysis and T-cells. The

adrenocorticotropic hormone (ACTH) can suppress the immune response to both

T-dependent and T-independent antigens, lymphokines and IFN-gamma production by

T-cells, and Igs synthesis and macrophage activation by IFN-gamma (Blalock, 1989).

The effect of opioids on the immune system is very variable dependently on the type of

opioid peptide and immune cell. For example, alpha-endorphins are potent suppressors of

antibody production, but not beta or gamma-endorphins (Johnson et al. 1982).

Alpha-endorphins have multiple effects on the immune system dependently on the cell

type: they block T helper cells from producing and/or secreting interleukins needed for

antibody production, they inhibit B-cells from producing antibody and transforming

plasma cells. Beta-endorphins enhance T-cell proliferation and differentiation to

cytotoxic T-cells, stimulate cytotoxicity in NK cells, and they are chemotactic factors for

monocytes and neutrophils. Neuropeptide Y (NPY) can have a stimulatory or inhibitory


radicals in peritoneal monocytes, and macrophage phagocytic activity; increases

monocytes mobilization, leukocytes adherence capability, and immune cell redistribution.

Neuropeptide Y can produce a shift in Th2 response by suppressing Thl cells release of

IFN-gamma and stimulating the production of IL-4 by Th2 cells, thus causing an increase

in serum antibodies. Neuropeptide Y can also influence the immunomodulatory effects of

other neurotransmitters, thereby acting as a neuroimmune co-transmitter (Ahmed et al.

2001; Bedoui et al. 2003).

Gastrointestinal hormones can also modulate the immune system. Substance P

increases T-cell proliferation, degranulation of basophils and mast cells, enhances

macrophage phagocytosis, stimulates 02, H202 and tromboxane B2 production (Goetzi et

al. 1985; Payan and Goetzi, 1985). Substance P and somatostatin can also modulate cells

involved in delayed-type hypersensitivity reactions and cell-mediated immunity.

Somatostatin stimulates release of histamine from mast cells, but suppresses histamine

and leukotriene D4 release from basophils. It also suppresses T-cell proliferation. These

two hormones can have similar effects, such as stimulating macrophage phagocytosis, but

they can also have opposite effects: substance P can increase thymidine and leucine

uptake from T-cells in the presence or absence of mitogens, while somatostatin inhibits

this process. The effects of these hormones can be different even on the same cell type

located in different tissues. For example, tissue mast cells and basophils in the circulation

do not release mediators in response to neuropeptides. Instead, nanomolar to

subpicomolar concentrations of somatostatin can suppress Ig-E mediated histamine and

(Blalock, 1989).

Vasoactive intestinal peptide can influence the immune system

stimulating cytotoxic T-cell, but not B-cell proliferation. This effect can be explained by

the presence of VIP receptors on T, but not in B-cells.

The immunoregulatory function of hormones has also been demonstrated in fish.

Sex steroids, especially androgens, can suppress the immune system (Hou et al. 1999;

Yamaguchi et al. 2001). Melatonin also appears to modulate the immune system (Yada

and Nakanishi, 2002).

Prolactin and growth hormone enhance both the cellular and

humoral immune responses, and are important for the proliferation of B-cells (Sakai et al.

1996a, b, c; Yada et al. 1999). The decrease in immune responses of hypophysectomized

fish demonstrated the importance of hormonal modulation of the immune system (Yada

et al. 1999; Yada and Azuma, 2002).

As in mammals, gastrointestinal hormones, such as

substance P, somatostatin, and calcitonin gene-related peptide, may regulate the immune

functions of fish (McGillis and Figueiredo, 1996; Ndoye et al. 1991).

Lymphokines are usually associated with the immune system, but they can also

influence the neuroendocrine system.

When IL-1 enters the circulation, it can act as a

hormonal signal. IL-1 may pass through the fenestrated endothelium in the brain (median

eminence) and stimulate corticotrophin-releasing hormone (CRH) secretion from the

CRH nerve-terminals that will influence the physiological activity of other parts of the

body. In addition, IL-1 may activate the brain endothelial cells to produce IL-1, IL-6, and

prostaglandins, and to secrete other mediators into the brain (Tilders et al. 1994). As for

hormones, lymphokines can also be produced by the neuroendocrine system, for example


axis in response to infection and inflammation (Cohen and Kinney, 2001; Tumbull and

Rivier, 1995; Yada and Nakanishi, 2002).

Hormone and lymphokine receptors have been identified in cells of both immune

and neuroendocrine systems. IL-1 has two distinct receptors, Type I and II. Type I is

present on T-cells, fibroblasts and endothelial cells; Type II on B-cells and macrophages.

IL-1 receptors are present also in the choroid plexus, pituitary gland, and hippocampus.

Even if some hormones and lymphokines are produced and interact with both

immune and neuroendocrine systems, some differences are present. For example, ACTH

is produced by the hypothalamus, but lymphocytes and macrophages can also synthesize

it. The hypothalamus can synthesize two distinct forms of ACTH that differ in their

amino acid sequence (1-39; 1-24). Interestingly,

both forms of ACTH have the same

steroidogenic activity, but only the ACTH 1-39 can regulate the immune system (Johnson

et al. 1982). At least in this context, cells of the immune system, but not cells of the

neuroendocrine systems, can distinguish among closely related or truncated peptides

(Blalock, 1989).

The interaction between the nervous and immune system is also evident from the

cell types common to both systems: astrocytes-macrophages. Even if astrocytes are

considered a type of macrophage, there are differences between these two cell groups.

During inflammation, astrocytes are involved in interacting with invading T-cells.

Activated T-cells are capable of passing the blood-brain barrier and thus infiltrate the

brain parenchyma. In the CNS, T-cells can recognize peptides presented by astrocytes via


than Thi (Aloisi et al. 1998). Astrocytes, even if they belong to the macrophage cell

group, play an important role in nerve regeneration and function. Astrocytes are the

major source of nerve growth factor (NGF). NGF constitutes one of the members of the

neurotrophin family, and it influences growth, differentiation, survival, and maintenance

of peripheral and central neurons. NGF can also act as a protective agent in

neuroinflammatory diseases (Oren et al. 2004). In addition to astrocytes, NGF can also be

released by oligodendrocytes and microglia, and to a certain extent by Thl cells

(Santambrogio et al. 1994). Cells from both the nervous and immune system express the

functional NGF receptor TrkA, suggesting that NGF participates in the interactions

between these two systems. Astrocytes restimulate preferentially Th2 cells because these

cells releasing IL-4 and IL-10 induce NGF secretion by astrocytes; while IL-2 and

IFN-gamma released by Th2 do not stimulate NGF secretion (Oren et al. 2004).

