Harry Dickerson Ichthyophthirius multifiliis: a model

Theodore Clark
of cutaneous infection and
imniunity in fishes

Authors' addresses Summary: The parasitic ciliate Ichthyophthirius muitifiiiis offers a tiseful system
Harry Dickersan', Theodore Clork', for the study of cutaneous immunity against an infectious microorganism.
'Depanmeni of Medical Microbiology and Naive fish usually die following infection, but animals surviving sublethal
Parasitology, College of Veterinary Medicine, parasite exposure become resistant to subsequent challenge. This resis-
University oi Georgia, Athens. Georgia. USA, tance correlates v^ith the presence of humoral antibodies in the sera and
'Deparimcni of Microbiology aiid Immunology. cutaneous tnucus of immune fish. A mechanistn of immunity has recently
College of Veterinary Medicine, CorneO been elucidated that involves andbody binding to surface proteins
University, Ithaca, New York. USA. (referred to as immobilization antigens or i-antigens) located on the par-
asite cell and ciliary membranes. Antibody-mediated cross-liiiking of i-
Correspondence to: antigens triggers a response by the parasite resulting in its exit from tbe
Harry Dickerson host. These effects can be observed directly on the surface of live fish. In
Department of Medical Microbiology and addition to allowing the observation of effector responses in vivo, khihyoph-
ParasHology thirius also provides a means to study the ontogeny of the mucosal immune
College of Veterinary Medicine response. The sites of antigen capture and presentation, and the sites of
University of Georgia antibody production, are unknown wiih regard to cutaneous immunity.
Athens GA 30605 Because the external epithelial surfaces of fish are often the points of
USA pathogen entry, a basic utiderstanding of the itiductive immune mecha-
Fax: 1 706 542 8254 nisms and immune cell interactions in tbe skin and gills is extremely
e-maii: hwd@ca!c.vet.uga,edu important with regard to vaccine development. The development of Ich-
thyophthirius as an experimental system and how it might be used to address
Acknowledgements these issues are discussed in this review.
The authors thank Mr K. Carter, College of
Viiierinary Medicine, University of Georgia, for
the graphic iliusUations seen in Figs 1 & 2,
The authors' research was supported by grant
9S-37 204-2139 from the US Department of Introduction
Agriculttire National Research Inidative Grant
Program. The obligate parasite, Ichthyophthirius multifiliis, is a ciliated proto-
zoan with a hfe-cycle that includes a fish-associated feeding
stage (trophom) and a free-swimming infective stage (theront)
(Fig, ]). Because the number of parasites increases by 2-3 orders
of magnitude with each round of infection, fish in closed envi-
ronmeiits, such as ponds and aquaria, are rapidly overwhelmed.
While the disease is usually lethal, it has long been known that
fish surviving epizootics become resistant to subsequent para-
site infection {1-3). Fish can also be immunized in the labora-
tory by exposure to limited numbers of parasites, followed by
Immunotogical Reviews 1998
Vol. 166; 377- 3S-I treatment of aquaria with low concentrations of formalin to kill
Printed in Denmark AM rights reserved free-swimming theronts that develop from the initial infection
Copyright © Munksgaard 1998 (4—12). Resistance becomes manifest within 3-4 weeks and
Immunological Reviews lasts for a period of at least 12 months (4),
ISSN 0)05-2896

