Mucosal Igm Antibody with D-Mannose Affinity in Fugu Takifugu

Home , Takifugu, Takifugu rubripes

Mucosal IgM Antibody with d-Mannose Affinity in Fugu Takifugu rubripes Is Utilized by a Monogenean Parasite Heterobothrium okamotoi for Host Recognition This information is current as of October 5, 2021. Kento Igarashi, Ryohei Matsunaga, Sachi Hirakawa, Sho Hosoya, Hiroaki Suetake, Kiyoshi Kikuchi, Yuzuru Suzuki, Osamu Nakamura, Toshiaki Miyadai, Satoshi Tasumi and Shigeyuki Tsutsui

J Immunol 2017; 198:4107-4114; Prepublished online 12 Downloaded from April 2017; doi: 10.4049/jimmunol.1601996 http://www.jimmunol.org/content/198/10/4107 http://www.jimmunol.org/ Supplementary http://www.jimmunol.org/content/suppl/2017/04/12/jimmunol.160199 Material 6.DCSupplemental References This article cites 42 articles, 10 of which you can access for free at: http://www.jimmunol.org/content/198/10/4107.full#ref-list-1

Why The JI? Submit online. by guest on October 5, 2021

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2017 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Mucosal IgM Antibody with D-Mannose Afﬁnity in Fugu Takifugu rubripes Is Utilized by a Monogenean Parasite Heterobothrium okamotoi for Host Recognition

Kento Igarashi,* Ryohei Matsunaga,† Sachi Hirakawa,‡ Sho Hosoya,† Hiroaki Suetake,‡ Kiyoshi Kikuchi,† Yuzuru Suzuki,† Osamu Nakamura,* Toshiaki Miyadai,‡ Satoshi Tasumi,† and Shigeyuki Tsutsui*

How parasites recognize their definitive hosts is a mystery; however, parasitism is reportedly initiated by recognition of certain molecules on host surfaces. Fish ectoparasites make initial contact with their hosts at body surfaces, such as skin and gills, which are covered with mucosa that are similar to those of mammalian guts. Fish are among the most primitive vertebrates with immune systems that are equivalent to those in mammals, and they produce and secrete IgM into mucus. In this study, we showed that the Downloaded from monogenean parasite Heterobothrium okamotoi utilizes IgM to recognize its host, fugu Takifugu rubripes. Oncomiracidia are infective larvae of H. okamotoi that shed their cilia and metamorphose into juveniles when exposed to purified D-mannose–binding fractions from fugu mucus. Using liquid chromatography–tandem mass spectrometry analysis, proteins contained in the fraction were identified as D-mannose–specific IgM with two D-mannose–binding lectins. However, although deciliation was significantly induced by IgM and was inhibited by D-mannose or a specific Ab against fugu IgM, other lectins had no effect, and IgM without D-mannose affinity induced deciliation to a limited degree. Subsequent immunofluorescent staining experiments showed that fugu http://www.jimmunol.org/ D-mannose–specific IgM binds ciliated epidermal cells of oncomiracidium. These observations suggest that deciliation is triggered by binding of fugu IgM to cell surface Ags via Ag binding sites. Moreover, concentrations of D-mannose–binding IgM in gill mucus were sufficient to induce deciliation in vitro, indicating that H. okamotoi parasites initially use host Abs to colonize host gills. The Journal of Immunology, 2017, 198: 4107–4114.

ish were the ﬁrst animals to emerge with functional adap- lectins, lysozymes, and antimicrobial peptides (5–9). Hence, tive immune systems equivalent to those in mammals. The teleost pathogens require strategies for breaching mucosal gills, skin, and digestive tracts are covered with mucous barriers.