The interaction between immune and neuroendocrine systems is more complex

than expected because there may be a bidirectional relation among immune cells and

behaviors. Mast cells can reside in the brain or they can migrate within the brain after

inflammation. In mammals it has been demonstrated that the migration of mast cells in

the brain can be triggered not only by inflammatory processes, but also by specific

behaviors (Silver et al. 1996). For example in sexually active doves, after a brief period

of courtship, there is a marked increase in mast cells in the medial habenula (brain)

compared to control birds that have a relative absence of mast cells. During courtship

there is an increase of gonadal steroids that could influence mast cell migration. In fact,


elevated also during the 4-5 week period when parents take care of the eggs and young.

There is a reason why mast cells migrate specifically into the medial habenula: this

structure in the brain is a major integrative link between septal nuclei and the midbrain.

Another example where mast cell migration and activity is influenced by behavior

is seen in rats where stress was induced by social isolation (Silver et al. 1996). In stressed

rats the mean number of mast cells in the thalamus was reduced by 90%. Sex hormones

can directly and indirectly influence the synthesis and release of the granular content of

mast cells. However, under these physiological conditions the secretion of granular

content occurs without inflammation or tissue damage. Also, the interactions between

mast cells and elements of the peripheral nervous system have been documented for

peripheral ganglia, connective tissue, and gut mucosa. In these tissues and in the brain,

mast cells could mediate alterations in blood flow, neurotransmission, and local immune

responses such as switching of IgE in B-cells. Mast cells can produce up to 290

mediators, including histamine, interleukins 1-6, granulocyte-macrophage colony

stimulating factor, a variety of proteoglycans, neuropeptides, nitric oxide, and

chemoattractants for eosinophils and monocytes (Silver et al. 1996). A more clear effect

of mast cells on behavioral changes is given in an experiment of induction of an anxiety

status in rat (Ikarashi and Yuzurihara, 2002). In this experiment, several drugs that

stimulate histamine release from mast cells and neurons, or that block histamine receptors

with a histamine release inhibition effect, were used in combination and injected

intracerebroventricularly. After 60 minutes from injection, the time spent in light areas

a dark-light boundary. The results of the experiment show that the release of histamine

due to drug administration causes an increase in the anxious state of rats.

The effect of

the different drugs shows also that histamine can be released by mast cells or neurons.

Table 5-1. First appearance of immune organs of different fish species.
Organ first appearance
Species TempC Thymus Kidney Spleen Reference
Salmo gairdneri 14 5 dprh Grace and Manning, 1980;
Tatner and Manning, 1983;
Tatner, 1986
Cyprinus carpio 22 2 dph 2 dph 5 dph Botham and Manning, 1981
Danio rerio 28 54 hpf 72 hpf Trede et al. 2001;
Willett et al. 1999
Zoarces viviparous 3 months 3 months 2 months Bly, 1985

Abbreviations: dprh= days pre-hatch, dph= days post-hatch, hpf= hours


First appearance of lymphocytes in immune organs of different fish


First appearance of lymphocytes
Species TempC Thymus Kidney Spleen GALT Reference
Salmo salar 4-7 22 dprh 14 dprh 42 dph Ellis, 1977
Salmo gairdneri 14 3 dph 5 dph 6 dph 13 dph Grace and Manning, 1980;
Tatner and Manning, 1983
Tatner, 1986
Ictalurus punctatus 5 dph 7 dph 14 dph Petrie-Hanson and
Ainsworth, 2001
Cyprinus carpio 22 3 dph 6 dph 8 dph Botham and Manning, 1981
Barbus conchonius 23 4 dph 4 dph 7 dph Grace, 1981
Danio rerio 28 65 hpf 3 weeks Trede et al. 2001;
Willett et al. 1999
Zoarces viviparous 2 mprep I mprep After birth Bly, 1985
Abbreviations: GALT= Gut Associated Lymphoid Tissue, dprh= days
pre-hatch, dph= days post-hatch, hpf= hours post-fertilization, mprep= months


Time of maturation of the immune organs.

Development to mature lymphocytic organ
Species TempoC Thymus Kidney Spleen Reference.
Salmo gairdneri 14 5 dph Grace and Manning, 1980;
Tatner and Manning. 1983;
Tatner, 1986
Ictalurus punctatus 10 dph Petrie-Hanson and
Ainsworth, 2001
Cyprinus carpio 22 5 dph Botham and Manning, 1981
Danio rerio 28 1 week Trede et al. 2001;
Willett et al. 1999

Abbreviations: dph= days



Fish Vaccine

The purpose of vaccination is to provide an individual with resistance to a

particular pathogen without having to undergo a potentially risky infection.


confers a protection against disease, not necessarily infection. Thus, vaccines protect

against outbreaks of disease, but not necessarily against persistent infection in an

asymptomatic carrier state.

Vaccination works by inducing a protective immune response

that,, through memory cells persists for a relatively long period of time, though the

precise longevity varies. The term "immunity"

refers to this memory state. This memory

state varies among different pathogens, and also varies among fish species against the

same pathogen. Natural exposure to infection acts as a booster to the immunity produced

by vaccination, provided the memory cells are still present. In the absence of natural

exposure, booster vaccination is necessary to maintain the level of immunity. Specificity

and memory are two of the key elements exploited by vaccination since the adaptive

immune response is stronger after the second encounter with antigen. The vaccine may

stimulate non-specific defense mechanisms that provide a degree of protection over a

short time, but these mechanisms lack a memory component.

Vaccines must be safe, have no side effects, and should induce a high level of


most cases, some degree of antigen purification or enrichment is necessary. In this last

case it is important to identify the antigens that stimulate a protective immune response.

Vaccines can be classified as dead, live, and those that are genetically engineered.

Dead vaccines are composed of inactivated pathogens or extracts. Live vaccines are

composed of attenuated pathogens with low or no virulence. Live vaccines are developed

by serial passage of a pathogen in culture, or by using naturally occurring mutants and

cross-reacting strains, or using recombinant DNA technology deleting specific genes

coding for virulence factors. Compared to dead vaccines, live vaccines produce a higher

stimulation especially of the cellular branch of the immune system, and for this reason in

general, they produce a greater immune response (Marseden et al. 1996). However, a

problem associated with live vaccines is the possibility that the pathogen can revert to a

virulent state or that it can still possesses a residual virulence. In these cases, infection

and mortality will be the result of fish exposed to the pathogen. Genetically-engineered

vaccines are generally composed of a purified highly immunogenic antigen that has been

expressed in a microorganism through insertion of DNA coding for that specific antigen.

Generally, these vaccines provide superior protection because they are able to stimulate a

more complete immune response resulting in a broader and more long-lasting immunity

(Winton, 1998). Another significant advantage of genetically-engineered vaccines is the

ability to construct multivalent preparations that can protect fish against several

pathogens or different strains of the same pathogen.

All microorganisms and their products are composed of many different antigens,


those associated with virulence, could be caused by an adaptation of the pathogen to

avoid destruction by the host's immune response. Antigens may induce antibody

production, but, unless these antibodies can neutralize the components of pathogenicity,

they are valueless in giving protection. For this reason, assays of blood antibody levels

following vaccination may give no indication of the degree of protection unless the

antibodies are directed against pathogenic antigens. For example, when salmonids have

been vaccinated with specific Aeromonas salmonicida bacterins, antibodies against the

pathogen can be detected in the blood, but the fish are not protected (Ellis, 1985).

