190

J. Parasitol., 93(1), 2007, pp. 190–191� American Society of Parasitologists 2007

Balb/Cj Male Mice Do Not Feminize after Infection with Larval Taenia crassiceps

Jerry R. Aldridge, Jr., Mary A. Jennette, and R. E. Kuhn*, Department of Biology, Wake Forest University, Box 7325, Reynolda Station,Winston-Salem, North Carolina 27109. *To whom correspondence should be addressed. e-mail: kuhnray@wfu.edu

ABSTRACT: Balb/cJ mice fail to mount an immune response capable ofclearing infection with larval Taenia crassiceps. Additionally, maleBalb/cJ mice display a lag in larval growth of approximately 3 wk ascompared to growth in female mice. It has been reported that male Balb/cAnN mice generate a protective immune response early in infection,and become permissive to larval growth after they feminize (200-foldincrease in serum estradiol and 90% decrease in serum testosterone).To determine if a different strain of Balb/c mice (Balb/cJ) also feminize,serum was collected from infected male mice for 16 wk and levels of17-�-estradiol and testosterone were measured via ELISA. In addition,the mounting responses of 12- and 16-wk infected male mice, as wellas uninfected control mice, were determined after isolation with a fe-male mouse. The results of these experiments show that male Balb/cJmice do not feminize during infection with larval T. crassiceps. Therewas no significant change in serum levels of either 17-�-estradiol ortestosterone during the course of infection (�16 wk). Moreover, therewas no significant decrease in the number of times infected male micemounted the female mouse as compared to uninfected controls. Theseresults suggest that there may be variances between the substrains ofBalb/c mice that lead to the phenotypic differences reported for maleBalb/cJ and Balb/cAnN mice.

Balb/cJ mice infected intraperitoneally (i.p.) with larval Taenia cras-siceps fail to mount an immune response sufficient to clear the infection.The rate of larval growth, however, is notably different between thegenders of Balb/cJ mice early in infection. Larvae reproduce more rap-idly in the female host, with female mice harboring significantly morelarvae than infected males at wk 3, 9, and 12 postinfection (PI) (datanot shown). Larval expansion within the male host appears to lag ap-proximately 3 wk behind the female host. This sexual dimorphism inparasite growth has been observed in other parasitic infections (Zuk andMcKean, 1996; Klein, 2004).

It has been reported that male Balb/cAnN mice infected with larvalT. crassiceps feminize between 8 and 12 wk PI (Bojalil et al., 1993;Terrazas et al., 1994; Larralde et al., 1995; Morales-Montor et al., 2001),and that male mice display a 90% decrease in serum testosterone anda 200-fold increase in serum estradiol at 2–3 mo PI. In addition, themounting response of males in the presence of a female mouse ceasedduring chronic infections, and these males reportedly become sterile(Bojalil et al., 1993; Terrazas et al., 1994; Larralde et al., 1995; Morales-Montor et al., 2001). The latter authors suggest that the male host dis-plays a protective phenotype against larval expansion and only becomespermissive to larval reproduction after feminization. Here, we reportthat male Balb/cJ mice do not feminize after larval infection as wasreported for male Balb/cAnN mice, indicating possible differences be-tween the substrains of Balb/c mice.

Balb/cJ mice were from The Jackson Laboratory (Bar Harbor, Maine)and were bred and maintained in the animal facilities of Wake ForestUniversity in compliance with institutional animal care guidelines. Micewere infected at 6–8 wk of age. The ORF strain of T. crassiceps (Free-man, 1962) was used for infections. Larvae were obtained from theperitoneal cavity of chronically infected (�4 mo) female Balb/cJ miceand were washed 5 times with an equal volume of ice-cold phosphate-buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4,1.47 mM KH2PO4 [pH 7.4]) under sterile conditions. Mice were infectedby injecting 10 nonbudding (�2 mm) larvae in 0.7 ml PBS into theperitoneal cavity using a 20-gauge needle.

Three-week and 6-wk infected male mice, as well as chronically in-fected male mice, were bled from the orbital plexus following etheranesthesia. Serum was also collected from age-matched, uninfectedmale Balb/cJ mice. To collect the serum, blood was allowed to clot for1 hr at room temperature, followed by 2 hr on ice. Serum was isolatedafter centrifugation at 1,000 g and stored at �80 C.

Serum collected from infected and uninfected male Balb/cJ mice wasanalyzed for the presence of testosterone and 17-�-estradiol via ELISA(ALPCO Diagnostics; Salem, New Hampshire) according to manufac-turer’s instructions.

Male Balb/cJ mice (at either 12 or 16 wk PI) were placed individuallyin a cage with a mature female Balb/cJ mouse. Uninfected, age-matchedmale Balb/cJ mice were treated similarly. Male mice were allowed tostay in the cage with the female mouse for 15 min, and the number oftimes the male mounted the female in an attempt to breed was recorded.Additionally, the individual female mice were observed to determine ifthe females were inseminated and pups were delivered.

The data from these experiments were analyzed with 1-way analysisof variance (ANOVA). Tukey’s honestly significant difference (HSD)post-hoc test was used in cases where there was homogeneity of vari-ances, and the Games–Howell post-hoc test was used when there wereunequal variances. SPSS� 14.0 for Windows (Chicago, Illinois) wasused for analysis.

To determine if male Balb/cJ mice feminize after long-term infectionwith larvae, serum steroid levels of infected and uninfected male micewere determined at 3–6 wk PI and at 4 mo PI. The was no significantchange in either 17-�-estradiol or testosterone levels in serum (ANO-VA; Games–Howell post-hoc test) at any time PI as compared to age-matched, uninfected control male mice (Figs. 1, 2).

To determine if male Balb/cJ mice lose their mounting response afterlong-term infections, 12- and 16-wk infected male mice, as well as age-matched, uninfected male mice were placed in a cage with individualfemale mice, and the number of times the male mounted the femaleduring a 15-min time period was recorded. The results of these exper-iments show that the number of times that 12-wk infected male Balb/cJ mice (n � 4) mount the female mouse (average of 2.25 mounts/mouse) is significantly higher than uninfected controls (average 0.5mounts/mouse; Table I; P � 0.05: ANOVA, Tukey’s HSD post-hoctest). There was no significant difference in the number of times 16-wkinfected male mice (n � 4) mounted the female (average of 0.5 mounts/mouse) mouse as compared to uninfected, control male mice (n � 4)(Table I). Additionally, 2 of the 4 female mice that bred with eitheruninfected or 12-wk infected male mice had pups 3 wk after breeding.One of the 4 female mice that bred with 16-wk infected male miceproduced pups.

In contrast to previously published studies using Balb/cAnN mice(Bojalil et al., 1993; Terrazas et al., 1994; Larralde et al., 1995; Morales-Montor et al., 2001, 2004), the results of these experiments show thatmale Balb/cJ mice do not feminize after long-term infection with larvalT. crassiceps. There was no significant change in serum 17-�-estradiollevels through 16 wk of infection.

In addition to serum levels remaining constant during infection, maleBalb/cJ mice infected for 12 wk mounted the female mouse more fre-quently than uninfected controls. There was no significant difference inthe number of times 16-wk infected male mice mounted the femalemouse as compared to uninfected male mice. Moreover, there was nosignificant difference between infected and uninfected male mice intheir capacity to breed successfully. Infected male Balb/cJ mice, there-fore, do not become sterile as was reported for Balb/cAnN males (Lar-ralde et al., 1995) as a result of chronic infection. These results, com-bined with the steroid levels from infected males, demonstrate that maleBalb/cJ mice do not feminize during the course of infection.

It is unlikely that these differences are due to variations in the larvaebecause the same ORF strain of T. crassiceps was used in studies onBalb/cAnN mice (Terrazas et al., 1994; Larralde et al., 1995; Moraleset al., 1996; Morales-Montor et al., 2001) and the present study. It ispossible, however, that the differences are because of variability be-tween the 2 strains of mice used. We used Balb/cJ mice, whereas studiesshowing feminization used Balb/cAnN mice (Bojalil et al., 1993; Ter-

RESEARCH NOTES 191

FIGURE 1. Male Balb/cJ mice were infected i.p. with 10 nonbuddinglarvae of Taenia crassiceps. Total serum 17-�-estradiol levels from 3-and 6-wk infected male mice, as well as chronically infected (�4 mo)male mice was determined via ELISA. The results show that there areno significant differences in serum estradiol levels between normal andinfected Balb/cJ male mice.

TABLE I. Balb/cJ male mice were infected i.p. with 10 nonbuddinglarvae. Infected (12 wk or 16 wk PI) or normal male mice were indi-vidually placed in a cage with a female mouse, and the number of timesthe male mounted the female in 15 min was recorded. The results showthat there was not a significant decrease in the number of times thatinfected male Balb/cJ (both 12 wk and 16 wk PI) mice mounted thefemale mouse.

Number of times male mounted female in 15 min

Uninfected 12 wk PI 16 wk PI

Male 1 1 3 0Male 2 1 3 2Male 3 0 1 0Male 4 0 2 0Average 0.5 2.25 0.5

FIGURE 2. Male Balb/cJ mice were infected i.p. with 10 nonbuddinglarvae of Taenia crassiceps. Total serum testosterone levels from 3- and6-wk infected male mice, as well as chronically infected (�4 mo) malemice was determined via ELISA. The results show that infection withlarval T. crassiceps does not lead to a decrease in serum testosteronelevels.

razas et al., 1994; Larralde et al., 1995; Morales-Montor et al., 2001,2004). There are notable differences between the strains of mice. Onedisparity between these substrains of Balb/c mice is that Balb/cAnNmice lack the expression of the major histocompatibility class I–likeprotein Qa-2 (Fragoso et al., 1996, 1998). Other possible genetic dis-tinctions between the substrains are poorly defined. Therefore, it is pos-sible that there are uncharacterized genetic differences between the sub-strains of mice that are inducing the phenotypic variability observed.

It also is worth noting that host gender has been shown to influenceH-2 haplotype-dependent immune responses during infection with Mo-loney leukemia virus (M-MuLV). Storch and Chused (1984) showedthat H-2d positive Balb/cJ and Balb/cAnN display a sexual dimorphism,with the male mice demonstrating higher resistance to M-MuLV–in-duced lymphomas than female mice. In contrast, H-2k positive CBA/Nand CBA/CaHN fail to display a sexual dimorphism in susceptibility toM-MuLV induced lymphomas. This may partially explain the differencein parasite growth between the genders of host during infection withlarval T. crassiceps. Together, these findings suggest that certain MHChaplotypes may allow for variations in the efficacy of immune responsesbetween the genders, with some haplotypes promoting gender-specificdifferences in immune responses, whereas others do not. Further studiescharacterizing the differences between the substrains of Balb/c mice arenecessary to better compare the course of infection in each.

Funding for this project was provided in part by the Grady Britt Fundof the Department of Biology, Wake Forest University.

LITERATURE CITED

BOJALIL, R., L. I. TERRAZAS, T. GOVEZENSKY, E. SCIUTTO, AND C. LAR-RALDE. 1993. Thymus-related cellular immune mechanisms in sex-associated resistance to experimental murine cysticercosis (Taeniacrassiceps). Journal of Parasitology 79: 384–389.

FRAGOSO, G., E. LAMOYI, A. MELLOR, C. LOMELI, T. GOVEZENSKY, AND

E. SCIUTTO. 1996. Genetic control of susceptibility to Taenia cras-siceps cysticercosis. Parasitology 112(Pt 1): 119–124.

———, ———, ———, ———, M. HERNANDEZ, AND E. SCIUTTO.1998. Increased resistance to Taenia crassiceps murine cysticer-cosis in Qa-2 transgenic mice. Infection and Immunity 66: 760–764.

FREEMAN, R. S. 1962. Studies on the biology of Taenia crassiceps.Canadian Journal of Zoology 40: 969.

KLEIN, S. L. 2004. Hormonal and immunological mechanisms mediatingsex differences in parasite infection. Parasite Immunology 26: 247–264.

LARRALDE, C., J. MORALES, I. TERRAZAS, T. GOVEZENSKY, AND M. C.ROMANO. 1995. Sex hormone changes induced by the parasite leadto feminization of the male host in murine Taenia crassiceps cys-ticercosis. Journal of Steroid Biochemistry and Molecular Biology52: 575–580.

MORALES, J., C. LARRALDE, M. ARTEAGA, T. GOVEZENSKY, M. C. RO-MANO, AND G. MORALI. 1996. Inhibition of sexual behavior in malemice infected with Taenia crassiceps cysticerci. Journal of Para-sitology 82: 689–693.

MORALES-MONTOR, J., S. BAIG, R. MITCHELL, K. DEWAY, C. HALLAL-CALLEROS, AND R. T. DAMIAN. 2001. Immunoendocrine interactionsduring chronic cysticercosis determine male mouse feminization:Role of IL-6. Journal of Immunology 167: 4527–4533.

———, A. CHAVARRIA, M. A. DE LEON, L. I. DEL CASTILLO, E. G.ESCOBEDO, E. N. SANCHEZ, J. A. VARGAS, M. HERNANDEZ-FLORES,T. ROMO-GONZALEZ, AND C. LARRALDE. 2004. Host gender in par-asitic infections of mammals: An evaluation of the female hostsupremacy paradigm. Journal of Parasitology 90: 531–546.

STORCH, T. G., AND T. M. CHUSED. 1984. Sex and H-2 haplotype controlthe resistance of CBA-BALB hybrids to the induction of T celllymphoma by Moloney leukemia virus. Journal of Immunology133: 2797–2800.

TERRAZAS, L. I., R. BOJALIL, T. GOVEZENSKY, AND C. LARRALDE. 1994.A role for 17-beta-estradiol in immunoendocrine regulation of mu-rine cysticercosis (Taenia crassiceps). Journal of Parasitology 80:563–568.

ZUK, M., AND K. A. MCKEAN. 1996. Sex differences in parasite infec-tions: Patterns and processes. International Journal for Parasitology26: 1009–1023.

192 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

J. Parasitol., 93(1), 2007, pp. 192–196� American Society of Parasitologists 2007

The Phylogenetic Position of Allocreadiidae (Trematoda: Digenea) From PartialSequences of the 18S and 28S Ribosomal RNA Genes

Anindo Choudhury, Rogelio Rosas Valdez, Ryan C. Johnson, Brian Hoffmann, and Gerardo Perez-Ponce de Leon*, Division of NaturalSciences, St. Norbert College, 100 Grant Street, De Pere, Wisconsin 54115; *Instituto de Biologıa, Universidad Nacional Autonoma de Mexico,Ap. Postal 70-153, C.P. 04510 Mexico, D.F., Mexico. e-mail: anindo.choudhury@snc.edu

ABSTRACT: Species of Allocreadiidae are an important component ofthe parasite fauna of freshwater vertebrates, particularly fishes, and yettheir systematic relationships with other trematodes have not been clar-ified. Partial sequences of the 18S and 28S ribosomal RNA genes from3 representative species of Allocreadiidae, i.e., Crepidostomum cooperi,Bunodera mediovitellata, and Polylekithum ictaluri, and from 79 othertaxa representing 78 families of trematodes obtained from GenBank,were used in a phylogenetic analysis to address the relationships ofAllocreadiidae with other plagiorchiiforms/plagiorchiidans. Maximumparsimony and Bayesian analyses of combined 18S and 28S rRNA genesequence data place 2 of the allocreadiids, Crepidostomum cooperi andBunodera mediovitellata, in a clade with species of Callodistomidae andGorgoderidae, which, in turn is sister to a clade containing Polylekithumictaluri and representatives of Encyclometridae, Dicrocoelidae, and Or-chipedidae, a grouping supported by high bootstrap values. These re-sults suggest that Polylekithum ictaluri is not an allocreadiid, a conclu-sion that is supported by reported differences between its cercaria andthat of other allocreadiids. Although details of the life cycle of callod-istomids, the sister taxon to Allocreadiidae, remain unknown, the rela-tionship of Allocreadiidae and Gorgoderidae is consistent with theirlarval development in bivalve, rather than gastropod, molluscs, and withtheir host relationships (predominantly freshwater vertebrates). The re-sults also indicate that, whereas Allocreadiidae is not a basal taxon, itis not included within the suborder Plagiorchiata. No support was foundfor a direct relationship between allocreadiids and opecoelids either.

The Allocreadiidae (Looss, 1902) (Trematoda: Digenea: Plagiorchii-formes) is an important component of the parasite fauna of freshwatervertebrates, particularly fishes (Yamaguti, 1971; Bykovskaya and Ku-lakova, 1987; Caira, 1989; Thatcher, 1993; Perez-Ponce de Leon et al.,1996; Hoffman, 1999). Adults of the family typically possess an un-spined tegument, a well-developed cirrus sac, tandem gonads, and rel-atively large eggs. The cercariae are typically ophthalmoxiphidocercar-iae that develop in bivalves rather than in gastropods (Hopkins, 1934;Yamaguti, 1975; Gibson, 1996; Caira and Bogea, 2005; Cribb, 2005).The Allocreadiidae has had a checkered systematic past (see Caira andBogea, 2005; Cribb, 2005). A phylogenetic analysis using morpholog-ical characters (Brooks and McLennan, 1993) placed it in a basal po-lytomy with other trematode families united under the Plagiorchiifor-mes. According to their revised classification (Brooks and McLennan,1993), Allocreadiidae is placed as the sole family under the suborderAllocreadiata. This classification, and especially the nomenclature, hasnot been universally accepted. Other workers have preferred to refer tothe order as a much larger Plagiorchiida La Rue, 1957 (Gibson 1996;Olson et al., 2003), which includes many of the orders established byBrooks and McLennan (1993). Gibson (1996) placed Allocreadiidae,along with the Opecoelidae, in the suborder Allocreadioidea within thePlagiorchiida, a scheme followed with some modification (superfamilyAllocreadiodea) in more recent reference works (Cribb, 2005; Jones etal., 2005). Irrespective of the ordinal nomenclature used, the availableclassification schemes suggest, at the very least, that the allocreadiidsare plagiorchiidans/plagiorchiiforms.

Despite the fact that the allocreadiids are widely distributed in fresh-water fishes, the phylogenetic analyses by Brooks et al. (1985, 1989)and Brooks and McLennan (1993), based on morphology, remain theonly hypothesis of the family’s systematic relationships with other di-geneans. The plagiorchiiforms have been the subject of several molec-ular analyses (Tkach et al., 1999, 2000; Tkach, Pawlowski et al., 2001;Tkach, Snyder et al., 2001; Olson et al., 2003), but these analyses didnot include the Allocreadiidae. The present study investigates the re-

lationships of the family using partial sequences of the 18S and 28Sribosomal RNA genes from representatives of the family and from pub-lished work.