377
 DiLkerson S Cldi< • Parasite ^y for studic? o*" fish

As .shown by Hines & Spira (10) almost 25 years ago, sera type lead.s to the development of cross-immunity between bet-
from immune fish immobilize the parasiie in vitro. With the idea erologous strains (30)- Because there is substantial evidetice fbr
that ihc antigens rcsponsibJe for immobilization play a role in the involvement of i-autigens in protective iminuuity, the pos-
eliciting acquired immunity, considerable effort has been spent sibility that shared antigenic determiuants are responsible for
in attempts to defiiie the target aniigens responsible for this cross-protection clearly exists.
phenomenon (5-6, 13-14). These antigens (referred to as To characterize the i-antigens in more detail, a 1.2 kb
immobilization antigens, or i-antigens) have been identified as cDNA and its corresponding gene encoding the 48 kDa antigen
a class of highly abundant surface proteins on the plasma and oflclilliyoplithirius isolate GI have been isolated (23) (T. G. Clark,
cihary membranes of I. multifiliis. Structurally, they are related to manuscript submitted). Northern blotting studies using the
analogous, well-studied stirface antigens of the free-living cili- 1.2 kb cDNA probe have revealed that i-antigen transcripts are
ates, Parami^cittm and Tclraii/mtna (I 5-24). Antibody binding to developmentally regulated and undergo a massive increase in
these proteins causes cilia to chimp together and inhibits cell abundance during transition from the fish-associated tropliout
motility A strong argument for the involvement of the i-anii- stage to the infective theront stage (23). Steady state levels of i-
gens of rcbthyophiliirius in eliciting protective inmiunity can be antigen niRNA reach extraordinarily higb levels equivalent to
made on the basis of four observations. First, fish mount a about 6%ofthe totalpnlyA+ of the cell (23). Based on Soutbern
humoral antibody response against these antigens (S-7, 24, hybridization analysis, the Gl isolate contains m o closely
25). Second, both Western blotting data, and in vilro immobili- related genes which encode the 48 kDa and 60 kDa i-antigens
zation assays indicate that antibodies against the i-autigens are of this isolate (23, 24, 29). The gene for the 48 kDa antigen has
present in mucus at the site of infection (that is, at the surface recently been cloned and sequenced in its entirety (T. G. Clark,
of the fish) (26). Third, mcnise monoclonal antibodies against unpublished data). It contains a single uninterrupted reading
these antigens provide virtually complete passive protection frame of 1,326 nt encoding a protein of predicted mass of
against infection (24, 27, 28). Finally, and most importantly, 45,023 Da that has a number of characteristic features.
the i-antigcns themselves stimulate protective immunity when
The deduced protein begins with a stretch of 20 mostly
injected into fish (X. Wang, H. W Dickerson, ttnpublished
hydrophobic amino acids that are indicative of a signal peptide
data).
for a membrane or secretory protein (a predicted cleavage site
between an alanine and valine residue at positions 2 0 - 2 1 ( 3 1 )
has been verified by N-terniinal amino acid sequence analysis
The surface immobilization antigens of Ichthyophthirius
of tbe protein itself). A second stretch of 14 mostly hydropho-
in wild-type isolates of Ichthyophthiriui, the i-antigens exist as bic amino acids is present at the carboxy terminus of the
cither one or two antigenically related polypeptides which deduced protein. These are separated by a short spacer from
migrate on SDS-PAGE gels with a M, ranging between 40-60 three small amino acids ending in a cysteine (22 residues
kDa depending on serolype. These surface proteins are upstream of the predicted stop codon). This type of structure is
extremely abundant aud comprise up t(3 60% of the total ciliary highly characteristic of an addition site for a glycosylphospho-
membrane protein (13). Based on immobilization with spe- tidyhnositol (GPl) anchor (32), and is consistent with the fact
cific antisera, five i-antigeu serotypes of Iclnliyophtliirius (desig- that the analagous i-antigens of the free-living ciliates, Puramc-
nated A-E) are currently known. The majority of isolates (eight cium and Tamhymtna, are also known to be bound to the plasma
often) fall within three discrete serotype groups (A, C, or D). membrane through a GPl-moiety (33, 34). Amino acid resi-
Polyclonal antisera against any one serotype fails to immobilize dues 316-324 (GNFEAGKSQ) encode a putative ATP/GTP
heterologous strains, and the epitopes responsible fbr immobi- binding site (or P-loop domain) (35). Finally, within the cen-
lization are distinct for the i-antigens of different serotypes tral region of the protein, there are five homolog(nis segmeuts
(29). Nevertheless, the proteins tiieniselves share antigenic (of ~80 residues each) that are repeated in tandem. The inosL
determinants in common, since they all cross-react in Western striking feature of these repeats is the presence of periodic cys-
blots with polyclonal antisera against the i-antigens of any teine residues which fall into register when tbe repeats are
given strain (29). Southern hybridization analysis of genomic aligned. The spacing of these cysieines is precisely what would
DNA from different isolates bas revealed that the i-antigens are, be expected for zinc-binding (or "zinc-finger") domains. A
in fact, products of a related gene family which shares approx- large number of proteins have been shown to contain such
imately 85% homology at the nucleotide sequence level (29). motifs, the vast majority of which are involved in DNA binding
Exposure of channel catfish (ktalmus piiiiLtatus) to a giveu sero- and/or transcriptional activation (36). An interesting excep-

378 Immuuolagicul Reyievv^ 166/ 1998

 Dickcreon & Clark • Parasite ^yrtcn for studies of fish immunrtv

Fig. 1. Life cycle of Ichthyophthirius muilifiliis, Frcc-

Tomite swiinniin^ llieroius (-40 um tn length) actively pcnctralc
(driven by ihe aclioii of ciiia) inio ihe surface epJtIielia of
Encystment
the skin and giik of,•susceptible hosts. While on the fish,
and division
they transform into trophonts that sustain nourishment
from hdsi lissuc, I, mtiltifiJiis is an obligate parasite and
growth o}-i the fish is required lo complete (he life cycle.
Tomont Theront
Over a period oi'4- to 7 da)',s (ll^C.) they grow to
-S00-800 iJin in diameter and are visible to the naked eye
as individual while ';p()is widiiii the skin. In response EO
unidentified signals, trophonts leave ihe fish and become
free-swimming loiiionts that attach to a substrate \a the
water where they eneysi and undergo 9-10 divisions t{)
form 500-1000 tomites (18-2+h al Z2°C). Tomites
differentiate uito theronts and penetrate the cyst wall to
escape into ihe waier column, thus completing the life
cycle. All stages of ihe parasite are ciliated. In immune fish,
[heronts can enter ihe epilheiium, but in the presence o[
antibodies iliey apparently are prevented from developing
further.