The form in which the antigen is administered may be very important. Generally,

soluble antigens are administered in adjuvant to elicit antibody production. Oil-based

vaccines have proven to give high levels of protection (Midtlyng et al. 1996a; Nordmo

and Ramstad, 1997). In oil-based vaccines, it is important that the dispersed phase is

homogenous to guarantee a reproducibility of injection. In the case of large oil droplets

the dispersion tends to phase separate, with the possible sedimentation of droplets, and

formation of a high-density lower phase or flotation of oil droplets.

This will result in

non-uniform dosing during the vaccination period. Phase separation can be avoided or

minimized by producing vaccines with a small droplet size. Small droplet size is also

preferable in vivo because fine dispersion will lead to an even distribution of the vaccine

intracoelomically (Muller and Harnisch, 1998). There is also an optimum size for

macrophage uptake and lymphatic absorption, the parameters determining vaccine

efficacy and antigenic response. In in vitro studies, highest particle uptake from

because the particles approach the size of the macrophage itself. Studies on lymphatic

absorption of particulates demonstrate that optimum size is below 1 pm (Davis et al.


Ideally, the vaccine dispersion should be stable after shaking the vaccine solution,

at least for one day. In vaccine production, it is also important to reduce the viscosity of

the formula. A low viscosity spreads more easily in the coelom and requires less injection

force, thereby reducing the chances of possible lesions in the coelom. However, in some

case, a lower viscosity may cause a too fast release of antigen, and a wider diffusion of

the vaccine causing a too generalized inflammation.

The route of antigen administration is also important. Generally, intramuscular

injection of antigen elicits higher antibody production than bath exposure (Fukuda and

Kusuda, 1981; Midtlyng et al. 1996b). Another consideration is that the immunogenicity

of antigens may vary among fish species.

In bacteria, serotype differences are frequently associated with variation in the

structure of the lipopolysaccharides, LPS. If antigens which represent different

pathogenic groups or strains are involved in eliciting a protective immune response, then

the different serotypes must be incorporated into the vaccine to protect against the

different serotypes of the pathogen. Such vaccines are called multivalent.

6.2. Adjuvants

Adjuvants are substances that enhance the immune response. They comprise a

variety of compounds whose mechanisms of action are variable and in many cases not

well understood. Administering antigen with adjuvant can considerably enhance the

adjuvant, and it may be significant that in both cases the protective antigens are

lipopolysaccharide, considered to be a T-independent antigen. Protection against other

diseases has been achieved only when antigen was given in adjuvant (Aeromonas

vaccine). In these cases the adjuvant may either improve the immunonogenicity of the

antigens or have a non-specific stimulatory effect on the non-specific defense

mechanisms, or both (Ellis, 1988).

One of the problems in using adjuvants, is that they can cause side effects or

severe local tissue responses (Lillehaug; Midtlyng, 1996; Midtlyng et al. 1996b).

The most effective adjuvant and one of the most widely used by experimental

immunologists is Freund'

Complete Adjuvant, FCA, which is a mixture of killed

Mycobacterium tuberculosis in a mineral oil in which the antigen in saline solution is

emulsified. The emulsion can only be administered by injection. FCA can cause

undesirable side effects, such as local granuloma formation, autoimmune diseases, and

tuberculin sensitization. By intramuscular injection, it induces sterile abscesses with

extensive local muscle necrosis. Intracoelomic injection is less traumatic, but visceral

adhesions and granuloma formation may result. Slight reduction in fish growth rate can

also occur.

Freund's Incomplete Adjuvant, FIA, lacks the Mycobacterium. It is used in the

same way as FCA and has only slightly less traumatic side effects, but it is not as

effective an immunopotentiator as FCA.

Other injectable adjuvants are aluminum hydroxide gel mixed with saponin,


6.3. Injection Vaccination

The most effective way to immunize fish is by intracoelomic injection. This

method permits the use of adjuvant that enhances the magnitude of the immune response.

This method has some disadvantages, principally by requiring anesthetization and

handling, both of which can be quite stressful for the fish. It is also very labor intensive

and can be difficult to apply to small fish. Nevertheless, the overall smaller amount of

vaccine required for this method, makes it more economical for fish larger than 40 g

(Lillehaug, 1989).

Intramuscular injection on the dorsal surface of the fish is rarely used mainly

because of the possibility of creating unsightly scarring, and because of potential leakage

of the inoculums before absorption. However, salmon broodstock are usually injected

intramuscularly in the "dorsal sinus", a potential space between musculature in front of

the dorsal fin (Mitchell, personal communication).

The swim bladder has also been used as a site for vaccine administration (Endo et

al. 1995). In some species of fish, the swim bladder is surrounded by well-vascularized

thin collagenous membranes that facilitate the absorption of the vaccine in to the blood

stream without causing as much damage to the body as intracoelomic or intramuscular

injections. Moreover, the swim bladder has a large spaced lumen and the surrounding

membrane can prevent the backflow and spread of the injected agents. Absorption of

injected agents can take up to 7 days. This slow ratio of absorption can be useful in some

treatments against diseases, but it can also cause pathological damages to the organ wall.


6.4. Immersion Vaccination

The first methods developed for immersion vaccination required a hyperosmotic

immersion step. The concept behind this technique was that a hyperosmotic media would

have induced a greater absorption of antigen through the gills and skin of the fish. It was

quickly realized that the hyperosmotic step was unnecessary and often very stressful. For

this reason, direct immersion methods were developed. There are two main immersion

methods, the dip and the bath. The substantial differences among the methods are the

time of exposure to the antigen and the concentration of antigen in the water, the higher

the concentration, the shorter the time. In dip vaccination fish are kept in the water with a

vaccine dilution of 1:3 or 1:10 or 1:100 for 5-30 seconds. In the bath method fish are kept

for an hour in the water with a vaccine dilution of 1:500-1:5000. A further variation of

these methods is the spray method in which fish are sprayed with the vaccine after being

removed from the water (Fukuda and Kusuda, 1981).

The uptake of some bacterial antigens, such as Aeromonas sp., by fish following

immersion has been demonstrated to be quite effective.

The mechanism of uptake is not

precisely understood, but the gills appear to be the main route of entry, with the skin and

lateral line system possibly playing a role. The failure to immunize fish against many

diseases by the immersion method is probably not mainly associated with inadequate

uptake of the antigens, but with the nature of the ensuing immune response. There is also

the probability that mucosal immunity, and not systemic immunity, is produced. This

hypothesis seems to be proven by the fact that while in injection vaccination there are


immersion vaccinations stimulate integumentary immunity, which is quite sufficient to

protect against natural exposure, but not to challenge by injection.

A disadvantage of vaccination by immersion compared to injection is that much

more vaccine is required. In addition, immersion vaccinations done at high concentration

can be toxic.