Partial sequences of the 18S and 28S rRNA genes from 82 taxa wereused in our analyses (see Fig. 1). These include a representative speciesfrom each family (see Fig. 2) appearing in Olson et al. (2003), for atotal of 79 species from that study, plus 3 allocreadiids, i.e., Crepidos-tomum cooperi Hopkins, 1931, Bunodera mediovitellata, and Polylek-ithum ictaluri, for which sequences were generated in this study. Rep-resentatives of Crepidostomum and Bunodera were chosen because theirstatus as allocreadiids is not in doubt (Hopkins, 1934; Caira, 1989; Cairaand Bogea, 2005). The status of Polylekithum ictaluri as an allocreadiidhas been controversial because of its reportedly possessing a cercariathat is very different from that of other allocreadiids (Cable, 1952; Cairaand Bogea, 2005). It was included in this study with the expectationthat the analyses would throw some light on its classification as well.Crepidostomum cooperi was collected from the bluegill Lepomis ma-crochirus in Mud Lake, Chambers Island, Wisconsin, during the sum-mer of 2004; B. mediovitellata was collected from the three-spine stick-leback, Gasterosteus aculeatus in British Columbia (see Choudhury andLeon-Regagnon, 2005); and P. ictaluri was collected from ictalurid cat-fishes, Ameiurus melas and Ictalurus punctatus, in southern Manitoba(see Platta and Choudhury, 2006).

Samples were washed with saline and stored refrigerated in 95% or100% ethanol. The worms were digested and the genomic DNA ex-tracted using the Qiagen DNEasy extraction kit (Qiagen Inc., Valencia,California). Genes of interest were amplified using polymerase chainreaction (PCR) techniques. The 28s rRNA gene was amplified using 1of 2 primer combinations: the digl2 forward primer (5� AAG CAT ATCACT AAG CGG 3�) and the LSU1500 reverse primer (5� GCT ATCCTG AGG GAA ACT TCG 3�) (Tkach et al., 1999, 2000; Snyder andTkach, 2001) for Crepidostomum cooperi or the forward primer 29sy(5� CTA ACC AGG ATT CCC TCA GTA ACG GCG AGT 3�) andthe reverse primer 28sz (5� AGA CTC CTT GGT CCG TGT TTC AAGAC 3�) (Leon-Regagnon et al., 1999) for Polylekithum ictaluri and Bun-odera mediovitellata. The 18S rRNA gene was amplified using the for-ward primer ‘Worm A’ (5� GCG AAT GGC TCA TTA AAT CAG 3�)and the reverse primer ‘Worm B’ (5� CTT GTT ACG ACT TTT ACTTCC 3�) (Littlewood and Olson, 2001; Olson et al., 2003). PCR reac-tions were performed on a PerkinElmer GeneAmp 9700 thermocycler(PerkinElmer Inc., Wellesley, Massachusetts) using Ex-taq DNA poly-merase (TaKaRa Mirus Corporation, Madison, Wisconsin) in a totalreaction volume of 50 �l. The amplification protocol consisted of aninitial denaturing cycle of 5 min at 94 C; 40 cycles of the following:94 C for 30 sec, 50 C or 54 C for 30 sec for primer annealing, and 72C for 2 min for replication; and a final hold for elongation at 72 C for5 min. PCR products were purified using the QIAquick PCR Purifica-tion Kit (Qiagen Inc.).

Purified products were sent to MCLab (South San Francisco, Cali-fornia) for automated sequencing. Each sequence was manually editedfor accuracy using ABI Editview (Perkin-Elmer) or FinchTV (GeospizaInc., Seattle, Washington). Sequences have been deposited to Genbank(www.1ncbi.nih.gov) with the following accession numbers: C. cooperi(28S: EF202098; 18S: EF202097), P. ictaluri (28S: DQ189999, 18S:EF202096), and B. mediovitellata (28S: EF202573, 18S: 202095). Se-quences were aligned using Clustal W alignment software in MEGA3.1 (Molecular Evolutionary Genetic Analysis, Kumar et al., 2004,www.megasoftware.net), checked by eye, and edited to remove extraportions (overhang) due to the unequal lengths of the various sequences.

The aligned 18S dataset consisted of 947 characters (including spac-

RESEARCH NOTES 193

FIGURE 1. Maximum parsimony phylogram from the phylogenetic analysis of the combined 18S and 28S rRNA gene (partial sequence) datasetof digenean taxa including representatives of Allocreadiidae (bold). Numbers following the taxa are GenBank accession numbers for their 18Sand 28S rRNA gene sequences. Numbers at the internodes are bootstrap values (above) and Bremer decay values (below).

194 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

FIGURE 2. Tree from the Bayesian analysis on the combined 18S and 28S rRNA gene (partial sequence) dataset of digenean taxa includingrepresentatives of Allocreadiidae (bold), with posterior probability values. Family names follow the species names.

RESEARCH NOTES 195

es) and the aligned 28S dataset consisted of 680 characters (includingspaces). The combined aligned 18S 28S dataset comprised 1,627characters (including spaces), of which 777 were parsimony informa-tive. The combined dataset was analyzed using maximum parsimony asimplemented in PAUP* 4.0 b10 (Phylogenetic Analysis Using Parsi-mony, Swofford, 2002) and by Bayesian methods (Huelsenbeck andRonquist, 2001). Unweighted maximum parsimony analysis was per-formed with character states unordered, using a heuristic search, withaddseq�random, nreps�100, swap�tbr. A nonparametric bootstrapwith 1,000 replicates was used to evaluate the robustness of the cladesthrough nodal support. Bremer decay indices were also calculated. Forthe analysis in PAUP, the ‘Diplostomata’ (represented by 9 species inBrachylaimidae, Leucochloridiidae, Diplostomidae, Strigeidae, Clinos-tomidae, Sanguinicolidae, Schistosomatidae, Spirorchiidae) of Olson etal. (2003) were used as outgroups in keeping with their basal positionin previous phylogenetic studies (Tkach et al., 2000; Olson et al., 2003).Bayesian estimates of phylogeny were generated using MrBayes 3.04b(Huelsenbeck and Ronquist, 2001) with a substitution model that wasselected as best-fit model by Modeltest (Posada and Crandall, 1988).Because the18S and 28S sequences are linked but actually representdifferent loci, they were entered into Modeltest separately and the mod-els were unlinked prior to running the analysis. The maximum likeli-hood model employed 6 substitution types (‘‘nst � 6’’) and a gammadistribution for variation across sites. The Markov chain Monte Carlosearch was run with 4 chains for 1,000,000 generations, sampling theMarkov chain every 1,000 generations, and the sample points of thefirst 50,000 generations were discarded as ‘‘burn-in,’’ after which thechain reached stationarity. A 50% majority rule consensus tree wascomputed in PAUP 4.0 b10 (Swofford, 2002).

The maximum parsimony analysis yielded 1 tree, 7,970 steps long,with a CI � 0.24, RI � 0.45, RC � 0.108, and HI � 0.76. The analysisreturned this tree 66 times out of the 100 random replicates. The tree(Fig. 1) indicates that 2 species of Allocreadiidae, C. cooperi, and B.mediovitellata, form a clade that is sister to Prosthehystera obesa (Cal-lodistomidae), whereas the third allocreadiid, P. ictaluri, is in a separateclade as a sister taxon to the encyclometrid, Encyclometra colubrimu-rorum. These relationships are supported by high (�99) bootstrap nodalsupport. The 2 allocreadiids C. cooperi and B. mediovitellata, and thecallodistomid, P. obesa, are in turn related to the gorgoderid, Gorgoderasp., a relationship supported by lower bootstrap and Bremer decay val-ues (Fig 1.). Polylekithum ictaluri belongs to a clade that includes, inaddition to its sister taxon, E. colubrimurorum, a dicrocoelid, Brachy-lecithum lobatum, and an orchipedid, Orchipedum tracheicola, a group-ing that is very weakly supported (Fig. 1). All of these stated relation-ships from the maximum parsimony analysis (Fig. 1) are supported byBayesian analysis (Fig. 2) that shows an identical tree topology for thesetaxa. The results from the parsimony analysis also indicates that thesister taxa, Macivaria macassarensis (Opecoelidae) and Opistholebesamplicoelus (Opistholebetidae), form a clade that is sister to the cladecomprising the allocreadiids and the representatives of Encyclometri-dae, Dicrocoelidae, Orchipedidae, Gorgoderidae, and Callodistomidae,but with low support (Fig. 1).

The results suggest that the 3 allocreadiids (C. cooperi, B. mediovi-tellata, and P. ictaluri) used in this study do not constitute a naturalgroup. Although this may raise questions about the monophyly of Al-locreadiidae, there is already other evidence to suggest that P. ictaluridoes not belong in Allocreadiidae. The status of P. ictaluri as an allo-creadiid has been in some doubt ever since Seitner (1951) described thecercaria of P. ictaluri as being gymnocephalous, oculate, lacking a sty-let, and developing in gastropods. This is in contrast to other allocrea-diid cercariae, which are ophthalmoxiphidocercariae and develop insphaeriid bivalves (Hopkins, 1934; Caira, 1989). This was discussed byCable (1952), who suggested, however, that Seitner (1951) had erred indescribing the cercariae of some other digenean, such as Skrjabinop-solus manteri, a deropristiid of the lake sturgeon. The status of P. ic-taluri was also discussed by Caira and Bogea (2005), who suggestedthat either the allocreadiids do not share a common cercaria morphologyand first intermediate bivalve host or P. ictaluri is not an allocreadiid.The results of the present study suggest the latter and also suggest thatthe species should be added to Encyclometridae. It is not surprisingthen that P. ictaluri also placed basally in the more restricted study byPlatta and Choudhury (2006). For the remainder of the discussion, we

regard the clade C. cooperi B. mediovitellata as representative ofAllocreadiidae.

The major finding of this study, i.e., the sister relationship of 2 well-established species of Allocreadiidae with a representative of the Cal-lodistomidae (P. obesa) is surprising, given its morphology, but hasbeen suggested before based on previous unpublished molecular anal-yses (S. Curran, pers. comm.). Prosthenhystera obesa is a widespreadparasite of the gall bladder of mainly characid fishes in the middleAmerican and neotropical region (Thatcher, 1993; Kohn et al., 1997;A. Choudhury, unpubl. obs.). The genus is also represented farther northby ‘Prosthenhystera oonastica’, a species named by Wilmer Rogersfrom the channel catfish, Ictalurus punctatus, in Alabama (USNPC75499, 75500), but apparently not formally described. Unfortunately,details of the life cycle of Prosthenhystera spp. or of others in thisfamily remain unknown (to our knowledge), which prevents furthercomparison with the biology of allocreadiids. However, the typical adulthabitat of the callodistomids, the gall bladder of fishes (Bray, 2002), isnot unknown in the biology of allocreadiids; immature Bunodera lu-ciopercae and mature Crepidostomum wikgreni are found in the gallbladders of perch and whitefish, respectively (Cannon, 1971; Gibsonand Valtonen, 1988).

The implication that the clade Callodistomidae Allocreadiidae to-gether is closely related to the Gorgoderidae is supported by some as-pects of the biology of gorgoderids, if not their typical adult habitat(urinary bladders). Like allocreadiids, the gorgoderids also use bivalvemolluscs, rather than gastropods, as first intermediate hosts (Yamaguti,1975). Cable (1952) also noted the similarities between the cercariaexcretory vesicles of allocreadiids and gorgoderids. Both families arealso particularly common in freshwater environments (Yamaguti, 1971;Cribb, 1987; Hoffman, 1999) and both are parasites of aquatic verte-brates, mainly fishes, although some are parasitic in amphibians as well.This latter fact is particularly true of the Gorgoderidae, where numerousspecies parasitize the urinary bladders of ranid frogs. The Allocreadi-idae is not as speciose in amphibians, from which 2 monotypic generarepresented by Bunoderella metteri Schell, 1964 and Caudouterina rhy-acotritoni Martin, 1966, are known.

This study also raises questions about the composition of the super-family Allocreadioidea Looss, 1902 and the relationships of the typefamily Allocreadiidae with other families that have been placed in thissuperfamily, particularly Opecoelidae (Cribb, 1987). The close relation-ship of the Opecoelidae and Allocreadiidae has long been assumedbased on adult morphology but Cribb’s (2005) succinct and insightfulreview of the topic, in conjunction with Cable’s (1956) detailed discus-sions question such an assumption. The results of the present studyseem to provide more evidence that the 2 families are not closely re-lated. The idea (Cable, 1956; see also Cribb, 2005) that the Opecoelidaeand Opistholebetidae warrant a superfamily separate from Allocreadioi-dea is also well supported by the results of the present study, particularlybecause Allocreadiidae is arguably more closely related to Callodistom-idae, Gorgoderidae, Encyclometridae, and Dicrocoelidae than it is toOpecoelidae.

We thank Russ Feirer, Division of Natural Sciences, St. Norbert Col-lege, for advice regarding techniques and material support. We are alsograteful to Virginia Leon-Regagnon, Universidad Nacional Autonomade Mexico, Mexico City, Mexico, for her help and advice on variousaspects of molecular systematics. A.C. is also thankful for a discussionon this subject with Tom Cribb, University of Queensland, Australia,and with Stephen S. Curran, Gulf Coast Research Laboratory, OceanSprings, Mississippi. A.C. wishes to thank Patrick Nelson, North-SouthConsultants, Winnipeg, Manitoba, Canada, and Kevin Campbell, De-partment of Zoology, University of Manitoba, Winnipeg, Manitoba,Canada, for help with field work leading up to this study. This researchwas supported by an SNC Student-Faculty collaborative grant to A.C.,B.H., and R.C.J., by an SNC Faculty Summer grant to A.C., and bygrants from the Programa de Apoyo a Proyectos de Investigacion eInnovacion Tecnologica (PAPIIT-UNAM) IN-220605 and Consejo Na-cional de Ciencia y Tecnologıa (CONACyT) 47233 to G.P.PD.L.

LITERATURE CITED

BRAY, R. A. 2002. Family Callodistomidae Odhner, 1910. In Keys tothe Trematoda, vol. 1, A. Jones, R. A. Bray, and D. I. Gibson (eds.).CABI International Press, Cambridge, U.K., p. 255–260.

196 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

BROOKS, D. R., R. T. O’GRADY, AND D. R. GLEN. 1985. Phylogeneticanalysis of the Digenea (Platyhelpmithes: Cercomeria) with com-ments on their adaptive radiation. Canadian Journal of Zoology 63:411–443.

———, S. M. BANDONI, C. A. MACDONALD, AND R. T. O’GRADY. 1989.Aspects of the phylogeny of the Trematoda Rudolphi, 1808 (Platy-helminthes: Cercomeria). Canadian Journal of Zoology 67: 2609–2624.

———, AND D. A. MCLENNAN. 1993. Parascript. Parasites and the lan-guage of evolution. Smithsonian Institution Press, Washington,D.C., 429 p.

BYKOVSKAYA, I. E., AND A. P. KULAKOVA. 1987. Klass Trematody—Trematoda Rudolphi, 1808. In Opredelitel’ Parazitov Presnovod-nykh Ryb Fauni SSSR, vol. 3, O.N. Bauer (ed.). NAUKA, Lenin-grad, Russia, p. 77–198.

CABLE, R. M. 1952. On the systematic position of the genus Deropristis,of Dihemistephanus sturionis Little, 1930, and of a new digenetictrematode from a sturgeon. Parasitology 42: 85–91.

———. 1956. Opistholebes diodontis n. sp., its development in the finalhost, the affinities of some amphistomatous trematodes from marinefishes, and the allocreadioid problem. Parasitology 40: 38.

CANNON, L. R. G. 1971. The life cycles of Bunodera sacculata and B.luciopercae (Trematoda: Allocreadiidae) in Algonquin Park, On-tario. Canadian Journal of Zoology 49: 1417–1429.

CAIRA, J. N. 1989. A revision of the North American papillose Allo-creadiidae (Digenea) with independent cladistic analyses of larvaland adult forms. Bulletin of the University of Nebraska State Mu-seum 11: 1–58.

———, AND T. BOGEA. 2005. Family Allocreadiidae. In Keys to theTrematoda, A. Jones, R. A. Bray, and D. I. Gibson (eds.). CABIInternational Press, Cambridge, U.K., p. 417–436.

CHOUDHURY, A., AND V. LEON-REGAGNON. 2005. Molecular phylogenet-ics and biogeography of Bunodera spp. (Trematoda: Allocreadi-idae), parasites of percid and gasterosteid fishes. Canadian Journalof Zoology 83: 1540–1546.

CRIBB, T. H. 1987. Studies on gorgoderid digeneans from Australianand Asian freshwater fishes. Journal of Natural History 21: 1129–1172.

———. 2005. Superfamily Allocreadiodea Looss, 1902. In Keys to theTrematoda, vol. 2, A. Jones, R. A. Bray, and D. I. Gibson (eds.).CABI International Press, Cambridge, U.K., p. 413–416.

GIBSON, D. I. 1996. Trematoda. In Guide to the parasites of fishes ofCanada. Part IV, L. Margolis and Z. Kabata (eds.). Canadian Spe-cial Publication of Fisheries and Aquatic Sciences, Ottawa, Canada,p. 1–373.

———, AND E. T. VALTONEN. 1988. A new species of Crepidostomum(Allocreadiidae: Digenea) from north-eastern Finland, with com-ments on its possible origin. Systematic Parasitology 12: 31–40.

HOFFMAN, G. L. 1999. Parasites of North American freshwater fishes,2nd ed. Cornell University Press, Ithaca, New York, 539 p.

HOPKINS, S. H. 1934. The papillose Allocreadiidae. University of IllinoisBiological Monographs 13: 1–80.

HUELSENBECK, J. P., AND F. RONQUIST. 2001. MrBayes: Bayesian infer-ence of phylogenetic trees. Bioinformatics 17: 754–755.

JONES, A., R. A. BRAY, AND D. I. GIBSON. 2005. Keys to the Trematoda,vol 2. CABI International Press, Cambridge, U.K., 768 p.

KOHN, A., B. M. M. FERNANDES, AND M. DE FATIMA D BAPTISTA-FARIAS.1997. Rediscription of Prosthenhystera obesa (Diesing, 1850) (Cal-lodistomidae, Digenea) with new host records and data on mor-

phological variability. Memorias do Instituta Oswaldo Cruz 92:171–179.

KUMAR, S., K. TAMURA, AND M. NEI. 2004. MEGA3: Integrated soft-ware for molecular evolutionary genetics analysis and sequencealignment. Briefings in Bioinformatics 5: 150–163.

LEON-REGAGNON, V., D. R. BROOKS, AND G. PEREZ-PONCE DE LEON.1999. Differentiation of Mexican species of Haematoloechus Loss,1989 (Digenea: Plagiorchiformes): Molecular and morphologicalevidence. Journal of Parasitology 85: 935–946.

LITTLEWOOD, D. T. J., AND P. D. OLSON. 2001. Small subunit rDNA andthe Platyhelminthes: Signal, noise, conflict and compromise. In In-terrelationships of the Platyhelminthes, D. T. J. Littlewood and R.A. Bray (eds.). Taylor and Francis, London, U.K., p. 262–278.