F by guest on October 5, 2021 membranes that are similar to those in mammalian guts. Tele- The monogenean ectoparasite Heterobothrium okamotoi colo- ost mucus contains Igs, such as IgM and IgT (1), the latter of nizes external surfaces of the Tetraodontiformes teleost fugu which was recently discovered as being unique to teleosts (2). Takifugu rubripes (10) and has high host specificity; no report has The epithelial poly-IgR transports mucosal IgM in teleosts in a shown parasitism on any species other than T. rubripes (11). This similar way to the mammalian counterpart that carries mucosal species has a monoxenous life cycle that is initiated as oncomir- IgA (3). Thus, external surfaces of teleosts could be regarded as acidia, which are free-swimming larvae with cilia that attach to ancient mucosal surfaces that elicit intestine-like immune external host surfaces and grow at the gills for ∼30 d. Subse- responses of higher vertebrates (4). In addition, external mu- quently, oncomiracidia move to branchial cavity walls of their cosa of teleosts contains various defense molecules, including hosts, where they mature and spawn eggs. Oncomiracidia make initial contact with mucosal surfaces of the gills or the skin and *School of Marine Biosciences, Kitasato University, Kanagawa 252-0373, Japan; rapidly shed their ciliated epidermal cells like other monogeneans †Fisheries Laboratory, The University of Tokyo, Shizuoka 431-0214, Japan; and ‡ (12). Although it remains unclear how these parasites recognize Faculty of Marine Bioscience, Fukui Prefectural University, Fukui 917-0003, Japan their hosts and avoid attack by host defenses, they distinguish Received for publication November 28, 2016. Accepted for publication March 13, 2017. physical, physiological, and/or chemical characteristics of their host from nonhosts, including other Takifugu species. This work was supported by Grant-in-Aids for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan (26450267 to S. Tsutsui and Several protein–protein (13), protein–lipid (14), and protein– 22780171 to S. Tasumi). S. Tasumi was also funded by the Towa Foundation for Food carbohydrate (13, 15, 16) interactions have been shown to mediate Research and the Lotte Shigemitsu Prize. host–parasite relationships. Among these, carbohydrate interac- Address correspondence and reprint requests to Dr. Satoshi Tasumi or Dr. Shigeyuki tions are known to trigger parasitism in some marine parasites. For Tsutsui, The University of Tokyo, Shizuoka 431-0214, Japan (S. Tasumi) or Kitasato University, 1-15-1 Kitasato, Kanagawa 252-0373, Japan (S. Tsutsui). E-mail ad- example, a tandem-repeat galectin that is produced by hemocytes dresses: [email protected] (S. Tasumi) or [email protected] of the eastern oyster Crassostrea virginica is used as an entry (S. Tsutsui) receptor by the protozoa Perkinsus marinus (17–19). Similarly, The online version of this article contains supplemental material. oncomiracidia of the monogenean species Neobenedenia girellae Abbreviations used in this article: CBB, Coomassie Brilliant Blue; FSW, filtration- Benedenia seriolae sterilized seawater; LC-MS/MS, liquid chromatography–tandem mass spectrometry; and are likely to recognize glycoproteins of ms-IgM, D-mannose-specific IgM; non-ms–IgM, IgM without D-mannose affinity; their hosts, as indicated by suppression of parasite attachment- QTL, quantitative trait locus; TBS (+), 25 mM Tris-HCl buffer (pH 7.5) containing inducing capacities of skin mucus extract by plant lectins (20, 150 mM NaCl, 10 mM CaCl , and 10 mM MgCl . 2 2 21). Furthermore, N. girellae reportedly recognizes T. rubripes Copyright Ó 2017 by The American Association of Immunologists, Inc. 0022-1767/17/$30.00 skin mucosa using the glycoprotein Wap65-2 (22). Hence, www.jimmunol.org/cgi/doi/10.4049/jimmunol.1601996 4108 A PARASITE UTILIZING HOST IgM FOR HOST RECOGNITION interactions via carbohydrates may be central to T. rubripes host recognition by H. okamotoi. In our previous studies, we identified two D-mannose–specific lectins in gill and skin mucus of T. rubripes.Amongthese,the 13-kDa lily-type lectin pufflectin binds adult H. okamotoi (6). Moreover, pufflectin protein and mRNA expression in gills is much higher in T. rubripes than in the closely related species T. niphobles (23), which is resistant to this parasite, suggesting participation of this lectin in parasitism. The other mucus lectin kalliklectin is homologous to the H chain of mammalian plasma kallikrein (24), although its role has not yet been clarified. Thus, in the current study, we focused on the D-mannose–binding fraction of fugu mucus and hypothesized that FIGURE 2. Effects of the purified D-mannose–binding fraction on it participates in the host recognition of H. okamotoi. Initially, we deciliation of H. okamotoi oncomiracidia and SDS-PAGE patterns of the determined whether oncomiracidia cause deciliation when exposed to fraction. (A) The D-mannose–binding fraction was purified from fugu the fraction and then identified the molecules involved. In these ex- mucus using D-mannose affinity chromatography. FSW or the purified periments, the D-mannose–binding fraction induced the deciliation of D-mannose–binding fraction was added to wells containing ∼10 oncomir- oncomiracidia, but the deciliation-inducing protein was a D-mannose- acidia, and percentages of deciliated individuals were calculated. Values specific IgM rather than the lectins. In this article, we show that this are presented as mean 6 SD of triplicate assays. Deciliation of the frac- important vertebrate defense molecule is exploited as a critical host- tion-treated group differed significantly from the control. (B) The binding Downloaded from recognition factor by parasites. fraction was mixed with equal volumes of SDS sample buffer containing reducing agent and was resolved on a 12.5% SDS polyacrylamide gel. Protein bands were visualized by staining with CBB. Arrows indicate Materials and Methods protein bands. *p , 0.05. Fish and parasites T. rubripes (.500 g) were purchased from fish farms in Japan, reared in Deciliation assays of H. okamotoi oncomiracidia recirculation tanks, and fed commercial pellets. Skin mucus was collected Skin mucus from T. rubripes was homogenized in three volumes of 25 mM http://www.jimmunol.org/ by scraping skin surfaces with a spatula. Blood samples were taken from Tris-HCl buffer (pH 7.5) containing 150 mM NaCl, 10 mM CaCl , and caudal blood vessels using syringes and needles. After clotting for several 2 10 mM MgCl [TBS (+)] and centrifuged at 15,000 3 g for 15 min at 4˚C. hours, sera were harvested from blood samples by centrifugation. Gill 2 To observe deciliation, .10 oncomiracidia were mounted in wells of a 96- mucus was carefully collected from three T. rubripes individuals using a well culture plate with 25 ml of seawater within 12 h after hatching. cotton swab, with care not to contaminate the samples with blood. All Identical volumes of skin mucus extracts were added to each well. samples were taken after anesthetization with 2-phenoxyethanol. Filtration-sterilized seawater (FSW) alone was used as a control. After 1 h Infection experiments were performed using 1-y-old T. rubripes fin- of incubation at 20˚C, oncomiracidia were mounted on a glass slide and 6 n gerlings with standard lengths of 7.51 4.23 cm ( = 33) that were observed under a microscope. For deciliation assays, the statuses of purchased from a commercial hatchery in Japan. T. niphobles (standard oncomiracidium were described according to Yoshinaga et al. (20). Briefly, lengths, 7.41 6 3.92 cm, n = 33) were generated by artificial insemination oncomiracidia with completely or partially shed ciliated epidermal cells and were reared for 2 y in our laboratory prior to infection experiments. by guest on October 5, 2021 and unfolded haptors were regarded as deciliated individuals, and swim- Large quantities of H. okamotoi eggs that were connected by long fil- ming individuals with cilia were counted as ciliated. Degrees of deciliation aments were collected from plastic net traps, using forceps, that were set in were expressed as numbers of deciliated individuals per numbers of ex- T. rubripes drainpipes from tanks containing heavily infected . Egg strings amined individuals 3 100. Experiments were conducted using three wells were washed with filtered seawater and transferred into 500-ml beakers per sample. Activities of purified proteins were determined using the same containing seawater. The eggs were incubated at 20˚C with aeration, and procedure as described below unless otherwise stated. seawater was replaced daily. Hatched oncomiracidia were collected and concentrated by filtration using a 20-mm nylon mesh. Affinity purification and deciliation assays of D-mannose–binding Observations of deciliation of H. okamotoi oncomiracidia. To observe proteins from skin and skin mucus deciliation in detail, oncomiracidia were mounted on glass slides with seawater and T. rubripes gill fragments. Images were captured using It was difficult to obtain sufficient quantities of mucus from the gills of a differential interference contrast light microscope equipped with a full T. rubripes, where H. okamotoi colonize. However, parasite oncomiracidia high-vision compact recorder (model 1R-100; Chunichi Denshi) and were also attach to fish skin (11, 25); although they disappear within 2 d (11), analyzed using Arcsoft TotalMedia Theater (SANYO) software. this observation suggests that host molecules that are recognized by oncomiracidia are also present in skin mucus. Thus, extracts of skin and mucus were used in analyses. Fractions containing D-mannose–binding proteins were collected from the skin, as described in our previous study (6). Fraction buffers were exchanged with FSW by ultrafiltration. Protein concentrations were determined using Quick Start protein assay kits (Bio-Rad), with BSA as a standard for all experiments. Protein concentrations were then adjusted to 300 mg/ml for deciliation assays.

