lable at ScienceDirect

Fish & Shellfish Immunology 30 (2011) 1152e1158

Contents lists avai

Fish & Shellfish Immunology

journal homepage: www.elsevier .com/locate/ fs i

Elicited cross-protection and specific antibodies in Mozambique tilapia(Oreochromis mossambicus) against two different immobilization serotypesof Cryptocaryon irritans isolated in Hawaii

Ichiro Misumi a,*, Teresa D. Lewis a, Akihiro Takemura b, Jo-Ann C. Leong a

aHawai‘i Institute of Marine Biology, School of Ocean & Earth Science & Technology, University of Hawai‘i, PO Box 1346, K�ane‘ohe, Hawai‘i 96744, USAb Sesoko Station, Tropical Biosphere Research Center, University of the Ryukyus, Motobu, Okinawa 905-0227, Japan

a r t i c l e i n f o

Article history:Received 28 February 2009Received in revised form1 March 2011Accepted 1 March 2011Available online 6 March 2011

Keywords:Cryptocaryon irritansOreochromis mossambicusSerotypeCross-protectionImmobilization

* Corresponding author.E-mail address: imisumi@hawaii.edu (I. Misumi).

1050-4648/$ e see front matter � 2011 Elsevier Ltd.doi:10.1016/j.fsi.2011.03.004

a b s t r a c t

The objective of this study was to determine whether immunization of Mozambique tilapia withdifferent Cryptocaryon irritans i-antigen serotypes elicited cross-protection against challenge infection byboth serotypes. Fish were directly exposed to live theronts of isolate W1 or isolate K1, that expressdifferent surface i-antigens. There was no significant difference in the number of trophonts infecting thefish between the two isolates, W1 and K1, following primary exposure. Serum from immunized fishexposed to live theronts showed higher immobilization titres and ELISA values against homologousisolates than to heterologous isolates after the primary exposure. However, mucus antibody did notimmobilize theronts although the ELISA results clearly indicated that mucus antibodies recognizingC. irritans were generated. In a study with Western blot analyses, serum antibodies recognized only anantigen of the corresponding serotype and no proteins common to both serotypes were identified.Sequence analyses of 754 bases of rDNA nucleotide sequence including complete nuclear ribosomal ITS-1e5.8S rDNAeITS-2 region were conducted and found to be identical for W1- and K1-isolates. Thesefindings confirmed that both isolates were members of the species, C. irritans, and that rDNA analysiswould not distinguish the two isolates. In conclusion, despite the fact that the immobilization assays andELISA detected two serotypes in vitro, challenge assays provided evidence for only one type of C. irritans.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

The parasitic ciliate, Cryptocaryon irritans, infects many speciesof marine fish and causes significant losses in the aquacultureindustry [1]. It can be found in temperate and tropical regionsworldwide [2,3] and thus, can be a major roadblock for marineaquaculture development. Intraspecific variations of C. irritans ingene sequence, morphology, and development have been reported[3e6]. However, there is little information on serotypic variationsbetween isolates of C. irritans.

The life cycle of C. irritans is similar to that of Ichthyophthiriusmultifiliis, a parasite belonging to a taxonomically different phylum[7,8]. I. multifiliis also has a direct life cycle consisting of an infectivetheront, a parasitic trophont and a reproductive tomont [9]. Themechanism of immunization against C. irritans is also similar to thatof I. multifiliis [10]. I. multifiliis expresses abundant glycosyl

All rights reserved.

phosphatidylinositol (GPI)-anchored membrane proteins referredto as immobilization antigens (i-antigens) [11,12]. Different sero-types of I. multifiliis have been identified using antibody-specificimmobilization assays [13]. Intraperitoneal (IP) injection withpurified i-antigens from I. multifiliis elicited serotype-specificprotection and serotype-specific antibodies that immobilized livetheronts in vitro [14,15]. Thus, it is clear that i-antigens themselveshave a direct role in eliciting protective immunity in fish and can bethe principal target antigens for vaccine development. However,when fish were immunized with live I. multifiliis theronts by eitherIP injection or direct exposure, these fish were conferred cross-serotype protection despite development of immobilization anti-bodies against the corresponding serotype [16,17]. These resultsstrongly indicate the existence of other antigens which arecommon to all I. multifiliis and can elicit cross-serotypeprotection [17].

Previous studies have also shown that fish surviving an infectionof C. irritans were protected against reinfection [18e20]. Themechanism of this protection appears to be mediated by serum andmucosal antibodies that recognize and immobilize C. irritans

I. Misumi et al. / Fish & Shellfish Immunology 30 (2011) 1152e1158 1153

[19e22]. Hatanaka et al. [22] isolated the integral membraneprotein expressed on the surface of theronts, and reported evidencethat this membrane protein may be an i-antigen. A more recentstudy by this same group has shown that serotypic variations basedon immobilizations assays exist among C. irritans isolates [23].However, no information was provided that confirmed serotypicdifferences based on challenge infection assays.

We report here the identification of two serotypes of C. irritansby immobilization assays in tilapia in Hawai‘i. Both isolates wereconfirmed as C. irritans by rDNA sequence analysis. We alsoconfirmed the identification of the two serotypes by fish challengeexperiments. Fish were immunized by direct exposure to either ofthe two parasite serotypes and challenged with homologous orheterologous parasite serotype. Immobilization of the parasite andthe specific antibody level were assessed for the fish serum andcutaneous mucus. The objective of this study was to determinewhether the immunization by the different parasite serotypes eli-cited the production of serotype-specific immobilizing antibodiesand cross-protective immune protection.

