J. Parasitol., 96(1), 2010, pp. 116–124 F American Society of Parasitologists 2010

MOLECULAR CHARACTERIZATION AND HISTOPATHOLOGY OF MYXOBOLUS KOI INFECTING THE GILLS OF A KOI, CYPRINUS CARPIO, WITH AN AMENDED MORPHOLOGICAL DESCRIPTION OF THE AGENT

Alvin C. Camus andMatt J. Griffin * Department of Pathology, College of Veterinary Medicine, University of Georgia, Athens, Georgia 30602. e-mail: [email protected]

ABSTRACT:AMyxobolus sp., morphologically resembling M. toyamai, M. longisporus, and M. koi, was isolated from the gills of a koi, Cyprinus carpio, that died in an ornamental pond. Large plasmodia were localized within lamellae, causing severe disruption of the normal branchial architecture, sufficient to compromise respiration. Although the case isolate shared several features with the aforementioned species, several key characteristics were most compatible with M. koi. In valvular view, spores were elongate and pyriform with a rounded posterior, 15.4 (14.5–16.5) mm long and 8.3 (7.1–9.0) mm wide. Polar capsules were pyriform and elongate, 10.1 (9.0–10.9) mm long and 3.1 (2.5–3.5) mm wide. Polar filaments were coiled perpendicular to the long axis of the spore making 10 turns (9–11). A BLAST search using a generated 18S SSU rDNA sequence resulted in no direct matches. Phylogenetic analysis of the sequence indicates a close relationship to M. longisporus and M. toyamai from C. carpio, within the clade of Myxobolus species infecting gills of cyprinid fishes (89% or greater bootstrap support by maximum parsimony and minimum evolution distance analysis). Although published descriptions are inconsistent, morphometric similarities to this isolate suggest a morphotype of M. koi. The description of M. koi is supplemented here with new data on pathogenicity and spore morphology and is characterized at the molecular level using 18S small subunit rDNA sequence data.

With its worldwide introduction to temperate waters as a food MATERIALS AND METHODS and ornamental fish, the common carp (koi), Cyprinus carpio, has Origin of parasite become one of the most widely distributed freshwater fish species in North America. As such, an in depth knowledge of its In July 2008, the body of a 13.3-cm, 53.5-g koi was received for necropsy by the Department of Pathology, University of Georgia College pathogens is of significant importance to fisheries biologists of Veterinary Medicine, Athens, Georgia. The fish was 1 of 25 koi that (Dykova et al., 2003). One such group of pathogens includes the had died from an original group of 45 over a 1-mo period. The 23,000-L primitive metazoans belonging to the , including pond contained multiple fish species, but mortalities had occurred only in representatives of Myxobolus. the koi. Before submission, the pond had been treated with furanace, a At present, there are 29 Myxobolus species known to parasitize nitrofuran antibiotic used for bacterial infections, but fish had continued to die following 1 to 2 days of anorexia. Water quality parameters were C. carpio, of which 17 infect the gills (Eiras et al., 2005; Lom and reported as good, but specifics were not given. No additional history was Dykova´, 2006). Most reports of Myxobolus species in the provided, and the origin of the fish is unknown. A request for an on site common carp are from Europe and Asia, and at present only visit of the pond was declined by the owner. M. koi and M. toyamai have been reported from the United States (Kudo, 1917; Hensley, 1975; Eiras et al., 2005). Reports of M. koi Gross andhistological examination provide inconsistent morphometric descriptions (Crawshaw and The fish was placed in right lateral recumbency and the left opercular Sweeting, 1986; Rukyani, 1990; Yokoyama et al., 1997), and no flap removed. Gill clips and skin scrapings were performed using standard genetic sequence data are available for the parasite. Relatively methods (Stoskopf, 1993). Additional gill arches were removed in their entirety and placed in 70% ethanol. The left abdominal wall was aseptically little has been added to the body of knowledge for M. toyamai removed to permit examination of internal organs, but microbiological since its original description by Kudo (1917), although partial 18S samples were not collected due to the degree of autolysis present. SSU rDNA sequence data can be found in GenBank (accession Representative samples of gill, heart, liver, spleen, head and trunk DQ273268). Myxobolus longisporus has not been reported from kidney, intestine, brain, swim bladder, skin, and muscle were removed by sharp dissection and fixed for 24 hr in 10% neutral buffered formalin. the United States, but more extensive sequence data are available Tissue samples were processed routinely by dehydration in a graded series (GenBank accession AY364637). of ethanol solutions of increasing strength, followed by clearing in a series A koi that had died in an ornamental pond was found to have of xylene, embedding in paraffin, and sectioning at 5 mm. Prepared slides severe gill changes resulting from heavy myxozoan infection. were stained with hematoxylin and eosin (Allen, 1992). Further investigation revealed myxospores with morphometric Wet mount preparation andmorphology features similar to M. toyamai, M. longisporus, and M. koi, yet differing from these myxozoans in 1, or more, key characteristics. Individual pseudocysts (n 5 3) were excised from ethanol preserved gill The present article provides 18S small subunit (SSU) rDNA tissue by sharp dissection, placed on a microscope slide with a drop of physiological saline (0.85% [w/v] of NaCl), and mechanically ruptured. sequence data coupled with a morphological description and The released myxospores were further diluted with saline, coverslipped, comparisons to other morphologically similar Myxobolus species and examined using a BX-50 microscope (Olympus, Tokyo, Japan). isolated from koi, supporting the recognition of this isolate as a Representative images (n 5 25; light microscopy) of pooled myxospores morphotype of M. koi. from the 3 pseudocysts were captured using a SPOT Insight QE digital camera (Diagnostic Instruments, Sterling Heights, Michigan) and analyzed using SPOT Basic 3.1 image analysis software (Diagnostic Instruments). A separate preparation of myxospores was prepared and stained with Lugol’s iodine for visualization of the iodinophilous vacuole. Received 20 March 2009; revised 13 September 2009, revised 5 October 2009; accepted 7 October 2009. Molecular analysis *Thad Cochran National Warmwater Aquaculture Center, Mississippi State University, Mississippi State, Mississippi 38776. Individual pseudocysts (n 5 3) identified microscopically from ethanol DOI: 10.1645/GE-2113.1 preserved samples were excised and placed individually into 1.5-ml