6.5. Oral Vaccination

Oral vaccination is a preferred method since it is suitable for mass administration

to fish of all sizes, imposes no handling stress on the fish, and does not interfere with

routine husbandry practices. Oral vaccination is the only method suitable for extensive

pond rearing of fish where catching the fish prior to harvesting for injection or immersion

vaccination is impractical. However, oral vaccination has some intrinsic limitations. In

the case of killed vaccines large amounts are required, thus increasing the cost, and there

is the uncertainly of individual dosage concentration since each fish consumes a different

amount of food. Unfortunately, the most important limitation of oral vaccines is their

poor potency, and many trials with many different vaccines have resulted in low or

inconsistent levels of protection (Ellis, 1988). A major problem seems to be the

destruction of antigens in the stomach and foregut before they reach the immune sensitive

areas of the lower gut (Rombout et al. 1985). To overcome this problem, enteric-coated

spheres can be used. Microparticles are used as carrier systems and they are capable of

partially protecting and entrapping antigens against enzymatic degradation. Experiments

with rats (O'Hagan et al. 1989) and with coho salmon (Piganelli et al. 1994) have shown


Another reason for a low protection state following oral immunization is that oral

vaccine may induce a mucosal, but not a systemic response. A semi-autonomous mucosal

immune system exists in fish. Stimulation of any mucosa can lead to antibody production

in.all mucosal compartments (gills, skin, guts),

without necessarily stimulating a systemic

response and inducing a memory state. Induction of mucosal immune responses via the

oral route often requires multiple oral administrations (Joosten et al. 1997

Lin et al.

2000; O'Hagan, 1992).

6.6. Biomolecular Vaccines

The application of biomolecular techniques to vaccine design can help produce

vaccines that confer higher protection and are safer than vaccines produced by

conventional techniques. In fish, research into recombinant DNA technology has focused

mainly on viral vaccines. Currently, there are only a few vaccines available, including

one vaccine against Infectious Pancreatic Necrosis Virus (IPNV) (Leong, 1993; Leong

and Fryer, 1993

Leong et al. 1997

Lorenzen, 1999; Midtlyng, 1997). Expression-based

vaccines can be divided into several different groups: 1) those agents that replicate within

the host, such as attenuated strains; 2) recombinant viral and bacterial vectors; and

3) nonreplicating agents that serve only as vehicles to deliver nucleic acid sequences,

such as retroviral sequences and genetic immunization.

The first step in the production of a recombinant DNA vaccine is to identify the

proteins/antigens of the pathogen that stimulate a protective immune response in the host.

More than one antigen may be protective, and some antigens may stimulate some parts of


cultured so the protective antigens can be produced on a large scale. Because no

infectious agent is present, these types of vaccines have a high level of safety, and low

production costs after initial research and development (Winton, 1998).

Bacterial expression systems based upon E. coli may lack the ability to

glycosylate and correctly fold some types of proteins, resulting in less effective

protection. In these cases another expression system that is able to produce a protein in its

correct native conformation, such as baculovirus vector growth in insect cells, must be

chosen. However, these vectors can make the vaccines too expensive. These alternative

expression systems were used to produce a vaccine for rainbow trout against the viral

haemorrhagic septicemia virus (VHSV) (Lecocq-Xhonneux et al. 1994). The cloned gene

for one virus glycoprotein was inserted into a baculovirus expression system and the

recombinant protein was produced in insect cells. The recombinant protein had a slighter

lower molecular weight than the native viral protein and displayed a different pattern of

glycosylation. The recombinant protein was transported to the plasma membrane and

induced cell fusion in a manner similar to the authentic protein. The vaccine based on this

recombinant protein conferred a protection to rainbow trout similar to that conferred by

killed and attenuated vaccines, and neutralizing antibodies recognized it.

With recombinant DNA technology, besides inserting sections of DNA into the

genome of a cell, it is also possible to delete sections. Specific deletion of a gene that

codes for a virulence factor in a pathogen, while preserving other protective antigens,

allows the production of attenuated forms where reversion to virulent wild type is


In addition to traditional genetic immunization and recombinant DNA technology,

there are several other biomolecular techniques that have been applied to vaccine


Genetic immunization using naked DNA is the most recent approach in vaccine

design. This technology is based on the observation that skeletal muscle cells injected

with purified plasmid DNA express plasmid-encoded proteins. In mammals the specific

immune response after DNA vaccination is caused by antibodies, T-helper cells and

cytotoxic cells.

Protein engineering. The aim of this technique is to improve immunogenicity of

an antigen or to mass produce it cheaply. Protective antigens may be made more

immunogenic by cleavage of epitopes that stimulate the

T-cell suppressor. The protective

epitopes of an antigen may be composed of a short sequence of amino acids that can be

artificially synthesized quite cheaply for use as vaccines in combination with appropriate

carrier antigens.

Anti-idiotype vaccines. These are antibodies, the antigen-binding site of which

mimics the structure of a protective antigen. They are produced by first raising antibodies

to the appropriate antigen in a suitable animal. The specific antibodies produced

(idiotypes) are then used to raise further antibodies that react with the antigen-binding

site of the first antibody (anti-idiotypic antibodies). The antigen-binding site of the anti-

idiotype can be used as a vaccine since it resembles stoichiometrically the original

antigen and will induce formation of antibodies which cross reacts with it. Such vaccines

Genomics and Proteomics Applied to Vaccine Design

New approaches to vaccine development are important because there are several

infectious diseases for humans and animals for which the traditional approaches have

failed to produce effective vaccines. Moreover, especially for humans, regulatory

authorities often require that vaccines meet high standards of safety. For these reasons,

there is increased interest in the application of the new technologies known as genomics,

proteomics, and transcriptomics to human and animal vaccine design.

Genomics, proteomics and transcriptomics have become important tools for

increased understanding of the biology of pathogens (viral, bacterial and parasitic) on a

large scale, giving information on which genes and proteins are expressed and under

which conditions. This information can help assign functions to unknown genes and

proteins, decipher the mechanism of pathogenicity of organisms, reveal new possible

targets for drugs, and assist in the design of new vaccine candidates for humans and

animals (Barret et al. 2000; Betts, 2002; Cordwell et al. 2001; Grandi, 2001; Klade, 2002;

Moxon and Rappuoli, 2002; Nilsson, 2002; Potter and Babiuk, 2001; Tarleton and

Kissinger, 2001; Teixeira et al. 2002; Wallace and Wallace, 2002). For example, antigens

present only in virulent bacteria strains may make good vaccine candidates.

DNA microarrays are one of the most useful tools for the study of the

transcriptome, the complete set of transcripts produced by an organism. In this technique

DNA elements are arranged as spots in a grid pattern on a solid support. These arrayed

targets are hybridized with complex probes prepared from an RNA population for a


Gene expression profiling is performed by two-color hybridization. DNA labeled with

different fluorochromes is produced from RNA isolated from two experimental

conditions. The two populations are combined in equal quantities and hybridized to the

array. Scanning the array gives the quantitative fluorescence intensities for each fluor in

each spot, from which ratios are calculated to give relative expression levels for each

gene. This analysis can reveal even shifts of hundreds of genes expressed. Such

techniques have been used to study the whole genome of Streptococcus pneumoniae to

identify suitable antigens for vaccine production (Wizemann et al. 2001).