2. Materials and methods

2.1. Fish

All fish were tagged with T-bar anchor tags inserted at the baseof the dorsal fin and kept in 60-L glass aquaria that received aera-tion and flow-through seawater (120 L/h). All seawater wascontinuously pumped in from Kaneohe Bay and filtered with 5 mmfilters and treated with UV before use in the fish aquaria. The watertemperature was 27 � 1 �C, and the salinity of the seawater was34&. Brood stocks of Mozambique tilapia (Oreochromis mossam-bicus) raised in freshwater were provided by Dr. Gordon Grau at theHawaii Institute of Marine Biology, University of Hawaii. A brace ofthe brood stockwas initially kept in fresh tap water in the SEA lab atHIMB for 2 days and acclimated to seawater by gradually increasingthe salinity up to 34& throughout 2 weeks. All fish were fedapproximately 2% of the body weight per day once daily with SilverCup Trout Chow (Nelson and Sons, Murray, UT). All experimentswere conducted in accordance with the principles and proceduresapproved by the Institutional Animal Care and Use Committee,University of Hawaii.

2.2. Parasite isolation and culture

Two different isolates of C. irritans were used in this study. Thefirst isolate, designated W1, was obtained from an infectedMozambique tilapia kept for two weeks with coral sand collectedfrom a display tank at the Waikiki Aquarium (Honolulu, Hawaii).The second isolate, designated K1, was obtained from an infectedMozambique tilapia kept for one week with bottom depositscollected from a flow-through circular fish tank at the Hawai’iInstitute of Marine Biology, Coconut Island (Kaneohe, Hawaii). Bothisolates were obtained in April, 2008 and passaged separately andcontinuously on naïve Mozambique tilapia in UV treated andfiltered seawater at 27 � 1 �C in 60-L fish aquaria. A naïve tilapiawas added and replaced twice a month for propagation of tomontsin the aquarium. The health of the fish was observed daily and anydead fish or fish with lesions were removed from the tanks. Pro-tomonts were collected from heavily infected fish with 1000 mlmicropipettor under a stereo-microscope (American Optical,Model 41). One hundred protomonts were collected and washedfour times and placed in sterile 120 ml specimen collectioncontainers (Starplex Scientific, Ontario, Canada) with filtered(0.2 mm) seawater and supplementedwith 100 I.U. ml�1 Penicillin Gpotassium and 100 mg/ml streptomycin sulphate [24]. After

confirmation of encystment of the tomont, cultures were incubatedfor 2e3 weeks at 22e23 �C. Newly hatched theronts from theculture were used for DNA extraction, immobilization assay, andfuture propagation, and the remaining theronts were stored insterile phosphate-buffered saline (PBS) at �80 �C until processing.

2.3. Experimental design

A total of 36 tilapia (mean weight 97.0 � 2.5 g/length16.7 � 0.15 cm) were used for this study. They were randomlydivided into two challenge groups. Each challenge of this study wasrun in triplicate (6 fish per tank). Each challenge groupwas exposedto either W1 or K1 strain for the primary exposure. Newly hatchedtheronts in concentrations of 20,000 theronts per fish were trans-ferred to each aquarium. Two days after the initial exposure, thenumber of infected trophonts on the left pectoral fin was countedunder an inverted microscope (Olympus IX; Olympus Optical,Tokyo) in order to measure the infection level for each fish. All fishwere transferred into new, clean tanks every 5 days to avoid acci-dental reinfection by C. irritans. Fifteen days after primary exposure,each challenge group was subdivided into two subgroups. Onesubgroup from each challenge group was mixed together in onetank. These mixed groups were subsequently exposed to W1 or K1strain (20,000 theronts per fish) for the secondary challenge study.As a result, depending on the order of exposures, 4 different chal-lenge groups were formed (designated as fish exposed to W1W1,W1K1, K1W1, and K1K1), comprising 9 individuals each. Two daysafter the secondary exposure, the number of infected trophonts onthe right pectoral fin was counted. And again, all fish were trans-ferred into new clean tanks every 5 days to avoid accidental rein-fection by C. irritans.

2.4. Sampling mucus and serum

Cutaneous mucus and blood were collected three times totalthroughout this study from all individual tilapia 2 days prior toinitial and secondary theront exposures and 14 days after thesecondary exposure. Fish were sampled at random from eachchallenge group and were anaesthetized by 0.2 ppm of tricainemethanesulfonate (MS 222: Finquel, Redmond, WA).Mucus wassampled from the anaesthetized fish by lightly swabbing one side ofthe fish 10 times from head to tail except fins with a sterile cottonswab. The cotton part of the swab was cut and placed in a 1.5 mlmicrocentrifuge tube containing 100 ml of PBS. The tube wasvigorously vortexed for 2 min to elute the mucus, and excess liquidwas removed from the swab by pressing against the side of thetube. The resulting liquid was centrifuged at 1000 g for 10 min, andthe collected supernatant was distributed into aliquots of 20 ml in200 ml microcentrifuge tubes and stored at �20 �C. Followingmucus sampling, blood (200 ml) was withdrawn from caudal vein ofthe anaesthetized fish using a 1 ml syringe with a 23 gauge needleand transferred into a 1.5 ml microcentrifuge tube. Blood sampleswere allowed to clot overnight at 4 �C. Serum was obtained bycentrifugation of blood at 1000 g for 10 min. Serum samples weredistributed in aliquots of 20 ml in 200 ml microcentrifuge tubes andstored at �20 �C for later analysis. In this study, the serum andmucus collected from 4 different challenge groups (W1W1, W1K1,K1W1, and K1K1) following the secondary exposures were desig-nated as W1W1-, W1K1-, K1W1-, and K1K1-antiserum or mucusantibody.

2.5. Immobilization assay

Blood serum and cutaneous mucus from individual fish weretested for immobilizing antibodies against both isolatesW1 and K1.