OLSON, P. D., T. H. CRIBB, V. V. TKACH, R. A. BRAY, AND D. T. J.LITTLEWOOD. 2003. Phylogeny and classification of the Digenea(Platyhelminthes: Trematoda). International Journal for Parasitolo-gy 33: 733–755.

PEREZ-PONCE DE LEON, G., L. GARCıA-PRIETO, D. OSORIO-SARABIA, AND

V. LEON-REGAGNON. 1996. Helmintos parasitos de peces de aguascontinentales de Mexico. Serie Listados Faunısticos de Mexico VI.Instituto de Biologıa, Universidad Nacional Autonoma de Mexico,Mexico City, Mexico, 100 p.

PLATTA, C., AND A. CHOUDHURY. 2006. Systematic position and rela-tionships of Paracreptotrematina limi, based on partial sequencesof 28S rRNA and cytochrome c oxidase subunit 1 genes. Journalof Parasitology 92: 411–413.

POSADA, D., AND K. A. CRANDALL. 1998. Modeltest: Testing the modelof DNA substitution. Bioinformatics 14: 817–818.

SEITNER, P. G. 1951. The life history of Allocreadium ictaluri Pearse,1924 (Trematoda: Digenea). Journal of Parasitology 37: 223–244.

SNYDER, S., AND V. TKACH. 2001. Phylogenetic and biogeographicalrelationships among some Holarctic frog lung flukes (Digenea:Haematoloechidae). Journal of Parasitology 87: 1433–1440.

SWOFFORD, D. L. 2002. PAUP*: Version 4.0 beta 10. Software distrib-uted by Sinauer Associates Inc., Sunderland, Massachusetts.

THATCHER, V. E. 1993. Trematodeos neotropicais. Instituto Nacional dePesquisas da Amazonia (INPA), Manaus, Brazil, 553 p.

TKACH, V., B. GRABDA-KAZUBSKA, J. PAWLOWSKI, AND Z. SWIDERSKI.1999. Molecular and morphological evidences for close phyloge-netic affinities of the genera Macrodera, Leptophallus, Metalepto-phallus, and Paralepoderma (Digenea, Plagiorchioidea). Acta Par-asitologica 44: 170–179.

———, J. PAWLOWSKI, AND J. MARIAUX. 2000. Phylogenetic analysisof the suborder Plagiorchiata (Platyhelminthes: Digenea) based onpartial lsrDNA sequences. International Journal for Parasitology30: 83–93.

———, J. PAWLOWSKI, J. MARIAUX, AND Z. SWIDERSKI. 2001. Molecularphylogeny of the suborder Plagiorchiata and its position in the sys-tem of Digenea. In Interrelationships of the Platyhelminthes, D. T.J. Littlewood and R. A. Bray (eds.). Taylor and Francis, London,U.K., p. 186–193.

———, S. D. SNYDER, AND Z. SWIDERSKI. 2001. On the phylogeneticrelationships of some members of Macroderoididae and Ocheto-somatidae (Digenea, Plagiorchiodea). Acta Parasitologica 46: 267–275.

YAMAGUTI, S. 1971. Synopsis of digenetic trematodes of vertebrates.Vols. 1, 2. Keigaku Publishing Co., Tokyo, Japan, 1,074 p.

———. 1975. A synoptical review of life histories of digenetic trem-atodes of vertebrates with special reference to the morphology oftheir larval forms. Keigaku Publishing Co., Tokyo, Japan, 590 p.

RESEARCH NOTES 197

J. Parasitol., 93(1), 2007, pp. 197–198� American Society of Parasitologists 2007

Flotillin-1 Localization on Sporozoites of Eimeria tenella

E. del Cacho M. Gallego, F. Lopez-Bernad, C. Sanchez-Acedo, and J. Quilez, Department of Animal Pathology, Faculty of VeterinarySciences, University of Zaragoza, Miguel Servet 177, Zaragoza 50013, Spain. e-mail: edelcach@unizar.es

FIGURE 1. Immunoblot showing the protein identified by the mono-clonal antibody anti–flotillin-1 in E. tenella sporozoites (lanes 1 and 2)and in chicken brain (positive control, lane 3). Lane 4: negative control(Hsp70). Molecular mass (kDa) is indicated.

ABSTRACT: In an attempt to identify parasite surface components in-volved in the interaction with the host cell, the present research focuseson the rafts of Eimeria tenella that might be involved in the host cellinvasion process. To that end, this study was undertaken to investigatethe expression of flotillin-1, which is an important component and mark-er of lipid rafts at the plasma membrane of sporozoites of E. tenella.The expression of this plasma membrane protein was identified by anantibody that specifically reacts with flotillin-1 and was studied by elec-tron microscopy. Flotillin-1 was found to occur in patches on the surfaceof E. tenella sporozoites. Immunoblot analysis of the total proteins ofthe sporozoites showed only 1 band of approximately 48 kDa. Thisindicates that the antibody exclusively recognized the molecules of flo-tillin-1 expressed on the surface of E. tenella sporozoites. The presenceof flotillin-1 on the cellular membrane of sporozoites predominantly atthe apical tip suggests that flotillin-1 belongs to the invasion machineryof E. tenella.

Rafts are membrane compartments that are rich in glycosphingolipid,cholesterol, and acylated proteins and are resistant to detergent solubi-lization. Lipid rafts are involved in important cell processes, such assignal transduction, apoptosis, cell migration, synaptic transmission, or-ganization of the cytoskeleton, and protein sorting (Ding et al., 2004).In an attempt to identify parasite surface components involved in theinteraction with the host cell, we have focused in this research on therafts of Eimeria tenella that might be involved in the host cell invasionprocess. For this purpose, an antibody that specifically reacts with flo-tillin-1 was used. Flotillin-1 is known as a key structural componentand a marker of lipid rafts (Wakasugi et al., 2004). It is now clear thatlipid raft microdomains containing flotillin-1 act as platforms for con-ducting a variety of cellular mechanisms, and the proteins of the flotil-lin/reggie family participate in cell processes that have been activatedin the rafts. Rafts have been described in a variety of cells, but theyhave never been described in Eimeria. The authors demonstrate by im-munoelectron microscopy and immunoblotting techniques that flotil-lin-1 is expressed on the surface membrane of E. tenella sporozoites.

To conduct immunoelectron microscopy studies, sporozoites werefixed in 0.05% glutaraldehyde in Bouin solution for 60 min at roomtemperature and then embedded in Durcopan water soluble kit (ElectronMicroscopy Sciences, Hatfield, Pennsylvania) according to the manu-facturer’s instructions. Ultrathin sections (40–60 nm) were made withan LKB ultramicrotome (LKB, Bromma, Sweden). Grids were thenplaced on droplets of normal horse serum (blocking reagent, VectastainABC Kit, Vector Laboratories, Burlingame, California) for 10 min toblock nonspecific binding sites. Samples were incubated in the mouseanti–flotillin-1 monoclonal antibody (clone 18, Pharmingen, San Diego,California) as primary antibody at a 1:100 dilution for 18 hr. Afterseveral washes in PBS, samples were incubated in the secondary anti-body, a 40-nm gold-conjugated goat anti-mouse IgG antibody (BritishBioCell International, Cardiff, U.K.) for 2 hr. Thereafter, ultrathin sec-tions, double stained with uranyl acetate and lead citrate, were examinedby electron transmission microscopy. As a control, sections were in-cubated with mouse normal serum instead of the primary antibody, withthe remaining procedure being the same. In these control sections, spo-rozoites were devoid of positive reaction for flotillin-1.

Blot analyses were applied to the sporozoites, which were purifiedfollowing the procedure described in Raether et al. (1995). Extractionof proteins was performed by sonication to disrupt all sporozoites, fol-lowed by centrifugation at 100,000 g for 20 min. The supernatant wasstored in liquid nitrogen until use. Protein samples from sporozoiteswere subjected to SDS-PAGE in a discontinuous gel system on the basisof a standard protocol (Laemmli, 1970). An SDS-reducing sample buff-er containing SDS and 2-mercaptoethanol was used for SDS-PAGE.Electrophoresis was performed at 40 mA. After electrophoresis, the pro-

teins in the gel were transferred electrophoretically to nitrocellulosepaper in a Trans-Blot cell (Bio-Rad Laboratories, Richmond, Califor-nia). Electrophoresis was performed with a transfer buffer at room tem-perature for 18 hr at 40 mV. After the transfer of proteins, excess bind-ing sites on the nitrocellulose paper were blocked by washing the paperin PBS containing 4% blocking reagent (normal horse serum, VectastainABC Kit, Vector Laboratories). The blots were then treated with themouse anti–flotillin-1 monoclonal antibody for 1 hr at room tempera-ture. The blots were rinsed in PBS, then exposed to peroxidase-conju-gated goat anti-mouse IgG. They were then rinsed 3 times in PBS andtreated for peroxidase activity with DAB. The control lines used in theblots were loaded with chicken brain extract (lane 3, positive control)and human heat shock protein 70 (Hsp70; lane 4, negative control).

Immunoblot analysis of the total proteins of the sporozoites showeda band of approximately 48 kDa (Fig. 1). In addition to signaling pro-teins, lipid rafts also have some specific proteins that are generally usedas markers of lipid rafts. These evolutionarily conserved proteins arecomposed of 2 family members, flotillin-1 and flotillin-2, which are alsoknown as reggie-2 and reggie-1, respectively (Volonte et al., 1999).Flotillin-1 and flotillin-2 are encoded by highly conserved genes, withorthologs in mice, rats, humans, fishes, and Drosophila sp. (Lopez-Casas and del Mazo, 2003). Because both proteins show molecular sim-ilarities, it is not surprising that a monoclonal antibody raised against 1of these 2 types of flotillin cross-reacts with the other (Kokuboa et al.,2000). It is known that flotillin-1 protein consists of 427 amino acidresidues, with a predicted molecular weight of 48 kDa, and that flotil-lin-2 has a molecular mass of 42 kDa (Edgar and Polak, 2002). Ourimmunoblotting analysis showed that the anti–flotillin-1 antibody rec-ognized a 48-kDa band corresponding to flotillin-1. No other bandswere seen, which indicates that the antibody exclusively recognized themolecules of flotillin-1 expressed on the surface of E. tenella sporo-zoites.

Flotillin-1–labeled areas were patchy and located on the plasma mem-brane of the sporozoites, with prominent staining in the apex of themicroorganisms (Fig. 2). Eimeria is well known for its invasion process

198 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

FIGURE 2. Immunoelectron micrograph of E. tenella sporozoites.(A) Negative control, wherein no specific antibody was applied, show-ing a sporozoite section devoid of positive reaction for flotillin-1. (B)Sporozoite showing flotillin-1–positive patches on the parasite mem-brane. Note that the expression of flotillin-1 on the cellular membraneof sporozoites concentrates at the apical tip. Bar � 2 �m.

via the apical complex present at the apex of invasive forms. As soonas sporozoites come into contact with a host cell surface, a signal istransduced from the surface to the apex. The signal induces reorienta-tion, microneme exocytosis, apical binding to the host cell, and for-mation of the parasitophorous vacuole (Dubremetz et al., 1998). There-fore, many biological processes are involved in the host cell invasion,including numerous signal transduction pathways, cell migration, or-ganization of the cytoskeleton, protein sorting, and membrane traffick-ing. It has been proposed that flotillin-1 plays a structural role in all ofthese cellular mechanisms (Edidin, 2003). Therefore, the presence offlotillin-1 on the cellular membrane of sporozoites predominantly at the

apical tip suggests that flotillin-1 belongs to the invasion machinery ofE. tenella.

This work was supported by grant A46 from the Research Councilof Aragon, Spain.

LITERATURE CITED

DING, Y., W. H. JIANG, Y. SU, H. Q. ZHOU, AND Z. H. ZHANG. 2004.Expression and purification of recombinant cytoplasmic domain ofhuman erythrocyte band 3 with hexahistidine tag or chitin-bindingtag in Escherichia coli. Protein Expression and Purification 34:167–175.

DUBREMETZ, J. F., N. GARCIA-REGUET, V. CONSEIL, AND M. N. FOUR-MAUX. 1998. Apical organelles and host-cell invasion by Apicom-plexa. International Journal for Parasitology 28: 1007–1013.

EDGAR, A. J., AND J. M. POLAK. 2002. Flotillin-1: Gene structure: cDNAcloning from human lung and the identification of alternative poly-adenylation signals. International Journal of Biochemical and CellBiology 33: 53–64.

EDIDIN, M. 2003. The state of lipid rafts: From model membranes tocells. Annual Review of Biophysics and Biomolecular Structure 32:257–283.

KOKUBO, H., C. A. LEMEREB, AND H. YAMAGUCHIA. 2000. Localizationof fotillins in human brain and their accumulation with the pro-gression of Alzheimer’s disease pathology. Neuroscience Letters290: 93–96.

LAEMMLI, U. K. 1970. Cleavage of structural proteins during the assem-bly of the head of bacteriophage T4. Nature 227: 680–685.

LOPEZ-CASAS, P. P., AND J. DEL MAZO. 2003. Regulation of flotillin-1 inthe establishment of NIH-3T3 cell–cell interactions. FEBS Letters555: 223–228.

RAETHER, W., J. HOFMANN, AND M. UPHOFF. 1995. In vitro cultivationof avian Eimeria species: Eimeria tenella. In Biotechnology.Guidelines on techniques in coccidiosis research, J. Eckert, R.Braun, M. W. Shirley, and P. Coudert (eds.). European Commission,Brussels, Belgium, p. 79–84.

VOLONTE, D., F. GALBIATI, S. LI, K. NISHIYAMA, T. OKAMOTO, AND M.P. LISANTI. 1999. Flotillins/caveolins are differentially expressed incells and tissues and form a heterooligomeric complex with cav-eolins in vivo: Characterization and epitope-mapping of a novelflotillin-1 monoclonal antibody probe. Journal of Biological Chem-istry 274: 12702–12709.

WAKASUGI, K., T. NAKANO, C. KITATSUJI, AND I. MORISHIMA. 2004. Hu-man neuroglobin interacts with Xotillin-1, a lipid raft microdomain-associated protein. Biochemical and Biophysical Research Com-munications 318: 453–460.

J. Parasitol., 93(1), 2007, pp. 198–202� American Society of Parasitologists 2007

Cryptosporidium and Giardia in Marine-Foraging River Otters (Lontra canadensis) Fromthe Puget Sound Georgia Basin Ecosystem

J. K. Gaydos, W. A. Miller*, K. V. K. Gilardi†, A. Melli*, H. Schwantje‡, C. Engelstoft§, H. Fritz*, and P. A. Conrad*†, Orcas Island Office,University of California–Davis Wildlife Health Center, 1016 Deer Harbor Road, Eastsound, Washington 98245; *University of California–Davis,Department of Pathology, Microbiology, and Immunology, School of Veterinary Medicine, One Shields Avenue, Davis, California 95616;†University of California–Davis Wildlife Health Center, School of Veterinary Medicine, One Shields Avenue, Davis, California 95616; ‡BritishColumbia Ministry of Environment, P.O. Box 9338 Stn Prov Govt, Victoria, British Columbia, V8W 9M1 Canada; §Alula Biological Consulting,1967 Nicholas Road, Saanichton, British Columbia V8M 1X8, Canada. e-mail: jkgaydos@ucdavis.edu

ABSTRACT: Species of Cryptosporidium and Giardia can infect humansand wildlife and have the potential to be transmitted between these 2groups; yet, very little is known about these protozoans in marine wild-life. Feces of river otters (Lontra canadensis), a common marine wild-life species in the Puget Sound Georgia Basin, were examined for spe-cies of Cryptosporidium and Giardia to determine their role in the ep-

idemiology of these pathogens. Using ZnSO4 flotation and immuno-magnetic separation, followed by direct immunofluorescent antibodydetection (IMS/DFA), we identified Cryptosporidium sp. oocysts in 9fecal samples from 6 locations and Giardia sp. cysts in 11 fecal samplesfrom 7 locations. The putative risk factors of proximate human popu-lation and degree of anthropogenic shoreline modification were not as-

RESEARCH NOTES 199

sociated with the detection of Cryptosporidium or Giardia spp. in riverotter feces. Amplification of DNA from the IMS/DFA slide scrapingswas successful for 1 sample containing �500 Cryptosporidium sp. oo-cysts. Sequences from the Cryptosporidium 18S rRNA and the COWPloci were most similar to the ferret Cryptosporidium sp. genotype. Riverotters could serve as reservoirs for Cryptosporidium and Giardia spe-cies in marine ecosystems. More work is needed to better understandthe zoonotic potential of the genotypes they carry as well as their im-plications for river otter health.

Cryptosporidium parvum and Giardia duodenalis are protozoans thatcan infect humans as well as numerous species of wildlife and domesticanimals (Fayer et al., 2000; Thompson, 2000). Little is known aboutthe epidemiology of these pathogens at the human–wildlife interface;yet, often wildlife are assumed to be transmitting pathogens to humans,especially Giardia species (Dykes et al., 1980; Erlandsen et al., 1988).Human recreational water quality standards are being developed andenforced for beaches in Washington and other states; yet, virtually noth-ing is known about the epidemiology and the potential for zoonotictransmission of Cryptosporidium and Giardia species in marine eco-systems. Oocysts of C. parvum can survive for at least 1 yr in salt water(Tamburrini and Pozio, 1999) and can be concentrated by filter-feedingbivalves, including mussels (Tamburrini and Pozio, 1999) and oysters(Fayer et al., 1999). Similarly, G. duodenalis cysts can persist in aquaticenvironments and are concentrated by bivalves such as mussels andclams (Graczyk et al., 2003).

Cryptosporidium and Giardia spp. have been identified in Californiasea lions (Zalophus californianus) (Deng et al., 2000) and ringed seals(Phoca hispida) (Olson et al., 1997; Fayer et al., 2004; Hughes-Hankset al., 2005; Santin et al., 2005) as well as in North Atlantic right whales(Eubalaena glacialis) and bowhead whales (Balaena mysticetus)(Hughes-Hanks et al., 2005). Cryptosporidium sp. has been identifiedin dugong (Dugong dugon) (Hill et al., 1997) and Giardia spp. havebeen identified in harp seals (P. groenlandica), grey seals (Halichoerusgrypus), and a harbor seal (P. vitulina) from eastern coastal Canada(Measures and Olson, 1999), but not from other mammals that usemarine waters.