Through exploring the effects of drug action using DNA microarrays, it is

possible to identify genes within pathways that may have not been previously realized,

and may provide alternative drug targets. Moreover, the generation of signature profiles

of drug action may allow prediction of the mode of action of novel compounds.

Improvements in the sensitivity of this technique may be achieved through amplifying

mRNA, thus facilitating analysis of small amounts of RNA recovered from tissues or


Another technique for studying genome expression is In Vivo Expression

Technology (IVET). Plasmid libraries are created with random fragments of bacterial

chromosomes fused to promoterless genes that encode an antibiotic resistance gene. A

single crossover between the plasmid and the corresponding homologous region on the

chromosome produces mutant bacteria. The bacteria are then injected in to antibiotic

treated animals. At this point only the mutant bacteria that have the antibiotic resistance


Salmonella typhimurium, Pseudomonas aeruginosa, Yersinia enterocolitica, Vibrio

cholerae and Staphylococcus aureus (Grandi, 2001).

In Signature Tagged Mutagenesis (STM), bacterial mutants are created by random

insertion of a set of transposomes differing by a short synthetic nucleotide sequence

(TAG), so that each mutant can then be recognized by its TAG. The bacteria mutants are

then injected in to animals and the bacterial clones with a mutation in a gene needed for

pathogenesis will not survive. The missing mutant clones can be detected by recovering

bacteria and doing a TAG-specific DNA hybridization, in this way all the genes required

for infection can be identified. For example, in a study on Group B Streptococcus,

out of 1600 transposome mutants were defective for survival in the animal host, and only

50% of these mutants have insertions in known genes (Jones 2000).

Proteomics, in addition to describing the protein complement to the genome, can

provide important information on proteins expressed at defined points in time or under

defined conditions, their location within the cell, and the existence of post-translational

modifications. Protein analysis can be more useful than mRNA analysis for several

reasons. The final products of genes are proteins; mRNAs can be present in a cell but not

translated; and from a single mRNA molecule, several proteins can be produced, as in the

case of some hormones such as calcitonin and calcitonin-related gene. Both hormones are

translated by the same mRNA, but by different splicing occurring in the brain or in the

thyroid. The simple analysis of mRNA would not detect all these different proteins

produced by the same mRNA. Stability, half-life, post-transcriptional, co-translation and


environmental stimuli may not completely represent the expression of a corresponding

protein in vivo.

Protein analysis, however, can present some difficulties that are not encountered

in genome analysis, such as target abundance and molecular recognition. Another

difference is that target DNA or RNA can be amplified by PCR. There are no similar

techniques that easily allow amplification of target proteins. In genome studies,

sequences can be recognized by oligonucleotide probes, but for protein analysis,

recognition is mediated by antibodies, other proteins or ligands, and each of them is very

specific for only one or a few proteins.

In proteomic analysis, several techniques are utilized, with the most common

being two-dimensional gel electrophoresis (2DE). Two-dimensional gel electrophoresis is

a technique that can be applied to a complex protein mixture, such as that extracted from

whole cells and tissues. The proteins migrating in a pH gradient gel are first separated

based on their charge (isoelectric point).

Then, the proteins migrate according to their

molecular mass, in a direction perpendicular to the first migration axis. Size fractionation

is achieved by first equilibrating the isoelectric focusing gel in a 2% solution of the

detergent sodium dodecylsulphate (SDS) which binds non-specifically to all proteins and

confers a uniform negative charge. After fractionation, the protein gel is stained using, for

example, silver staining or Coomassie brilliant blue or non-covalent fluorescent dyes.

The outcome is a unique pattern of dots, each dot representing a protein. This pattern can

be used as a protein fingerprint of the cell. Typically more than 1,000 distinct proteins

separation of up to 10,000 protein spots.

Western blotting of the gel and detection of

antigens with patient sera can help distinguish immunorelevant proteins.

This technique can be utilized to easily compare protein expression of a particular

celH in different situations by just comparing the 2DE gels and looking for a spot absent

or present in one of the gels. However, a problem for this type of analysis is that the

conditions in which the gels are prepared and run should be the same in each experiment,

and this can be difficult to accomplish.

To enhance separation of proteins, instead of 2DE, reverse phase microcapillary

liquid chromatography can be utilized. Multidimensional chromatography can also be

applied. This technique consists of serial chromatographic separation according to

different properties.

Identification of proteins on the 2DE gel can be achieved. First, proteins inside

the gel are digested with trypsin. The resultant peptide mixtures can be analyzed with

mass spectrometry, measuring the mass of each peptide obtained from the digestion of

the same protein. Each protein produces a characteristic peptide mixture that can be used

as a fingerprint. The measure of the mass of the peptide can be correlated to the amino

acid composition of each peptide, and then the amino acid sequence of the protein can be

determined. This fingerprint can be compared with theoretical digests of proteins in

databases through a bioinformatics tool in order to identify the protein. Otherwise,

peptides can be separated by size-exclusion chromatography (SEC) and reversed-phase

(RP) chromatography, then directly analyzed by mass spectrometry (MS).


resolving power of 2DE in conjunction with narrow-range immobilized pH gradients and

subcellular prefractionation has been estimated to reveal up to 75% of protein genome,

while the remaining 20-30% cannot be resolved due to technical limitations associated

with this technology. Moderate or low abundant proteins could not be observed in the

2DE, especially if a nearby spot representing a high abundant protein obscures their

signal. Scarce proteins may be completely undetected. Fractionating the total protein

sample and running a gel specific for particular types of proteins, such as membrane

proteins or proteins with a particular isoelectric pH, can overcome this problem.

Prefractionating complex mixtures into specific cellular compartments, such as

membrane or extracellular or intracellular proteins, is important because for pathogenic

organisms, proteins of interest are most likely confined to the membrane or extracellular

fractions. However, due to their intrinsically hydrophobic nature, membrane proteins can

be difficult to solubilise in gel. Recently, several new solubilising agents have been

developed and have allowed the visualization of significantly more hydrophobic

membrane-associated proteins (Cordwell et al. 2001). This type of analysis was

performed on Streptococcus agalactiae to identify the major outer surface proteins that

can be used for vaccine production (Hughes et al. 2002).

Bacterial proteins secreted during growth may be important for several pathogenic

processes, including toxicity or cell-signaling, but the proteomic analysis of such proteins

can be difficult due to the presence of contaminating proteins from undefined sources or

from supplement animal sera. In some bacteria, the analysis of such proteins could be


important for vaccine production were identified analyzing the secreted proteins of group

A Streptococcus (Lei et al. 2000).