Tropbont

tion lo this, however, is a diverse family of anrigcns, referred to immunity against the parasite. These studies have important
as variant-specific surface proteins (VSPs), on the surface of imphcations for the nature of the inmiune response in fish. In
Giardia lamblia, a mammalian gut parasite. These antigen.s have this regard, we have found that while mouse IgG class mAhs
tandemly repeated domains that inchide characteristically confer strong protection against the parasite (as little as 20 pg
spaced cysteine residues that can bind zinc in vitro (37, 38), of inouse antihody administered intraperitoneally (i,p-) pro-
While Ichtfiyophthirius and Giardia are highly diverged from an tects juvenile channel catfish weighing 10 g against lethal chal-
evolutionary standpoint, it is intriguhig that they express sur- lenge), IgM fails to protect (28). Similarly, tetrameric
face antigens with this novel structural featnre in common, Ii (IgM-hke) antihc')dies from the sera of Ichlhyophthiri us-immune
should he noted that the precise functions of the i-antigens and channel catfish also fail to protect despite the fact that these
VSPs are unknown. antibodies strongly immohilize the parasite in vitro. Based on
In addition to the genetic elements that encode the 48-kDa ELISA and in vitro immohilization assays, we have found that IgG
i-antigen of the Gl isolate, a partial cDNA (~ 1 kh) correspond- class mAhs reach both the serum and the mucus of test animals
ing to the 55-kDa i-antigen of an isolate designated G,S folkjwing i.p, injection (28), In contrast, serum antihodies
(T. L, Lin, T. G, Clark. H. W Dickerson, unpuhlished) has heen from immune fish administered i.p, to naive fish enter the
cloned and sequenced. Because the G1 and G5 isolates are rep- serum hut not the mucus, while IgM marine mAbs enter tiei-
resentative of different serotypes (A and D respectively), com- ther comparttiient (28), Nevertheless, in actively immune fish
parison of the dednced amino acid sequences should he antibodies against kiithyophtfiirius are present in cutaneous
extremely informative in terms of the antigenic relationship mucus. Taken together, these results strongly suggest that pro-
hetween them. While the complete sequence of the G5 cDNA tective antihodies are produced locally in the skin as well as the
is not yet availahle, preliminary data indicates that extensive serum (27, 28), This work is highly relevant to an understand-
homologies between the Gl and GS proteins exist. ing of immunity in fish against surface pathogens in general
and is entirely consistent with biochemical and metabolic evi-
dence that serum and mucus antibodies are distinct (39-42),
The mechanisms of immunity against Ichthyophthirius Whether such antihodies arise locally from lymphocytes within
the skin, or are produced elsewhere and enter the epithehum
Although a considerahle amount is known ahout the structure
through some, as yet, iinidentilied pathway is unclear (7, 24,
of the i-antigens and the genes that encode them, their hioltig-
39, 41, 42),
ical function remains ohscure. Nevertheless, passive immuniza-
tion experiments with mouse monoclonal antihodies (mAhs) With regard to mechanisms of immunity, passive immuni-
against these proteins have shed light on the mechanisms of zation studies using fish experimentally infected with Ichthyoph-