116 CAMUS AND GRIFFIN—AMENDED DESCRIPTION OF MYXOBOLUS KOI 117

TABLE I. Primer sequences.

Primer Sequence (59–39) Direction Sequence location (bp)* Reference

H9 TTACCTGGTCCGGACATCAA Forward 1331–1350 Hanson et al., 2001 H2 CGACTTTTACTTCCTCGAAATTGC Reverse 2021–2024 Hanson et al., 2001 MyxospecF TTCTGCCCTATCAACTWGTTG Forward 301–322 Fiala, 2006 MyxospecR GGTTTCNCDGRGGGMCCAAC Reverse 1214–1195 Fiala, 2006 Genmyxo3 TGATTAAGAGGAGCGGTTGG Forward 982–1001 Griffin et al., 2008 Genmyxo4 GGATGTTGGTTCCGTATTGG Forward 957–976 Griffin et al., 2008 Genmyxo5 TAAGCGCAGCAACTTTGAGA Reverse 617–598 Griffin et al., 2008 ERIB1 ACCTGGTTGATCCTGCCAG Forward 2–20 Barta et al., 1997 ERIB10 CCTCCGCAGGTTCACCTACGG Reverse 2079–2059 Barta et al., 1997

* Sequence location based on the 18S SSU rDNA sequence of Henneguya exilis (AF021881). microcentrifuge tubes containing 600 ml of PuregeneH cell lysis solution highly similar sequences of the NCBI nonredundant nucleotide (nr/nt) (Gentra Inc., Minneapolis, Minnesota). After an initial incubation of database. Preliminary analysis (data not shown) included 75 of the most 10 min at 95 C, 3 ml of proteinase K (20 mg/ml) was added to the lysate, similar sequences (excluding duplicates) identified by the BLAST search. mixed by inversion, and incubated at 56 C overnight. The remainder of the From this initial data set, the number of sequences was reduced to 48, isolation was carried out according to the manufacturer’s suggested which included the 25 most similar Myxobolus species sequences, as well as protocol. The purified genomic DNA was then suspended in 30 mlof sequences from representative members of well-defined groups of closely Puregene DNA hydration solution (10 mM Tris and 1 mM EDTA, related myxobolids (Andree et al., 1999; Salim and Desser, 2000; Kent et pH 7.0–8.0). al., 2001; Molna´r et al., 2002; Eszterbauer, 2004; Fiala, 2006; Ferguson et al., 2008; Iwanowicz et al., 2008; Sze´kely et al., 2009). 18S SSU rDNA gene amplification The following sequences were downloaded and aligned using the Clustal W application of the MEGA4 (Molecular Evolutionary Genetics Analysis, The 18S SSUrDNA gene was first amplified with the universal 4.0) software package (Tamura et al., 2007) and used in the construction eukaryotic primers ERIB1 and ERIB10 (Barta et al., 1997; Fiala, 2006) of a phylogram: Myxobolus alburni (EU567313), Myxobolus algonquinen- (Table I). Nested polymerase chain reaction (PCR) reactions were carried sis (AF378335), Myxobolus ampullicapsulatus (DQ339482), Myxobolus out in duplicate using the generic myxozoan primer sets H2/H9 described by Hanson et al. (2001); MyxospecF-MyxospecR from Fiala (2006); and arcticus (AB353129), Myxobolus bilobus (DQ008579), Myxobolus bramae Genmyxo3, Genmyxo4, and Genmyxo5 from Griffin et al. (2008) (AF507968), Myxobolus carassii (DQ452012), Myxobolus csabai (EU643628), Myxobolus cycloides (DQ439810), Myxobolus cyprini (Table I). The initial 25-ml PCR reaction mixture contained 2.5 mlof TaKaRa TaqH Hot Start Version 10X PCR Buffer (TaKaRa Bio Inc., (AF380140), Myxobolus cyprinicola (DQ439805), Myxobolus dispar Otsu, Shiga, Japan), 5 mM of each deoxynucleotide triphosphate, 10 pmol (AF507972), Myxobolus dogieli (EU003978), Myxobolus dujardini of each primer, 0.625 U of TaKaRa Taq Hot Start Taq polymerase, (DQ439803), Myxobolus ellipsoides (DQ439812), Myxobolus fryeri template, and nuclease-free water to volume. The reaction mixture was (EU346370), Myxobolus gayerae (DQ439809), Myxobolus guanqiaoensis cycled on a PTC-100 thermal cycler (MJ Research, Watertown, (EF690300), Myxobolus hungaricus (AF448444), Myxobolus insidiosus Massachusetts), with an initial denaturation step of 95 C for 10 min (U96494), Myxobolus intimus (AY325285), Myxobolus kisutchi followed by 30 cycles of 95 C for 1 min, 48 C for 1 min, 72 C for 2 min, (EF431919), Myxobolus longisporus (AY364637), Myxobolus macrocapsu- and a final extension step of 72 C for 10 min. Two ml of PCR product from laris (AF507969), Myxobolus margitae (EU598803), Myxobolus martini the initial reaction was used in the nested PCR with the following primer (AF186836), Myxobolus muellericus (DQ439808), Myxobolus muelleri combinations: H2/H9, MyxospecF/MyxospecR, Genmyxo3/H2, and (AY325384), Myxobolus musculi (AF380141), Myxobolus neurotropus Genmyxo4/H2. All reaction components remained the same, except the (DQ846661), Myxobolus obesus (AY325286), Myxobolus parviformis volume of nuclease free water was increased to bring the reaction volume (AY836151), Myxobolus pavlovskii (AF507973), Myxobolus pellicides to 25 ml, and the annealing temperature was increased to 52 C from 48 C. (AF378339), Myxobolus pendula (AF378340), Myxobolus pseudodispar The PCR product was run on a 1.2% agarose gel and stained with (EF466088), Myxobolus pseudokoi (AF186839), Myxobolus squamalis GelstarH nucleic acid stain (Cambrex, East Rutherford, New Jersey) to (U96495), Myxobolus stanlii (DQ779996), Myxobolus tasikkenyirensis confirm the presence of DNA product. (EU643626), Myxobolus terengganuensis (EU643629), Myxobolus toyamai (DQ273628), Henneguya adiposa (EU492929), Henneguya cutanea Sequencing of the PCR product (AY676460), Henneguya exilis (AF021881), Henneguya gurleyi (DQ673465), Henneguya ictaluri (AF195510), and Henneguya sutherlandi The PCR products were purified using the QIAquickH PCR purification (EF191200). Tetracapsuloides bryosalmonae (U70623) served as an kit (QIAGEN, Valencia, California), resuspended in 20 ml Puregene DNA outgroup to root the phylogenetic analysis. Phylogenetic and molecular hydration solution and quantified using a NanoDropH spectrophotometer evolutionary analyses were conducted using maximum parsimony analysis and the accompanying software (NanoDrop Technologies, Inc., Wilming- (MP) with a close-neighbor-interchange (CNI) search level 3 and the ton, Delaware). Each product was sequenced directly 3 times in each direction by dideoxy chain termination sequencing (Sanger et al., 1977) with an ABI Prism Dye Termination Cycle Sequencing kit (Applied Biosystems, Foster City, California) using approximately 40 ng of template per reaction. The products were purified using Centri-Sep Spin columns (Princeton Separations, Adelphia, New Jersey) and analyzed at the USDA Mid-South Area Genomics Laboratory (Stoneville, Mississippi). Excess dye terminators were removed by ethanol/EDTA precipitation as per the ABI protocol. The obtained sequence fragments were assembled using the SeqMan utility of the Lasergene software package (DNASTAR, Inc., Madison, Wisconsin) and a contiguous sequence was generated.

Phylogenetic analysis FIGURE 1. Wet mount preparations of spores of Myxobolus koi in The obtained sequence was compared with similar sequences of valvular (A–C) and sutural (D) views. Line drawing of a typical spore (E) described myxozoans identified using a BLAST (megablast) search for in valvular view is also shown. (Scale for B–D as for A. Bar 5 10 mm). 118 THE JOURNAL OF PARASITOLOGY, VOL. 96, NO. 1, FEBRUARY 2010

TABLE III. Identity of Myxobolus koi SSU rDNA sequence to other Myxobolus spp. SSU rDNA sequences available on GenBank. Sequence ambiguities were removed before calculation. M. . toyamai

Kudo, 1917 DNA sequence identity GenBank Species (%)toM. koi accession no.

M. longisporus 1880/1968 (95.5) AY364637 M.

Ma (1998) M. toyamai 709/750 (94.5) DQ273268 Cyprinus carpio longisporus Chen and M. carassi 662/702 (94.3) DQ452012 M. guanqiaoensis 1488/1580 (94.2) EF690300 M. ampullicapsulatus 1455/1581 (92.0) DQ339482

M. M. bilobus 1792/1956 (91.6) DQ008579 6.5–7.0 6.5–7.0 7–8 M. pellicides 1788/1962 (91.1) AF378339 longisporus 1992

Nie and Li, M. pendula 1784/1965 (90.1) AF378340

random addition of 100 trees. Minimum evolution (ME) distance analysis M. was also performed using a CNI search level 3, with evolutionary distances longisporus et al., 2003 determined by maximum composite likelihood (Rzhetsky and Nei, 1992;