I. Misumi et al. / Fish & Shellfish Immunology 30 (2011) 1152e11581154

Assays were done as described by Clark et al. [25] with modifica-tions. Serum and mucus were heat inactivated at 56 �C for 30 min.Serum or mucus samples from individual fish were serially dilutedwith filtered (0.2 mm) seawater and 50 ml from each dilution wereadded into each well of a 96-well plate (Corning Costar, Corning,NY) containing 50 ml of theront suspension (approximately 200theronts per well) in filtered seawater. After 30 min incubation atroom temperature, the number of immobilized theronts wascounted under the inverted microscope (Olympus IX; OlympusOptical, Tokyo) to find the percentage that was no longer motile.The immobilization titre was defined as the reciprocal value of thedilution at which 95% of the parasites were unable to move.

2.6. Enzyme-linked immunosorbent assay (ELISA)

ELISA was conducted to assess the presence of specific anti-bodies in the serum and the cutaneous mucus against both isolates,W1 or K1. Sonicated C. irritans theront homogenates were dilutedto 5 mg/ml in carbonate coating buffer (0.05 M Na2CO3 and 0.05 MNaHCO3, pH 9.6), and 100 ml of the homogenate was added intoeachwell of a 96-well ELISA plate (Corning Costar, Corning, NY) andincubated overnight at 4 �C. All subsequent steps were performedat room temperature. Wells were washed 3 times with low saltwash buffer (20 mM Tris, pH 7.3; 380 mM NaCl; 0.05% Tween 20).Non-specific binding sites were blocked with PBS (250 ml/well)containing 1% (w/v) bovine serum albumin (BSA) for 2 h at 22 �C.After 3washes with low salt wash buffer, diluted fish serum (1:500)or mucus (1:16) in PBS were added to the plate at 100 ml/well andincubated for 3 h. PBS alone served as the control. The plate waswashed 5 times with high salt wash buffer (20 mM Tris, pH 7.3;500 mM NaCl; 0.1% Tween 20). At sequential 1 h intervals, wellswere incubated with a rabbit anti-tilapia immunoglobulin (diluted1:2500; a generous gift of Dr. Akihiro Takemura, University ofRyukyu, Okinawa, Japan) and commercial goat anti-rabbit IgGhorseradish peroxidase conjugate (diluted 1:2500; JacksonImmunoResearch Laboratories, Inc.,West Grove, PA). Then,100 ml of2,20-azino-bis 3-ethylbenzthiazoline-6-sulfonic acid (ABTS) perox-idase substrate (KPL, Gaithersburg, MD) was added to each well toallow for chromogenic development. Between additions of eachnew reagent, the ELISA plate was washed 5 times with high saltwash buffer. After 30min, the absorbance at 405 nmwas read usinga microplate reader (Molecular Devices Co., Sunnyvale, CA). In thisstudy, ELISA results were reported as optical density (OD) values ofgiven optimal dilutions of serum and mucus due to the largenumber of samples. Optimal dilutions were determined by thepreliminary ELISA with serial dilutions (twofold) of serum andmucus samples collected from naïve and hyper-immunized fish.The optimal dilutions allowed for maximum OD for positivecontrols with minimal back ground reading in negative controls.

2.7. Western blot analyses

Thepooled sera from4different challengegroups (W1W1,W1K1,K1W1, and K1K1) following the secondary exposures were analyzedbyWestern blot analysis for reactivity against any antigenic proteinsin the homogenates of C. irritans theronts. Briefly, sonicated therontproteins (0.5 mg per lane)were resolved on 10% SDS-PAGE gels undernon-reducing conditions and transferred to a polyvinylidenedifluoride (PVDF) membrane using a Mini-Protean 3 transblotapparatus (BioRad, California, USA). The Precision Plus proteinstandards (dual color) were used as molecular weight markers(BioRad). Equal protein loading was confirmed in a duplicated gelstained with silver stain. The PVDF membranes were blocked withTween-PBS containing 1% (w/v) BSA overnight for 1 h at roomtemperature. The membranes were washed three times with

Tween-PBS. Fish sera (diluted 1/75 in PBS), rabbit anti-tilapia IgMantibody (1:3000), and goat anti-rabbit IgG horseradish peroxidaseconjugate (1:5000; Jackson ImmunoResearch Laboratories) wereapplied in sequential order to the nitrocellulose membranes andincubated for 1 h at room temperature. The serawere pooledwithineachgroup (9fisheach) andwere analyzed. Pooled serumfrom9fishthat had never been exposed to C. irritans was used as negativecontrol. Immunoreactive bands were visualized with West Durachemiluminescent substrate (Pierce, Rockford, IL).

2.8. DNA isolation, amplification, and sequence

Intraspecific variation between two isolates, W1 and K1, wasexamined at the level of rDNA sequence analysis. DNA extractionfrom newly hatched theronts was performed with the QiagenDNeasy kit (Qiagen, California, USA) following the manufacturer’sprotocol. The polymerase chain reaction (PCR) was used to amplifythe rDNA fragment corresponding to the 30 portion of the 18S, theITS-1, the 5.8S, the ITS-2, and the 50 portion of the 28S using primersP1 (forward 50-GTTCCCCTTGAACGAGGAATTC-30) [6,23] and NC2(reverse 50-TTAGTTTCTTTTCCTCCGCT-30) [6,26]. PCR amplificationwas carried out in a final volume of 50 ml containing: 100 ng oftemplate, 2.5 U Taq-Polymerase (Bioline, Massachusetts, USA);0.3 ml 100 mMof each primer; 10� PCR buffer; 4 ml of 10 mM dNTPsin a thermocycler iCyclerTM (BioRad). PCR was performed at 94 �Cfor 5 min; 35 cycles at 94 �C for 30 s, 53 �C for 30 s, and 72 �C for90 s; followed by a final extension at 72 �C for 5 min [6]. Theamplified product was observed under UV light after ethidiumbromide staining of a 1.5% agarose gel. Sequencing reactions werecarried out using ABI BigDye reagents and the manufacturer’sprotocol and automated dideoxy sequencing was carried out on anABI 3130XL at the HIMB molecular core facility. Partial sequences(part of 18S, entire ITS-1, and part of 5.8S) of amplified regions werealigned with representative C. irritans sequences from GenBankwith BLAST (http://www.ncbi.nlm.nih.gov/BLAST/), and sequencesimilarities were calculated based on pairwise comparisons usingMacVector v. 8.1 software (Accelrys Inc., California, USA). Thefollowing accession numbers and sources were representatives ofC. irritans strains which were obtained from GenBank andcompared to nucleotide differences with isolates, W1 and K1:AF490382 (Penghu, Taiwan isolate), AF490383 (Kaoshiung, Taiwanisolate), AF490385 (Malaysia isolate), AY029269 (Israel), AY029270(USA), AY029273 (Australia), DQ270008 (China), DQ270010(China), AB381933 (Japan), AB381934 (Japan).