Immu/n)l»j|itul Rtvitvv.s i 6 6 / 1 9 9 8 379

Dickerson & Clark • Parasite system fo; iLudses offish tmmunrty

thirius have shown that i.p, injection of i-antigen-specific mouse theronts rapidly penetrate the superficial mucus barrier (jf the
mAhs results in complete clearance of parasites from the skin skin w-here the antibody concentration is too low to cause
within a period of about 8 h (beginning 3-5 h after incjcula- immobilization. Antibodies in the skin or deeper layers of
tion) (24, 27). The response to mouse niAhs requires i-antigen mucus bind to the surface i-antigens of the parasite cau,sing
cross-hnking on the parasite snrface since Fab, fragments have either exiting from the host (at antibody concentradons that are
the same effect as intact antibody, while monovalent Fah fail to subthreshold for immobihzation) or immobilization and cell
protect. The activity of Fah (i.e. protection) can be restored by deatii (facilitated by antihody at higher concentration). Where
subseqnent i,p. injection of bivalent goat anti-mouse antibody cutaneous antibodies originate is still conjectural. Their source
(27). Parasite clearance in response to mAb hinding has been could be either lymphocytes present locally in the skin or lym-
shown to result from a mass exiting of parasites from the host phoid tissues located outside the skin from which antibody is
(24, 27). Quite remarkably almost identical conclusions have transported to the epithelium through an unidentified pathway.
heen drawn from experiments with actively immune fish (7),
Working with carp. Cross & Matthews examined the fate of par-
asites on fish immunized by previous exposure to Ichtfeyophthir- Secretory cutaneous immunity in teleosts:
ius. While theronts were capable of invading the epithelium, studies in the channel catfish {Ictalurus punctatus}
roughly 80% disappeared within 2 h of infection. Their disap-
The common channel catfish has been used as a comparative
pearance was too rapid to he accounted for by cell-mediated
immimological research model for more than I 5 years result-
events, and it was concluded that active immunity was due to a
ing in the accumulation of a substantial amount of basic infor-
rapid exit of parasites from the host in response to antibody
mation on immunoglobulin structure and immune celi func-
binding (7). Furthermore, because some studies revealed only
tion (46). The predominant serum immunoglobuhn in this
weak correlations between serum antihody levels and protec-
species is an IgM-like tetramer with a molecular mass of
tion (II), it was argued that such antihodies must originate
700,000 Da (47-49). Under denaturing conditions (for in-
within the skin (7, 43), Together with the passive immuniza-
stance, in solvents such as sodium dodecyl sulfate or guani-
tion experiments cited ahove. these studies provide further sup-
dine-HCl) the molecule reveals an unusual architecture, where
port for the involvement of cutaneous antibodies in protection
various comhinations of eovalently linked heavy polypeptide
against khthyophthirius. Although the signahng pathway that trig-
chains (70,000 Da) and light chains (22.000-25,000 Da)
gers the parasite to leave the host is unknown, the phenomenon
form eight discernable antibody subpopulations (50, 51). The
itself represents a novel instance among protozoa in which
functional significance of this variable covalent couphng in vivo
immunity results from an effect of antibody on parasite behav-
is unknown. The structures of serum and cutaneous antibody
ior (27, 44). This discovery was possible hecause Ichthyophthiritis
appear to be identical (39),
can he directly ohserved in the skin.
Recent work (52) has shown the presence of a channel cat-
While the mechanisms responsihle for the parasite response fish imniunoglohulin analogoiis in structure to the IgD isotype
are unknown, at least one possibility is that i-antlgen cross-link- found in mammals. Its existence in a teleost species suggests
ing at the cell surface triggers a signal transduction cascade lead- that IgD is an evolutionarily ancient immunoglobulin with a
ing to the release of lytic enzymes from secretory vesicles within potentially important (and as yet unknown) function. Its pres-
the parasite. In this regard, it is known that the contents of cor- ence in cutaneous mucus has not been investigated
tical mucocysEs are ejected at the time of host invasion and it has (W L. Clem, personal communication).
been argued that tissue hydrolases released hy the parasite play Studies indicate that the thymus and head kidney are the
a role in ingress and egress from the host during the normal primary T and B-lympliocyte differentiative organs, while the
course of the hfe cycle (45). Furthermore, in addition to immo- spleen serves as a secondary or peripheral lymphoid tissue,
hilizing cells in vitro, binding of antihodies to i-antigens triggers B-cell progenitors and their site of origin have remained elusive
a secretory response in which mucus is shed from stored com- and the mammalian bone marrow equivalent (i,e, central lym-
partments within the cortical cytoplasm. Indeed, parasites phoid organ) remains in question (53), Thymus-derived cell-
impelled to leave fish after antihody administration are usually dependent immune functions such as hapten carrier, graft
surrounded by sloughed epithelial cells (44), rejection and delayed hypersensitivity have heen demonstrated,
Based on the above studies, a model of antibody-mediated although there is limited data regarding T-cell surface markers
protection against Ichthyophthirius is proposed which provides a (46. 54). B and T-celi proliferation occurs in vitro in response to
foundation for future research (Fig, 2), In this model, infective lipopolysaccharide and concanavalin A respectively (46). Anti-

380 Immunulogitfll Revieivs 166/1998

 ([)ickei'son S C\ii\-k • P.u.isite system for studies offish immunity

Invading Fig. 2. Mode! of antibody-mediated protection against

theront Ichthyophdiirius multifiliis in the skin of immune fish. Tlieronts
Exicing
peiietrjitoihrou^ii ihe eutaiieotismiicusinto theepitheiimnof the
theront
skin. Tetrameric antibodies in the mucus and/or skin bind to
parasite stirface i antigens. At inimunoglobulin concentrations
ihat are siibthrfshuld fbr innnobilization (i.e. in superficial
iTiLictis), antibody binding c'licit,s a signal thai results in exiting of
ihe parasite trom the host. At higher antibody concentrations
postulated to occur deeper within the epithelium, binding causes
immobilization and facilitaies kilhng of the parasite. The source
ofauiihodies present at the stirface offish is conjectural. They
may be secreted locally horn lymphocytes within die skin (as
sliovvn in the iigure), or be produced elsewhere and enter the
epithelium through ^n. as yet. uaidentined pathway, LeUers to the
right of the figure indicate the following: A, cutaneous niticus;
B, epithelium: C, dermis. Also depicted are lymphocytes (Ltn)
located wiihin the epithelium, and capillaries (Cp) looping from
[he dermis intc-> the epidermis.