Dykova Nei and Kumar, 2000). The initial tree for ME analysis was generated using the neighbor-joining algorithm with pairwise gap deletion (Saitou 2) 8.5 7.5–8.2 7.5–8.2 7–8

2.3) 2.5 2.0 2.0and Nei, 3–4 1987; Tamura et al., 2004). Clade support for both analyses was assessed by bootstrapping (1,000 replicates for ME; 100 replicates for MP) isolated from the gills of common carp, (Felsentstein, 1985). Analysis was performed using the MEGA4 (Mole-

M. koi cular Evolutionary Genetics Analysis, 4.0) software package (Tamura et 3.5 (12–15) 15.7 (15.5–16.5) 17.5 16.0–17.5 15 et al., 1997 (small type) al., 2007). Yokoyama Sequence identities to the SSU rDNA sequence of the case isolate were Myxobolus determined for the 8 best sequence matches identified from the aforementioned BLAST search. Sequences were aligned in a pairwise manner using the Clustal W utility of the MEGA4 (Molecular

M. koi Evolutionary Genetics Analysis, 4.0) software package (Tamura et al., et al., 1997

(large type) 2007), and the percent similarities between the case isolate the most similar Yokoyama Myxobolus species were determined.

REDESCRIPTION

Myxobolus koi Kudo, 1920

and Sweeting, 1986 Spore morphology: Fresh spores (n 5 25; light microscopy) elongate in Crawshaw rphologically similar species of valvular view, with pointed anterior end and rounded posterior, 15.4 (14.5–16.5) mm long and 8.3 (7.1–9.0) mm wide. Polar capsules both pyriform and elongate, slightly uneven in length, measuring 10.1 (9.0– 10.9) mm long and 3.1 (2.5–3.5) mm wide. Polar filaments coiled, perpendicular to long axis of spore body, making 10 turns (9–11) 1952 from Yokoyama et al., 1997 (Fig. 1). Polar capsules converged anteriorly but did not reach spore apex. Hoshina, Intercapsular appendix, iodinophilous vacuole and mucous coat not observed.

Taxonomic summary

from Yokayama et al., 1997 Type host: Cyprinus carpio. Nakai, 1926 Type locality: Gray, Georgia (33u579190N, 83u229590W). Site of infection: Gills.

1920 Prevalence: Unknown. M. koi M. koi M. koi M. koi Type specimens: Histological sections are deposited in the parasitology collection, and spores preserved in 95% ethanol will be deposited in the parasitology collection of the Queensland Museum, Brisbane, Australia.

Remarks

Case isolate The case isolate most closely resembles M. toyamai, M. longisporus, and M. koi in that they develop within the gills of C. carpio and possess similar pyriform-shaped spores that overlap in ranges of measurements of 2, or more, critical morphological characteristics. Polar capsules of the case

m) 15.4 (14.5–16.5) 14–16 10–13 10–13.4 14.1 (11.7–16.1) 13.3 (12.5–15) 1 isolate, although slightly uneven in length, are still relatively equal in size, m) 8.3 (7.1–9.0) 8–9 6–7 6–7.6 7.1 (6.0–8.0) 7.9 (7–9) 6.3 (5–7.5) 6.7 (6–8) m m m)m) 10.1 (9.0–10.9) 8–9 3.1 (2.5–3.5) 2.5–3 5–7 2–2.5 5.4–7.2 2.4–2.9unlike 7.5 (5.8–8.6) 3.0 (2.4–3.6)M. 6.9 (6.7–7.4) toyamai 2.2 6.8 (2.0–2.7) (5.9–7. 2.1 (1.6– , which possesses 2 polar capsules that are significantly m m different in size (Shul’man, 1966). Myxobolus toyamai also has a prominent iodinophilous vacuole and concave spore body in sutural view II. Morphological features and measurements (range in parentheses) of mo (Kudo, 1920; Shul’man, 1966), neither of which was observed in this case. capsules ( filament coils 10 — — — — 7–8 7–8 9 (8–10) — — — capsules ( The polar capsules of this isolate are significantly longer than most ABLE Spore length ( T Spore width ( Width of polar No. of polar GenBank accession no.Reference FJ841887 — This study — Kudo, — — — — AY364637 — — DQ273268 Length of polar reported for M. koi (Nakai, 1926; Hoshina, 1952; Crawshaw and CAMUS AND GRIFFIN—AMENDED DESCRIPTION OF MYXOBOLUS KOI 119

FIGURE 2. Phylogenetic tree gener- ated by maximum parsimony analyses of the 18S SSU rDNA sequences of myxosporeans, rooted at Tetracapsu- loides bryosalmonae. Numbers at nodes indicate bootstrap confidence values (100 replicates). Host species are listed in parentheses; black dia- monds indicate isolation from the gills of cyprinid fishes (M., Myxobolus; H., Henneguya; T., Tetracapsuloides).