SDS-PAGE Aliquots of all purified samples were subjected to SDS-PAGE to confirm purity. Samples were then mixed with equal volumes of 23 SDS sample buffer containing 2-ME and were loaded onto 12.5% gels. Protein bands were visualized by staining with Coomassie Brilliant Blue (CBB). FIGURE 1. Effects of mucus extracts on deciliation of H. okamotoi Identification of D-mannose–binding proteins by Western oncomiracidia. H. okamotoi oncomiracidia were microscopically exam- ined after 1 h of exposure to FSW (A) and skin mucus extracts from blotting and liquid chromatography–tandem mass T. rubripes (B). More than 10 oncomiracidia were observed, and images of spectrometry representative results are shown. Arrows show falling ciliated epidermal Three proteins were identified in D-mannose–binding fractions using West- cells. Other ciliated epidermal cells had been shed previously. Scale bars, ern blotting and liquid chromatography–tandem mass spectrometry 50 mm. (LC-MS/MS). Western blotting analyses were performed using anti-pufflectin The Journal of Immunology 4109 rabbit Ab (dilution 1:500) and HRP-conjugated anti-rabbit IgG goat Ab (dilution 1:5000; Sigma-Aldrich), as we described earlier (7). For LC- MS/MS analyses, bands on SDS-PAGE gels were excised and digested with trypsin (Wako) in the gel at 37˚C for 20 h. Digested peptides were then loaded on an LC-MS/MS system comprising an HPLC system (NANOSPACE SI-2; Shiseido Fine Chemicals) and an ion trap mass spectrometer (LCQ Deca; Thermo Finnigan). Individual tandem mass spectrometry spectra were processed using TurboSEQUEST software (Thermo Quest), and the generated peak list files were used to query Swiss-Prot using the Mascot program (http://www.matrixscience.com). Purification and deciliation assays of pufflectin, kalliklectin, and D-mannose–specific IgM Pufflectin was purified according to our previously described method (6), with some modifications. Briefly, pufflectin-rich fractions were purified by affinity chromatography with D-Mannose Agarose (Sigma-Aldrich) and were subjected to gel filtration using a Superose 6HR 10/30 column (Amersham Biosciences), with FSW as a running buffer. Kalliklectin and D-mannose–specific IgM (ms-IgM) were purified as described in our previous studies (24, 26), with modifications. These proteins were purified from the serum because it contains higher concentrations than skin mucus. Briefly, serum was incubated with D-Mannose

Agarose (Sigma-Aldrich) as described above, and proteins were eluted Downloaded from with 200 mM D-mannose and were separated using gel filtration, also as described above. Concentrations of purified pufflectin, kalliklectin, and ms-IgM were adjusted to 600, 200, and 200 mg/ml, respectively, and 50-ml aliquots were used in deciliation assays.

Tests of inhibition by D-mannose and antisera http://www.jimmunol.org/ To test inhibition by sugar, ms-IgM was incubated with 0.1 M D-mannose at 4˚C overnight. Untreated ms-IgM and 0.1 M D-mannose alone were used FIGURE 4. Inhibition of ms-IgM–induced deciliation by D-mannose as controls. Degrees of deciliation were estimated as described above. To and anti-fugu IgM antisera. (A) ms-IgM was preincubated with 0.1 M test inhibition by antiserum, ms-IgM was incubated with anti-fugu IgM D-mannose and subjected to deciliation assays. Untreated ms-IgM and rabbit antiserum (27) or with normal rabbit serum at a final total protein B concentration of 700 mg/ml at 4˚C overnight. Deciliation assays were D-mannose alone were used as controls. ( ) ms-IgM was incubated with conducted as described above. anti-fugu IgM rabbit antiserum or normal rabbit serum, and deciliation- Characterization of sugar specificity of ms-IgM. To determine whether inducing activities were analyzed. Untreated ms-IgM was used as a con- 6 , ms-IgM recognizes sugars other than D-mannose, affinity chromatograph trol. Bars represent mean SD (n = 3). ***p 0.001. analyses were performed using N-Acetyl-D-galactosamine–Agarose, by guest on October 5, 2021 N-Acetyl-D-glucosamine–Agarose (both from Sigma-Aldrich), L-Fucose– Purification and deciliation assays of IgM without D-mannose Agarose (J-Oil Mills), and a-Lactose–Agarose (Sigma-Aldrich). In these experiments, 200-ml aliquots of each sugar-conjugated agarose were in- affinity cubated with 2.5 ml of ms-IgM at 4˚C overnight, and unbound fractions IgM without D-mannose affinity (non-ms–IgM) was purified from serum as were collected. After washing agarose samples with 30 ml of TBS (+), we described earlier (26), with a minor modification. Briefly, IgM fractions bound proteins were eluted with 2.5 ml of the same buffer containing the were purified using Sephacryl S-300 16/60 and Superose 6HR 10/30 gel respective sugars at 0.2 M. Bound and unbound fractions were concen- filtration columns (Amersham Biosciences). Because these fractions con- m trated to 100 l using Vivaspin (Sartorius) and analyzed using SDS-PAGE. tained both ms-IgM and non-ms–IgM, the former was removed by incubation with D-Mannose–Agarose at 4˚C overnight. The resulting unbound fraction was used as non-ms–IgM. The buffer was replaced with FSW, and the concentration was adjusted to 300 mg/ml for use in deciliation assays. ms-IgM (200 mg/ml) was used as a positive control. Estimation of minimum concentrations of ms-IgM to induce deciliation Serially diluted ms-IgM was prepared at 400, 200, 100, or 50 mg/ml with FSW. Deciliation assays were conducted as described above.