Another technology used to study the proteome of a target cell or tissue is protein

arrays. Protein arrays are in theory very similar to DNA arrays being miniature devices

that have proteins or molecules that recognize other proteins. However, protein

microarrays have more limitations than DNA microarrays. First there is no procedure that

amplifies the sample proteins as there is for DNA. Second, while all DNA sequences are

constituted by the same four nucleotides that are similar in their chemical properties,

proteins are comprised of

amino acids with different chemical properties. The

principles of molecular recognition are very different among amino acids and also vary

among different proteins. For this reason the same reaction conditions can never be

applied for all proteins. Third, in a DNA array, labels are incorporated evenly and do not

interfere with hybridization. Fourth, solid substrate, such as nylon or glass, does not

interfere with hybridization. In contrast, for proteins, labeling is much more variable and

both labeling and attachment to a substrate may interfere with protein binding by

affecting protein folding or by blocking the binding site. In these arrays, antibodies are

attached to the array or chip surface.

Antibody arrays can detect specific proteins. Because antibodies are highly

discriminatory, they are suitable for detailed analysis of protein profiles and expression

levels. There are three different types of antibody arrays. The first is a standard

immunoassay in which the antibodies are immobilized and are used to capture labeled


allows exposure of two different protein samples at the same time, each labeled with a

different fluorophore. The second technique consists of a miniature sandwich assay in

which unlabelled proteins are captured from the solution by antibodies on the array after

which the proteins are detected by a secondary antibody. This technique requires that the

two antibodies recognize different parts of the proteins. The advantage of this technique

is that it is not necessary to label the proteins, a process that is variable and sometimes

inefficient. The third technique involves a tertiary detection system, and therefore it is

more sensitive. The principle of this technique is that the immobilized antibody captures

a protein. After, a secondary antibody linked to a circular oligonucleotide will recognize

the protein. In the presence of a circular DNA template and dNTPs, DNA polymerase

will amplify the circular DNA template producing a long DNA concameter containing

hundreds of copies of the template. Then a fluorescent-labeled oligonucleotide probe can

be added to detect the DNA template.

More recent techniques in proteomics use a microfluid device instead of a gel.

These devices should have a higher sensitivity because of the faster kinetics of binding in

solution than in a solid phase, and they should have a higher throughput because the

arrays are constructed in three dimensions.

Another technique to identify a new possible targets for vaccine production is in

silico analysis. This technique employs algorithms involving known gene sequences to

predict genes encoding secreted, surface-associated antigens and virulence factors. This

technique analyzes the genomic sequence of pathogens to detect proteins with


virulence factors can be detected by searching for unknown genes that are co-regulated

with known virulence genes. The selected genes are expressed in E. coli and the

expressed proteins are purified and tested for their effect on the immune system. This

technique is very fast because it uses robotic stations. Five scientists can produce and

purify 100 proteins each month. However, a limitation of this technique is that it can not

be used to detect new families of virulence factors. In a study on Chlamydia pneumoniae,

147 membrane-associated proteins were identified, 91% of them were successfully

expressed in E. coli, and 58 proteins induced Igs that bind C. pneumoniae (Grandi, 2001).

Collectively, these technologies will allow the identification of most of the

antigens targeted by the immune system under certain pathological situations. Once the

target antigens are identified, their genes must be cloned and expressed, then the

synthesized proteins must be purified and tested in vitro and in vivo for identification of

candidates for vaccine production. Moreover, the comparison of multiple immunomes

may allow the discovery of immunogenic structural features shared and conserved among

different pathogens, which could form the basis of broadly protective vaccines.

6.8. Herd Immunity

Concomitant with research on vaccine development, there is a need for an

improved understanding of how best to use vaccines to protect the community as well as

the individual. Epidemiological and economic issues are at least as important as

technological ones (Anderson and May, 1985; Waltner-Toews, 1989).

Herd immunity is the ability of groups of animals to resist becoming infected or to


The persistence of an infectious disease within a population requires that the

density of susceptible individuals exceed a critical value such that, on average, each

primary case of infection generates at least one secondary case. It is therefore not

necessary to vaccinate everyone within a community to eliminate infection, but it is

sufficient to vaccinate a percentage of the population so as to reduce the susceptible

fraction below the critical point. The central problem for the epidemiologist then is to

calculate the proportion of the population that should be vaccinated. This calculation is

not easy because many factors must be considered, such as desired level of vaccine

protection, primary infection rate, transmission rate, incubation time, duration of acquired

immunity, age-related changes in infection susceptibility, best age for immunization,

genetic factors, population density, and behavioral factors. For example a more dense

population requires higher rates of vaccination (Anderson and May, 1985; Waltner-

Toews, 1989).

When the concept of herd immunity is kept in mind, it becomes clear why the use

of vaccines that confer low or moderate protection is still beneficial. In fact, these

vaccines may slow down or stop the rate of disease transmission. A reduction in the rate

of transmission may be helpful especially when the production periods are not too long,

as in aquaculture where the production period is generally of one or two years.


Immunostimulants and Experiment Design

The term immunostimulants is applied to several compounds of diverse chemical

structure that have been shown to stimulate the immune system (Anderson, 1992; Raa,

1996; Sakai,


The exact mechanisms of the immune system stimulation are still not

well understood, and there are still doubts on the real efficacy of some of these

compounds. For many immunostimulants, such as vitamin C,

different investigations

have shown diverse and even conflicting efficacy results. However, comparing the results

of different studies is not an easy task due to differences in experimental protocols

including variations in doses, methods, and length of administration.

During experimental analyses, the body's

response to immunostimulant

administration over time is often not considered. This class of compounds can stimulate

the immune system, but prolonged dosing will produce a homeostatic response with

consequent decreases in stimulation. In addition, different immune parameters can be

stimulated at different doses and times. One example of these issues is evident in research

examining vitamin C administered to gilthead seabream (Spaurus aurata) (Ortuno et al.

1999). In this study, fish were fed diets containing 0.5

and 3 g/kg ascorbic acid for

8 and 10 weeks. Some of the non-specific immune parameters were enhanced by dietary

4, 6,


not affected. After 8-10 weeks, the serum complementary activity was almost the same in

all the trials. Even when high levels of ascorbic acid were maintained in the serum, the

immune responses did not improve, but returned to their normal levels. These data

suggests that after a transient enhancement of the non-specific immune response as a

result of vitamin C supplementation, an adjustment of the immune response to the new

conditions of higher ascorbic acid availability occur. If this hypothesis is correct, it might

explain the absence of effect described in some previous studies with this vitamin when

long administration periods were tested.

Body storage capability for different immunostimulants is another consideration

when experimental results are analyzed. In studies where different doses of an

immunostimulant that can be stored in the body, are tested, lack or a non-significant

difference in results among diets may be due primarily to body storage of the compound.

For this reason before the experiment it is important to feed diets without


Additional factors include storage capabilities and retention period. These may

differ, for a particular compound, among organs. For example, in rainbow trout after the

fish were switched for two weeks to the control diets, plasma and liver ascorbic acid

levels were strongly reduced meanwhile the level was still high in the head kidney. The

longer retention of vitamin C in the head kidney is related to its lymphoid function:

leukocytes are able to store elevated amounts of ascorbic acid in their cytosol and,

therefore, require a longer period of time to become depleted (Verlhac et al. 1996).


development, reproduction, wound healing, metal ion metabolism, response to stressors,

immune response, and perhaps lipid metabolism through the action of carnitine synthesis.