Sweeting, 1986; Yokoyama et al., 1997) but average only slightly longer than wider and possesses much longer polar capsules (Nie and Li, 1992; Dykova et that originally reported by Kudo (1920). This isolate also lacks the prominent al., 2003). A comparison of all descriptions of these morphologically similar iodinophilous vacuole noted by Shul’man (1966) and possesses 10 turns of Myxobolus species is presented in Table II. the polar filament, in contrast to the 7–8 reported by Yokoyama et al. (1997). Phylogenetic analysis of the 1962-bp sequence of the 18S SSU rDNA Compared with M. longisporus, the spore body of this species is significantly gene clearly distinguishes this isolate from all other Myxobolus species 120 THE JOURNAL OF PARASITOLOGY, VOL. 96, NO. 1, FEBRUARY 2010

FIGURE 3. Phylogenetic tree gener- ated by minimum evolution analyses of the 18S SSU rDNA sequences of myxosporeans, rooted at Tetracapsu- loides bryosalmonae. Numbers at nodes indicate bootstrap confidence values (1,000 replicates). Units of evolutionary distance are in number of base substitutions per site. Host species is listed in parentheses; black diamonds indicate isolation from the gills of cyprinid fishes (M., Myxobolus; H., Henneguya; T., Tetracapsuloides). sequenced to date. The obtained sequence was no more than 95% similar phylogenetic inference placed the case isolate within a distinct clade of to any available 18S SSU rDNA sequences available from GenBank Myxobolus species that includes 2 well-recognized groups parasitizing the (Table III), including M. toyamai and M. longisporus. There are no 18S gills of cyprinid fishes (100% bootstrap support by ME and MP analysis). SSU rDNA sequence data available for M. koi. Both methods of This isolate is within a smaller group containing M. longisporus and M. CAMUS AND GRIFFIN—AMENDED DESCRIPTION OF MYXOBOLUS KOI 121

FIGURE 4. Alignment of 18S SSU rDNA sequences emphasizing 2 of 3 diagnostic variable regions. No data are available for Myxobolus toyamai and M. carassii for variable region 3. Periods (.) designate conserved regions; dashes (2) designate gaps or missing data. Nucleotide positions are based on the 18S SSU rDNA gene sequence of M. longisporus (AY364637). toyamai from the gills of koi and M. ampullicapsulatus, M. guanqiaoensis, revealed small numbers of colonial protozoans, consistent with species of and M. carassii from the gills of goldfish, Carassius auratus (97% Epistylus. Internal organs were moderately autolyzed, and no gross lesions bootstrap support by ME; 89% support by MP) (Figs. 2, 3). were evident. A multiple alignment of sequences from Myxobolus species obtained Microscopically, approximately 85% of the normal gill lamellar from GenBank revealed 3 variable regions within the 18S SSU rDNA architecture was effaced by large numbers of plasmodia that markedly locus, similar to those described previously by Iwanowicz et al. (2008). expanded the width of affected filaments (Fig. 6A). The irregularly Pairwise comparisons revealed multiple bases differing between the case shaped, closely apposed plasmodia ranged in size from approximately 50 isolate and sequences of the closely related Myxobolus species (Fig. 4). to 750 mm in their longest dimension and often appeared to coalesce (Fig. 6B). Plasmodia were limited by amorphous, 4–10-mm, pale Gross andhistological examination eosinophilic capsules separated by thin fibrovascular septa interpretedas remnants of lamellar capillaries. Although typically filled by mature The fish was in fair physical condition. Gills were characterized grossly spores, multinucleated pansporoblasts were occasionally noted immedi- by widespread, multifocal to coalescing, 2–3 mm, nodular white foci ately internal to the capsule. Isolated plasmodia were rare, making precise (Fig. 5). Gill clip wet mounts revealed a myriad of pyriform and bipolar localization of early development difficult to determine, but appeared to myxospores, both free and within plasmodia. Additional findings included proceed in an intralamellar manner such that entire plasmodia were a 2.5 3 2.0 cm focus of shallow ulceration, surrounded by a narrow enveloped by a capillary border (Dykova and Lom, 2007) (Fig. 6C). margin of hyperemia, immediately caudal to the anus. The anal fins were Remaining areas of unaffected lamellae were congested and character- contained within the ulcer and had frayed margins. Scrapings of the fins ized by mild-to-moderate epithelial hyperplasia that partially filled lamellar troughs. Subepithelial tissues in gill filaments were infiltrated by scattered small mixed populations of eosinophils, lymphocytes, and plasma cells. Additional findings included the presence of rare gill monogeneans. Examination of internal organs revealed no significant findings, but interpretation was hindered by post-mortem autolysis.

DISCUSSION

The death of this koi is attributed to respiratory compromise, resulting from massive of the gills by a myxobolid-type myxozoan with morphological features most compatible with M. koi, a species for which no sequence data are available in GenBank. Although differences in key morphological character- istics exist between this isolate and several descriptions of M. koi (Nakai, 1926; Hoshina, 1952; Crawshaw and Sweeting, 1986; Yokoyama et al., 1997), it varies only slightly from the original report on the species (Kudo, 1920). The variability seen in earlier descriptions suggests that many morphotypes may exist for M. FIGURE 5. Gross appearance of a koi gill affected by Myxobolus koi. There are multifocal 2–3-mm white nodules composed microscopically of koi. Alternatively, the disparities may have resulted from artifacts aggregates of myxosporidial plasmodia. induced by the method of spore preservation or the identification 122 THE JOURNAL OF PARASITOLOGY, VOL. 96, NO. 1, FEBRUARY 2010