• -V--
- .--• i f -- -
Cutaneous Ab

Ab-producing
lymphocyte

gen processing requires accessory antigen-presenting cells, Only a few basic studies have been conducted on the cuta-
such as monocytes and macrophages. that interact with specific neous mucosal immune respcnise in channel catfish. These have
lymphocytes in an allogenic restricted fashion (55), By far the involved immunization with antigens administered by surface
majority of experimental studies involving lymphoid ceils of or systemic routes with subsequent evaluation of antihody pro-
catfish have nsed cells recovered from peripheral blood, head duction in sera and surface mucus. In an early study, Ourth (57)
kidney, or spleen. In contrast, less is known about the lympho- injected channel catfish i,p, with killed SQimondla paratyphi cells in
cytes and accessory cells associated with the secretory immune Freund's complete adjuvant (two injections at a 7 I day interval)
system of the skin and gut. and measured agglutinating antibody tilers in surface mucus at
100 days. He found that antigen given systemically elicited the
production of surface antihody (tuhe agghitination antihody
Cutaneous immunity titer of 1:80). Agglutinating antihody was identified in gel
The cutaneous surface of channel catfish consists of a delicate immunodiffusion tests using rabbit antisera against channel
and thin epidielial cell layer covered with mucus. The skin is catfish immunoglohuhn. He hypothesized that mucus antibody
non-keratinized and contains large numbers of mucus-produc- originates from the serum by transudation. hut that it might
ing goblet cells (56), It is enervated and sensitive to environ- also be produced locally by lymphocytes in the skin,
mental stimuli. The deeper layers contain meianophores and A more extensive investigation was carried out hy Lobb in
alarm cells that release proteinaceous suhstances when injured. 1987, and provided a number of important observations on the
Capillaries extend into the epidermis from the dermal layers, nature of the cutaneous immune response (39), First, when
and come within 10 cell thicknesses of the surface (39), Lym- fish were carefully exposed to soluble antigen (dlnitrophenol
phocytes are found scattered throughout the epidermis and coupled to horse serum albumin) by hath exposure, the pre-
appear most numerous at the junction with the underlying der- dominant immune response occurred at ihe surface, as mea-
mis (39). No organized lymph areas have heen ohserved, sured by the production of cutaneous antibodies. The produc-
Immunoglobulin is present in surface mucus as well as the epi- tion of secretory antibodies without a concomitant systemic
dermal and submucosal tissues (C. Xu, unpublished data). serum antihody response indicated a compartmentalization of

Revievv,s 166/1998 381

Dickerson & Clark • Parasite 'jv for studies of fish imnrjnry

the secretory and systemic immune responses. This finding nically challenging hecause of the difficulties associated with
supports earlier and snhsequent research in a number of fish the collection and measurement of antibodies and immune
species suggesting the existence of a separate secretory immune cells from the skin,
system (40. 42). Second, the molecular structure of antibodies Ichthyophthirius and the channel catfish provide a good
isolated from the cutaneons mucus was identical to that of host-parasite experimental system to address the above ques-
serum iminunoglohulin. The heavy chain of mncus antihody ti(nis heeause the parasite's activity is restricted to the snrface
migrated at the same relative position as the heavy chains found epithelial layers of the skin, and the effector response of the
in serum antibody, that is - 7 0 . 0 0 0 Da, The light chains showed host is known to he antihody mediated. In addition, the target
three molecular mass variants of approximately 26.000. antigens (i.e, i-antigens) have been identified and character-
24,000, and 22.000 Da. The light chains of cutaneous mucus ized. Thus, with existing reagents (mouse mAbs against i-anti-
antihody comprised tw^j different isotypes as occurs in channel gens and channel catfish immunogkjhulin). it is possible to
catfish serum antibody (58), One class, designated F. is com- design in vivo and in vitro experinrients that can be used to eluci-
posed of tw(D mass variants of 24.000 and 22.000 Da (58), The date the basic components and kinetics of the cutaneous
other class, designated G. is represented by the 26,000 Da immune response. In mammals, the production of enteric
form. Both classes of light chain were found in cutaneous mucosal antibody is initiated locally either by direct antigen
mucus. There was no apparent immunoglobulin-associated stimulation of resident lymphoid cells or throtigh migration to
secretory component protein present, mucosal surfaces of antigen-specific lymphocytes generated in
the gut-associated lymphoid tissue (59). Some evidence exists
for such a meelianism in fish as well (60), By analogy, we
Channel catfish and Ichthyophthirius as a host-parasite model hypothesize that the skin offish is also an active participant In
for studies of cutaneous immunity the immune response, serving as hoth an indnctive and an
effector site, and not merely as an end target tissue t)f antihody
As the field of comparative immunology developed, research
deposition. Thus, methods used to evaluate the trafficking of
on the immune system of fishes progressed from simple
mammahan ceils among mncosal surfaces and their appropriate
descriptive studies to sophisticated lahoratory experimentation.
systemic immune organs should be applicable to similar studies
The channel catfish has emerged as an important model for
in the channel catfish. For instance, using ELISPOT assays (61,
such studies, because reagents such as mAhs to specific immii-
62) with mAhs specific to Ichthyophthirius i-antigens and channel
noglobulin antigenic determinants and other cell-surface
catfish immunoglohuhn, it should be possible to isolate and
markers are availahle and are freely shared among workers in
count i-antigen specific antibody-secreting cells in the skin,
the research community As mentioned above, certain reagents,
head kidney, and spleen following parasite challenge. Informa-
such as definitive T-cell surface markers, are still scarce. Never-
tion on the ontogeny of the cutaneous immune response
theless, even withottt all the tools availahle to mammalian
against the parasite can be obtained hy following the develop-
immimologists, a great deal has been accomplished regarding
ment of populations of Ichthyophthiiius-specific antibody-secret-
elucidation of the fundamental mechanisms of (eleost immu-
ing cells in these tissues. Indeed, such studies are currently
nity Much of this work has heen done through basic studies
underway (J, Maki, J, R, McGhee, unpnhlished).
involving in vitro experiments with isolated cells and molecules,
and has recently been extended to in vivo stndies of host-patho- Importantly, because Ichthyophthirius is confined to the sur-
gen interaction (27, 44, 46), Important questions still remain face epithelial ceil layers, it can be visuahzed by light micros-
unanswered, however, particularly wath regard to cutaneons copy from the time it invades fish until the time it exits. Con-
secretory immunity. For instance, where and how does antigen sequently, it provides a unique system for the direct study of
presentation occur and what cells are involved? How do mucus host-parasite interactions. Because of this, it has heen possible
antihodies get to the animal's surface, and where are they pro- to directly visuahze the effect of antibody on the live parasite
duced? Are antibodies secreted locally by resident lymphocytes, within the host as described earlier. We know of no other host-
and, if so, where do these cells normally reside? If antibodies parasite system in which this is possible. Finally, hecause Ich-
are produced at remote sites such as the head kidney, spleen, or thyophthirius is an important pathogen in its own right, informa-
other tissues, what are the mechanisms hy which they are tion regarding j>rotective immunity can be applied directly to
transported to the surface of the skin? These questions are tech- the development of vaccines for nse in the aquacultnre industry.