Semiquantification of ms-IgM concentrations in fugu gill mucus using Western blotting Gill mucus extracts were obtained by homogenization with three volumes of TBS (+), followed by centrifugation, as described above. Subsequently, 50-ml gill mucus extracts from each fish were incubated with 50 mlofD-Mannose– Agarose at 4˚C overnight. After three washes in TBS (+), 50 mlofSDSsample buffer was added to the agarose, and 20-ml aliquots were subjected to SDS- PAGE. As standards, 100, 50, 25, and 12.5 mg/ml IgM were diluted with three FIGURE 3. Percentage deciliation of oncomiracidia after exposure to volumes of TBS (+), and 20-ml aliquots were loaded onto gels. Separated purified D-mannose–binding proteins. Pufflectin-rich fractions were puri- proteins were transferred onto nitrocellulose membranes and were stained with fied from fugu mucus extracts using D-mannose affinity chromatography anti-fugu IgM antiserum, as described above. and were further purified using gel filtration. For purification of kalliklectin Immunofluorescence microscopy of H. okamotoi and ms-IgM, fugu serum was subjected to D-mannose affinity chromatography, and the proteins were separated using gel filtration. Three can- oncomiracidia using ms-IgM didate proteins were subjected to deciliation assays. Data are presented as H. okamotoi oncomiracidia were fixed with 4% paraformaldehyde at 4˚C mean 6 SD (n = 3). ***p , 0.001. for 2 h. After three washes in PBS, oncomiracidia were incubated with 4110 A PARASITE UTILIZING HOST IgM FOR HOST RECOGNITION

100 mg/ml ms-IgM in PBS at 4˚C overnight. As controls, 100 mg/ml ms-IgM was preincubated with 0.1 M D-mannose at 4˚C overnight and added to oncomiracidia instead of ms-IgM. Further controls included 100 mg/ml non-ms–IgM and PBS alone and were added likewise. After three washes in PBS, oncomiracidia were incubated at 4˚C overnight with anti-fugu IgM rabbit antiserum (27) that was diluted 1:200 in PBS. Oncomiracidia were washed three times in PBS and were mixed with FITC-conjugated anti- rabbit IgG goat F(ab9)2 (Santa Cruz Biotechnology), which was diluted 1:100 in PBS. Following a 1-h incubation at room temperature and washing with PBS, oncomiracidia were observed using a confocal laser scanning microscope (LSM 510; Carl Zeiss). Infection experiment In vivo experimental H. okamotoi challenges of T. rubripes and T. niphobles were conducted as described by Ohhashi et al. (11), with some FIGURE 6. Sugar-binding properties of ms-IgM. ms-IgM was subjected modifications. Within 24 h after hatching, oncomiracidia were labeled with to affinity chromatography analyses using agarose beads conjugated with CFSE (Dojindo), as we described earlier (28). N-acetyl-D-galactosamine (GalNAc), N-acetyl-D-glucosamine (GlcNAc), Experimental challenges were performed using 33 individuals of the two L-fucose, or lactose. Bound (B) and unbound (U) fractions were resolved species in triplicate. Initially, 22 fish (11 of each species) were acclimated on 12.5% SDS polyacrylamide gel that was subsequently stained with together in the same tank (three 500-l tanks were used) for 24 h. Before the CBB. The arrow and arrowhead indicate H and L chains, respectively. challenge test, the volume of seawater in the tank was reduced to 22 l (1 l per fish). The fish were then exposed to ∼2200 CFSE-labeled oncomiracidia

(100 oncomiracidia per fish) for 3 h. At the end of the exposure period, fish miracidia swam freely in sea water, but made contact and attached Downloaded from were transferred to three new 500-l tanks, and two individuals of each to gill filaments with haptors immediately afterward. Parasites species were sampled at 3, 24, 48, 72, and 96 h postexposure. Sampled fish were immediately anesthetized, and bilateral gills were dissected and then moved around on filaments, with expansions and contractions washed extensively with FSW in a 500-ml beaker to remove unattached of their bodies (Supplemental Video 1), until they found suitable oncomiracidia. The number of parasites attached to gill filaments was places to settle (Supplemental Video 2). Within 1 min after set- counted using a fluorescence stereomicroscope (MZ16F; LEICA). Gills tlement, body surfaces appeared fluffy, and deciliation was initi- were photographed using a digital camera, and surface areas were calcu-

ated (Supplemental Video 3). Oncomiracidia actively twisted their http://www.jimmunol.org/ lated using ImageJ software (https://imagej.nih.gov/ij/). The numbers of parasites were normalized to gill areas (square millimeters) of individuals. bodies, and the remaining cells were finally released from the surface of the parasite (Supplemental Video 4) after ∼15 min of Statistical analyses deciliation. All numerical data, with the exception of those from infection experiments, Partially purified D-mannose–binding proteins induce deciliation 6 are presented as mean SD, and all statistical analyses were performed of H. okamotoi oncomiracidia. In a further experiment, we incu- using R. Deciliation rates were compared using the Student t test, followed by the Kolmogorov–Smirnov test for normality and F tests of distribution bated H. okamotoi oncomiracidia with skin mucus extracts from normality. Differences were considered significant when p = 0.05. Deci- fugu and investigated the roles of the mucus components in liation rates were compared between treatments using a generalized linear deciliation. This experiment showed that a subset of oncomir- model, with treatment as the fixed effect. Changes in the numbers of acidia shed ciliated epidermal cells (Fig. 1). Then, we investigated by guest on October 5, 2021 H. okamotoi oncomiracidium on gill surfaces were compared between whether the D-mannose–binding fraction from skin mucus extracts T. rubripes and T. niphobles using the generalized linear model, with species and time postinfection as fixed effects. Multiple comparisons of could lead to the same phenomenon; we observed similar induc- significant effects were performed using the glht function “multcomp” tion of deciliation (Fig. 2A), with significantly higher percentages (p , 0.05), with false discovery rate correction. of deciliated individuals than in controls treated with FSW (p = 0.01778). Results ms-IgM induces deciliation of H. okamotoi oncomiracidia. Using Deciliation of H. okamotoi oncomiracidium SDS-PAGE analysis, we identified major proteins with molecular We carefully observed the behavior of oncomiracidia after the masses ∼ 78, 40, and 13 kDa in the D-mannose–binding fraction, exposure to gill filaments of T. rubripes in vitro. Initially, onco- which induced deciliation (Fig. 2B). Among these, the 13-kDa protein was identified as pufflectin in a Western blotting experiment using anti-pufflectin antiserum (Supplemental Fig. 1). To identify the other proteins in the deciliation-inducing fraction, we excised corresponding bands from SDS-PAGE gels and performed LC-MS/MS analyses, which indicated that the 40- and 78-kDa proteins were kalliklectin (matching score, 966) and IgM H chain (ms-IgM; matching score, 562), respectively. Subsequently, we purified the three proteins using affinity chromatography and gel filtration (Supplemental Fig. 2) and tested their deciliation activities. In these experiments, ms-IgM induced 100% deciliation, whereas the other proteins had only small effects on oncomiracidia morphology (Fig. 3). In contrast, deciliation rates were dramatically decreased to 3% after pretreatment of ms-IgM with 0.1 M D-mannose (p , 0.001, Fig. 4A), indicating that ms-IgM induces deciliation of H. okamotoi oncomiracidia through a pro- FIGURE 5. Concentration-dependent deciliation-inducing activities of ms-IgM. Oncomiracidia were exposed to ms-IgM at 12.5, 25, 50, and tein–carbohydrate interaction. In addition, antisera against 100 mg/ml, and the percentages of individuals that shed their cilia were T. rubripes IgM significantly inhibited the deciliation-inducing estimated. Values are presented as mean 6 SD (n = 3 each). Different activity of ms-IgM (p , 0.001, Fig. 4B), whereas normal sera letters indicate significant differences; p , 0.05 between 12.5 and 25 mg/ml, did not affect deciliation. Finally, ms-IgM–induced deciliation and p , 0.001 among the other treatments. was concentration dependent at .25 mg/ml (Fig. 5). The Journal of Immunology 4111