of some isolates may simply have been erroneous. The physical similarities of Myxobolus species myxospores make the identifica- tion and differentiation between morphologically similar species difficult. This is especially true for those isolated from the same tissue and type host, eliminating tissue tropism and host specificity as discriminatory characteristics. In such instances, molecular or serological techniques must be used to differentiate between morphologically similar species (Yokoyama et al., 1997; Kent et al., 2001; Eszterbauer, 2002; Hogge et al., 2004; Sze´kely et al., 2009), although molecular data are preferred because closely related species may share similar antigenic determinants. As indicated, classifications based strictly on morphology can result in ambiguous descriptions because many myxozoans have few defining characteristics (Kent et al., 2001), resulting in exaggerated host and tissue ranges for some species. As such, molecular techniques are often the only means for differentiating species that are virtually indistinguishable from a morphological perspective (Salim and Desser, 2000; Eszterbauer, 2002; Hogge et al., 2004, 2008). However, reconciling sequence data with older, and in this instance inconsistent, morphological descriptions can be problematic. At present, there are no universal criteria for discriminating between myxozoan species. Pairwise comparisons and phyloge- netic analysis of this isolate demonstrated the 18S SSU rDNA sequence to cluster within a clade that includes 2 well-recognized groups of gill-infecting Myxobolus species isolated from cyprinid fishes. This supports previous work suggesting tissue tropism correlates to molecular phylogeny (Salim and Desser, 2000; Molna´r et al., 2002; Eszterbauer, 2004; Ferguson et al. 2008; Hogge et al., 2008; Sze´kely et al., 2009). Myxobolus toyamai, M. longisporus, M. carassi, and M. guanqiaoensis all present greater than 94%, but no more than 96%, sequence similarity to the isolate. Although small in scale, these differences are probably significant, considering they are localized within the aforementioned diagnostic variable regions and in excess of the 1–3% variation that is normally attributed to intraspecific variation (Schlegel et al., 1996; Whipps et al., 2003; Cone et al., 2005; Easy et al., 2005). The genetic differences demonstrated by this isolate compared with other Myxobolus species is also significantly greater than the sequence similarity shared by other closely related groups of myxozoans. Molna´r et al. (2002) reported .99% sequence similarities between SSU rDNA sequences from the morphologically similar M. pseudodis- par, M. cyprini, and M. musculi. Similarly, Kudoa minithyrsites and share .98% sequence similarities, as do Kudoa alliaria, Kudoa funduli, and Kudoa clupeidae (Whipps et al., 2003; Whipps and Diggles, 2006); M. pendula and M. pellicides differ by only 8 bases across nearly 2000 bp of SSU rDNA sequence (Kent et al., 2001). The common carp has been introduced throughout temperate freshwaters of North America, and koi continue to be imported as a popular ornamental fish. The importation of exotic fish species, and potentially their parasites, can have significant consequences on indigenous populations. Well-known examples include Myx- FIGURE 6. Hematoxylin- and eosin-stained histological sections of koi gill. obolus cerebralis (Bartholomew and Reno, 2002), Bothriocephalus (A) A gill filament expanded by discrete and coalescing plasmodia that obliterate acheilognathi (Heckmann et al., 1987), and Centrocestus formo- the normal lamellar architecture. (B) Irregular plasmodia limited by thick amorphous capsules separated by thin fibrovascular remnants of lamellar sanus (Mitchell et al., 2005). It is therefore important to chronicle capillaries. (C) Individual plasmodium filled by mature myxospores and associated disease agents of koi for surveillance and diagnostic surrounded by an outer capillary border suggestive of intralamellar development. purposes. CAMUS AND GRIFFIN—AMENDED DESCRIPTION OF MYXOBOLUS KOI 123