382 Immunnloyicol Keviewi 166/1998

 Difkerion & Cla'"k • PEirasitc system for studies of fisn •mmunity

References
Biuer ON. The immune; respim.se of clianiiei i 3. Dickerson HW Clark TG, Findly RC. 28. Tin TL, Clark TG, Dickerson HW Passue
catfish ktuluni.s pLncfcutis Rafinesquc lo khthyopfitliidui niiiltifiiii; has membrane immunization of cliaiiiicl catfish (ictalLinis
iclithyoplithirius multjfiiiK FouqitijL associated immobilization antigens. pLiictutu^) against the ciliated protozoan parasite
Dokl Akiid Naiik SSSR I 9 53;93:37 7-3 79, JProtozool 1989;36:1 59-164. klifhyophlhinui multifiliis by use of murine
Becke.rt H, Allison R, Some iiosi responses of 14, lin TL, Dickerson HW Purification and pardal monoclonal antibodies.
white catfish to Ithlliyophihinus muItifiliK characterization of immobilization andgens Infect Immun 1996,64:4085-4090.
R)iiquet. from klithyophtfiinus rauilililiis. 29. Dickerson HW Clark TG, Lefl AA. Serotypic
Proceedings of the Southsjastern Association JProtozool 1992,39;4S7-463. variadon among isolates of khtliyophtliirius
of Game and Fisli Commissioners 15. Preer JR Surface antigens of Puiumecium hi: multifiliis based on immobilization.
I964,18:438-44i. Gall JG, ed. The molecular biolog) ofciifatcd JFukaryot Mkrohiol 1 993;40;8 16-820.
Bntc.lier AD, Oulbreaks of white spot or protozoa. Orlando; Academir; 1986. 30. lelT AA, YoshinagaT, Dickerson HW
Ichihyophthiriasis (khthyophtiiiriiis muilifiJii? p, 3 0 i - 3 3 9 . Cross immunity in cliaimel catfish against two
Fouqnet, i 876) at the hatcheries of Ballarat 1 6. Preer JR, Preer LB, Rudinan B, Barnett A. immobilization serotypes of khthyophtliiriu.^
Fish Acciimati/ation Society with notes on Molecular biology of die genes fbr mulfifilib.
lahoratory experiments. immobilization antigens in Pa tamed u tii. JFish Dis 1994;17 429-432.
Proc R Soc Victoria 1941 ;53: i 26-144, JProto/ooi l9B7;34;4]8-423 31 Nielsen H, Engelbrecht J, Brmiak S.
Burkart MA, Clark TG. Dickerson HW. 17. Caron F, Meyer E. Molecular basis of surface von Heijne G Identilicatioii of prokaryoiic
Imintmizalion of channel catfish, Jtfulurus antigen variation in paramecia. and etikaryoilc signal pepddes and prediction
pimctdtus Rafinesqtic, against khthyopiithiriui rVniin Rev Microbiol 1 989;+3:23-42, of their cleavage sites.
mukifilii.s (Fouquct): killed versns live 18. Doerder VV. ilallberg RL. Identification of a Protein Eng l 9 9 7 ; 1 0 ; l - 6 ,
vaccines, cDNA coding fbr the SerH3 Siirface protein of 32. Enghind P TIK* structure and biosynthesis of
JFish Dis 1 9 9 0 ; n : 4 O I - 4 I O . Tcttiih)'moio thetmopliilfl. glycosylphospatidyhnositoi protein anchors.
Clark TG, Dickerson HVv: GratzekJB, J Protozo*;^! 1989,36:304-307 AnntiRev Biochem 1993;62; 1 21-1 38.
Findly RC. In vitto response of Idithyophtliirius 19, Epstein LM, Forney JD, Mendelian and non- 33. Capdeville Y. Cardoso de .Ahneida ML,
muilifiliis to sera from immune channel Mendelian mutations affecting surface antigen Dexegnaucourt G. The membrane-anchor of
catfish. expression in ParunieciujTi leitaurdla. PurumeciLiTi temperatnre-specific surface
JFish Biol ]987;31:203-208. Mol Cell Biol I984;4:1 583-1 590. andgens ts a glycosylinositol phospholipid
Clark TG, Dickerson HW, Findly RC, Immune 20, Ron A, Williams KE, Doerder FP The Btochcm Biophys Res Commun
response of channel catfish to ciliary andgens immobihzadon antigens of Tttiafiytrsnii )987;147;12I9-I22S,