The Puget Sound Georgia Basin (PSGB) marine ecosystem is a high-ly productive inland sea in the transboundary region between Washing-ton state, U.S.A., and British Columbia, Canada (4830�N, 12340�W).Nearly 7 million people reside along the shores of this ecosystem wherehumans, wildlife, and domestic animals share habitat and marine re-sources (Fraser et al., 2006). To determine whether Cryptosporidiumand Giardia species are present in marine wildlife from this ecosystem,we examined the feces of free-ranging marine-foraging river otters(Lontra canadensis) for Cryptosporidium and Giardia spp. River ottersare widely distributed throughout the marine waters of the PSGB wherethey primarily feed on numerous species of marine fish and inverte-brates (Stenson et al., 1984; Jones, 2000). They have a high metabolicrate and defecate at latrine sites on docks and on shorelines (Ben-Davidet al., 1998), presenting the opportunity to evaluate fecal samples col-lected at the land–sea interface around the region. Specifically, our ob-jectives were to (1) determine whether river otters in the region wereinfected with these protozoans, (2) characterize the isolates by usingmolecular tools, and (3) evaluate whether Cryptosporidium and Giardiaspp. infections in river otters are associated with otter proximity tolarger human populations or degree of adjacent shoreline modification.

In Washington state, river otter fecal samples were collected fromdocks and along the marine shoreline between February and June 2003.Fecal samples were identified as river otter based on size, conformation,content, and location, and were collected individually for refrigerationuntil testing. Only samples less than 24 hr old were collected as deter-mined by degree of desiccation relative to weather conditions over thepast few days. To take replicate samples at a spatial location, but avoidrepeat sampling of the same animal, samples were collected from com-munal latrine sites when 3 to 5 fresh fecal samples were depositedduring the last 24 hr and could be individually identified as being fromdifferent animals by gross examination of fish bones and shellfish re-mains in the samples. Latrine study sites were spaced at least 20 kmapart by water (shoreline or open water) or 15 km apart over land.Adjacent sites that were less than 40 km apart were sampled on thesame day or the following day to prevent resampling of the same animalat 2 adjacent sites. Sites were classified by their percentage of shoreline

modification as determined by the Washington state ShoreZone Inven-tory (http://www.sharedsalmonstrategy.org/images/maps/shoreline.jpg)and size of the proximal resident human population by using the 2000U.S. Census (http://quickfacts.census.gov/).

In British Columbia, latrines were identified from land and water.Using the same criteria for identification and freshness, fecal sampleswere collected along the marine shoreline of southern Vancouver Islandfrom the city of Victoria and outlying islets to the community of PortRenfrew between June and November 2004. Individual samples werecollected opportunistically from small as well as communal latrine sitesand stored chilled until testing. No efforts were made to prevent resam-pling of the same animal.

Fresh fecal samples collected in Washington state were tested forCryptosporidium sp. oocysts and Giardia sp. cysts by using 2 tech-niques, i.e., double centrifugation flotation with ZnSO4 (specific gravity1.2) and immunomagnetic separation followed by direct immunofluo-rescent antibody (IMS/DFA) detection (Pereira et al., 1999; Zajac et al.,2002). When positively labeled Cryptosporidium sp. oocysts or Giardiasp. cysts were identified by IMS/DFA, the slide contents were scrapedand washed for DNA extraction and amplification. Cryptosporidium sp.DNA in positive samples was amplified using established primers de-signed specifically to amplify genomic DNA sequences for a segmentof the 18S rRNA gene (Xiao et al., 1999) and a segment of the Cryp-tosporidium sp. oocyst wall protein (COWP) gene (Spano et al., 1997).The primers used on Giardia sp.-positive samples amplified a segmentof the 18S rRNA gene (McGlade et al., 2003). Amplified products weresequenced and aligned using GeneDoc software (Nicholas et al., 1997)for comparison with GenBank reference genotypes.

Fresh fecal samples collected in British Columbia were tested onlyfor Cryptosporidium sp. oocysts and Giardia sp. cysts by using ZnSO4

flotation (specific gravity 1.2) and sucrose flotation (Gajadhar, 1994).Molecular characterization was not performed.

Putative risk factors that could be associated with detection of Cryp-tosporidium sp. or Giardia sp. in river otter feces were identified, andthe data were analyzed for all Washington study sites and time points.The sample collection month was recorded, ranging from Februarythrough June 2003. The coastal human population living adjacent toeach river otter fecal collection site was categorized as low (�5,000people), medium (5–15,000 people), or high (�15,000 people), basedon data from the 2000 Census (http://quickfacts.census.gov/). The extentof shoreline modified for human habitation in the region of each riverotter fecal collection site was categorized as low (�40% modification),medium (41–60% modification), or high (�60% modification). Logisticregression methods were used to assess the strength of association be-tween each putative risk factor and the separate outcomes of detectingspecies of Cryptosporidium or Giardia in river otter feces, while ad-justing for repeated sampling within sites by using a cluster variable.Statistical analyses were conducted using Stata software (Stata Corpo-ration, College Station, Texas), and significant P values were definedas �0.1.

In Washington state, 57 fresh river otter fecal samples were collectedfrom 13 locations throughout the Puget Sound region (Fig. 1). Cryp-tosporidium sp. oocysts were detected in 4 samples from 4 locations(Fig. 1). Giardia sp. cysts were detected in 11 samples from 7 locations(Fig. 1). Fecal samples coinfected with species of Cryptosporidium andGiardia were not detected, although animals infected with Cryptospo-ridium sp. and animals infected with Giardia sp. were detected con-currently at 3 locations (Seattle, Union, and Shelton; Fig. 1). ThreeCryptosporidium sp.-positive samples were detected by IMS/DFA butnot fecal flotation, and 1 positive was detected by flotation but not byIMS/DFA. Of the IMS/DFA-positive samples, 2 positive samples hadfewer than 5 oocysts, and 1 sample contained �500 oocysts. Giardiasp. cysts were detected by IMS/DFA but not by fecal flotation in 6samples and by flotation but not IMS/DFA in 5 samples. Cysts per IMS/DFA sample ranged from as few as 2 to as many as 203. Frank bloodwas noted in the 1 sample containing �200 Giardia sp. cysts.

Amplification of DNA from the IMS/DFA slide scrapings was suc-cessful for the sample containing �500 Cryptosporidium sp. oocystsbut not for the other Cryptosporidium sp. or Giardia sp.-positive sam-ples that contained fewer parasites. For both the Cryptosporidium sp.18S rRNA and the COWP loci, polymerase chain reaction (PCR) prod-ucts were sequenced in 2 directions and found to be most similar to theferret, Cryptosporidium sp., genotype in a nBLAST search and by using

200 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

FIGURE 1. Location and number of Cryptosporidium sp.- (in square) and Giardia sp. (in circle)-positive fecal samples and the total numberof river otter fecal samples collected per site (in triangle). Due to differences in sampling, the number of fecal samples collected is not given forCanadian sites. Actual locations of sample are represented by circles; squares, and triangles are off set slightly for easier reading. Inset identifiesPuget Sound Georgia Basin location relative to Oregon, Washington, and British Columbia.

GeneDoc alignment software. Differences along the 816-bp 18S rRNAlocus were noted when comparing the river otter sequence (GenBankDQ288166) and the ferret Cryptosporidium sp. genotype (AF112572).Specifically, in the region from bp 198 to bp 1017, the river otter 18Ssequence differed from the ferret genotype at bp 229 and in the bp 690-700 variable region. Similarly, differences were noted between the riverotter sequence (GenBank DQ288167) and ferret genotype (AF266267)along the 538-bp COWP locus. The river otter COWP locus differedfrom the ferret genotype with T-C conversions at bp 354 and bp 747bp as well as a C-T conversion at bp 819.

Risk factor analysis of the variables sample collection months, ad-jacent human population, and extent of shoreline modification for as-sociation with detecting species of Cryptosporidium or Giardia in riverotter feces were assessed using the results of all 57 fecal samples andby using a cluster variable to adjust for repeated sampling within sites.None of the putative risk factors was found to be significantly associated(P � 0.1) with detection of Cryptosporidium or Giardia spp. in riverotter fecal samples.

In British Columbia, 36 river otter fecal samples were collected from30 sites along the southern tip of Vancouver Island. Cryptosporidiumsp. oocysts were detected in 5 samples from 2 locations (Fig. 1). Giar-dia sp. cysts were not detected.

Human, companion animal, and agricultural-related fecal material aredischarged, dumped, or carried in runoff into marine waters all over theworld (Fayer et al., 2004). In the PSGB, for example, untreated sewageeffluent from an estimated 210,000 people living on the southern endof Vancouver Island is discharged from 2 marine outfalls averaging80,000 m3/day and 50,000 m3/day in wet winter months (Hodgins etal., 1998), highlighting the need to better understand the epidemiologyof Cryptosporidium and Giardia spp. in marine wildlife and their eco-systems. We identified species of Cryptosporidium and Giardia in ma-rine-foraging river otters throughout the PSGB, highlighting for the first

time that this widely distributed species can be infected with and shedthese protozoans throughout this marine ecosystem. River otters drinkfresh water from streams and culverts that drain urbanized shorelinesand empty into the marine waters of the PSGB. They also defecate ondocks and boats and on shorelines adjacent to where people grow andharvest shellfish, which have been shown to concentrate these potentialpathogens (Fayer et al., 1999; Tamburrini and Pozio, 1999; Graczyk etal., 2003). Because some Cryptosporidium and Giardia spp. have healthimplications and can be transmitted amongst humans and animals, it isimportant to consider whether human discharge could be infecting riverotters or whether river otters could be contaminating shellfish beds andthereby infecting humans.

Four of 57 or 7% of river otters in the marine waters of the PugetSound region were shedding Cryptosporidium sp. at the time of sam-pling, and 11 of 57 or 19% were shedding Giardia sp. In Washington,attempts were made to minimize resampling of the same individual bygross evaluation of fecal contents at each site, by selecting adjacentsites that were at least 20 km apart by water or 15 km apart over land,and by sampling adjacent sites on the same or subsequent days. Dataon marine-foraging river otter home range in Washington are not avail-able, but home range for marine-foraging river otters in Alaska variesbetween 10 and 40 km of shoreline, depending on the sex of the animaland site (Bowyer et al., 2003). Assuming that home range is similar inWashington state, it is highly unlikely that the same animal was sampledat adjacent sites due to the distances between selected sites and thatadjacent sites that were less than 40 km apart were sampled on thesame or following days. The British Columbia river otter survey showedthat a Cryptosporidium sp. also is present in otters using similar habitatsnearby in Canada. Using different laboratory methods and varying sam-ple sizes, other work has identified species of Giardia and Cryptospo-ridium infection of varying prevalence in different species of marinemammals. Giardia sp. prevalence was estimated at 20% (Olson et al.,

RESEARCH NOTES 201

1997) and 64.5% (Hughes-Hanks et al., 2005) in ringed seals, 50% inharp seals (Measures and Olson, 1999), 71.4% in right whales (Hughes-Hanks et al., 2005), and 33.3% in bowhead whales (Hughes-Hanks etal., 2005). Cryptosporidium sp. prevalence ranges from 5.1% in bow-head whales to 22.6% in ringed seals and 24.5% in right whales(Hughes-Hanks et al., 2005).

River otters can be found in urban and rural areas throughout thePSGB. We used human population and the degree of anthropogenicshoreline modification as putative risk factors in an effort to determinewhether human proximity or development could be influencing riverotter infection with species of Cryptosporidium or Giardia. We foundno significant association, but this does not negate the possibility thatanthropogenic factors not evaluated in this study could be involved inmarine mammal infections. Using mussels (Mytilus sp.) and oysters(Crassostrea spp.) as bioindicators of marine and estuarine contami-nation, Miller et al. (2005) and Fayer et al. (2002) found that freshwateroutflow and precipitation events were associated with increased oddsfor detecting Cryptosporidium sp. Due to the relatively large homerange of river otters compared with sessile organisms and lack of in-formation about Giardia and Cryptosporidium spp. infection and shed-ding duration, these environmental risk factors were not evaluated forriver otters. More work is needed to characterize the Cryptosporidiumand Giardia species genotypes found in river otters and other wildlifespecies to better evaluate anthropogenic and environmental factors be-fore this issue can be resolved in the Pacific Northwest.

Analysis of the river otter Cryptosporidium sp. isolate at 2 loci re-vealed a novel genotype that most closely resembles the C. parvumferret genotype (AF266267 and AF112572). Ferrets and river otters areclosely related mustelids, and it is possible that the Cryptosporidium sp.identified represents a genotype unique to river otters. If this genotypeis host adapted to river otters and is the genotype most often associatedwith marine-foraging river otters in the PSGB, then it is less likely thatriver otter infection is zoonotic and associated with human habitation,shoreline modification, or fecal waste discharged into marine waters.Additional biological and molecular characterization is needed to de-termine whether the river otter genotype can be transmitted to humansor other animals. Other Cryptosporidium sp. genotypes have beenshown to be host adapted and only present a major health risk to im-munocompromised humans outside the host range (Gajadhar, 1994;Fayer et al., 2000).

Mere detection of infectious agents does not imply their ability tocause morbidity or mortality in the host, and the significance of Cryp-tosporidium and Giardia spp. infection to the health of marine-foragingriver otters is unknown. Although the potential exists for these organ-isms to cause disease in river otters, neither parasite has been previouslyidentified as a pathogen in this species (Kimber and Kollias, 2000).Frank blood was present in the 1 sample containing �200 Giardia sp.cysts, but this could have been coincidental and does not prove thatGiardia sp. causes disease in river otters. To better understand the roleof Cryptosporidium and Giardia spp. in causing disease, otters fromzoological collections with gastrointestinal disease should be tested forspecies of Cryptosporidium and Giardia by using IMS/DFA as well asZnSO4 flotation. Isolates should be characterized molecularly and theremission of clinical signs should follow treatment with appropriateantiprotozoal medication to confirm protozoal involvement in causingdisease.

This study used the most sensitive methods currently available fordetection of Cryptosporidium sp. oocysts and Giardia sp. cysts in fecalsamples. Traditional methods such as fecal flotation and DFA alonehave been shown to detect 1,000 or more oocysts or cysts/g feces (Xiaoand Herd, 1993). Immunomagnetic separation is a concentration methodthat allows for analysis of a 0.5-ml fecal pellet before DFA quantitation,in contrast to the 10 �l that is analyzed when DFA is used alone. TheIMS method has been shown to improve sensitivity by 1–2 log10 units,from 1,000 oocysts to 10 oocysts/g feces (Pereira et al., 1999). Withthe increased sensitivity of IMS/DFA detection, we can expect to detectboth newly infected high shedding animals as well as chronic low shed-ding animals. In addition, this study used a serial fecal processing pro-tocol to directly combine IMS concentration with quantitative fluores-cent microscopy, followed by DNA amplification, so that molecularcharacterization could be carried out without requiring a separate IMSconcentration step. Amplification of samples positive by fecal flotationwas not attempted because the IMS/DFA preparation was expected to

be more sensitive, specific, and efficient for providing clean DNA tem-plate for PCR analysis. It is interesting that 5 samples were positive forGiardia sp. cysts by fecal flotation but negative by IMS/DFA testing.The Giardia sp. cysts seen in the flotation slide could have been anantigenically different Giardia species than the G. duodenalis againstwhich the IMS antibodies were designed.

Adding marine-foraging river otters to the list of marine species thatcan be infected with and shed Cryptosporidium and Giardia spp. pro-vides support that these protozoan parasites are probably more commonin marine waters and in marine wildlife than was previously thought.Additional work is warranted to better understand the taxonomy, epi-demiology, and zoonotic potential of Cryptosporidium and Giardia spp.in river otters and other marine wildlife.

We thank H. Brunner-Gaydos, S. Campbell, C. Jones, A. Shaffer, B.Waddell, B. Waddington, marina owners, and others who helped collectsamples or granted permission to scout for latrine sites and collect sam-ples on private property. J. Slocomb provided geographic informationsystem assistance. This work was conducted in accordance with allappropriate state, provincial, and federal regulations and policies. It waspartially funded by the Nestucca Trust Fund (Project 64), the MorrisAnimal Foundation (Grants D02ZO-90 and D04ZO-04), and theSeaDoc Society, a marine ecosystem health program of the Universityof California–Davis Wildlife Health Center; a center of excellence atthe University of California–Davis School of Veterinary Medicine.

LITERATURE CITED

BEN-DAVID, M., R. T. BOWYER, L. K. DUFFY, D. D. ROBY, AND D. M.SCHELL. 1998. Social behavior and ecosystem processes: River otterlatrines and nutrient dynamics of terrestrial vegetation. Ecology 79:2567–2571.

BOWYER, R. T., G. M. BLUNDELL, M. BEN-DAVID, S. C. JEWETT, T. A.DEAN, AND L. K. DUFFY. 2003. Effects of the Exxon Valdez oil spillon river otters: Injury and recovery of a sentinel species. WildlifeMonographs 153: 1–53.

DENG, M., R. P. PETERSON, AND D. O. CLIVER. 2000. First findings ofCryptosporidium and Giardia in California Sea Lions (Zalophuscalifornianus). Journal of Parasitology 86: 490–494.

DYKES, A. C., D. D. JURANEK, R. A. LORENZ, S. SINCLAIR, W. JAKU-BOWSKI, AND R. DAVIES. 1980. Municipal waterborne giardiasis: Anepidemiologic investigation. Annals of Internal Medicine 92: 165–170.

ERLANDSEN, S. L., L. A. SHERLOCK, M. JANUSCHKA, D. G. SCHUPP, R.W. SCHAEFER, W. JAKUBOWSKI, AND W. J. BEMRICK. 1988. Cross-species transmission of Giardia spp.: Inoculation of beavers andmuskrats with cysts of human, beaver, mouse, and muskrat origin.Applied and Environmental Microbiology 54: 2777–2785.

FAYER, R., J. P. DUBEY, AND D. S. LINDSAY. 2004. Zoonotic protozoa:From land to sea. Trends in Parasitology 20: 531–536.

———, E. J. LEWIS, J. M. TROUT, T. K. GRACZYK, M. C. JENKINS, J.HIGGINS, L. XIAL, AND A. A. LAL. 1999. Cryptosporidium parvumin oysters from commercial harvesting sites in the Chesapeake Bay.Emerging Infectious Diseases 5: 706–710.

———, U. MORGAN, AND S. J. UPTON. 2000. Epidemiology of Cryp-tosporidium: Transmission, detection and identification. Interna-tional Journal for Parasitology 30: 1305–1322.

———, J. M. TROUT, E. J. LEWIS, L. XIAO, A. LAL, M. C. JENKINS, AND

T. K. GRACZYK. 2002. Temporal variability of Cryptosporidium inthe Chesapeake Bay. Parasitology Research 88: 998–1003.

FRASER, D. A., J. K. GAYDOS, E. KARLSEN, AND M. S. RYLKO. 2006.Collaborative science, policy development and program implemen-tation in the transboundary Georgia Basin/Puget Sound. Environ-mental Monitoring and Assessment 113: 49–69.

GAJADHAR, A. A. 1994. Host specificity studies and oocyst descriptionof a Cryptosporidium sp. isolated from ostriches. Parasitology Re-search 80: 316–319.

GRACZYK, T. K., D. B. CONN, D. J. MARCOGLIESE, H. GRACZYK, AND Y.DE LAFONTAINE. 2003. Accumulation of human waterborne para-sites by zebra mussels (Dreissena polymorpha) and Asian fresh-water clams (Corbicula fluminea). Parasitology Research 89: 107–112.