It has been suggested repeatedly that molecular sequence data GRIFFIN, M. J., D. J. WISE,A.C.CAMUS,M.J.MAUEL,T.E.GREENWAY, be included when new species are described or when new data are AND L. M. POTE. 2008. A novel Henneguya sp. from channel catfish (Ictalurus punctatus) described by morphological, histological and presented for previously established species (Kent et al., 2001; molecular characterization. Journal of Aquatic Health 20: Lom and Dykova´, 2006; Whipps and Diggles, 2006). The 127–135. morphological, histological, and molecular information provided HANSON, L. A., D. LIN,L.M.POTE, AND R. SHIVAJI. 2001. Small subunit here offers a definitive identification of a morphotype of M. koi, rRNA gene comparisons of four actinosporean species to establish a polymerase chain reaction test for the causative agent of proliferative eliminating ambiguities that can result from identifications based gill disease in channel catfish. Journal of Aquatic Animal Health 13: on morphology alone. For the first time, 18S SSU rDNA 117–123. sequence data and type specimens have been submitted to an HECKMANN, R. A., P. D. GREGER, AND J. E. DEACON. 1987. New host international depository. Submitted to GenBank under the records for the Asian fish tapeworm, Bothriocephalus acheilognathi,in accession FJ841887, this genetic information has been used to endangered fish species from the Virgin River, Utah, Nevada, and Arizona. Journal of Parasitology 73: 226–227. establish phylogenetic relationships with other similar myxobolid HENSLEY, G. H. 1975. Parasites of fishes from Sacramento-San Joaquin species. Delta, California. California Fish and Game 61: 201–208. HOFFMAN, G. L. 1970. Intercontinental and transcontinental dissemina- LITERATURE CITED tion and transfaunation of fish parasites with emphasis on whirling disease (Myxosoma cerebralis). In A symposium on diseases of fish ALLEN, T. C. 1992. Hematoxylin and eosin. In Armed Forces Institute of and shellfish, S. F. Snieszko (ed.). American Fisheries Society, Pathology Laboratory methods in histotechnology, E. B. Prophet, B. Washington, D.C., p. 69–81. Mills, J. B. Arrington, and L. H. Sobin (eds.). American Registry of HOGGE, C., M. CAMPBELL, AND K. JOHNSON. 2004. Discriminating between Pathology, Washington, D.C., p. 53–58. a Neurotropic Myxobolus sp. and M. cerebralis, the causative agent ANDREE, K. B., C. SZE´ KELY,K.MOLNA´ R,S.J.GRESOVIAC, AND R. P. of salmonid whirling disease. Journal of Aquatic Animal Health 16: HEDRICK. 1999. Relationships among members of the 137–144. Myxobolus (Myxozoa: ) based on small subunit ribosomal ———, ———, AND ———. 2008. A new species of myxozoan DNA sequences. Journal of Parasitology 85: 68–74. () from the brain and spinal cord of rainbow trout BARTA, J. R., D. S. MARTIN,P.A.LIBERATOR,M.DASHKEVICZ,J.W. (Oncorhynchus mykiss) from Idaho. Journal of Parasitology 94: 218– ANDERSON,S.D.FEIGHNER,A.ELBRECHT,A.PERKINS-BARROW,M.C. 222. JENKINS,H.D.DANFORTH, ET AL. 1997. Phylogenetic relationships HOSHINA, T. 1952. Notes on some myxosporidian parasites of fishes of among eight Eimeria species infecting domestic fowl inferred using Japan. Journal of Tokyo University of Fisheries 39: 69–89. complete small subunit ribosomal DNA sequences. Journal of IWANOWICZ, L. R., D. D. IWANOWICZ,L.M.POTE,V.S.BLAZER, AND W. B. Parasitology 83: 262–271 SCHILL. 2008. Morphology and 18S rDNA of Henneguya gurlei BARTHOLOMEW, J., AND P. RENO. 2002. The history and dissemination of (Myxosporea) from Ameiurus nebulosus (Siluriformes) in North whirling disease. American Fisheries Society Symposium 29: 3–24. Carolina. Journal of Parasitology 94: 46–57. CHEN, C., AND C. MA. 1998. Fuana Sinica. Myxozoa, Myxosporea. KENT, M. L., K. B. ANDREE,J.L.BARTHOLOMEW,M.EL-MATBOULI, Science Press, Beijing, China, 993 p. S. S. DESSER,R.H.DEVLIN,S.W.FEIST,R.P.HEDRICK,R.W. CONE, D. K., J. YANG,G.SUN, AND R. EASY. 2005. and HOFFMAN,R.W.KHATTRA, ET AL. 2001. Recent advances in our molecular phylogeny of Myxobolus bilobus n. sp. (Myxozoa) knowledge of the Myxozoa. Journal of Eukaryotic Microbiology 48: parasitizing Notemigonus crysoleucas (Cyprinidae) in Algonquin 395–413. Park, Ontario, Canada. Diseases of Aquatic Organisms 66: 227–232. KUDO, R. 1917. Contributions to the study of parasitic protozoa. II. CRAWSHAW, M. T., AND R. A. SWEETING. 