of kiithyopbliiiriu.s inulnfiliii, liiermopliiki are glycoproteins. 34. Ko Y-G, Thompson GA )r, hnmobilization
Dev Comp hnmimol 1988; 12:.S8 1-S94, JProtozool 1992;39:.S08-510. antigens froiTi Tctruhymeiia thermophik are
Cross ML, Matthews RA, Ichthyophthiriasis in 2i Bruns PJ. Immobilization antigens of glycosyl phosphaddylinosttol-liiiked proteins,
carp, Cyprmtii tarpio L.: fate of parasites in Tetraiiymma pyriiotmh. 1, Assay and extraction, IProtozool l992;39;719-723
immunized lkh, bxpCehRes ) 97 1 ;6S;44S-4.S3. 35. SarasteM. Sibbald PR, Wittinghofer A. The
J Fish Dis 1992:15;4')7-S05. 22. Smith DL, et al. Characterization of the T, L, i, P-loop - a connnoii modf in ATP- and
Ooss MI, Matthews RA. Localized lenkr)cyte S, M, and P cell suriace (immobilization) GTP-bindmg proteins
response to kinhj'oplithiriiii muilihiiis andgens ol Tetruhymenii thermophiia; molecular Trends Eiochem Sci 1 990;15;430-434,
establishment in immune carp weights, isoforms, and cross-reaetivity of 36. BergjM, Shi Y. The galvanization of biology: a
Cypdnus curpio L. aniisera. grov.ing appreciation for the role of zinc.
Vet Immunol Jmjtiiiiiopathol J Prolozool i992;39;420—4-28. Science 1 996;271; 1 08 I - ] 085,
l993;38:34i-358. 23. Clark TG, McGraw RA, Dickerson HW: 5 7. Zhang Y, ,Aley SB, Statile> SL Jr, Gillin FD,
Cross ML, Matthews RA. khlhyophfliiiiu'; Developmental expressicjn of surface antigen Gystciiie-dependent zmc binding by
inulrifilit^'Fouquet (Cihophora): the genes in the parasitic, cihate, Ichthynplithiriu.^ memhrane proteins of Giurdia kimWiu.
localization of sites immunogenic to the host mullililiis. Inieetlmmnn I993;6!; S20-5 24,
Cyprinii.'. curpio (L,) Proc Natl Acad Sci USA 1992:89:6363-6367. 38. Nash TO, Mowatt MR. Variant-specific surface
Fish Shellfish hnnnuiol 1993;3:1 3-24. 24. Clark TG, Lin T, Dickerson HW Suriace proteins of Giatdiu lamliim are zinc-binding
10. Fliiies RS, Spira DT. Jclithyophthtnasis in the immobilization antigens of khlhyopbtliiiius proteins,
mirror carp Cyprinui" larpio (L.) V Acquired tnuitifiliii; their role in protective immunity. Proe Nad Acad,Sci USA I 993;90:5489-S493.
immunity .JjinuRev I'ish Dis 1995;.5; 1 I 3-1 3 I. 39. I.ubb CJ. Secretory unmunity induced in
JFish Biol i974;6:373-378. 25. Cross MI,. .Antibody binding following channel catfish, ktiilunis puncmtui, following
Houghton G, Matthews KA exposiire of hve khtliyophthinui niiilrifilii; bath immunization.
Immunosuppression of carp (Cyptinui- curpio I.) (Ciiiophora) to serum frc")m immune carp Dev Comp Immunol I 987;11 ;727-738.
to ichthyophtlnriasis using the corlicosteroid Cyprinus curpio. 40. Lobb GJ, Clem LW The metabolic relationship
triamcinolone acetonide. Dis Aquatic Organisms 1993; 17:1 59-1 64 of the immunoglobuhiis in fish serum,
Vet Jnimunol hnmunopathol 26. XuC. Immunity of channel cathsh to the cutaneous mucus, and bile.
1986,12:413-419. ciliated protist, Iciitliyoplithitius multitilijs [PliD JImmunol I98I;127;] 52 5-iS29.
12. McCallum HT AcqLiired resistance of hiack thesis]. Athens (GA): University of Georgia;
mollies Pciecilia iutipinmi to infection by
khtiiyophihinus niukifilii.?. 27 Clark TG, LmT, Dickerson HW Surface antigen
Parasitolog) 1 986;93.2S ] - 2 6 i . cross-linking triggers forced esist of a
protozoan parasiie irom its host.
Proc Natl Acad Sci USA 1 996;93;6825-6829.