HILL, B. D., I. R. FRASIER, AND H. C. PRIOR. 1997. Cryptosporidium

202 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

infection in a dugong (Dugong dugon). Australian Veterinary Jour-nal 75: 670–671.

HODGINS, D. O., S. W. TINIS, AND L. A. TAYLOR. 1998. Marine sewageoutfall assessment for the capital regional district, British Colum-bia, using nested three-dimensional models. Water Science andTechnology 38: 301–308.

HUGHES-HANKS, J. M., L. G. RICKARD, C. PANUSKA, J. R. SAUCIER, T.M. O’HARA, L. DEHN, AND R. M. ROLLAND. 2005. Prevalence ofCryptosporidium spp. and Giardia spp. in five marine mammalspecies. Journal of Parasitology 91: 1225–1228.

JONES, C. 2000. Investigations of prey and habitat use by the river otter,Lutra canadensis, near San Juan Island, Washington. M.S. Thesis,Western Washington University, Bellingham, Washington, 47 p.

KIMBER, K. R., AND G. V. KOLLIAS. 2000. Infectious and parasitic dis-eases and contaminant-related problems of North American riverotters (Lontra canadensis): A review. Journal of Zoo and WildlifeMedicine 31: 452–472.

MCGLADE, T. R., I. D. ROBERTSON, A. D. ELLIOT, AND R. C. THOMPSON.2003. High prevalence of Giardia detected in cats by PCR. Vet-erinary Parasitology 110: 197–205.

MEASURES, L. N., AND M. OLSON. 1999. Giardiasis in pinnipeds fromEastern Canada. Journal of Wildlife Diseases 35: 779–782.

MILLER, W. A., M. A. MILLER, I. A. GARDNER, E. R. ATWILL, M. HARRIA,J. AMES, D. JESSUP, A. MELLI, D. PARADIES, K. WORCESTER, P. OLIN,N. BARNES, AND P. A. CONRAD. 2005. New genotypes and factorsassociated with Cryptosporidium detection in mussels (Mytilusspp.) along the California coast. International Journal for Parasi-tology 35: 1103–1113.

NICHOLAS, K. B., H. B. NICHOLAS JR., AND D. W. DEERFIELD, II. 1997GeneDoc: Analysis and visualization of genetic variation. EMB-NEW.NEWS 4: 14.

OLSON, M. E., P. D. ROACH, M. STABLER, AND W. CHAN. 1997. Giardiasisin ringed seals from the Western Arctic. Journal of Wildlife Dis-eases 33: 646–648.

PEREIRA, M. D. G. C., E. R. ATWILL, AND T. JONES. 1999. Comparison

of sensitivity of immunofluorescent microscopy to that of a com-bination of immunofluorescent microscopy and immunomagneticseparation for detection of Cryptosporidium parvum oocysts inadult bovine feces. Applied and Environmental Microbiology 65:3236–3239.

SANTIN, M., B. R. DIXON, AND R. FAYER. 2005. Genetic characterizationof Cryptosporidium isolates from ringed seas (Phoca hispida) innorthern Quebec, Canada. Journal of Parasitology 91: 712–716.

SPANO, F., L. PUTIGNANI, J. MCLAUCHLIN, D. CASEMORE, AND A. CRI-SANTI. 1997. PCR-RFLP analysis of the Cryptosporidium oocystwall protein (COWP) gene discriminates between C. wrairi and C.parvum isolates, and between C. parvum isolates of human andanimal origin. FEMS Microbiology Letters 150: 209–217.

STENSON, G. B., G. A. BADGERO, AND H. D. FISHER. 1984. Food habitsof the river otter Lutra canadensis in the marine environment ofBritish Columbia. Canadian Journal of Zoology 62: 88–91.

TAMBURRINI, A., AND E. POZIO. 1999. Long-term survival of Cryptospo-ridium parvum oocysts in seawater and in experimentally infectedmussels (Mytilus galloprovincialis). International Journal for Par-asitology 29: 711–715.

THOMPSON, R. C. A. 2000. Giardiasis as a re-emerging infectious diseaseand its zoonotic potential. International Journal for Parasitology 30:1259–1267.

XIAO, L., U. M. MORGAN, J. LIMOR, A. ESCALANTE, M. ARROWOOD, W.SHULAW, R. C. A. THOMPSON, R. FAYER, AND A. A. LAL. 1999.Genetic diversity within Cryptosporidium parvum and relatedCryptosporidium species. Applied and Environmental Microbiolo-gy 65: 3386–3391.

———, AND R. P. HERD. 1993. Quantitation of Giardia cysts and Cryp-tosporidium oocysts in fecal samples by direct immunofluorescenceassay. Journal of Clinical Microbiology 31: 2944–2946.

ZAJAC, A. M., J. JOHNSON, AND S. E. KING. 2002. Evaluation of theimportance of centrifugation as a component of zinc sulfate flota-tion examinations. Journal of the American Animal Hospital As-sociation 38: 221–224.

J. Parasitol., 93(1), 2007, pp. 202–204� American Society of Parasitologists 2007

Evidence of Experimental Postcyclic Transmission of Bothriocephalus acheilognathi inBonytail Chub (Gila elegans)

Scott P. Hansen‡, Anindo Choudhury†, and Rebecca A. Cole*, U.S. Geological Survey–National Wildlife Health Center, 6006 Schroeder Rd,Madison, Wisconsin 53711; †Division of Natural Sciences, St. Norbert College, 100 Grant St., De Pere, Wisconsin 54115; ‡ Current affiliation:Wisconsin Department of Natural Resources, 101 S. Webster St., Madison, Wisconsin 53706; * To whom correspondence should beaddressed. e-mail: rebecca�cole@usgs.gov

ABSTRACT: We examined the role that predation of infected conspecificfish and postcyclic transmission might play in the life cycle of the Asianfish tapeworm, Bothriocephalus acheilognathi (Cestoda: Pseudophyllid-ea) Yamaguti, 1934. Young-of-the-year (YOY) bonytail chub (Gila ele-gans) were exposed to copepods infected with B. acheilognathi andsubsequently fed to subadult bonytail chub. Within 1 wk after con-sumption of the YOY chub, subadults were necropsied and found in-fected with gravid and nongravid tapeworms. This study provides evi-dence that postcyclic transfer of B. acheilognathi can occur. Postcyclictransmission may be an important life history trait of B. acheilognathithat merits consideration when studying the impact and distribution ofthis invasive and potentially pathogenic tapeworm.

Postcyclic parasite transmission, as proposed by Bozhkov (1969), oc-curs when an adult parasite is ingested indirectly by its definitive host’spredator and subsequently survives to parasitize the predator (Odening,1976; Nickol, 1985). The phenomenon has been demonstrated experi-mentally in fish for a number of acanthocephalan species (Lassiere andCrompton, 1988; Kennedy, 1999; Rauque et al., 2002; McCormick andNickol, 2004). This mode of parasite transmission was also demonstrat-

ed through mechanical transplantation for different stages of severalspecies of Proteocephalus (Willemse, 1969) and has been suggested tooccur in several other species of this genus (Scholz and Hanzelova,1998). It is also a likely route of transmission for the fish tapewormBothriocephalus pearsei (Scholz et al., 1996). Reports of postcyclictransmission are often suggestive and derive from observations of par-asites infecting fish that are not likely to consume the proper interme-diate host to acquire the infection.

Pseudophyllidean tapeworms utilize a copepod first intermediate hostfor the larval procercoid stage and generally require a fish as the secondintermediate host for the plerocercoid stage. Plerocercoids typically em-bed and develop in muscle or visceral mass of the fish. Larger fish or,in some cases, a mammal is host to the adult stage of the worm. Thelife cycle of the Asian fish tapeworm (Bothriocephalus acheilognathiYamaguti, 1934) (Bothriocephalus gowkongensis Yeh, 1955), however,does not require a second intermediate or paratenic host, and transmis-sion to the final fish host takes place directly via ingestion of copepodsinfected with the procercoid stage (Korting, 1975). Postcyclic transmis-sion has not been reported as a mode of transmission for B. acheilo-gnathi, although small fish have been reported as ‘‘carriers’’ (Hoffman,1999).

RESEARCH NOTES 203

Bothriocephalus acheilognathi is a species native to Japan; however,its distribution has expanded to 5 continents because of its predilectionfor widely transplanted cyprinid fish such as carp and grass carp, aswell as for mosquitofish and several baitfish species (Hoffman, 1999).In the United States, it was discovered in portions of the Colorado Riverwatershed (Heckman et al., 1986; Brouder and Hoffnagle, 1997) andnow infects the federally endangered humpback chub (Gila cypha) asits main host in the Little Colorado River (Choudhury et al., 2004).Pathogenicity has been well documented in juvenile, hatchery-rearedcarp (Liao and Shih, 1956; Bauer et al., 1973). Hansen et al. (2006)demonstrated that infections in bonytail chub (Gila elegans) decreasegrowth by approximately 8% when compared to controls and that 62%of infected fish on a limited food ration suffered mortality. While theseexperimental exposures targeted young fish that select copepods as prey,questions arose as to whether older fish could become infected as aresult of piscivory, including cannibalism. Herein we present results thatdocument postcyclic transmission of B. acheilognathi via cannibalismas a transmission route.

Bonytail chub were acquired from Dexter National Fish Hatchery(DNFH), Dexter, New Mexico, and reared in 76-L tanks at the NationalWildlife Health Center (NWHC), Madison, Wisconsin, until they wereapproximately 26 mo old. These fish were not exposed to B. acheil-ognathi at NWHC and to our knowledge were not exposed at DNFH.Fish were fed Biokyowa� (Kyowa Hakko, New York, New York) com-mercial fish food and some Artemia sp. nauplii ad libitum during thisperiod. Four fish from this group of 22 large chub (hereafter referredto as predator chub) were placed in a 114-L tank for the exposure trial.The 18 remaining chub were withheld from the experimental processand served as controls.

Eight young-of-the-year (YOY) bonytail chub (age 69–71 days; ap-proximately 22-mm fork length; hereafter referred to as prey chub) wereexposed to cyclopoid copepods (Acanthocyclops robustus) infected withB. acheilognathi (as described in Hansen et al., 2006). At 11 days post-exposure, the prey chub were removed and released into the tank con-taining the 4 predator chub. Small-diameter Biokyowa and Artemia sp.nauplii were occasionally added for the prey chub. The predator chubwere not fed throughout the course of the trial until all small chub wereconsumed, after which they were fed Biokyowa ad libitum.

All prey chub released into the predator tank were consumed within23 days. Approximately 1 wk after the last fish was consumed, thepredator fish were necropsied and examined for evidence of postcyclictapeworm infections. Three of the 4 predator chub were infected with1, 1, and 36 B. acheilognathi worms, respectively. Four of the wormsfrom the group of 36 had mature segments containing tanned eggs.Fifteen of the 18 control fish were examined for infections of B. achei-lognathi, and no worms were found.

The present study provides strong evidence that B. acheilognathi canbe transmitted postcyclically from one fish to another. Although smallhumpback chub feed primarily on zooplankton and smaller inverte-brates, the feeding habits become generalist and opportunistic with age,and cannibalism has been reported (Kaeding and Zimmerman, 1983;Stone, 2004; Stone and Gorman, 2006), thereby promoting the chancefor postcyclic transmission. Furthermore, it appears that parasite devel-opment may be important for successful postcyclic transmission sincepredator chub that consumed 5-day, postexposure prey chub were notfound to be infected (data not shown), whereas predator chub that con-sumed 11-day, postexposure prey chub, as reported here, were found tobe infected.

Postcyclic transmission is not necessarily important in the life historyof the parasite unless the worms can be transferred in the mature stateor become sexually mature after transfer (Nickol, 2003). Our observa-tion of worms with tanned eggs from the second postcyclic exposureof 1 predator fish is suggestive of the potential for postcyclic transmis-sion in the life cycle of this tapeworm. It remains unknown if the gravidworm matured in a prey or predator fish since the last prey chub wasconsumed 34 days after being exposed to infected copepods allowing,at the experimental temperatures of approximately 23 C, enough timefor B. acheilognathi to become patent before the postcyclic transmissionevent (Bauer et al., 1973).

Small tributary streams to the Little Colorado River likely act asimportant reservoirs for B. acheilognathi since they provide a viablehabitat for copepod production and are home to an abundance of smallcyprinid species such as speckled dace (Rhinichthys osculus) and fat-

head minnows (Pimephales promelas) (Choudhury et al., 2004). Post-cyclic infections in adult and subadult fish can play an important func-tional role in expanding the distribution of B. acheilognathi outsidethese areas. Larger chub in general will have a greater range than smallcyprinids, including larval and postlarval humpback chub, thus pro-moting the potential for spread of B. acheilognathi to other thermallysuitable portions of the Colorado River watershed.

Results of this study provide insight into the life history evolution ofparasite transmission and reproductive strategies. Greater parasite trans-mission success is generally associated with increased life cycle com-plexity (but see Poulin, 1998), as evidenced in a study of 2 sympatricBothriocephalus species (Morand et al., 1995). If this hypothesis canbe supported for B. acheilognathi, the question of why it maintains asimple life cycle remains. Postcyclic transmission may be an adaptationthat offsets the need for a more complex life cycle and other charac-teristics such as this tapeworm’s low host specificity (Freeman, 1973;Dove et al., 1997; Hoffman, 1999; Salgado-Maldonado et al., 2003).The ‘‘crowding effect’’ from increasing parasite density and competitionin the limited space of host tissue can negatively impact cestode de-velopment inhibiting growth, maturity, and fecundity (Liao and Shih,1956; Jones and Tan, 1971; Davydov, 1978; Kennedy, 1983). In theevent of postcyclic transmission, the relocation of worms to older andlarger fish with greater gut area, higher food intake, and reduced met-abolic rate may alleviate these crowding effects.

Although the impact of B. acheilognathi infections in terms of re-duced survival and growth retardation appear to most heavily affect theearlier developmental stages of fishes, including bonytail chub (Liaoand Shih, 1956; Hansen et al., 2006), the effects on older and largerfish may be more discrete. The prevalence and intensity of B. acheil-ognathi infections have been found to decrease in older carp (Liao andShih, 1956; Bauer et al., 1973; Choudhury et al., 2004), which may beattributed to a change in diet or a more developed immune system.Meanwhile, the piscivorous nature of older humpback chub, unlike carp,enhances the potential for tapeworm infections in mature fish. Our studydemonstrates an additional route for tapeworm acquisition in humpbackchub and emphasizes the need for future research on the potential neg-ative affects this parasite may have on older fish.

We thank Jennifer Phillips for her excellent animal care and technicalassistance. We are also very grateful to Stanislav Stankevitch for hisassistance in translating Russian literature. We also thank Roger Ham-man for providing the bonytail chub and for his expertise in fish hus-bandry.

LITERATURE CITED

BAUER, O. N., V. A. MUSSELIUS, AND Y. A. STRELKOV (EDS.). 1973.Diseases of pond fishes. Israel Program for Scientific Translations,Jerusalem, Israel, p. 115–122.

BOZHKOV, D. K. 1969. Postcyclic parasitism and postcyclic hosts inhelminths. Bulgarian Academy of Sciences 29: 183–189.

BROUDER, M. J., AND T. L. HOFFNAGLE. 1997. Distribution and preva-lence of the Asian fish tapeworm, Bothriocephalus acheilognathi,in the Colorado River and tributaries, Grand Canyon, Arizona, in-cluding two new host records. Journal of the Helminthological So-ciety of Washington 64: 219–226.

CHOUDHURY, A., T. L. HOFFNAGLE, AND R. A. COLE. 2004. Parasites ofnative and non-native fishes of the Little Colorado River, GrandCanyon, Arizona. Journal of Parasitology 90: 1042–1053.

DAVYDOV, O. N. 1978. Growth, development and fecundity of Bothrio-cephalus gowkongensis (Jen, 1955), a cyprinid parasite. Hydrobi-ological Journal 14: 60–66.

DOVE, A. D. M., T. H. CRIBB, S. P. MOCKLER, AND M. LINTERMANS.1997. The Asian fish tapeworm, Bothriocephalus acheilognathi, inAustralian freshwater fishes. Marine and Freshwater Resources 48:181–183.

FREEMAN, R. S. 1973. Ontogeny of cestodes and its bearing on theirphylogeny and systematics. Advances in Parasitology 11: 481–557.

HANSEN, S. P., A. CHOUDHURY, D. HEISEY, J. A. AHUMADA, T. L. HOFF-NAGLE, AND R. A. COLE. 2006. Experimental infection of the en-dangered bonytail chub (Gila elegans) with the Asian fish tape-worm (Bothriocephalus acheilognathi): Impacts on survival,growth, and condition. Canadian Journal of Zoology 84: 1383–1394.

204 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

HECKMAN, R. A., J. E. DEACON, AND P. D. GREGER. 1986. Parasites ofthe woundfin minnow, Plagopterus argentissimus, and other en-demic fishes from the Virgin River, Utah. Great Basin Naturalist46: 662–676.

HOFFMAN, G. 1999. Parasites of North American freshwater fishes, 2nded. Cornell University Press, Ithaca, New York, 539 p.

JONES, A. W., AND B. D. TAN. 1971. Effect of crowding upon growthand fecundity in the mouse bile duct tapeworm Hymenolepis mi-crostoma. Journal of Parasitology 57: 88–93.

KAEDING, L. R., AND M. A. ZIMMERMAN. 1983. Life history and ecologyof the humpback chub in the Little Colorado and Colorado Riversof the Grand Canyon. Transactions of the American Fisheries So-ciety 112: 577–594.

KENNEDY, C. R. 1983. General ecology. In Biology of the Eucestoda,vol. I, C. Arme and P. W. Pappas (eds.). Academic Press, NewYork, New York, p. 27–80.

———. 1999. Post-cyclic transmission in Pomphorhynchus laevis(Acanthocephala). Folia Parasitologia 46: 111–116.

KORTING, W. 1975. Larval development of Bothriocephalus sp. (Ces-toda: Pseudophyllidea) from carp (Cyprinus carpio L.) in Germany.Journal of Fish Biology 7: 727–733.

LASSIERE, O. L., AND D. W. T. CROMPTON. 1988. Evidence for post-cyclic transmission in the life-history of Neoechinorhynchus rutili(Acanthocephala). Parasitology 97: 339–343.

LIAO, H., AND L. SHIH. 1956. Contribution to the biology and controlof Bothriocephalus gowkongensis Yeh, a tapeworm parasitic in theyoung grass carp (Ctenopharyngodon idellus C. & V.). Acta Hy-drobiologica Sinica 2: 129–185.

MCCORMICK, A. L., AND B. B. NICKOL. 2004. Postcyclic transmissionand its effect on the distribution of Paulisentis missouriensis (Acan-thocephala) in the definitive host Semotilus atromaculatus. Journalof Parasitology 90: 103–107.