1986. Myxobolus koi Kudo, 1919: Myxoboolus toyamai nov. spec., a new myxosporidian parasite in A new record for Britain. Journal of Fish Diseases 9: 465–467. Cyprinus carpio L. Journal of Parasitology 3: 163–170. DYKOVA, I., I. FIALA, AND P. NIE. 2003. New data on Myxobolus ———. 1920. Studies on Myxosporidia: A synopsis of genera and species longisporus (Myxozoa: ), a gill infecting parasite of carp, of Myxosporidia. Illinois Biological Monographs 5: 1–265. Cyprinus carpio haematopterus, from Chinese lakes. Folia Parasito- LOM, J., AND I. DYKOVA´ . 2006. Myxozoan genera: Definition and notes on logica 50: 263–268. taxonomy, life cycle, terminology and pathogenic species. Folia ———, AND J. LOM. 2007. Histopathology of protistan and myxozoan Parasitologica 53: 1–36. infections in fishes: An atlas. Academia, Praha, Czech Republic, MITCHELL, A. J., R. M. OVERSTREET, AND T. M. BRANDT. 2005. Spread of 219 p. an exotic fish-gill trematode: A far-reaching and complex problem. EASY, R. H., S. C. JOHNSON, AND D. K. CONE. 2005. Morphological and Fisheries 30: 11–16. molecular comparison of Myxobolus procerus (Kudo 1934) and M. MOLNA´ R, K., E. ESZTERBAUER,C.SZEKELY,A.DAN, AND B. HARRACH. intramusculi n. sp. (Myxozoa) parasitizing muscles of the trout-perch 2002. Morphological and molecular biological studies on intramus- Percopsis omiscomaycus. Systematic Parasitology 61: 115–122. cular Myxobolus spp. of cyprinid fish. Journal of Fish Diseases 25: EIRAS, J. C., K. MOLNA´ R, AND Y. S. LU. 2005. Synopsis of the species of 643–652. Myxobolus Butschli, 1882 (Myxozoa: Myxosporea: Myxobolidae). NAKAI, N. 1926. Eine neue Myxosporidienart aus den Kiemem des Systematic Parasitology 61: 1–46 Karpfens. Japanese Journal of Zoology 1: 173. ESZTERBAUER, E. 2002. Molecular biology can differentiate between NEI, M., AND S. KUMAR. 2000. Molecular evolution and phylogenetics. morphologically indistinguishable myxosporean species: Myxobolus Oxford University Press, New York, New York, 333 p. elegans and M. hungaricus. Acta Veterinaria Hungarica 50: 59–62. NIE, D. S., AND L. X. LI. 1992. On the myxosporidians of freshwater ———. 2004. Genetic relationship among gill-infecting Myxobolus species fishes from lake Huama, Hubei province: II. Descriptions of new (Myxosporea) of cyprinids: Molecular evidence of importance of species (Myxosporea: Bivalvulida). Acta Zootaxonomica Sinica 17: tissue-specificity. Diseases of Aquatic Organisms 58: 35–40. 133–150. FIALA, I. 2006. The phylogeny of Myxosporea (Myxozoa) based on small RUKYANI, A. 1990. Histopathological changes in the gills of common carp subunit ribosomal RNA gene analysis. International Journal for (Cyprinus carpio L.) infected with the myxosporean parasite Parasitology 36: 1521–1534. Myxobolus koi Kudo, 1920. Asian Fisheries Science 3: 337–341 FELSENSTEIN, J. 1985. Confidence limits on phylogenies: An approach RZHETSKY, A., AND M. NEI. 1992. A simple method for estimating and using the bootstrap. Evolution 39: 783–791 testing minimum evolution trees. Molecular Biology and Evolution 9: FERGUSON, J. A., S. D. ATKINSON,C.M.WHIPPS, AND M. L. KENT. 2008. 945–967. Molecular and morphological analysis of Myxobolus spp. of SAITOU, N., AND M. NEI. 1987. The neighbor-joining method: A new salmonid fishes with the description of a new Myxobolus species. method for reconstructing phylogenetic trees. Molecular Biology and Journal of Parasitology 94: 1322–1334. Evolution 4: 406–425. 124 THE JOURNAL OF PARASITOLOGY, VOL. 96, NO. 1, FEBRUARY 2010