2.9. Statistical analysis

Data are expressed as mean � S.E.M. (standard error of mean).When the normality test indicated a normal distribution, para-metric statistical test (the Student’s t-test and ANOVA) was used forstatistical comparisons of data. Otherwise, non-parametric statis-tical test (the ManneWhitney U-test and the KruskaleWallace test)was used. StudenteNewmaneKeuls multiple range test was usedto determine which treatments were significantly different fromothers. Statistical analyses were performed using Sigma Plot 11software (Systat Software Inc., San Jose, CA). Measurements wereconsidered significant when P values of both parametric and non-parametric tests were below 0.05.

3. Results

3.1. Experimental infection

Challenge experiments by direct exposures with live therontswere carried out to compare infectivity between W1- and

Table 2Antibody responses 2 days before the primary exposure. Immobilization (IMM)titres represent the reciprocal value of the dilution at which 95% of the parasiteswere unable to move. ELISA results were reported as OD values of given optimaldilutions of serum (1:500) and mucus (1:16). The values are mean (S.E.M.) for eachexperimental group.

Experimental group n Utilizedisolate

IMM assay ELISA

Serum Mucus Serum Mucus

a) Assays against W1Naïve 1 18 W1 0.0 (0.0) 0.0 (0.0) 0.24 (0.01) 0.28 (0.01)Naïve 2 18 W1 0.0 (0.0) 0.0 (0.0) 0.25 (0.01) 0.26 (0.01)

b) Assays against K1Naïve 1 18 K1 0.0 (0.0) 0.0 (0.0) 0.26 (0.01) 0.27 (0.01)

I. Misumi et al. / Fish & Shellfish Immunology 30 (2011) 1152e1158 1155

K1-isolates and to determine if immunized fish showed protectionagainst the secondary infection with either isolate. There was nosignificant difference in the number of trophonts infecting the fishbetween the two isolates, W1 and K1, following the primaryexposure (Table 1a). When fish were primarily exposed to andrecovered from W1-isolates, those W1-immunized fish showedcomplete protection against both homologous (W1) and heterolo-gous (K1) isolates following the secondary exposure (Table 1b).However, K1-immunized fish showed protection only againsthomologous (K1) isolates on the secondary exposure, although theinfection with heterologous (W1) isolate following the secondaryexposure was milder than the primary infection.

Naïve 2 18 K1 0.0 (0.0) 0.0 (0.0) 0.27 (0.01) 0.26 (0.01)

3.2. Antibody responses prior to primary exposure

Serum and mucus antibody responses against two isolates fornaïve fish weremeasured by immobilization assay and ELISA. Therewas no statistical difference between two groups, naïve 1 and naïve2 in immobilization assay and ELISA (Table 2). Sera andmucus fromnaïve tilapia before infection with C. irritans did not immobilizeW1- or K1-theronts. The sera andmucus of naïve tilapia had ameanantibody absorbance values of 0.24 � 0.01e0.28 � 0.01 (opticaldensity at 405 nm).

3.3. Antibody responses after primary exposure

Serum and mucus samples were collected 13 days after theprimary exposure (Table 3). Serum from immunized fish exposed tolive theronts showed higher immobilization titres and ELISA valuesagainst homologous isolates than heterologous isolates after theprimary exposure. Antisera from fish that recovered from infectionwith W1-isolate immobilized only W1-theronts, and antisera fromfish that recovered from the infection of K1-isolates immobilizedonly K1-theronts. The mean immobilization titre against W1-theronts (129.6 � 25.6) was significantly higher than that againstK1-theronts (40 � 6.0) for antisera from fish immunized withhomologous isolates. Fish serum antibody levels measured by ELISAcorrelated with this activity. ELISA values of serum samples weresignificantly higher for fish immunized with homologous isolatesthan heterologous isolates. Compared to naïve fish serum as shownin Table 2, ELISA values for all groups increased significantly afterthe primary exposure except for K1-immunized fish serum againstthe W1-isolate.

All mucus samples did not immobilize any live theronts.However, the ELISA value of mucus from W1-immunized fishagainst its homologous isolate was significantly higher than those

Table 1Comparison of the number of infected trophonts in a pectoral fin following theprimary and the secondary exposures to either two different C. irritans isolates, W1or K1. Fish were re-exposed to theronts 15 days after the primary exposure.Trophonts were counted 2 days after each exposure. The number of trophonts is themean number (S.E.M.) for each experimental group. Fish were exposed by bathimmersion to live theronts at the dose of 20,000 theronts per fish. Mean valuesfollowed by the same superscript letters are not statistically different (Stu-denteNewmaneKeuls test, P > 0.05).

Experimental group # of fish Challenge strain # of trophonts (S.E.M.)

a) Primary exposureNaïve 1 18 W1 112.3 (8.4)Naïve 2 18 K1 103.3 (6.1)

b) Secondary exposure (15 days after the primary exposure)W1 immune 9 W1 0.0 (0.0)a

K1 immune 9 W1 21.3 (5.1)b

W1 immune 9 K1 0.2 (0.2)a

K1 immune 9 K1 0.0 (0.0)a

of other experimental groups. Compared to naïve fish mucus inTable 2, ELISA values for all groups increased significantly after theprimary exposure.