MORAND, S., F. ROBERT, AND V. A. CONNORS. 1995. Complexity in par-

asite life cycles: Population biology of cestodes in fish. Journal ofAnimal Ecology 64: 245–264.

NICKOL, B. B. 1985. Epizootiology. In Biology of the Acanthocephala,D. W. T. Crompton and B. B. Nickol (eds.). Cambridge UniversityPress, Cambridge, U.K., p. 307–346.

———. 2003. Is postcyclic transmission under estimated as an epizoo-tological factor for acanthocephalans? Helminthologia 40: 93–95.

ODENING, K. 1976. Conception and terminology of hosts in parasitology.Advances in Parasitology 14: 1–93.

POULIN, R. 1998. Evolutionary ecology of parasites: From individualsto communities. Chapman and Hall, London, U.K., 212 p.

RAUQUE, C. A., L. G. SEMENAS, AND G. P. VIOZZI. 2002. Post-cyclictransmission in Acanthocephalus tumescens (Acanthocephala:Echinorhynchidae). Folia Parasitologica 49: 127–130.

SALGADO-MALDONADO, G., AND R. F. PINEDA-LOPEZ. 2003. The Asianfish tapeworm (Bothriocephalus acheilognathi): A potential threatto native freshwater species in Mexico. Biological Invasions 5:261–268.

SCHOLZ, T., AND V. HANZELOVA. 1998. Tapeworms of the genus Proteo-cephalus Weinland, 1858 (Cestoda: Proteocephalidae), parasites offishes in Europe. Studies of the Academy of Sciences, No. 2, Ac-ademia, Prague, Czech Republic, 120 p.

SCHOLZ, T., J. VARGAS-VAZQUEZ, AND F. MORAVEC. 1996. Bothrioceph-alus pearsei n. sp. (Cestoda: Pseudophyllidea) from cenote fishesof the Yucatan Peninsula, Mexico. Journal of Parasitology 82: 801–805.

STONE, D. M. 2004. Differential detection of ingested items evacuatedfrom genus Gila cyprinids by two nonlethal alimentary tract lavagetechniques. Journal of Freshwater Ecology 19: 559–565.

———, AND O. T. GORMAN. 2006. Ontogenesis of endangered hump-back chub (Gila cypha) in the Little Colorado River, Arizona.American Midland Naturalist 155: 123–135.

WILLEMSE, J. J. 1969. The genus Proteocephalus in the Netherlands.Journal of Helminthology 43: 207–222.

J. Parasitol., 93(1), 2007, pp. 204–208� American Society of Parasitologists 2007

Helminth Community of Scaled Quail (Callipepla squamata) From Western Texas

Jill N. Landgrebe, Barbara Vasquez, Russell G. Bradley, Alan M. Fedynich*, Scott P. Lerich†, and John M. Kinsella‡, Caesar KlebergWildlife Research Institute, Texas A&M University-Kingsville, 700 University Boulevard, MSC 218, Kingsville, Texas 78363-8202; †National WildTurkey Federation, P.O. Box 4126, Amarillo, Texas 79116; ‡Department of Pathobiology, College of Veterinary Medicine, University of Florida,Gainesville, Florida 32610-0880; *To whom correspondence should be addressed. e-mail: alan.fedynich@tamuk.edu

ABSTRACT: Forty-eight scaled quail (Callipepla squamata) were col-lected during August 2002 at Elephant Mountain Wildlife ManagementArea in Brewster County, Texas, and examined for helminths. Eightspecies of helminths were found (5 nematodes and 3 cestodes), repre-senting 2,811 individuals. Of these species, Gongylonema sp., Procyr-nea pileata, and Choanotaenia infundibulum are reported from scaledquail for the first time. Prevalence of Aulonocephalus pennula, Gon-gylonema sp., Oxyspirura petrowi, Physaloptera sp., P. pileata, C. in-fundibulum, Fuhrmannetta sp., and Rhabdometra odiosa was 98, 2, 56,4, 60, 2, 25, and 35%, respectively. Aulonocephalus pennula numeri-cally dominated, accounting for 88% of total worms. Statistical analyseswere performed on the 5 species with �25% prevalence using the after-hatch-year host sample (n � 38). Prevalence of P. pileata was higher(P � 0.049) in females than in males and higher (P � 0.037) in thesample collected from the site that had spreader dams (berms 1–2 mhigh and 4–55 m long constructed in varying sizes to catch and retainrainfall) than the control site (no spreader dams). Higher rank meanabundance of A. pennula and O. petrowi (P � 0.0001 and P � 0.0052,respectively) was found in the host sample collected from the site thathad spreader dams than the control site. A host gender-by-collectionsite interaction (P � 0.0215) was observed for P. pileata. Findingsindicate that scaled quail are acquiring indirect life cycle helminths inarid western Texas habitats.

Arid regions may pose a particularly challenging environment forpersistence of helminth populations, especially for direct life cycle spe-cies in which high evaporation rates may decrease survival of exposedinfective stages (Gray et al., 1978). Additionally, helminths that havecomplex life cycles requiring threshold densities of both intermediateand definitive hosts may be at a disadvantage in regions where envi-ronmental conditions limit host densities (Dobson, 1989).

In arid regions of western Texas, spreader dams (berms 1–2 m highand 4–55 m long constructed in varying sizes to catch and retain rain-fall) are being used to increase available surface water and create moistsoil environments for wildlife. However, it is unknown how these hab-itat modifications influence host–parasite systems. For example, con-centrating susceptible hosts around spreader dams could facilitate trans-mission of direct and indirect life cycle helminths. To examine furtherthe influence of arid environments on parasite communities, we assessedthe helminth community in scaled quail (Callipepla squamata) fromwestern Texas. Our specific objectives were to determine the helminthfauna of scaled quail from Elephant Mountain Wildlife ManagementArea (EMWMA) in Brewster County, Texas; to examine the structureand patterns of observed helminth communities; and to relate thesefindings to host–parasite interactions.

Elephant Mountain Wildlife Management Area (Fig. 1; 302�1.5�N,10334�2.9�W) was acquired in 1985 by the state of Texas, with stew-

RESEARCH NOTES 205

FIGURE 1. Location of Elephant Mountain Wildlife ManagementArea, Brewster County, Texas.

ardship responsibilities by Texas Parks and Wildlife Department. Twomajor drainages occur on the study area. Calamity Creek is an inter-mittent stream that occurs on the western side of EMWMA. It flowsonly during episodes of flooding resulting from intense rainfall (Lerich,2002). Chalk Valley is a shallow drainage with a restricted watershedoccurring on the east side of Elephant Mountain. Chalk Valley does nottypically contain free water on a large scale on the EMWMA (Lerich,2002). Spreader dams were constructed on this portion of the EMWMAbeginning in the 1950s.

The EMWMA is characterized as arid, averaging 367 mm (1986–2003) of precipitation at the headquarters located on the Calamity Creekwatershed (Texas Parks and Wildlife Department, unpubl. obs.). Precip-itation data have not been collected from the Chalk Valley area. Annualprecipitation the year before host collections was 205 mm, which wasbelow the 15 yr (1986–2000) average of 373 mm. During 2002 beforehost collections, approximately 177 mm of the total annual precipitation(286 mm) occurred. Wildlife access to other water resources on theEMWMA includes livestock watering troughs on both portions of theEMWMA.

Major plant communities of the EMWMA have been described byLerich (2002); vegetation on the Chalk Valley watershed varies fromgrama (Bouteloua spp.) and tobosa (Pleuraphis [Hilaria] spp.)-domi-nated grasslands to Chihuahuan Desert scrubland, whereas plant com-munities on the Calamity Creek watershed include dropseed (Sporo-bolus spp.) and tobosa-dominated grasslands to mesquite (Prosopisglandulosa)-dominated shrublands. Descriptions of the EMWMA’s cli-mate and hydrology are presented in Lerich (2002).

Scaled quail were collected by shooting during August 2002. Thiscollection method was chosen to avoid biases often associated withusing baited trap stations (Weatherhead and Ankney, 1984), which canartificially concentrate hosts and reduce sampling distribution of thehost population. Our collection method, however, samples ‘‘normal’’birds, i.e., those that are capable of flight, maintaining covey member-ship, and pair-bond relationships. Consequently, it does not sample in-dividuals too weak to maintain social relationships. Therefore, data pre-

sented and interpretation should be viewed within this context. Collec-tions were conducted under permits from Texas Parks and Wildlife De-partment (permit SPR-0498-949) and the EMWMA (entry permit02-10). This study was conducted in accordance with Texas A&M Uni-versity-Kingsville’s Institutional Animal Care and Use Committee (au-thorization Y2K-6-4).

Scaled quail were aged as hatch-year (fledged juveniles; HY) or after-hatch-year (subadults and adults; AHY) according to Dimmick and Pel-ton (1996) and sexed by gonadal examination. Each bird was placedinto an individually marked freezer bag, fast-frozen in the field by im-mersion into a container of ethyl alcohol and dry ice, and stored infreezers until necropsy.

Each scaled quail was examined for helminths according to proce-dures outlined in Glass et al. (2002). Nematodes were fixed in glacialacetic acid, stored in a mixture of 70% alcohol and 8% glycerin, andexamined with a microscope and/or dissection scope in alcohol–glycerinwet mounts. If necessary for identification, nematodes were cleared inglycerin. Cestodes were fixed in acid–formalin–ethyl alcohol solution,stored in 70% alcohol, stained with Semichon’s acetocarmine, mountedin Canada balsam, and examined with a microscope.

Specimens of Aulonocephalus found in this study are reported as A.pennula, based on the following information. Canavan (1929) describedSubulura pennula from the scaled quail in Texas and Arizona. Basedon nematodes also collected from the scaled quail in Texas, Chandler(1935) created a new genus, Aulonocephalus, with A. lindquisti as thetype species, making no mention of S. pennula in his diagnosis. Theprincipal character of the new genus was ‘‘six conspicuous trough-likegrooves radiating from the mouth’’ (Chandler, 1935). Although Canavan(1929) did not describe cervical grooves in S. pennula, they are clearlyillustrated in his Fig. 35, and Inglis (1958) moved this species to thegenus Aulonocephalus. Besides being described from the same host andthe same locality, A. pennula and A. lindquisti were similar in generalsize and shape, length of spicules, and shape of the gubernaculum. Twoprincipal differences are that the vulva of A. pennula was described asanterior to the middle of the body and the male had 2 pairs of papillaebeside the ventral sucker, whereas the vulva of A. lindquisti was ‘‘at orslightly behind middle of body’’ (Chandler, 1935) and the male had 3pairs of papillae beside the sucker. Examination by 1 of us (J.M.K.) ofa gravid female paratype of A. lindquisti from the U.S. National ParasiteCollection (USNPC 39556) revealed that the vulva was anterior to themiddle of the body at about 42% of the body length. A male paratypeclearly had only 2 pairs of papillae beside the ventral sucker. In additionto the specimens from the present study, a large series of specimensfrom scaled quail in the USNPC identified as A. lindquisti (76543,79280), A. pennula (48604, 68684), Subulura brumpti (68658, 68659),and Subulura strongylina (32340, 32289, 32291) were examined, andall were determined to be the same species, A. pennula. Therefore, thereseems to be only one species of Aulonocephalus in the scaled quail,and A. lindquisti is here declared a synonym of A. pennula (Canavan,1929) Inglis, 1958.

The following helminths found in the present study were depositedin the U.S. National Parasite Collection, Beltsville, Maryland: A. pen-nula (097443), Gongylonema sp. (097445), Oxyspirura petrowi(097442), Physaloptera sp. (097446), Procyrnea pileata (097444), Cho-anotaenia infundibulum (097440), Fuhrmannetta sp. (097441), andRhabdometra odiosa (097439).

Prevalence, intensity, mean intensity, and abundance follow the def-initions of Bush et al. (1997). Common species were defined as thosewith �75% prevalence across the collective host sample, intermediatespecies �25 and �75%, and rare species �25%. Infracommunity refersto all infrapopulations of parasite species occurring within a single host,and component community refers to all infrapopulations of parasitesoccurring within a particular subset of a host species (Bush et al., 1997).The term microhabitat refers to anatomical localities within the host.

Statistical analyses were performed in SAS version 6.12 (SAS Insti-tute, Cary, North Carolina) using the AHY sample. Chi-square analysiswas used to determine whether prevalence in the AHY sample variedby the main effects of host gender and collection site (Chalk Valley sitewith spreader dams, treatment; Calamity Creek site with no spreaderdams, control) for species of helminths with �25% prevalence. In caseswhere expected cell sizes were �5, Fisher’s exact test was used to assesssignificance. The frequency distribution pattern of abundance for eachspecies with �25% prevalence was tested for normality. Non-normally

206 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

TABLE I. Descriptive statistics of helminths from 48 scaled quail (Callipepla squamata) collected during August 2002 at Elephant MountainWildlife Management Area, Brewster County, Texas.

Species

Prevalence

No. %

Intensity

x � SE Range

Abundance

x � SE Total

Nematoda

Aulonocephalus pennula (Canavan, 1929) Inglis, 1958 (C)* 47 98 52.8 � 6.5 2–173 51.7 � 6.5 2,483Gongylonema sp. Molin, 1857 (PR) 1 2 1 1 0.02 � 0.02 1Oxyspirura petrowi Skrjabin, 1929 (E) 27 56 5.6 � 0.8 1–15 3.2 � 0.6 152Physaloptera sp. Rudolphi, 1819 (CW, T) 2 4 4.5 � 3.5 1–8 0.2 � 0.2 9Procyrnea pileata (Walton, 1928) Chabaud, 1975 (G) 29 60 3.1 � 0.4 1–10 1.9 � 0.3 91

Cestoda

Choanotaenia infundibulum (Bloch, 1779) (LI, SI) 1 2 5 5 0.1 � 0.1 5Fuhrmannetta sp. Stiles and Orlemann, 1926 (LI, SI) 12 25 2.2 � 0.6 1–8 0.6 � 0.2 27Rhabdometra odiosa (Leidy, 1887) Jones, 1929 (LI, SI) 17 35 2.5 � 0.8 1–12 0.9 � 0.3 43

* Microhabitats: C � ceca, CW � carcass wash, E � eye surface and nictitating membrane, G � gizzard koilin lining, LI � large intestine, PR � proventriculus, SI� small intestine, T � trachea.

FIGURE 2. Number of helminth species found in 48 scaled quail(Callipepla squamata) collected during August 2002 at Elephant Moun-tain Wildlife Management Area, Brewster County, Texas.

distributed abundance data were rank transformed (Conover and Iman,1981) and examined for the main and interaction effects variables withthe general linear model (GLM) of analysis of variance. The GLMprocedure allows for statistically valid comparisons for datasets withunbalanced samples (SAS Institute). The least squares mean option inGLM was used to separate ranked means for significant interaction ef-fects variables. Statistical significance was determined at P � 0.05.Descriptive statistics are presented as a mean � 1 SE.

For component community assessments, the numerical dominanceindex (DI; Leong and Holmes, 1981) was used to rank species of hel-minths by the number of individuals that each species contributed tothe total number of helminth individuals across the collective sample,by host age, host gender, and collection site. The percentage similarityindex (PSI; Pielou, 1975) and the Jaccard’s coefficient of similarity in-dex (JI; Krebs, 1989) were used to assess similarity of component com-munities by host age, host gender, and collection site.

Eight species of helminths were found in 48 (10 HY, 38 AHY) scaledquail (Table I). One bird was uninfected, whereas the remaining 47individuals had 1–6 species (Fig. 2). Helminth infracommunities aver-aged 2.9 � 0.2 species and 59.8 � 7.2 individuals. Prevalences rangedfrom 98% for A. pennula to 2% for C. infundibulum and Gongylonemasp. (Table I).

Of the 5 species in which chi-square analysis was conducted on the

AHY sample, prevalence of P. pileata was higher (P � 0.049) in thefemale sample (79%) than in the male sample (46%) and higher (P �0.037) in the sample collected from the treatment site (69%) than thecontrol site (33%). Higher rank mean abundance of A. pennula and O.petrowi (P � 0.0001 and P � 0.0052, respectively) was found in thehost sample collected from the site that had spreader dams than the sitewithout spreader dams. A significant (P � 0.0215) host gender-by-col-lection site interaction was observed for P. pileata, in which rank meanabundance in the male sample from the site without spreader dams(control site) was lower than the male sample from the treatment site(P � 0.0064), female sample from the control site (P � 0.0005), andthe female sample from the treatment site (P � 0.0005).

Component community comparisons using DI values found A. pen-nula numerically dominated, accounting for 88% of total worms andwas numerically dominant across host age, host gender, and collectionsite variables (Table II). Variation was found in host age, host gender,and collection site variables for JI (Table III). However, PSI values ofall comparisons were similar (Table III).

Factors in arid environments that facilitate concentrations of suscep-tible hosts, either by host behavior, i.e., covey and family groups orartificial influences, i.e., increasing available water sources, could affecthelminth community structure. Human-altered habitats in which thereis a discrepancy between the cues used by an animal to assess habitatquality and the true quality of the habitat have been characterized as anecological trap (Kokko and Sutherland, 2001; Schlaepfer et al., 2002).This might be the case in our study if benefits of providing a waterresource are outweighed by increased negative impacts of parasitism.Based on occurrence, 5 species were acquired readily (�25% preva-lence) by scaled quail, irrespective of collection site. However, meanrank abundance values were higher in 2 (A. pennula and O. petrowi)of 5 species on the treatment site, with a third species (P. pileata)demonstrating a gender-by-site effect. This suggests that the site withspreader dams may facilitate increased acquisition of helminth individ-uals. High numbers of indirect life cycle helminths found in scaled quailfrom the site with spreader dams suggest increased availability of in-termediate hosts, which are likely attracted to the enhanced vegetationresource surrounding spreader dams. It was beyond the scope of thisstudy to assess intermediate or definitive host density between treatmentand control sites; however, given equivalent collection effort on bothsites, scaled quail were more easily acquired on the site with spreaderdams, suggesting attraction to these locations.

Finding 8 species was surprising given that the few studies conductedin Texas found 3–4 species (Wallmo, 1956; Howard, 1981; Dancak etal., 1982). Of the 8 species found in the present study, Gongylonemasp., P. pileata, and C. infundibulum have not been reported in scaledquail. Although 2 of these were found only on the treatment site, bothwere rare species (Tables I, II). Consequently, it could solely be anartifact of low sample size (Walther et al., 1995) on the control siteinstead of a site effect. Findings do suggest a more complex helminth

RESEARCH NOTES 207

TA

BL

EII

.N

umbe

rof

helm

inth

indi

vidu

als

(N)

and

dom

inan

cein

dex

(DI)

valu

esge

nera

ted

acro

ssth

eho

stsa

mpl

e,an

dby

host

age,

host

gend

er,

and

coll

ecti

onsi

tefo

rhe

lmin

ths

foun

din

scal

edqu

ail

(Cal

lipe

pla

squa

mat

a)co

llec

ted

duri

ngA

ugus

t20

02at

Ele

phan

tM

ount

ain

Wil

dlif

eM

anag

emen

tA

rea,

Bre

wst

erC

ount

y,T

exas

.