evieivs 166/1998 383

Dickerson Parasite system for studies offish immunity

41, Lumsden JS, Ostland VE, Byrne PJ, 47. Lobb CJ, Olseii MOJ, hnmunoglobulin heav^ 56, Grizzle JM, Rogers WA, Anatomy and
Ferguson HW Detection of a distinct H chain isotypes in a teieost fish, liistologyoi the channel catfish. Aubnrn {AL);
gill-stirfacc antibody response following Jhnmmiol 1988;141;I236-1245. Auburn University F.xporiment Station; 1976,
horizontal infection and hath challenge of 48. Klesius P Immune system of channel catfish; 57, Ourth DD, Secret(jry IgM, lysozyme and
brook tront, Saivclinus fontindb with Fluvohuttenum an overture on immunity to Edwardsidlu iduluti. lymphocytes in die skin mucus ol the channel
brachinphiium, the causative agent of bacterial Annu Rev Fish Dis i992;2;325-338. catfish, k!durui piinctutu.s.
gill disease. 49, Wilson MR, Warr GW FLsh immunoglobtilins Dev CoiTip Imnutnol I 980;4;6S-74.
Dis Aquadc Organisms 199 3; 16:21-2 7, and the genes that encode them, 58 Lobb CJ, Olson MO, Clem IW
42, Rombont HWM, Taverne H van de Kamp M, Annu Rev Fish Dis I 992;2 2Oi-22 i. Immnnoglobulin hght chain classes in a
Taverne-Tlieile AJ, Differences in mucus and 50, Lobb CJ, Clem LW Distinctive subpopulations leleost fish
serum immnnoglobulin of carp of catfish sernm antihody and J hmiinnol 1 984;132; 19 I 7-1923,
(Cyprinu.? carpi0 L.) immunoglobnliii, 59, McGhec J, Mestecky J, Deri/baugh M,
Dev Comp Immunol 1993; 1 7:309-3 I 7. Mol Immunol I983;2O;81 1-81 S, EldridgeJ, Ilirasawa M, Kiyono II. The
43, Matthews RA. khtfiyophthmus muitifiliis fouquct 51. l o b b GJ. Covalent structure and aflhiity ol mucosai immune system; from fnndanieiital
1876: Infection and protective responses channel catfish aiiti-dinitropiienyl antibodies, concepts to vacciiie <levelopment.
wilhin the fish host, hi: Pike AW Lewis JW eds. Mollmmunoi i985;22;993 999. Vaccine l992;10;7S-88,
Parasitic diseases of fish. Dyfed, UK; Samara 52, Wilson M, Bengten E, Miller NW Clem IW 60. Jcnkms PG, Wrathmel AH, Harris JE,
Publishing; 1994. p, 17-42. DuPasquier L, Warr GVV A novel chimeric Ig Pulsford AL. Systeiiiic and mncosal responses
44, Clark TG, Dickerson HW Andbody-mediated iieavy chain from a telcost fish shares to enterically derived antigen in Oreociiromis
effects on parasite behavior; evidence of a similarities to IgD.
novel mechanism of immunity against a Proc Natl Acad SCI USA 1997;94;4S93-4597, Fish Shellfish Immiinol 1994.4;25S-27I,
parasitic protist. ')3. Kaatari SL. Fish B lymphocytes; defining their 61, Czorinsky C. Nilsson LA, Nygre 11,
Parasitol Today 1 9 9 7 ; 1 3 ; 4 4 9 ^ 9 0 , form and function. Ouchterlony O, A solid phase eii/yme-hnketl
45, Ewing MS, Kocan KM. khthyophthirius miiitifiliis Annu Rev Fish Dis 1992;2; l 6 l - i 8 0 , immunospot (ELISPOT) assay lor enumeration
(Ciiiophora) development in gill epithelium. 54, Passer BJ, Chen C, Miller NW, Cooper MD. of specific antihody secreting cells.
JProtozool 1986;33:369-374, Identification of a T lhieage antigen in the J hiHiiunol Methods 1983 ;65; 1 09-1 2 I.
46, Clem LW; Bly JE, Ellsaesscr CF, Lobb CJ, catfish, 62. Davidson GA, Elhs M , Secombes CJ. An
Miller KW Chaimet caiflsli as an DevComp Immunol 1996;20;44 1-450, FLISPOT assay for the quantification of specific
unconventional mode! for immunological S S. Vallejo AN, Miller NW, Clem LW Pjiylogeny ol antibody secreting cells to AeromunM Milmoniciild
sttidics, immune recognition processing and in rainbow tront. Oncodiyijchas myki.»
JExpZool 1990; Stippi 4; 123-12,5. presentation of structurally defined protems in (Walbaii];n),
cliannei catilsh immune responses JFish Dis 1992;1S;85"89
Devhnmunol 199 1 ;1; 137-1 48,

384 Immsinolojicdl Reviews 166/1998