SALIM, K., AND S. DESSER. 2000. Descriptions and phylogenetic systematics TAMURA, K., J. DUDLEY,M.NEI, AND S. KUMAR. 2007. MEGA4: of Myxobolus spp. from cyprinids in Algonquin Park, Ontario. Molecular evolutionary genetics analysis (MEGA) software version Journal of Eukaryotic Microbiology 47: 309–318. 4.0. Molecular Biology and Evolution 24: 1596–1599. SANGER, F., S. NICKLEN, AND A. R. COULSON. 1977. DNA sequencing with ———, M. NEI, AND S. KUMAR. 2004. Prospects for inferring very chain-terminating inhibitors. Proceedings of the National Academy large phylogenies by using the neighbor-joining method. Pro- of Sciences USA 74: 5463–5467. ceedings of the National Academy of Sciences USA 101: 11030– SCHLEGEL, M., J. LOM,A.STECHMAN,D.BERNHARD,D.LEIPE,I.DYKOVA, 11035. AND M. L. SOGIN. 1996. Phylogenetic analysis of complete small WHIPPS, C. M., R. D. ADLARD,M.S.BRYANT,R.J.LESTER,V.FINDLAY, subunit ribosomal RNA coding regioni of Myxidium lieberkuehni: AND M. L. KENT. 2003. First report of three Kudoa species from Evidence that Myxozoa are Metazoa and related to the Bilateria. eastern Australia: Kudoa thyrsites from Mahi mahi (Coryphaena Archiv fu¨r Protistenkunde 147: 1–9. hippurus), Kudoa amamiensis and Kudoa minithyrsites n. sp. from SHUL’MAN, S. S. 1966. Myxosporidia of the USSR. Nauka Publishers, Sweeper (Pempheris ypsilychnus). Journal of Eukaryotic Microbiol- Moscow, Russia, 631 p. ogy 50: 215–219. STOSKOPF, M. 1993. Fish medicine. W. B. Saunders Company, Philadel- ———, AND B. J. DIGGLES. 2006. Kudoa alliaria in flesh of Argentinian phia, Pennsylvania, 882 p. hoki Macruronus magellanicus (Gadiformes; Merlucciidae). Diseases SZE´ KELY, C. S., F. SHAHAROM-HARRISON,G.CECH,G.OSTOROS, AND K. of Aquatic Organisms 69: 259–263. MOLNA´ R. 2009. Myxozoan infections in fishes of the Tasik Kenyir YOKOYAMA, H., D. INOUE,A.KUMAMARU, AND H. WAKABAYASHI. 1997. Water Reservoir, Terengganu, Malaysia. Diseases of Aquatic Myxobolus koi (Myxozoa: Myxosporea) forms large- and small-type Organisms 83: 37–48. ‘cysts’ in the gills of common carp. Fish Pathology 32: 211–217.