Spe

cies

Tot

al(n

�48

)

ND

I

Hat

ch-y

ear

(n�

10)

ND

I

Aft

er-h

atch

-yea

r(n

�38

)

ND

I

Mal

e(n

�34

)

ND

I

Fem

ale

(n�

14)

ND

I

Con

trol

site

(n�

14)

ND

I

Tre

atm

ent

site

(n�

34)

ND

I

Aul

onoc

epha

lus

penn

ula

2,48

388

.333

688

.42,

147

88.3

1,49

189

.999

286

.124

582

.82,

238

89.0

Oxy

spir

ura

petr

owi

152

5.4

71.

814

56.

078

4.7

746.

413

4.4

139

5.5

Pro

cyrn

eapi

leat

a91

3.2

246.

367

2.7

452.

746

4.0

237.

868

2.7

Rha

bdom

etra

odio

sa43

1.5

102.

733

1.3

191.

124

2.1

82.

735

1.4

Fuh

rman

nett

asp

.27

1.0

30.

824

1.0

120.

715

1.3

72.

320

0.8

Phy

salo

pter

asp

.9

0.3

00

90.

48

0.5

1�

0.1

00

90.

3C

hoan

otae

nia

infu

ndib

ulum

50.

20

05

0.2

50.

30

00

05

0.2

Gon

gylo

nem

asp

.1

�0.

10

01

�0.

11

�0.

10

00

01

�0.

1

TABLE III. Comparative analyses of helminth communities by usingpercentage similarity (PSI) and Jaccard’s (JI) indices for scaled quail(Callipepla squamata) collected during August 2002 at Elephant Moun-tain Wildlife Management Area, Brewster County, Texas.

Comparison PSI* JI†

Hatch-year (n � 10) and after-hatch-year (n � 38) 94.9 0.63Males (n � 34) and females (n � 14) 95.4 0.75Control site (n � 14) and treatment site (n � 34) 92.1 0.63

* Values range from 0 to 100, where 0 � completely dissimilar communities and100 � completely similar communities.

† Values range from 0 to 1, where 0 � completely dissimilar communities and 1� completely similar communities.

community occurring in scaled quail at Elephant Mountain WildlifeManagement Area than reported in other geographic regions. Alterna-tively, it could reflect the intensity of necropsy examination among stud-ies (this study involved comprehensive examination of all possible hel-minth microhabitats). Overall, however, helminth infra- and componentcommunities were relatively species poor and were dominated numer-ically at the infra- and component community level by a single indirectlife cycle nematode (A. pennula). This species had a dramatic effect onPSI values; it contributed 86.1–89.9% of all worm individuals acrosshost age, host gender, and collection site comparisons (Table II). Rel-atively low JI values reflected the presence/absence of the 3 rare species(Gongylonema sp., Physaloptera sp., and C. infundibulum) among thehost age, host gender, and collection site comparisons. Absences werelikely a reflection of the low host sample sizes within certain host sub-groups. Walther et al. (1995) indicated that larger host sample sizesincreases species richness. However, increased species richness is oftenthe result of the addition of rare species that contribute little to com-ponent community structure, e.g., numerical dominance. In this study,common and intermediate species across the AHY host sample werealso common and intermediate even for the smallest main effects var-iable (control site; n � 12).

All of the species found in the present study were indirect life cyclehelminths. Dominance of heteroxenous species has been reported inother helminth communities found in arid regions (Patrick, 1994; Fe-dynich et al., 2001), which suggests a selective advantage. Possibly, thisis related to temporal persistence of a parasitic species in hostile envi-ronments by maintaining infective life stages in intermediate hosts(Mackiewicz, 1988). Alternatively, Dobson (1989) suggested that in de-sert environments where host populations are typically low, communi-ties should be dominated by monoxenous parasite species, particularlythose parasites with asexual, viviparous, or autoinfection strategies. Re-gardless of life cycle strategies, it seems that harsh environments likelyincrease selective pressure on parasites to increase transmission effi-ciency by synchronizing their reproductive activity with that of theirhosts (Dobson, 1989). Based on our findings, additional research shouldbe directed at arid regions where anthropogenic influences may alterhost–parasite exposure probabilities.

We thank M. Mayfield and M. Smith for field assistance. This ismanuscript 06-116 of the Caesar Kleberg Wildlife Research Institute.Funding was provided by the Caesar Kleberg Wildlife Research Instituteand the University Research Council at Texas A&M University-Kings-ville.

LITERATURE CITED

BUSH, A. O., K. D. LAFFERTY, J. M. LOTZ, AND A. W. SHOSTAK. 1997.Parasitology meets ecology on its own terms: Margolis et al. re-visited. Journal of Parasitology 83: 575–583.

CANAVAN, W. P. N. 1929. Nematode parasites of vertebrates in the Phil-adelphia Zoological Garden and vicinity. Parasitology 21: 63–102.

CHANDLER, A. C. 1935. A new genus and species of Subulurinae (Nem-atodes). Transactions of the American Microscopical Society 54:33–35.

CONOVER, W. J., AND R. L. IMAN. 1981. Rank transformations as a bridgebetween parametric and nonparametric statistics. American Statis-tician 35: 124–129.

208 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

DANCAK, K., D. B. PENCE, F. A. STORMER, AND S. L. BEASOM. 1982.Helminths of the scaled quail, Callipepla squamata, from North-west Texas. Proceedings of the Helminthological Society of Wash-ington 49: 144–146.

DIMMICK, R. W., AND M. R. PELTON. 1996. Criteria of sex and age. InResearch and management techniques for wildlife and habitats, 5thed., T. A. Bookhout (ed.). The Wildlife Society, Bethesda, Mary-land, p. 169–214.

DOBSON, A. P. 1989. Behavioral and life history adaptations of parasitesfor living in desert environments. Journal of Arid Environments17: 185–192.

FEDYNICH, A. M., T. MONASMITH, AND S. DEMARAIS. 2001. Helminthcommunity structure and pattern in Merriam’s kangaroo rats, Di-podomys merriami Mearns, from the Chihuahuan Desert of NewMexico, U.S.A. Comparative Parasitology 68: 116–121.

GLASS, J. W., A. M. FEDYNICH, M. F. SMALL, AND S. J. BENN. 2002.Helminth community structure in an expanding white-winged dove(Zenaida asiatica asiatica) population. Journal of Wildlife Diseases38: 68–74.

GRAY, G. G., D. B. PENCE, AND C. D. SIMPSON. 1978. Helminths ofsympatric Barbary sheep and mule deer in the Texas Panhandle.Proceedings of the Helminthological Society of Washington 45:139–141.

HOWARD, M. O. 1981. Food habits and parasites of scaled quail insoutheastern Pecos County, Texas. M.S. Thesis. Sul Ross State Uni-versity, Alpine, Texas, 135 p.

INGLIS, W. G. 1958. The comparative anatomy of the subulurid head(Nematoda) with a consideration of its systematic importance. Pro-ceedings of the Zoological Society of London 130: 577–604.

KOKKO, H., AND W. J. SUTHERLAND. 2001. Ecological traps in changingenvironments: Ecological and evolutionary consequences of a be-haviorally mediated Allee effect. Evolutionary Ecology Research3: 537–551.

KREBS, C. J. 1989. Ecological methodology. Harper Collins Publishers,New York, 654 p.

LEONG, T. S., AND J. C. HOLMES. 1981. Communities of metazoan par-asites in open water fishes of Cold Lake, Alberta. Journal of FishBiology 18: 693–713.

LERICH, S. P. 2002. Nesting ecology of scaled quail at Elephant Moun-tain Wildlife Management Area, Brewster County, Texas. M.S.Thesis. Sul Ross State University, Alpine, Texas, 80 p.

MACKIEWICZ, J. S. 1988. Cestode transmission patterns. Journal of Par-asitology 74: 60–71.

PATRICK, M. J. 1994. Parasites of Dipodomys merriami: Patterns andprocesses. Ph.D. Dissertation. University of New Mexico, Albu-querque, New Mexico, 109 p.

PIELOU, E. C. 1975. Ecological diversity. John Wiley and Sons, Inc.,New York, 165 p.

SCHLAEPFER, M. A., M. C. RUNGE, AND P. W. SHERMAN. 2002. Ecologicaland evolutionary traps. Trends in Ecology and Evolution 17: 474–480.

WALLMO, O. C. 1956. Ecology of scaled quail in west Texas. TexasGame and Fish Commission, Austin, Texas, 58 p.

WALTHER, B. A., P. COTGREAVE, R. D. PRICE, R. D. GREGORY, AND D.H. CLAYTON. 1995. Sampling effort and parasite species richness.Parasitology Today 11: 306–310.

WEATHERHEAD, P. J., AND C. D. ANKNEY. 1984. Comment: A criticalassumption of band-recovery models may often be violated. Wild-life Society Bulletin 12: 198–199.

J. Parasitol., 93(1), 2007, pp. 208–211� American Society of Parasitologists 2007

Modification of Host Erythrocyte Membranes by Trypsin and Chymotrypsin Treatmentsand Effects on the In Vitro Growth of Bovine and Equine Babesia Parasites

Masashi Okamura*, Naoaki Yokoyama†, Noriyuki Takabatake, Kazuhiro Okubo, Yuzuru Ikehara‡, and Ikuo Igarashi, National ResearchCenter for Protozoan Diseases, Obihiro University of Agriculture and Veterinary Medicine, Obihiro, Hokkaido 080-8555, Japan; *Laboratory ofZoonoses, School of Veterinary Medicine and Animal Sciences, Kitasato University, Towada, Aomori 034-8628, Japan; †To whomcorrespondence should be addressed; ‡Research Center for Glycoscience, National Institute of Advanced Industrial Science and Technology,Tsukuba, Ibaraki 305-8568, Japan. e-mail: yokoyama@obihiro.ac.jp

ABSTRACT: In the present study, we investigated the effects of proteasepretreatments of host erythrocytes (RBC) on the in vitro growth ofbovine Babesia parasites (Babesia bovis and B. bigemina) and equineBabesia parasites (B. equi and B. caballi). The selected proteases, tryp-sin and chymotrypsin, clearly modified several membrane proteins ofboth bovine and equine RBC, as demonstrated by SDS-PAGE analysis;however, the protease treatments also modified the sialic acid contentexclusively in bovine RBC, as demonstrated by lectin blot analysis. Anin vitro growth assay using the protease-treated RBC showed that thetrypsin-treated bovine RBC, but not the chymotrypsin-treated ones, sig-nificantly reduced the growth of B. bovis and B. bigemina as comparedto the control. In contrast, the growth of B. equi and B. caballi was notaffected by any of these proteases. Thus, the bovine, but not the equine,Babesia parasites require the trypsin-sensitive membrane (sialoglyco)proteins to infect the RBC.

Babesia parasites, which are tick-transmitted hemoprotozoa in a widevariety of wild and domestic animals, cause a severe disease character-ized by fever, anemia, hematuria, and hypotensive shock syndrome,with occasional death (Kuttler, 1988; Homer et al., 2000). The disease,called babesiosis, often results in great economic loss in the livestockindustry and serious concern relative to the international trade of horsesthroughout the world (Brown and Palmer, 1999). For the successfuldevelopment of novel chemotherapeutics and vaccines against babesi-osis, the molecular mechanism by which the Babesia parasites infect

the host erythrocytes (RBC) should be correctly elucidated (Yokoyamaet al., 2005, 2006).

The mechanism of erythrocyte invasion by Plasmodium falciparum,which shows several similarities to Babesia spp. with regard to its path-ological nature (Bannister and Mitchell, 2003) and the process of eryth-rocyte invasion (Yokoyama et al., 2006), has been extensively studied.The effects of P. falciparum invasion by pretreatment of human RBCwith several kinds of proteases have been determined to identify theerythrocyte receptors required for the infection. For example, RBC havea high amount of sialoglycoproteins on the surface (Furthmayr, 1977;Siebert and Fukuda, 1986), and some of them, such as glycophorins Aand B, have been indicated as the host receptors. The glycophorin A–dependent invasion of P. falciparum is sensitive to the pretreatment ofhost RBC with trypsin, while the glycophorin B–dependent one is tryp-sin-resistant (Camus and Hadley, 1985; Sim et al., 1990; Orlandi et al.,1992; Dolan et al., 1994; Sim et al., 1994; Goel et al., 2003). In studieswith the bovine Babesia parasites, B. bovis, B. bigemina, and B. div-ergens, pretreating bovine RBC with trypsin and chymotrypsin was alsoreported to affect their erythrocyte invasion. The B. bovis invasion wassensitive to both the trypsin and chymotrypsin treatments (Gaffar et al.,2003), while that by B. bigemina was sensitive to the trypsin, but notthe chymotrypsin, treatment (Kania et al., 1995). However, the prolif-eration of B. divergens decreased after the pretreatment with chymo-trypsin, but not after that with trypsin (Zintl et al., 2002). In contrast,there have been no studies on protease sensitivities to equine Babesia

RESEARCH NOTES 209

FIGURE 1. Effect of protease treatments on bovine (A) and equine(B) erythrocyte membrane fractions. Bovine and equine RBC weretreated with 5 mg/ml trypsin (Tr) or chymotrypsin (Chtr) and lysed toobtain the ghost proteins, which were subjected to SDS-PAGE followedby CBB staining or lectin blot with MAL II. PBS-treated RBC werealso included as the control. Major detectable bands of the control sam-ples and molecular weight standards are indicated in kDa to the rightof the CBB and lectin blot panels, respectively. The results are repre-sentative of 5 separate experiments.

parasites, such as B. equi and B. caballi. To gain a better understandingof the erythrocyte infections by Babesia parasites, we evaluated thesensitivities of Babesia parasites to pretreatments of the correspondinghost RBC with trypsin and chymotrypsin on a panel of bovine andequine Babesia parasites including B. bovis, B. bigemina, B. equi, andB. caballi. For these parasites, in vitro culture methods have alreadybeen established (Levy and Ristic, 1980; Bork et al., 2003; Okamura etal., 2005). Moreover, the present study, in conjunction with the resultsof lectin blot analysis, also suggested the possible role of host sialogly-coproteins in their erythrocyte infections.

The USDA strains of B. equi and B. caballi, the Texas strain of B.bovis, and the Argentina strain of B. bigemina were cultured in vitroaccording to a microaerophilus stationary-phase culture method (Levyand Ristic, 1980). Briefly, B. equi and B. caballi were cultured withequine RBC at a 10% packed cell volume in M199 (Sigma, St. Louis,Missouri) and RPMI1640 (Sigma), respectively, containing 40% equineserum and an antibiotic-antimycotic (100 ) supplement (InvitrogenCorp., Carlsbad, California) in 24-well cell culture plates (Nalge NuncInternational, Roskilde, Denmark). Hypoxanthine (13.61 �g/ml) (Sig-ma) was added as a vital supplement to the medium only for B. equi.Babesia bovis and B. bigemina were cultured with bovine RBC in M199and RPMI, respectively, containing 40% bovine serum and an antibi-otic-antimycotic supplement. Blood was retrieved from a selectedhealthy cow or horse and defibrinated by shaking with glass beads for30 min at room temperature. RBC and sera for the in vitro cultureswere collected by centrifugation at 910 g for 7 min at 4 C, followedby removal of the buffy coat. RBC were then stored in a Vega y Mar-tinez (VyM) buffer (Vega et al., 1985) at 4 C before use. Cultures wereroutinely maintained by changing the media daily and adjusting theparasitemia between 1% and 12%. Parasitemia was determined bycounting more than 1,000 RBC on Giemsa-stained thin smears. Culturesof B. bovis, B. bigemina, and B. equi were incubated at 37 C in anatmosphere of 5% CO2, 5% O2, and 90% N2, while B. caballi wasincubated in an atmosphere of 5% CO2 (Bork et al., 2003).

For the trypsin or chymotrypsin treatment of RBC, freshly preparedbovine RBC were separately treated with 0.2, 1.0, and 5.0 mg/ml tryp-sin (EC 3.4.21.4) (equal to 2,640, 13,200, and 66,000 U/ml, respective-ly) or chymotrypsin (EC 3.4.21.1) (equal to 11, 55, and 275 U/ml,respectively) for 3 hr at 37 C on a rotator (all the enzymes were pur-chased from Sigma). The treated RBC were then thoroughly washed 5times in phosphate-buffered saline (PBS) by centrifuging at 910 g for7 min at 4 C and then suspended in a VyM buffer to store at 4 C beforeuse. Control RBC were treated with PBS and prepared in the VyMbuffer as above. The protease treatments at these concentrations did notshow a hemolytic effect (data not shown), indicating that the treatmentdid not affect the integrity of the erythrocyte membrane.

For sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and lectin blot analyses, the erythrocyte ghosts were preparedby hypotonic lysis of the enzymatically treated or control RBC in a 10mM phosphate buffer (Dodge et al., 1963). The ghost proteins of 2.0mg/ml were mixed with an equal volume of a 2 SDS sample buffer(62.5 mM Tris-HCl [pH 6.8], 10% glycerol, 5% 2-mercaptoethanol, 2%SDS, and 0.01% bromphenol blue), sonicated for 1 min, and heated at100 C for 5 min. The ghost samples (10 �g/lane) were subjected toSDS-PAGE on a 10% polyacrylamide gel followed by Coomasie Bril-liant Blue (CBB) (ICN Biochemicals Inc., Aurora, Ohio) staining. Inaddition, lectin blot analysis using biotinylated Maackia amurensis lec-tin II (MAL II; Vector Laboratories, Burlingame, California), whichspecifically recognizes �2-3-linked sialic acids, was also performed withthe protein-transferred membrane to observe the effects of proteases onthe sialic acid content (Okamura et al., 2005).

An in vitro growth assay was carried out as described previously(Okamura et al., 2005). Briefly, B. equi–, B. caballi–, B. bovis–, or B.bigemina–infected RBC whose parasitemia had been adjusted to 5%with intact RBC were mixed with the trypsin- or chymotrypsin-treatedRBC in an appropriate medium in 96-well tissue culture plates (NalgeNunc International). This procedure yielded a final result of 80% pro-tease-treated RBC and 1% parasitemia in the cultured RBC. For thecontrol, PBS-treated RBC, rather than protease-treated cells, were used.The assay was carried out in 5 wells for each treatment in 3 separateexperiments. At 24 hr postinitiation of the culture, each parasitemia wasdetermined and translated to the percentage growth relative to the con-trol parasitemia in the PBS-treated RBC. Statistical analyses were per-

formed by 1-way ANOVA with Dunnett’s post-hoc test (InStat�,GraphPad Software, San Diego, California) for a comparison of thepercent growth. Differences were considered significant when p-valuesof less than 0.05 were obtained. In this assay, no differences were de-tected between the PBS-treated and untreated RBC (data not shown).