Sugar specificity of ms-IgM To investigate sugar specificity, ms-IgM was subjected to affinity chromatographs with four saccharides. In these experiments, ms-IgM was detected in unbound fractions in all chromatographs, but it was not detected in any bound fractions (Fig. 6), suggesting D-mannose specificity of ms-IgM. Deciliation-inducing activity of non-ms–IgM To determine whether this deciliation-inducing activity is specific to ms-IgM or is universal to IgM, we purified non-ms–IgM (Supplemental Fig. 3). In experiments with this protein, percentages of deciliated oncomiracidia were significantly lower than in those exposed to ms-IgM (p , 0.001, Fig. 7) but were significantly higher than in controls (p , 0.001, Fig. 7). Concentrations of ms-IgM in gill mucus. To determine whether FIGURE 8. Western blot semiquantification of ms-IgM concentrations ms-IgM is present in sufficient quantities to induce deciliation of in gill mucus. D-mannose–binding fractions of gill mucus extracts from oncomiracidia in vivo, we measured ms-IgM concentrations in three fish were subjected to Western blotting with anti-fugu IgM antiserum. T. rubripes gill mucus samples. Based on band intensities of the As standards, 100, 50, 25, and 12.5 mg/ml IgM were used.

standards, concentrations of ms-IgM in the mucus from the gills of Downloaded from . m three T. rubripes individuals were 25 g/ml (Fig. 8), indicating secretion of IgM into body mucus is a conserved feature of tele- that ms-IgM could induce deciliation of oncomiracidia on gill osts (1). Thus, to resolve this contradiction, we compared surfaces in vivo. H. okamotoi infectivity in T. rubripes and T. niphobles. Median Binding of ms-IgM to ciliated epidermal cells on H. okamotoi numbers of attached oncomiracidia on 1-mm2 gill areas did not oncomiracidia. Because T. rubripes ms-IgM induces shedding differ significantly between these species after 3 h (p = 0.09307, of cilia from H. okamotoi oncomiracidium, we investigated the Fig. 10), whereas significant differences were detected at the other http://www.jimmunol.org/ presence of an ms-IgM ligand in oncomiracidium. Immunofluo- time points (p , 0.001). In addition, the numbers of oncomir- rescence observations showed that ms-IgM binds to surfaces of acidia on T. rubripes did not change significantly throughout the H. okamotoi oncomiracidium (Fig. 9A), as indicated by intense experiment (Fig. 10). However, the numbers of oncomiracidia fluorescence on ciliated epidermal cells on oncomiracidia sur- attached to T. niphobles decreased by 24 h, and further decreases faces. In contrast, negligible fluorescence was observed on were significant at 48 h postexposure. Finally, no oncomiracidia oncomiracidia surfaces in the presence of PBS alone (data not were observed on the gills of T. niphobles at 96 h postexposure shown) or non-ms–IgM (Fig. 9B). In further experiments, binding (Fig. 10). of ms-IgM to oncomiracidia was inhibited by pretreatment of ms-IgM with D-mannose (Fig. 9C), suggesting that ms-IgM binds Discussion by guest on October 5, 2021 partner molecules on ciliated epidermal cells via interactions be- The present data demonstrate that the monogenean parasite tween IgM and saccharides containing D-mannose. H. okamotoi exploits a host defense system for invasion. Specif- In vivo infection experiment. H. okamotoi reportedly have strict ically, when free-swimming parasite larvae were attacked by host specificity for T. rubripes and cannot parasitize the closely ms-IgMs of their host T. rubripes, they shed their ciliated epi- related species, T. niphobles (11). However, it is unlikely that the dermal cells. Exploitation of host defensive molecules has been body mucus of T. niphobles and other fish lacks ms-IgM, because reported for pathogenic microorganisms, such as the schistosome worm Schistosoma mansoni, which causes schistosomiasis in humans and uses host TNF-a to induce replication (29), and HIV, which uses CD4 Ags on T cells as essential and specific receptor components (30). In addition to these characterized processes, the present fish parasite uses the key immune molecule IgM as a host- recognition protein. This finding suggests that vertebrate Igs may also be used by some pathogens. In mammals, a subset of natural IgM Abs is constantly produced by unique B cells (B1 cells) in the blood, regardless of Ag stim- ulation, whereas specific Abs are generated following Ag exposures (31, 32). Ags for these natural Abs are generally saccharides (33) that make up pathogen-associated molecular patterns. Hence, ms-IgM in gill mucus of T. rubripes may also be a natural Ab. IgMs that bind various saccharides have been identified in sera from teleosts, including T. rubripes (26, 34); although their counterparts in mammalian B1 cells have not been identified in fish, these IgMs FIGURE 7. Comparison of deciliation-inducing activity of IgM, with or likely act as natural Abs. ms-IgM may have provided the selective without D-mannose affinity. The IgM fraction was purified using two gel- pressures that led to their exploitation by H. okamotoi. filtration columns. From the fraction, ms-IgM was removed following In the current study, ms-IgM was shown to bind ciliated epi- incubation with the D-mannose affinity column, and the resulting unbound fraction was used as a non-ms–IgM control. Deciliation-inducing activities dermal cells of H. okamotoi oncomiracidia (Fig. 9A). However, of these IgMs were analyzed using FSW as a control. Three experiments ms-IgM binding was inhibited by the addition of D-mannose were performed for each sample, and data are expressed as mean 6 SD. (Fig. 9C), suggesting the presence of an ms-IgM ligand on the ***p , 0.001. surfaces of parasite cells that possesses oligosaccharides containing 4112 A PARASITE UTILIZING HOST IgM FOR HOST RECOGNITION