3.4. Antibody responses after secondary exposure

Serum and mucus samples were collected 14 days after thesecondary exposure (29 days after the primary exposure). In theimmobilization assay using W1-isolate (Table 4a), W1-therontswere immobilized by antisera only from fish exposed at least onceto the W1-theronts at the primary and secondary exposure.However, those W1-theronts were not immobilized by antiserafrom K1K1-immunized fish. There was no significant differencebetweenW1W1- (81.5�14.0) andW1K1- (59.7� 14.1) immunizedfish in the mean immobilization titre against W1-theronts, butthose titres were significantly higher than that for K1W1-immu-nized fish (6.9� 1.9). Fish serum antibody levels measured by ELISAcorrelated with this activity. There was no statistically significantdifference between W1W1- (1.35 � 0.11) and W1K1- (1.33 � 0.10)immunized fish in the mean ELISA value, but those values weresignificantly higher than that for K1W1-immunized fish(0.93 � 0.11). Fish exposed twice to K1 trophonts showed thelowest level of serum anti-W1 antibodies and the same level asnaïve fish. Similar results were obtained with ELISA measuring fishmucus antibody levels against the homogenate of the W1-isolate.Fish immunized twice with W1 showed higher levels of cutaneousanti-W1 antibodies than fish immunized once with W1 or twicewith K1. Levels of cutaneous anti-W1 antibodies of K1K1-immu-nized fish against the W1-isolate measured by ELISA were signifi-cantly lower than other challenge groups and at the same level asnaïve fish.

In the immobilization assay using K1-isolates (Table 4b), K1-theronts were immobilized only by antisera from fish exposed withthe K1-theronts on primary exposure. K1K1-immunized fish thatwere exposed twice with K1-theronts showed significantly higherimmobilization titres against K1-theronts than the fish exposedonce with K1-theronts. In the ELISA, both fish groups, K1W1 orK1K1 immune, showed higher levels of serum anti-K1 antibodiesthan other groups. There was no statistically significant differencebetween W1W1- (0.35 � 0.05) and W1K1- (0.40 � 0.05) immu-nized fish in themean ELISAvalue for anti-K1 serum antibodies, butthose values were significantly higher than the values for naïve fish.The ELISA using mucus antibody detected the highest level ofcutaneous anti-K1 antibodies from K1K1-immunized fish.However, there was no statistically significant difference in thelevel of cutaneous anti-K1 antibodies between fish exposed at leastonce to the K1-isolate and fish receiving either one or two treat-ments with K1-theronts. And fish exposed twice to the W1-isolate.

Table 3Antibody responses 13 days after the primary exposure. IMM titres represent the reciprocal value of the dilution at which 95% of the parasites were unable to move. ELISAresults were reported as OD values of given optimal dilutions of serum (1:500) and mucus (1:16). The values are mean (S.E.M.) for each experimental group. Significantdifferences are denoted * (P < 0.05, ManneWhitney U-test).

Experimental group (day 13) n Utilized isolate IMM assay ELISA

Serum Mucus Serum Mucus

a) Assays against W1W1 immune 18 W1 129.6 (25.6)* 0.0 (0.0) 1.31 (0.05)* 0.60 (0.05)*K1 immune 18 W1 0.0 (0.0) 0.0 (0.0) 0.27 (0.01) 0.30 (0.01)

b) Assays against K1W1 immune 18 K1 0.0 (0.0) 0.0 (0.0) 0.40 (0.02) 0.32 (0.01)K1 immune 18 K1 40.3 (6.0)* 0.0 (0.0) 1.50 (0.0)* 0.38 (0.04)

I. Misumi et al. / Fish & Shellfish Immunology 30 (2011) 1152e11581156

Nomucus samples immobilized any live theronts after primary andsecondary exposure.

3.5. SDS-PAGE and Western blotting analyses

Following silver staining, a protein profile for total therontproteins revealed a heterogeneous mixture of proteins withdifferent molecular weights and the dominant protein band wasapproximately 28 kDa for the W1-isolate and 29 kDa for the K1-isolate (Fig. 1). The pooled sera from 4 different challenge groups(W1W1,W1K1, K1W1 and K1K1) were used to probeWestern blotsof whole cell proteins from W1 or K1-theronts (Fig. 2). W1W1-antiserum antibodies bound W1 antigen (28 kDa), but they did notrecognize K1 antigen (29 kDa). Likewise, K1K1-antiserum anti-bodies recognized only K1 antigen and did not bind W1 antigen.While the K1W1-antiserum reacted with both antigens of the W1-and K1-isolates, W1K1-antiserum recognized only the antigen ofthe W1-isolate. No signals were detected when the proteins wereprobed with pooled serum from naïve fish (data not shown).

3.6. Sequence analysis

A total of 754 bases of rDNA nucleotide sequence comprising thecomplete nuclear ribosomal ITS-1e5.8S rDNAeITS-2 were used inthe sequence comparisons. Partial sequences (part of 18S, the entireITS-1, and part of the 5.8S segments; 291 bp) of the sequencedrDNA were aligned and compared with available sequences fromother isolates of C. irritans in GenBank with BLAST (Table 5). Bothisolates W1 and K1 had 100% identical sequences to isolates ofIsrael (AY029269) and Malaysia (AF490385).

Table 4Antibody responses 14 days after the secondary exposure (29 days after the primaryexposure). IMM titres represent the reciprocal value of the dilution at which 95% ofthe parasites were unable tomove. ELISA results were reported as OD values of givenoptimal dilutions of serum (1:500) and mucus (1:16). The values are mean (S.E.M.)for each experimental group. Mean values followed by the same superscript lettersare not statistically different (StudenteNewmaneKeuls test, P > 0.05).