Most teleost fish do not synthesize vitamin C because they lack the enzyme

L-gulonolactone oxidase. This vitamin is synthesized in other phylogenetic groups of

fish, such as sea lamprey, Petromyzon marinus, Atlantic dogfish, Squalus acanthias,

sterlet sturgeon, Acipenser ruthenus, white sturgeon, Acipenser transmontanus,

paddlefish, Polyodon spathula, and in some teleost such as Atlantic halibut, Hippoglossus

hippoglossus, and Atlantic salmon, Salmo salar, (Dabrowski, 1994; Dabrowski and

Clereszko 2001; Maeland et al. 1999; Maeland and Waagboe, 2000; Moreau, 1999).

Signs of vitamin C deficiency in fish include decreased growth, scoliosis,

lordosis, internal and fin hemorrhages, distorted gill filaments, fin erosions, anorexia, and

increased mortality (Halver et al. 1969; MacConnell, and Barrows, 1993).

Organ and tissue concentrations of ascorbic acid are correlated to dietary intake

levels (Hardie et al. 1991; Verlhac et al. 1996). Brain, thymus, head kidney, and

leukocytes concentrate more vitamin C and maintain its level for longer periods than

other organs, such as liver, in cases of dietary deficiencies. Liver and head kidney are the

primary storage organs for vitamin C in fish. The high level of vitamin C found in

thymus, brain, head kidney, and leukocytes may be related to the requirement for

antioxidant substances to maintain cellular membrane integrity to preserve vital organs

from oxidative processes, and to allow the optimal functioning of immune cells. The high

level of this compound in the head kidney could also be related to the lymphopoietic


Trunk kidney and spleen are also able to store large amounts of vitamin C.

Vitamin C is involved in the synthesis of cortisol and catecholamine hormones

involved in the stress response. In a stressed state, plasma cortisol levels increase and the

immune system efficiency decreases. Cortisol causes changes in numbers and proportions

of circulating leukocytes, generally resulting in lymphocytopenia and neutrophilia.

Cortisol also has an effect on the non-specific immune system causing inhibition of

respiratory burst activity, phagocytic activity, and leukocyte migration. Cortisol also

decreases the mitogen response and antibody production (Barton and Iwama 1991).

Vitamin C can improve the immune system by acting as a brake on steroidogenesis

through peroxidation of unsatured fatty lipids and so preventing their conversion into

cholesters that are important components of the cortisol. For this reason, during stressful

situations, the requirement of vitamin C increases.

Vitamin C plays a significant role in the immune response and resistance to

infectious diseases primarily by stimulating components of the non-specific immune

system, including the complement pathway, serum lysozyme, macrophages, and

neutrophils phagocytic activity (Hardie et al. 1991; Li and Lovell, 1985; Ortuno et al.

1999; Roberts et al. 1995; Verlhac et al. 1996). During phagocytosis, immune system

cells produce reactive oxygen radicals. These compounds have potent microbicidal

activity, but they are also auto toxic to fish macrophages (Secombes et al. 1988).

Administration of ascorbic acid has been shown to increase survival of fish challenged

with pathogens. Increased survival was observed in salmon challenged with Aeromonas


Labeo rohita, challenged with E. tarda (Sahoo and Mukherjee, 2002); in rainbow trout

challenged with the ciliate parasite Ichthyophthirius multifilis (Wahli et al. 1986).

However, in several other studies, supplementation of vitamin C did not improve

the immune system or the resistance of fish to bacteria or parasites (Bell et al. 1984; Lim

et al. 2000; Lygren et al. 1999; Sealey and Gatlin, 2002). In channel catfish,

administration of vitamin C increased macrophage migration, but did not increase

survival against E. ictaluri (Lim et al. 2000).

Ascorbic acid is also involved in mineral metabolism, influencing absorption and

decreasing metal toxicity.

Vitamin C transforms metals, such as cadmium, nickel, lead,

into their reduced forms, which are less efficiently absorbed and more rapidly excreted.

7.3. Vitamin E

Vitamin E, acting as an antioxidant, protects biological membranes and assists in

optimal functioning of the immune system. Phospholipids of mitochondria, endoplasmic

reticulum, and plasma membranes possess special affinities for alpha-tocopherol. Being a

lipid soluble antioxidant, vitamin E protects the highly unsaturated fatty acids in cells

against oxidative degeneration by terminating free-radical-initiated chain reactions and

decreasing the production of lipid peroxides and reactive oxygen species. Because these

chemicals are autotoxic and can destroy neutrophils and macrophages, they can adversely

affect the immune system (Lall and Olivier, 1991).

Alpha-tocopherol also stimulates the

phagocytic response, and total superoxide dismutase activity in liver and muscle (Lygren

et al. 1999).

Vitamin E significantly increases immunity and decreases mortality in Indian


Vitamin E deficiency results in decreased T and B lymphocyte function and

phagocytic index in rainbow trout challenged with Yersinia ruckeri (Blazer et al. 1984),

coelomic macrophage function in rainbow trout (Blazer, 1982), as well as serum

complement function and serum capability to opsonize bacteria in Atlantic salmon

(Hardie et al. 1990; Montero et al. 1999).

In some studies no positive response on the immune system was observed with

administration of vitamin E (Li et al. 1993; 1998) nor was there resistance to bacterial

challenge (Lall et al. 1988; Sealey and Gatlin, 2002).

Vitamin A

Vitamin A and other carotenoids can influence the innate and specific immune

responses. These compounds have high affinity for toxic oxygen radicals, which result

mainly from lipid peroxidation, but which are also generated during the respiratory burst

of white blood cells in the initial response to an infection (Benedich, 1989).

For this

reason the carotenoids can enhance the ability of macrophages, killer T-cells and

cytotoxic T-cells to kill tumor cells and bacteria. This capability is demonstrated in

homeothermic animals where p-carotene and canthaxanthin can enhance B and T-cell

proliferation in vivo (Benedich and Shapiro, 1986). However the effect of these

compounds on the immune system of fish has not been well investigated.

In a study on rainbow trout, vitamin A and astaxanthin were shown to have only

limited potential as immunomodulatory agents (Thompson et al. 2001). Fish fed diets

deficient in vitamin A and astaxanthin presented lower serum antiprotease and classical


phagocyte respiratory burst activity were unaffected by astaxanthin or vitamin A

deficient diets. The results of this research showed that deficiencies of vitamin A and

astaxanthin can cause a decrease of serum antiprotease activity and this could affect the

resistance of fish to external pathogens. In fish, the antiproteases play an important role in

neutralizing the extracellular proteases of bacterial pathogens, and a2-macroglobulin

level in particular has been reported to be an important correlate of protection against

infection by Aeromonas salmonicida (Ellis, 1985).