signaling receptors via epitopes that differ from D-mannose–containing oligosaccharides. In both cases, signaling pathways may not be highly activated, resulting in relatively low potency of deciliation induction. The lectins pufflectin and kalliklectin did not induce deciliation of H. okamotoi oncomiracidia (Fig. 3), despite binding to D-mannose. We also found that the D-mannose–specific plant lectin Con A did not induce deciliation (data not shown), likely reflecting differing binding affinities for these lectins and ms-IgM. In general, bind- 28 ing affinities (Kd values) of Ags to Abs are ,10 M per binding site, whereas those of lectins with oligosaccharides range from 1024 to 1025 M (35). In addition, IgM from teleosts, including T. rubripes, are usually tetramers (27, 36) with eight binding sites. Thus, high binding affinity for ligands may be a requirement of FIGURE 9. Binding of ms-IgM to H. okamotoi oncomiracidia in immunofluorescence microscopy analysis. Oncomiracidia were fixed with 4% deciliation signaling. Alternatively, additional associations be- paraformaldehyde and treated with ms-IgM (A), non-ms–IgM (B), and tween non–Ag-binding domains of IgM and other molecules on ms-IgM that had been preincubated with 0.1 M D-mannose (C). IgMs were oncomiracidia may be involved in the induction of deciliation detected using anti-fugu IgM rabbit antiserum and FITC-conjugated sec- signaling pathways. ondary Ab. Images of representative results are shown. Scale bars, 50 mm. H. okamotoi has been widely used as a model organism for fish Downloaded from DIC, differential interference contrast. parasitology because it has high host specificity for T. rubripes. In agreement, the present experimental infection showed that D-mannose. In addition, affinity chromatography analyses with oncomiracidia were excluded by the closely related species T. agarose beads that were conjugated with four sugars showed that ms- niphobles, and similar results were reported previously for IgM does not bind these other sugars (Fig. 6). However, although nontetraodontiform fish, including the Japanese flounder Para- these data suggest that ms-IgM binding is D-mannose specific, lichthys olivaceus (Pleuronectiformes) and the red seabream ms-IgM may bind other unidentified sugars to mediate deciliation of Pagrus major (Perciformes) (11). However, this high host speci- http://www.jimmunol.org/ oncomiracidium. Thus, further studies of ms-IgM specificities are ficity is likely established well after host recognition, because the required using high-throughput techniques, such as frontal affinity numbers of attached oncomiracidia did not differ significantly chromatography and glycoconjugate microarrays, and additional between 1-mm2 gill areas of T. niphobles and T. rubripes imme- investigations of glycans on oncomiracidium using mass spectrom- diately after exposure (p = 0.09307, Fig. 10). However, in etry may identify biologically important ligands of ms-IgM, which agreement with Ohhashi et al. (11), the parasite was exclusively may act as signal-transduction molecules that promote deciliation of rejected from T. niphobles within 96 h (Fig. 10). These observa- oncomiracidia. tions suggest that parasite recognition of hosts is unrestricted In contrast to ms-IgM, other IgMs with no affinity for D-mannose during the host-recognition stage, allowing attachment to various by guest on October 5, 2021 were poor inducers of deciliation (Fig. 7), indicating that decili- teleost species. Hence, IgM with mannose specificity may be ation is largely induced by binding of IgM to its ligand via widely present in skin and gill mucosa of teleosts. In agreement, Ag-recognition sites, rather than via Fc domains. However, the we confirmed that deciliation of oncomiracidia occurred in the percentages of deciliated oncomiracidia in the presence of IgM presence of gill filaments or skin mucus extracts from other fish fractions without mannose-binding affinity were significantly species in vitro (Supplemental Fig. 4), indicating that deciliation higher than were those of controls treated with FSW (Fig. 7). may be caused by ms-IgM in nonhost species. Potentially, these data reflect the presence of multiple signaling Establishment of parasite host recognition and specificity receptors that differ from the ms-IgM ligand on the surfaces of through coevolution and/or host switching is an area of great in- oncomiracidium cells, and subsets of polyclonal IgMs might bind terest. Parasitism of monogeneans is governed by various dynamic some of these. Alternatively, various IgMs may bind ms-IgM interactions between parasites and hosts, including host recogni-

FIGURE 10. In vivo experimental challenge of T. rubripes and T. niphobles with H. okamotoi. Thirty-three T. rubripes and T. niphobles individuals were acclimated together in the same tank for 24 h. The fish were then exposed to ∼2200 CFSE-labeled oncomiracidia for 3 h. Six individuals of each species of fish were sampled at 3, 24, 48, 72, and 96 h postexposure, and the numbers of oncomiracidia attached to 1-mm2 areas of gill filaments were counted. The box-plot graph represents lower and upper quartiles and medians of parasite burden. Upper and lower whiskers delineate data within positive and negative 1.5 interquartile ranges of the median, respectively. Different letters (a and b) represent significant differences in parasite numbers between sampling points for T. niphobles. No significant differences were identified between sampling points for T. rubripes. Parasite numbers were significantly lower for T. niphobles than for T. rubripes at the same sampling point. ***p , 0.001. The Journal of Immunology 4113 tion, subsequent attachment to the host, utilization of nutrition from 5. Cole, A. M., P. Weis, and G. Diamond. 1997. Isolation and characterization of pleurocidin, an antimicrobial peptide in the skin secretions of winter flounder. the host, and escape from host immunity (37). The present data J. Biol. Chem. 272: 12008–12013. indicate the presence of differing host molecules that are used 6. Tsutsui, S., S. Tasumi, H. Suetake, and Y. Suzuki. 2003. Lectins homologous to by H. okamotoi for host recognition and subsequent stages of those of monocotyledonous plants in the skin mucus and intestine of pufferfish, the life cycle and that rejection by nonhosts after attachment Fugu rubripes. J. Biol. Chem. 278: 20882–20889. 7. Tsutsui, S., S. Tasumi, H. Suetake, K. Kikuchi, and Y. Suzuki. 2005. Demon- contributes to the high host specificity of H. okamotoi for stration of the mucosal lectins in the epithelial cells of internal and external body T. rubripes. Specifically, nonhost fish likely shed H. okamotoi surface tissues in pufferfish (Fugu rubripes). Dev. Comp. Immunol. 29: 243–253. 8. Nakamura, O., Y. Inaga, S. Suzuki, S. Tsutsui, K. Muramoto, H. Kamiya, and from gill filaments using their defense systems and/or as T. Watanabe. 2007. Possible immune functions of congerin, a mucosal galectin, a result of molecular incompatibility. Given these possibili- in the intestinal lumen of Japanese conger eel. Fish Shellfish Immunol. 23: 683– ties, T. rubripes may have exclusive vulnerability to H. oka- 692. 9. Fernandes, J. M., G. D. Kemp, and V. J. Smith. 2004. Two novel muramidases motoi, which has acquired specific host immune-evasion from skin mucosa of rainbow trout (Oncorhynchus mykiss). Comp. Biochem. strategies. Future studies of parasite resistance and susceptibil- Physiol. B Biochem. Mol. Biol. 138: 53–64. ity mechanisms will likely provide solutions for the ensuing 10. Ogawa, K. 1991. Redescription of Heterobothrium tetrodonis (Goto, 1894) (Monogenea: Diclidophoridae) and other related new species from puffers of the diseases. genus Takifigu (Teleost: Tetraodontidae). J. Parasitol. 40: 388–396. The genetic distance between T. rubripes and T. niphobles is 11. Ohhashi, Y., T. Yoshinaga, and K. Ogawa. 2007. Involvement of host recognition sufficiently close (38) that interspecific F hybrids have been bred by oncomiracidia and post-larval survivability in the host specificity of Heter- 2 obothrium okamotoi (Monogenea: Diclidophoridae). Int. J. Parasitol. 37: 53–60. (39). Moreover, the full genome for T. rubripes has been solved 12. Roberts, L. S., and J. Janovy. 2009. Monogenoidea. L. S. Roberts, and J. Janovy, (40), and genetic maps have been generated (41, 42). Thus, eds. McGraw-Hill Higher Education, Boston, p. 295–311. T. rubripes has major advantages as a subject for further investi- 13. Sibley, L. D. 2011. Invasion and intracellular survival by protozoan parasites.