Experimentalgroup (day 29)

n Utilizedisolate

IMM assay ELISA

Serum Mucus Serum Mucus

a) Assays against W1W1W1 immune 9 W1 81.5 (14.0)c 0.0 (0.0)a 1.35 (0.11)c 1.23 (0.12)d

W1K1 immune 9 W1 59.7 (14.1)c 0.0 (0.0)a 1.33 (0.10)c 0.80 (0.18)c

K1W1 immune 9 W1 6.9 (1.9)b 0.0 (0.0)a 0.93 (0.11)b 0.45 (0.03)b

K1K1 immune 9 W1 0.0 (0.0)a 0.0 (0.0)a 0.26 (0.01)a 0.26 (0.01)a

b) Assays against K1W1W1 immune 9 K1 0.0 (0.0)a 0.0 (0.0)a 0.35 (0.05)a 0.36 (0.05)a

W1K1 immune 9 K1 0.0 (0.0)a 0.0 (0.0)a 0.40 (0.05)a 0.48 (0.12)a

K1W1 immune 9 K1 31.7 (10.2)b 0.0 (0.0)a 1.08 (0.14)b 0.43 (0.04)a

K1K1 immune 9 K1 100 (16.4)c 0.0 (0.0)a 1.32 (0.12)b 0.68 (0.07)b

4. Discussion

Immunization with one immobilization serotype of C. irritansprovided cross-protection against challenges by both homologousand heterologous serotypes. Thus, despite the fact that the immo-bilization assays detected two serotypes in vitro, challenge assaysdetected only one type of C. irritans. In the case of I. multifiliis,antibodies recognized not only epitopes on i-antigens but alsoepitopes on other cell surface antigens that were common totheronts of all serotypes [17]. In the study reported here, the anti-sera and mucus antibodies from immune fish reacted with wholeproteins of heterologous theronts in the ELISA assay, giving higherOD readings than serum and mucus from pre-exposed naïve fish.These results suggest that antisera and mucus antibody fromimmune fish also reacted with non-i-antigens of C. irritans in theELISA. Binding of those antibodies to other cell surface proteinsbesides i-antigen could trigger the forced premature exit ofinvading theronts from its host [27,28]. Cross and Matthews [27]reported that I. multifiliis could invade epithelium of immune fish,but 79% of them disappeared within 2 h of penetration with notrace at sites of initial infection.

We show here that after two consecutive immunizations withtwo different serotypes, the antibody response to the primaryimmunization serotype was always higher than the response to thesecondary immunization serotype. Compared to K1W1-immunized

Fig. 1. Silver stained SDS-PAGE. Lane 1: molecular weight standards; lane 2: sonicatedtotal proteins of W1-isolate; lane 3: sonicated total proteins of K1-isolate.

Fig. 2. Western blots probed with pooled sera from immunized fish following the secondary direct exposure with C. irritans theronts. Each lane was loaded with molecular weightstandards (1); total proteins of W1-isolate (2); or total proteins of K1-isolate (3). Antigen proteins were probed with W1W1- (A); W1K1- (B); K1W1- (C); or K1K1- (D) antiserum.

I. Misumi et al. / Fish & Shellfish Immunology 30 (2011) 1152e1158 1157

fish, W1K1-immunized fish did not produce as much specificantibody against the K1-isolate. Similarly, compared to W1K1-immunized fish, K1W1-immunized fish produced less specificantibody against W1-isolate. We hypothesize that K1-therontscould be forced to exit from W1-immunized fish before the K1-theronts could induce an immune response. On the other hand,a small number (6.9 � 1.9 theronts) of the W1-isolate could invadeK1-immunized fish as shown in Table 1b., so K1W1-immunized fishdeveloped immobilization antibody against W1-isolate eventhough the immobilization titre was much less than that of W1K1-immunized fish.

Antiserum collected from immune fish contained antibodiesagainst C. irritans and caused immobilization of theronts in vitro.However, mucus antibody did not immobilize theronts althoughthe ELISA results clearly indicated that mucus antibodies recog-nizing C. irritans were generated. This finding agrees with previousstudies on I. multifiliis and C. irritans [20,29,30]. Yambot and Song[20] suggested that the concentration of antibodies was too low toimmobilize the theronts in vitro. However, the ELISA was sensitiveenough to detect the antibody in the mucus.

In our study with Western blot analyses, serum antibodiesrecognized only an antigen of the corresponding serotype and noproteins common to both serotypes were identified. Hatanaka et al.[22] identified a 32 kDa polypeptide on the surface of the cilia oftheronts as an i-antigen using a C. irritans isolate designated as G32.In subsequent studies, fish antiserum against 37 kDa i-antigenpurified from another C. irritans isolate designated as G 37 werefound not to immobilize G32, indicating that multiple i-antigenserotypes exist among C. irritans isolates [23]. Similarly, ourWestern

Table 5Sequence similarity (%) between geographically different isolates of C. irritans in partcomparisons of 291 aligned base pairs (bp).