7.5. Folic Acid

The effect of folic acid on the immune system was investigated in channel catfish

(Duncan and Lovell, 1994).

Deficiency of folic acid caused binucleated erythrocytes,

spectacle cells (abnormally dividing erythrocytes with two nuclei joined by a strand of

nuclear material), hemocytoblasts (immature blood cells), cytoplasmic clearing,

poikilocytosis (variation in shape of erythrocytes), pyknosis (condensation of the nucleus

of erythrocytes), smudge cells (disintegrating blood cells), macrocytes, erythrocytes with

irregular nuclear borders, and segmented or spreading neutrophils. In the same research it

was observed that fish challenged with E. ictaluri showed maximum survival and

antibody production when folic acid was fed with a high level of vitamin C, but only high

supplementation of folic acid improved survival when the diet contained a marginal level

of vitamin C. In contrast, in Nile tilapia, Tilapia nilotica, administration of folic acid did

not increase antibody levels nor did it improve survival of fish challenged with

Streptococcus iniae (Lim and Klesius 2001).


7.6. Iron

Iron is an essential element for the functioning of organs and tissues of higher

animals, including fish, because of its important role in oxygen transport and cellular

respiration. Iron is also one of the most important micronutrients for its effect on immune

system functions and host defense against infection. Chemotactic response of coelomic

macrophages to Edwardsiella ictaluri exoantigen was depressed in iron-deficient channel

catfish (Barros et al. 2002; Lim and Klesius, 2001; Sealey et al. 1997), but this

abnormality was remedied by feeding an iron repleted diet for 4 weeks (Lim and Klesius,

2001). However, a high level of supplemental iron may be harmful to fish due to the

delicate balance between the requirement of iron for host defense mechanisms and the

requirement of iron to sustain microbial growth (Sealey et al. 1997).

Fish can absorb soluble iron from the water across the gill membrane and

intestinal mucosa, but feed is considered the major source of iron for fish due to low

concentrations of soluble iron in natural waters (Lim et al. 2000). The total dietary iron

requirement for optimum growth, feed efficiency, hematological values, and immune

response of juvenile channel catfish has been determined to be about 30 mg/kg diet

(Sealey et al. 1997).

Several studies have examined the combined effects of vitamin C and iron

supplementation on fish immune system function because ascorbic acid is known to be

involved in the metabolism of iron in animals.

Vitamin C enhances the absorption of iron

from the intestine by reducing ferric iron to ferrous state, a more soluble form that is


with iron-binding proteins, apoferritin and transferritin, into tissue ferritin. High levels of

dietary ascorbic acid in the absence of dietary iron may accelerate iron deficiency and

increase the severity of microcytic anemia in channel catfish (Lim et al. 2000).

Significant interactions between dietary levels of ascorbic acid and iron were

observed in channel catfish with regard to hematological values and survival of fish

challenged with E. ictaluri (Lim et al. 2000). Mean macrophage migration in the absence

or presence of E. ictaluri exoantigen was significantly higher for fish fed the iron

supplemented diets, but neither dietary level of iron nor ascorbic acid nor their interaction

influenced the survival level of juvenile channel catfish against E. ictaluri challenge.

However, the onset of mortality was earlier for fish fed the iron-deficient diet.

7.7. Selenium

Selenium is an essential trace element for all animals. It is a component of

enzymes such as glutathione peroxidase (GSH-Px). This enzyme catalyzes reactions

necessary for the conversion of hydrogen peroxide and fatty acid hydroperoxides into

water and fatty acid alcohols by using reduced glutathione, thereby protecting cell

membranes against oxidative damage. GSH-Px acts along with vitamin E as a biological

antioxidant to protect polyunsaturated phospholipids in cellular and subcellular

membranes from peroxidative damage. Tissues or cellular components that are inherently

low in glutathione peroxidase will not be affected by selenium, but will be still protected

by vitamin E, which acts as an antioxidant by a mechanism not involving GSH-Px. Wang

(1996) found that selenium deficiency in the presence of dietary vitamin E seriously


immune response is protecting cell membranes against peroxide damage during

phagocytosis. Production of hydrogen peroxide in phagocytes is necessary in the process

of killing microbes. However, the overproduction or leakage of peroxides into cytoplasm

can damage the cell membranes of macrophages when GSH-Px activity is low.

Selenium also has a protective effect against the toxicity of heavy metals such as

cadmium and mercury. In addition, selenium is important for normal humoral and

cellular immune response. Selenium plays a role in antibody production through the

proliferation and protection effects of GSH-Px on B-lymphocytes. Suppressed antibody

production appears to be a common result of selenium deficiency among animals.

Chinook salmon fed a selenium deficient diet showed greater mortality from

Renibacterium salmonarium infection (Thorarinsson et al. 1994).

The concentration and source of selenium can influence the immune response to

bacterial challenge (Wang,

1996; Wang et al. 1997). In channel catfish dietary selenium

concentrations for maximum survival from E. ictaluri challenge was 0.20 mg/kg for fish

fed Se-Methionine, and 0.40 mg/kg for fish fed Se-Yeast (Se-Y) or Na2SeO3. Antibody

production generally increased as dietary concentration of selenium increased. Antibody

titer was highest for fish fed Se-Y, intermediate for fish fed Se-M and lowest for fish fed

Na2SeO3. Macrophage chemotactic response was higher for fish fed Se-M and Se-Y than

for the control and Na2SeO3. A possible explanation for these results is that amino acid

bound trace minerals have higher biological values than inorganic forms because the

elements remain chelated after intestinal absorption, are transported intact to target


Supplementation of vitamin E or selenium to channel catfish at higher doses than

the recommended levels of one or both nutrients may enhance macrophage functions

(Wise et al. 1993). However, selenium and vitamin E do not complement each other,and

neither compensates for a deficiency of the other. Megadoses of selenium used for

disease therapy may involve risks because of the high toxicity of this mineral to animals.

Fish can absorb selenium present in the water across the gills and digestive tract

(Lim et al. 2001). The uptake of selenium as selenite across the gills is very efficient even

at low waterbone concentrations. At least in rainbow trout, the level of selenium in the

water can affect the dietary selenium requirement (Hodson and Hilton, 1983).

Selenium has been found to be required in the diet of Atlantic salmon, rainbow

trout and channel catfish (Lim et al. 2001), with the actual requirement dependent upon

the source ingested, polyunsaturated fatty acids, vitamin E content in the diet, and the

concentration of selenium in the water.

7.8. Zinc

Zinc is required for normal growth, development and function of all animal

species. The primary functions of zinc are based on its role as a cofactor in several

enzyme systems and as a component of a large number of metalloenzymes. Zinc plays an

important role in immune defense mechanisms. Zinc is fundamental to T-cell function

and a deficiency leads to thymus atrophy and decrease thymic hormone and lymphokine

production in mice (Fraker et al. 1977). Zinc may also affect resistance to parasites and

bacteria. For example, mice fed a zinc deficient diet for 8 days were significantly more