Immunol. Rev. 240: 72–91. Downloaded from gations of genes with causal links to phenotypes, such as host 14. Goldston, A. M., R. R. Powell, and L. A. Temesvari. 2012. Sink or swim: lipid specificity (43). Accordingly, to investigate disease mechanisms, rafts in parasite pathogenesis. Trends Parasitol. 28: 417–426. we (28) analyzed several quantitative trait loci (QTLs) that are 15. Boulanger, M. J., M. L. Tonkin, and J. Crawford. 2010. Apicomplexan parasite adhesins: novel strategies for targeting host cell carbohydrates. Curr. Opin. responsible for resistance and susceptibility to H. okamotoi in- Struct. Biol. 20: 551–559. fection using genome-wide QTL analyses of F2 hybrids. We de- 16. Brown, A., and M. K. Higgins. 2010. Carbohydrate binding molecules in malaria pathology. Curr. Opin. Struct. Biol. 20: 560–566. tected a significant QTL on linkage group 9 in fish that had been 17. Tasumi, S., and G. R. Vasta. 2007. A galectin of unique domain organization infected for 21 d and identified 214 annotated protein-coding from hemocytes of the Eastern oyster (Crassostrea virginica) is a receptor for the http://www.jimmunol.org/ genes. Although the gene for IgM was not among these, the protistan parasite Perkinsus marinus. J. Immunol. 179: 3086–3098. 18. Vasta, G. R. 2009. Roles of galectins in infection. Nat. Rev. Microbiol. 7: 424– molecule(s) responsible for exclusion of H. okamotoi is likely 438. present at this locus. 19. Vasta, G. R. 2012. Galectins as pattern recognition receptors: structure, function, In conclusion, we demonstrated that the monogenean parasite and evolution. Adv. Exp. Med. Biol. 946: 21–36. 20. Yoshinaga, T., T. Nagakura, K. Ogawa, and H. Wakabayashi. 2000. Attachment- H. okamotoi utilizes mannose-specific IgM of their host inducing capacities of fish tissue extracts on oncomiracidia of Neobenedenia T. rubripes during the first step of parasitism, although these girellae (Monogenea, Capsalidae). J. Parasitol. 86: 214–219. 21. Yoshinaga, T., T. Nagakura, K. Ogawa, Y. Fukuda, and H. Wakabayashi. 2002. observations alone are insufficient to explain the extremely high Attachment-inducing capacities of fish skin epithelial extracts on oncomiracidia host specificity. Hence, further studies of the molecular basis of of Benedenia seriolae (Monogenea: Capsalidae). Int. J. Parasitol. 32: 381–384. by guest on October 5, 2021 parasitism in H. okamotoi may comprehensively characterize 22. Ohashi, H., N. Umeda, N. Hirazawa, Y. Ozaki, C. Miura, and T. Miura. 2007. Purification and identification of a glycoprotein that induces the attachment of these relationships using combinations of molecular–biological oncomiracidia of Neobenedenia girellae (Monogenea, Capsalidae). Int. J. Par- and genetic comparisons of resistant T. niphobles and susceptible asitol. 37: 1483–1490. T. rubripes. 23. Tasumi, S., A. Yamaguchi, R. Matsunaga, K. Fukushi, Y. Suzuki, O. Nakamura, K. Kikuchi, and S. Tsutsui. 2016. Identification and characterization of pufflectin from the grass pufferfish Takifugu niphobles and comparison of its expression Acknowledgments with that of Takifugu rubripes. Dev. Comp. Immunol. 59: 48–56. 24. Tsutsui, S., N. Yamamura, T. Yoshida, and O. Nakamura. 2015. Fugu (Takifugu We thank Dr. K. Ogawa (President of the Meguro Parasitological Museum) rubripes) serum GlcNAc-binding lectin is a kalliklectin but has different prop- and Prof. T. Yoshinaga (The University of Tokyo) for their valuable com- erties from those of a reported homologue. J. Biochem. 158: 189–195. ments on deciliation and the metamorphosis of H. okamotoi oncomiracidia. 25. Chigasaki, M., M. Nakane, K. Ogawa, and H. Wakabayashi. 2000. Standardized We thank N. Mizuno, Y. Jo, and M. Fujita (The University of Tokyo) for method for experimental infection of tiger puffer Takifugu rubripes with oncomiracidia of Heterobothrium okamotoi (Monogenea: Diclidophoridae) with assistance with fish purchasing and rearing. Dr. E. Nagata (School of some data on the oncomiracidial biology. Fish Pathol. 35: 215–221. Medicine, Kitasato University) kindly assisted in LC-MS/MS analyses. 26. Tsutsui, S., T. Ariji, A. Sato, T. Yoshida, N. Yamamura, T. Odaka, K. Araki, Dr. T. Odaka (Fukui Prefectural University) and S. Yoshikawa (Nagasaki H. Suetake, T. Miyadai, and O. Nakamura. 2013. Serum GlcNAc-binding IgM of Prefectural Institute of Fisheries) provided fugu serum and parasite eggs, fugu (Takifugu rubripes) suppresses the growth of fish pathogenic bacteria: a novel function of teleost antibody. Dev. Comp. Immunol. 41: 20–26. respectively. 27. Miyadai, T., M. Ootani, D. Tahara, M. Aoki, and K. Saitoh. 2004. Monoclonal antibodies recognising serum immunoglobulins and surface immunoglobulin- positive cells of puffer fish, torafugu (Takifugu rubripes). Fish Shellfish Immu- Disclosures nol. 17: 211–222. The authors have no financial conflicts of interest. 28. Hosoya, S., S. Kido, Y. Hirabayashi, W. Kai, R. Kinami, T. Yoshinaga, K. Ogawa, H. Suetake, K. Kikuchi, and Y. Suzuki. 2013. Genomic regions of pufferfishes responsible for host specificity of a monogenean parasite, Hetero- bothrium okamotoi. Int. J. Parasitol. 43: 909–915. References 29. Amiri, P., R. M. Locksley, T. G. Parslow, M. Sadick, E. Rector, D. Ritter, and 1. Salinas, I., Y. A. Zhang, and J. O. Sunyer. 2011. Mucosal immunoglobulins and J. H. McKerrow. 1992. Tumour necrosis factor a restores granulomas and in- B cells of teleost fish. Dev. Comp. Immunol. 35: 1346–1365. duces parasite egg-laying in schistosome-infected SCID mice. Nature 356: 604– 2. Hansen, J. D., E. D. Landis, and R. B. Phillips. 2005. Discovery of a unique Ig 607. heavy-chain isotype (IgT) in rainbow trout: implications for a distinctive B cell 30. Dalgleish, A. G., P. C. Beverley, P. R. Clapham, D. H. Crawford, M. F. Greaves, developmental pathway in teleost fish. Proc. Natl. Acad. Sci. USA 102: 6919– and R. A. Weiss. 1984. The CD4 (T4) antigen is an essential component of the 6924. receptor for the AIDS retrovirus. Nature 312: 763–767. 3. Hamuro, K., H. Suetake, N. R. Saha, K. Kikuchi, and Y. Suzuki. 2007. A teleost 31. Gro¨nwall, C., J. Vas, and G. J. Silverman. 2012. Protective roles of natural IgM polymeric Ig receptor exhibiting two Ig-like domains transports tetrameric IgM antibodies. Front. Immunol. 3: 66. into the skin. J. Immunol. 178: 5682–5689. 32. Kearney, J. F., P. Patel, E. K. Stefanov, and R. G. King. 2015. Natural antibody 4. Xu, Z., D. Parra, D. Go´mez, I. Salinas, Y. A. Zhang, L. von Gersdorff Jørgensen, repertoires: development and functional role in inhibiting allergic airway dis- R. D. Heinecke, K. Buchmann, S. LaPatra, and J. O. Sunyer. 2013. Teleost skin, ease. Annu. Rev. Immunol. 33: 475–504. an ancient mucosal surface that elicits gut-like immune responses. Proc. Natl. 33. Takeuchi, Y., S. H. Liong, P. D. Bieniasz, U. Ja¨ger, C. D. Porter, T. Friedman, Acad. Sci. USA 110: 13097–13102. M. O. McClure, and R. A. Weiss. 1997. Sensitization of rhabdo-, lenti-, and 4114 A PARASITE UTILIZING HOST IgM FOR HOST RECOGNITION