Segment sequence similarity (%, 291 bp)

W1 K1 AF490385 AY029269 AB381933 AY029270W1K1 100AF490385 100 100AY029269 100 100 100AB381933 99.6 99.6 99.6 99.6AY029270 99.6 99.6 99.6 99.6 100AB381934 98.1 98.1 98.1 98.1 97.8 97.8DQ270010 97 97 97 97 96.7 96.7AY029273 96.7 96.7 96.7 96.7 96.3 96.3DQ270008 96.3 96.3 96.3 96.3 95.9 95.9AF490382 94.8 94.8 94.8 94.8 94.4 94.4AF490383 92.9 92.9 92.9 92.9 92.6 92.6

blotting analyses showed that the immunoreactive protein ofC. irritans could be the protein with a molecular weight of approx-imately 28 kDa for W1 and 29 kDa for K1. Those proteins were alsomost abundant on the silver staining analysis. W1W1-antiserumantibodies strongly bound W1 antigen (28 kDa) from theronts, butthey did not recognize the K1 antigen (29 kDa). Likewise, K1K1-antiserum antibodies strongly recognized only the K1 antigen anddid not recognize the W1 antigen. These results indicate that it ishighly likely that the 28 kDa and 29 kDa proteins from theW1- andK1-isolates, respectively, are the immobilization antigen (i-antigen)of C. irritans. The other proteins visualized by silver staining werenot recognized by fish immune antisera in Western blots. However,ELISA assays suggested that antisera from immunized fish con-tained antibodies to both non-i-antigens and i-antigens.We suggestthat the difference is due to the fact that the antisera titre for thenon-i-antigens was too low for Western blot detection, a findingsimilar to that of Swennes et al. [17] for I multifiliis.

Most studies on the protection mechanism against C. irritanshave been focused on the antibody responses [20,31]. However, thepotential involvement of innate immune responses during the earlystages of infection following the invasion of the theronts shouldalso be considered because innate immunity is the first line ofdefense against many pathogens [32]. Colorni et al. [33] reportedthat piscidin 2, an antimicrobial polypeptide, isolated from mastcells of hybrid striped bass was lethal to C. irritans theronts. It is notyet clear if other humoral factors such as complement, lysozymeand lectins play a role in the cross-protection. Further studies arenecessary to describe the innate immune responses againstC. irritans.

of 18S, entire ITS-1, and part of 5.8S segments. The matrix is based on pairwise

AB381934 DQ270010 AY029273 DQ270008 AF490382 AF490383

96.796.3 99.695.9 99.3 99.694.4 97.4 97.8 97.492.6 95.5 95.9 95.5 98.1

I. Misumi et al. / Fish & Shellfish Immunology 30 (2011) 1152e11581158

We performed sequence similarity searches using the rDNAsequence of different isolates of C. irritans in order to determinewhether the intraspecific serotypic variation was associated withrDNA sequence difference, and to determine the phylogeneticrelationships ofW1, K1, and other isolates fromdifferent geographiclocations. The non-coding ribosomal ITS-1 and -2 regions wereamplified in this study since they have been shown to exhibit highrates of evolution in protozoa and have been use to evaluate speciesrelationships for many taxa [34,35]. The extent of intraspecificvariation in C. irritans has been reported based on sequences of ITS-1 and -2 as well as on developmental and morphological characters[3e6]. Our analysis of sequence similarities confirmed that bothW1- and K1-isolates belonged to the C. irritans species. Moreover,the sequence analysis showed that the 754 bases of rDNA nucleo-tide sequence including complete nuclear ribosomal ITS-1e5.8SrDNAeITS-2 region were identical for W1- and K1-isolates. Hata-naka et al. [23] reported the existence of serotypic heterogeneity intwo different isolates of C. irritans, that had different ITS sequences(only 97.8% identity), so they suggested that this relationship couldbe exploited to identify different serotypes in the development offish vaccines. Our results indicate that there is no relationshipbetween serotypic variation and the ITS sequence and reliance onITS sequence analysis is not recommended.

We report here the first observation that immunization withdifferent C. iritans i-antigen serotypes elicited cross-protectionagainst both serotypes. Fish directly exposed to the correspondinglive theronts of C. irritans produced antiserum that immobilizedonly the homologous serotype. However, the immunization resul-ted in cross-protection against subsequent challenges by bothhomologous and heterologous serotypes. The result of cross-protection correlatedwith the serum antibody responses measuredby ELISA. These results suggest the presence of antigens that arecommon to all C. irritans and are involved in eliciting the cross-serotype protection, and i-antigens may not be involved in thatprotection. Therefore, i-antigens themselves may not be a goodtarget for the development of subunit vaccines for C. irritans. Sincean optimal vaccine must provide cross-protection against all sero-types, the identification of other antigen proteins that are commonin all serotypes and can elicit protective antibodies is needed forfuture vaccine development.

Acknowledgments

This research was supported by a grant from the Center forTropical and Subtropical Aquaculture #2002-38500-12039 (USDACREES) to Teresa Lewis and a grant from the National ScienceFoundation #EPS02-37065 to Jo-Ann Leong.

References

[1] Rigos G, Pavlidis M, Divanach P. Host susceptibility to Cryptocaryon sp.infection of Mediterranean marine broodfish held under intensive cultureconditions: a case report. Bull Eur Assoc Fish Pathol 2001;21:33e6.

[2] Colorni A, Burgess P. Cryptocaryon irritans brown 1951, the cause of white spotdisease in marine fish: an update. Aquar Sci Conserv 1997;1:217e38.

[3] Yambot AV, Song YL, Sung HH. Characterization of Cryptocaryon irritans,a parasite isolated from marine fishes in Taiwan. Dis Aquat Org2003;54:147e56.

[4] Diggles BK, Lester RJG. Variation in the development of two isolates of Cryp-tocaryon irritans. J Parasitol 1996;82:384e8.

[5] Diggles BK, Adlard RD. Intraspecific variation in Cryptocaryon irritans.J Eukaryot Microbiol 1997;44:25e32.

[6] Sun HY, Zhu XQ, Xie MQ, Wu XY, Li AX, Lin RQ, et al. Characterization ofCryptocaryon irritans isolates from marine fishes in mainland china by itsribosomal DNA sequences. Parasitol Res 2006;99:160e6.

[7] Diggles BK, Adlard RD. Taxonomic affinities of Cryptocaryon irritans and Ich-thyophthirius multifiliis inferred from ribosomal-RNA sequence data. Dis AquatOrgan 1995;22:39e43.

[8] Wright ADG, Colorni A. Taxonomic re-assignment of Cryptocaryon irritans,a marine fish parasite. Eur J Protistol 2002;37:375e8.