spumaviruses to human serum by galactosyl(alpha1-3)galactosylation. J. Virol. 39. Hosoya, S., W. Kai, M. Fujita, K. Miyaki, H. Suetake, Y. Suzuki, and K. Kikuchi. 71: 6174–6178. 2013. The genetic architecture of growth rate in juvenile Takifugu species. 34. Giang, D. T., E. Van Driessche, and S. Beeckmans. 2012. Serum carbohydrate- Evolution 67: 590–598. binding IgM are present in Vietnamese striped catfish (Pangasianodon hypo- 40. Aparicio, S., J. Chapman, E. Stupka, N. Putnam, J. M. Chia, P. Dehal, phthalmus) but not in North African catfish (Clarias gariepinus). Dev. Comp. A. Christoffels, S. Rash, S. Hoon, A. Smit, et al. 2002. Whole-genome shotgun Immunol. 36: 418–432. assembly and analysis of the genome of Fugu rubripes. Science 297: 1301– 35. Kuno, A., N. Uchiyama, S. Koseki-Kuno, Y. Ebe, S. Takashima, M. Yamada, and 1310. J. Hirabayashi. 2005. Evanescent-field fluorescence-assisted lectin microarray: a 41. Kai, W., K. Kikuchi, M. Fujita, H. Suetake, A. Fujiwara, Y. Yoshiura, new strategy for glycan profiling. Nat. Methods 2: 851–856. M. Ototake, B. Venkatesh, K. Miyaki, and Y. Suzuki. 2005. A genetic linkage 36. Kaattari, S., D. Evans, and J. Klemer. 1998. Varied redox forms of teleost IgM: map for the tiger pufferfish, Takifugu rubripes. Genetics 171: 227–238. an alternative to isotypic diversity? Immunol. Rev. 166: 133–142. 42. Kai, W., K. Kikuchi, S. Tohari, A. K. Chew, A. Tay, A. Fujiwara, S. Hosoya, 37. Buchmann, K., and T. Lindenstrøm. 2002. Interactions between monogenean H. Suetake, K. Naruse, S. Brenner, et al. 2011. Integration of the genetic map and parasites and their fish hosts. Int. J. Parasitol. 32: 309–319. genome assembly of fugu facilitates insights into distinct features of genome 38. Yamanoue, Y., M. Miya, K. Matsuura, S. Miyazawa, N. Tsukamoto, H. Doi, evolution in teleosts and mammals. Genome Biol. Evol. 3: 424–442. H. Takahashi, K. Mabuchi, M. Nishida, and H. Sakai. 2009. Explosive speciation 43. Kikuchi, K., W. Kai, A. Hosokawa, N. Mizuno, H. Suetake, K. Asahina, and of Takifugu: another use of fugu as a model system for evolutionary biology. Y. Suzuki. 2007. The sex-determining locus in the tiger pufferfish, Takifugu Mol. Biol. Evol. 26: 623–629. rubripes. Genetics 175: 2039–2042. Downloaded from http://www.jimmunol.org/ by guest on October 5, 2021