[9] Colorni A, Diamant A. Ultrastructural features of Cryptocaryon irritans, a ciliateparasite of marine fish. Eur J Protistol 1993;29:425e34.

[10] Matthews RA. Ichthyophthirius multifiliis fouquet and Ichthyophthiriosis infreshwater teleosts. Adv Parasitol 2005;59:159e241.

[11] Dickerson HW, Clark TG, Findly RC. Ichthyophthirius multifiliis has membrane-associated immobilization antigens. J Protozool 1989;36:159e64.

[12] Clark TG, Gao Y, Gaertig J, Wang XT, Cheng G. The i-antigens of Ichthyophthiriusmultifiliis are GPI-anchored proteins. J Eukaryot Microbiol 2001;48:332e7.

[13] Dickerson HW, Clark TG, Leff AA. Serotypic variation among isolates of Ich-thyophthirius multifiliis based on immobilization. J Eukaryot Microbiol 1993;40:816e20.

[14] Wang XT, Dickerson HW. Surface immobilization antigen of the parasiticciliate Ichthyophthirius multifiliis elicits protective immunity in channel catfish(Ictalurus punctatus). Clin Diagn Lab Immunol 2002;9:176e81.

[15] Wang XT, Clark TG, Noe J, Dickerson HW. Immunisation of channel catfish,Ictalurus punctatus, with Ichthyophthirius multifiliis immobilisation antigenselicits serotype-specific protection. Fish Shellfish Immunol 2002;13:337e50.

[16] Leff AA, Yoshinaga T, Dickerson HW. Cross immunity in channel catfish,Ictalurus punctatus (rafinesque), against 2 immobilization serotypes of Ich-thyophthirius multifiliis (fouquet). J Fish Dis 1994;17:429e32.

[17] Swennes AG, Findly RC, Dickerson HW. Cross-immunity and antibodyresponses to different immobilisation serotypes of Ichthyophthirius multifiliis.Fish Shellfish Immunol 2007;22:589e97.

[18] Burgess PJ, Matthews RA. Cryptocaryon irritans (ciliophora) - acquiredprotective immunity in the thick-lipped mullet, Chelon labrosus. Fish ShellfishImmunol 1995;5:459e68.

[19] Yoshinaga T, Nakazoe J. Acquired protection and production of immobilizationantibody against Cryptocaryon irritans (Ciliophora, Hymenostomatida) inmummichog (Fundulus heteroclitus). Fish Pathol 1997;32:229e30.

[20] Yambot AV, Song YL. Immunization of grouper, Epinephelus coioides, confersprotection against a protozoan parasite, Cryptocaryon irritans. Aquaculture2006;260:1e9.

[21] Luo XC, Xie MQ, Zhu XQ, Li AX. Protective immunity in grouper (Epinepheluscoioides) following exposure to or injection with Cryptocaryon irritans. FishShellfish Immunol 2007;22:427e32.

[22] Hatanaka A, Umeda N, Yamashita S, Hirazawa N. Identification and charac-terization of a putative agglutination/immobilization antigen on the surface ofCryptocaryon irritans. Parasitology 2007;134:1163e74.

[23] Hatanaka A,UmedaN,HirazawaN.Molecular cloning of a putative agglutination/immobilization antigen located on the surface of a novel agglutination/immobi-lization serotype of Cryptocaryon irritans. Parasitology 2008;135:1043e52.

[24] Yoshinaga T, Akiyama K, Nishida S, Nakane M, Ogawa K, Hirose H. In vitroculture technique for Cryptocaryon irritans, a parasitic ciliate of marine tele-osts. Dis Aquat Organ 2007;78:155e60.

[25] Clark TG, Dickerson HW, Gratzek JB, Findly RC. Invitro response of Ich-thyophthirius multifiliis to sera from immune channel catfish. J Fish Biol 1987;31:203e8.

[26] Chen W, Sun HY, Xie MQ, Bai JS, Zhu XQ, Li AX. Development of specific PCRassays for the detection of Cryptocaryon irritans. Parasitol Res 2008;103:423e7.

[27] Cross ML, Matthews RA. Ichthyophthiriasis in carp, Cyprinus carpio L.: fate ofparasites in immunized fish. J Fish Dis 1992;15:497e505.

[28] Clark TG, Lin TL, Dickerson HW. Surface antigen cross-linking triggers forcedexit of a protozoan parasite from its host. Proc Natl Acad Sci 1996;93:6825e9.

[29] Maki JL, Dickerson HW. Systemic and cutaneous mucus antibody responses ofchannel catfish immunized against the protozoan parasite Ichthyophthiriusmultifiliis. Clin Diagn Lab Immunol 2003;10:876e81.

[30] Xu DH, Klesius PH, Shelby RA. Immune responses and host protection ofchannel catfish, Ictalurus punctatus (rafinesque), against Ichthyophthiriusmultifiliis after immunization with live theronts and sonicated trophonts.J Fish Dis 2004;27:135e41.

[31] Bai JS, Xie MQ, Zhu XQ, Dan XM, Li AX. Comparative studies on the immu-nogenicity of theronts, tomonts and trophonts of Cryptocaryon irritans ingrouper. Parasitol Res 2008;102:307e13.

[32] Magnadottir B. Innate immunity of fish (overview). Fish Shellfish Immunol2006;20:137e51.

[33] Colorni A, Ullal A, Heinisch G, Noga EJ. Activity of the antimicrobial poly-peptide piscidin 2 against fish ectoparasites. J Fish Dis 2008;31:423e32.

[34] Schlegel M. Phylogeny of eukaryotes recovered with molecular data: high-lights and pitfalls. Eur J Protistol 2003;39:113e22.

[35] DeJonckheere JF. Molecular definition and the ubiquity of species in the genusNaegleria. Protist 2004;155:89e103.