SDS-PAGE analyses showed that trypsin and chymotrypsin clearlycleaved a number of the detectable membrane proteins of bovine andequine RBC (at least 87-, 71-, and 33-kDa and 88-, 76-, and 33-kDabands of bovine and equine RBC, respectively) (Fig. 1). In particular,the intensity of band 3 (87 and 88 kDa bands of bovine and equineRBC, respectively), which is the major integral membrane protein (In-aba and Maede, 1988; Inaba et al., 1996), was clearly decreased, andthe number of bands smaller than band 3 increased in both bovine andequine RBC. Lectin blot analyses with the bovine RBC revealed thatthe trypsin treatment particularly modified the sialic acid content andgenerated 3 major bands of 80, 37, and 28 kDa with high density, whilethe chymotrypsin treatment did not yield any clear modifications excepta sialic acid–positive band at less than 14 kDa (Fig. 1A). In contrast, 4major sialoglycoproteins of equine RBC (60–63 and 33–38 kDa) (Oka-mura et al., 2005) still remained after the protease treatments with tryp-sin and chymotrypsin (Fig. 1B).

Figure 2 shows the inhibition of B. bovis growth in the culture withthe treated RBC at 24 hr after the initiation of in vitro cultures. Thetrypsin treatment significantly decreased the parasite growth to 62.5%and 61.2% at the concentrations of 1.0 and 5.0 mg/ml, respectively,whereas the chymotrypsin treatment did not diminish the proliferation(Fig. 2A). A similar result was obtained in B. bigemina, in which theparasite growth was significantly decreased to 69.4%, 68.4%, and68.1% in the treated RBC with trypsin at concentrations of 0.2, 1.0,and 5.0 mg/ml, respectively (Fig. 2B). In contrast, the in vitro growthof the equine Babesia parasites, B. equi and B. caballi, was not at allaffected by these protease treatments (Figs. 2C, 2D).

In the present study, although the treatments with trypsin and chy-motrypsin resulted in clear modification of several membrane proteinsin both bovine and equine RBC, bovine and equine Babesia parasitesshowed different growth patterns in the presence of the modified RBC.The trypsin, but not the chymotrypsin, treatment significantly affectedthe in vitro growth of B. bovis and B. bigemina, while the treatmentswith these proteases did not influence the growth of B. equi and B.caballi. The present findings indicate that the bovine and equine Ba-besia parasites have trypsin-sensitive and chymotrypsin-resistant andtrypsin- and chymotrypsin-resistant pathways in their infections to thecorresponding host RBC, respectively.

The observed inhibition caused by the pretreatment of bovine RBCwith trypsin might be due to the incidental effects on the modificationof the sialic acid content of erythrocyte sialoglycoproteins because the

210 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

FIGURE 2. In vitro growth of Babesia bovis (A), B. bigemina (B),B. equi (C), and B. caballi (D) in corresponding host RBC that hadbeen treated with trypsin or chymotrypsin. The parasite growth in theprotease-treated RBC was expressed as the percentage relative to thecontrol growth in the PBS-treated RBC at 24 hr after the initiation ofcultures. The error bars indicate the standard deviation of 5 wells for arepresentative of 3 separate experiments. The asterisks represent signif-icant differences (P � 0.05) from the corresponding control growth.

sialic acid profile of bovine RBC was more largely altered by the treat-ment with trypsin than by that with chymotrypsin, while the profile ofequine RBC showed slight changes upon the use of either protease. Ourdata would, therefore, strongly support the importance of host sialogly-coproteins for erythrocyte infection by Babesia parasites, as describedrecently (Kania et al., 1995; Zintl et al., 2002; Gaffar et al., 2003;Okamura et al., 2005). Gaffar et al. (2003) reported that the pretreatmentof bovine RBC with trypsin or neuraminidase singly and in combinationshowed inhibitions of similar extents against erythrocyte invasions byB. bovis. These results suggest that unidentified sialoglycoprotein(s) ex-posing the trypsin cleavage sites existed on the surface of bovine RBCand might also function as the host receptor for the infection of B. bovisand B. bigemina. On the other hand, the possibility that equine parasitescan bind to intact trypsin sites on chymotrypsin-treated RBC or to intactchymotrypsin sites on trypsin-treated RBC cannot be excluded since acombined treatment with trypsin and chymotrypsin was not used in thepresent study because of its hemolytic effect confirmed by the prelim-inary experiments (data not shown).

The result obtained for B. bovis in the present study partially contra-dicts the result obtained by Gaffer et al. (2003), in whose study B. bovisinfection was sensitive not only to trypsin but also to chymotrypsin.This discrepancy might be explained by the difference in the concen-tration of chymotrypsin used: 11–275 U/ml for this study and 3,000U/ml for the study by Gaffar et al. (2003). Our preparation might appeartoo low to result in a significant inhibition of invasion. However, thepreliminary experiments showed that chymotrypsin at �1,000 U/ml hada hemolytic effect (data not shown). Since it is now unclear whetherthe decreased integrity affects the accurate evaluation of the inhibitoryeffect, care should be taken to interpret this discrepancy.

In P. falciparum, there are multiple pathways for erythrocyte infec-tion, and some sialoglycoproteins, such as glycophorins, are directlyinvolved in the infection. For example, the human glycophorins A andC are recognized by the corresponding ligands of P. falciparum,EBA175 and EBA140, respectively (Camus and Hadley, 1985; Sim etal., 1990; Orlandi et al., 1992; Goel et al., 2003). In addition, the asi-aloglycoprotein band 3 also serves as the host erythrocyte receptor forP. falciparum MSP-1 (Goel et al., 2003). Bovine and equine glyco-phorins and bovine band 3 have also been studied with regard to theirbiochemical properties (Murayama et al., 1982a, 1982b; Inaba andMaede, 1988; Inaba et al., 1996). Since 4 major sialoglycoproteins ofequine RBC observed to remain after the protease treatments with tryp-

sin and chymotrypsin (Fig. 1B) are likely glycophorins (Okamura etal., 2005), the involvement of glycophorins in the infection is suggested.Further study on the membrane components of animal RBC, includingthe glycophorins and band 3, would offer new information toward theelucidation of the mechanisms of host erythrocyte infections by Babesiaparasites.

In conclusion, the bovine Babesia parasites, B. bovis and B. bigemina,require trypsin-sensitive and chymotrypsin-resistant membrane (sial-oglyco) proteins to infect the bovine RBC. In contrast, the equine Ba-besia parasites, B. equi and B. caballi, utilize a trypsin- and chymo-trypsin-resistant pathway at the phase of their erythrocyte infections.This information will facilitate the examination of the molecular mech-anism whereby the Babesia parasites infect the host RBC.

We wish to thank all the staff and students of the National ResearchCenter for Protozoan Diseases for their critical comments and technicalassistance. This study was supported by Grants-in-Aid for ScientificResearch from the Japan Society for the Promotion of Science, by theIndustrial Technology Research Grant Program from the New Energyand Industrial Technology Development Organization (NEDO) of Ja-pan, by grants from Special Coordination Funds for Science and Tech-nology from the Science and Technology Agency and The 21st CenturyCOE Program (A-1), Ministry of Education, Culture, Sports, Science,and Technology, Japan, and by the Japan International CooperationAgency (JICA).

LITERATURE CITED

BANNISTER, L., AND G. MITCHELL. 2003. The ins, outs, and roundaboutsof malaria. Trends in Parasitology 19: 209–213.

BORK, S., N. YOKOYAMA, T. MATSUO, F. G. CLAVERIA, K. FUJISAKI, AND

I. IGARASHI. 2003. Growth inhibitory effect of triclosan on equineand bovine Babesia parasites. American Journal of Tropical Med-icine and Hygiene 68: 334–340.

BROWN, W. C., AND G. H. PALMER. 1999. Designing blood-stage vac-cines against Babesia bovis and B. bigemina. Parasitology Today15: 275–281.

CAMUS, D., AND T. J. HADLEY. 1985. A Plasmodium falciparum antigenthat binds to host erythrocytes and merozoites. Science 230: 553–556.

DODGE, J. T., C. MITCHELL, AND D. J. HANAHAN. 1963. The preparationand chemical characteristics of hemoglobin-free ghosts of humanerythrocytes. Archives of Biochemistry and Biophysics 100: 119–130.

DOLAN, S. A., J. L. PROCTOR, D. W. ALLING, Y. OKUBO, T. E. WELLEMS,AND L. H. MILLER. 1994. Glycophorin B as an EBA-175 indepen-dent Plasmodium falciparum receptor of human erythrocytes. Mo-lecular and Biochemical Parasitology 64: 55–63.

FURTHMAYR, H. 1977. Structural analysis of a membrane glycoprotein:Glycophorin A. Journal of Supramolecular Structure 7: 121–134.

GAFFAR, F. R., F. F. FRANSSEN, AND E. DE VRIES. 2003. Babesia bovismerozoites invade human, ovine, equine, porcine and caprineerythrocytes by a sialic acid–dependent mechanism followed bydevelopmental arrest after a single round of cell fission. Interna-tional Journal for Parasitology 33: 1595–1603.

GOEL, V. K., X. LI, H. CHEN, S. C. LIU, A. H. CHISHTI, AND S. S. OH.2003. Band 3 is a host receptor binding merozoite surface protein1 during the Plasmodium falciparum invasion of erythrocytes. Pro-ceedings of the National Academy of Sciences USA 100: 5164–5169.

HOMER, M. J., I. AGUILAR-DELFIN, S. R. TELFORD III, P. J. KRAUSE, AND

D. H. PERSING. 2000. Babesiosis. Clinical Microbiology Reviews13: 451–469.

INABA, M., AND Y. MAEDE. 1988. A new major transmembrane glyco-protein, gp155, in goat erythrocytes: Isolation and characterizationof its association to cytoskeleton through binding with band 3-ankyrin complex. Journal of Biological Chemistry 263: 17763–17771.

———, A. YAWATA, I. KOSHINO, K. SATO, M. TAKEUCHI, Y. TAKAKUWA,S. MANNO, Y. YAWATA, A. KANZAKI, J. SAKAI, A. BAN, K. ONO,AND Y. MAEDE. 1996. Defective anion transport and marked sphe-rocytosis with membrane instability caused by hereditary total de-ficiency of red cell band 3 in cattle due to a nonsense mutation.Journal of Clinical Investigation 97: 1804–1817.

RESEARCH NOTES 211

KANIA, S. A., D. R. ALLRED, AND A. F. BARBET. 1995. Babesia bigemina:Host factors affecting the invasion of erythrocytes. ExperimentalParasitology 80: 76–84.

KUTTLER, K. L. 1988. Worldwide impact of babesiosis. In Bebesiosisof domestic animals and man, M. Ristic (ed.). CRC Press, BocaRaton, Florida, p. 1–22.

LEVY, M. G., AND M. RISTIC. 1980. Babesia bovis: Continuous culti-vation in a microaerophilous stationary phase culture. Science 207:1218–1220.

MURAYAMA, J., M. TOMITA, AND A. HAMADA. 1982a. Glycophorins ofbovine erythrocyte membranes: Isolation and preliminary charac-terization of the major component. Journal of Biochemistry (To-kyo) 91: 1829–1836.

———, ———, AND ———. 1982b. Primary structure of horse eryth-rocyte glycophorin HA: Its amino acid sequence has a unique ho-mology with those of human and porcine erythrocyte glycophorins.Journal of Membrane Biology 64: 205–215.

OKAMURA, M., N. YOKOYAMA, N. P. WICKRAMATHILAKA, N. TAKABA-TAKE, Y. IKEHARA, AND I. IGARASHI. 2005. Babesia caballi and Ba-besia equi: Implications of host sialic acids in erythrocyte infection.Experimental Parasitology 110: 406–411.

ORLANDI, P. A., F. W. KLOTZ, AND J. D. HAYNES. 1992. A malaria in-vasion receptor, the 175-kilodalton erythrocyte binding antigen ofPlasmodium falciparum, recognizes the terminal Neu5Ac(�2-3)Gal-sequences of glycophorin A. Journal of Cell Biology 116:901–909.

SIEBERT, P. D., AND M. FUKUDA. 1986. Isolation and characterization of

human glycophorin A cDNA clones by a synthetic oligonucleotideapproach: Nucleotide sequence and mRNA structure. Proceedingsof the National Academy of Sciences USA 83: 1665–1669.

SIM, B. K., C. E. CHITNIS, K. WASNIOWSKA, T. J. HADLEY, AND L. H.MILLER. 1994. Receptor and ligand domains for invasion of eryth-rocytes by Plasmodium falciparum. Science 264: 1941–1944.

———, P. A. ORLANDI, J. D. HAYNES, F. W. KLOTZ, J. M. CARTER, D.CAMUS, M. E. ZEGANS, AND J. D. CHULAY. 1990. Primary structureof the 175K Plasmodium falciparum erythrocyte binding antigenand identification of a peptide which elicits antibodies that inhibitmalaria merozoite invasion. Journal of Cell Biology 111: 1877–1884.

VEGA, C. A., G. M. BUENING, T. J. GREEN, AND C. A. CARSON. 1985. Invitro cultivation of Babesia bigemina. American Journal of Veter-inary Research 46: 416–420.

YOKOYAMA, N., S. KUMAR, S. BORK, M. OKAMURA, AND I. IGARASHI.2005. Molecular approaches for the development of babesial vac-cines. In Asian parasitology, Vol. 4, A. Yano, H.-W. Nam, A. K.Anuar, and J. Shen (eds.). AAA Committee, Chiba, Japan, p. 214–223.

———, M. OKAMURA, AND I. IGARASHI. 2006. Erythrocyte invasion byBabesia parasites: Current advances in the elucidation of the mo-lecular interactions between the protozoan ligands and host recep-tors in the invasion stage. Veterinary Parasitology 138: 22–32.

ZINTL, A., C. WESTBROOK, H. E. SKERRETT, J. S. GRAY, AND G. MUL-CAHY. 2002. Chymotrypsin and neuraminidase treatment inhibitshost cell invasion by Babesia divergens (Phylum Apicomplexa).Parasitology 125: 45–50.

J. Parasitol., 93(1), 2007, pp. 211–213� American Society of Parasitologists 2007

Gastrointestinal Helminth Fauna of Enyalius perditus (Reptilia: Leiosauridae): Relation toHost Age and Sex

B. M. Sousa, A. Oliveira, and S. Souza Lima, Universidade Federal de Juiz de Fora, Departamento de Zoologia, Campus Universitario,Martelos, 36.036-900, Juiz de Fora, Minas Gerais, Brasil. e-mail: bernadete.sousa@ufjf.edu.br

ABSTRACT: The purpose of this study was to determine how the gas-trointestinal helminthofauna varies according to the age and sex of thelizard, Enyalius perditus, captured in Ibitipoca State Park in the stateof Minas Gerais, Brazil, and to discuss the ecological and behavioralsignificance of these relationships. Fifty-five specimens of E. perdituswere captured in drop traps, then killed, necropsied, and examined forthe presence of helminths in the gastrointestinal tract. Nematodes, in-cluding Strongyluris oscari, Oswaldocruzia subauricularis, and Aplec-tana vellardi, were found. This was the first record of the last-namedspecies in reptiles, and the first record of the first 2 species in E. per-ditus. The number of helminths increased with snout-vent length and,therefore, age of the lizards. Male E. perditus lizards were more heavilyinfected by nematodes than females; the largest numbers of nematodesoccurred in the caecum and large intestine.

Lizards are hosts to a wide variety of gastrointestinal nematodes(Goldberg and Bursey, 1992a), which may be acquired through infectedprey, ingestion of contaminated vegetable matter, coprophagy, geopha-gy, or integumental penetration (Goldberg et al., 1995; Ribas et al.,1995; Sanchis et al., 2000; Criscione and Font, 2001; Fontes et al.,2003). Thus, the level of infection in lizards may be related to materialsingested (Kennedy et al., 1986; Goldberg et al., 1993), habitat attributes(Goldberg and Bursey, 1992b), sex, or age of individual lizards (Pfaf-fenberger et al., 1986; Bundy et al., 1987; Vogel and Bundy, 1987;Ribas et al., 1995).

Ontogenetic differences in the level of infection occur between adultand juvenile lizards. An exponential increase in the mass of nematodesdirectly related to body size was observed by Ribas et al. (1995).

Lizard foraging strategy can also influence the intensity of infectionby nematodes. The higher infection intensity may, in part, be the result

of an active foraging strategy (Schoener, 1971). On the one hand, lizardssuch as Cnemidophorus ocellifer (Spix, 1825), which show an activeprey-seeking foraging pattern, generally have a higher associated hel-minth fauna than predators that have an ambush hunting strategy (Ribaset al., 1995). On the other hand, Bundy et al. (1987) postulated thathabitat attributes seem to be a more significant factor in determinationand occurrence of infection by nematodes than phylogenetic relation-ships or ecomorphs.

Many authors have highlighted the importance of using the parasitesas indicators of the diet and trophic interations of their hosts, in additionto movement patterns, reproduction sites, and social structure (Brooks,2000; Beron-Vera et al., 2001; Gomes et al., 2003; Ezenwa, 2004; Mar-cogliese, 2004). The purpose of this note is to determine the prevalence,intensity, and abundance of nematodes species present in Enyalius per-ditus Jackson, 1978 according to these lizards’ age and sex, and toexamine possible routes of infection.

Between March 1997 and March 1998, specimens of E. perditus werecaptured using pitfall traps at 3 forested sites in Ibitipoca State Park(2140�–2144�S, 4352�–4355�W), located in the municipality of LimaDuarte, Minas Gerais, Brazil.

We randomly selected 55 animals, which were killed with sodiumpentobarbital immediately after collection. At the field laboratory, wemeasured the snout-vent length (SVL) and examined the gonads to de-termine sexual maturity. Based on this information, we defined our E.perditus sample as adult males, i.e., individuals with SVL �47 mm,and adult females, with SVL �48 mm (Sousa, 2000). We then removedthe gastrointestinal tract and separated the stomach and large and smallintestines. We injected a 0.9% physiological solution into each part ofthe gut with the aid of a 1-ml throwaway syringe without a needle, thenplaced the gut contents into Petri dishes and analyzed them separately.

212 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

TABLE I. Abundance of nematodes per organ of the digestive tract andby sex of Enyalius perditus from Ibitipoca State Park, Minas Gerais,Brazil.

Nematode

Abundance perorgan*

S SI LI

Abundanceby sex

Male Female

Strongyluris oscari 0 6 236 210 32Oswaldocruzia subauricularis 12 26 12 41 9Aplectana vellardi 0 4 8 10 2

* S � stomach; SI � small intestine; LI � large intestine.

TABLE II. Abundance of nematodes per SVL.

Nematode

SVL � 60 mm*

LI SI S

SVL 35–60 mm*

LI SI S

SVL � 35 mm*

LI SI S

Strongyluris oscari 197 7 0 38 0 0 0 0 0Oswaldocruzia subauricularis 14 22 11 0 3 0 0 0 0Aplectana vellardi 8 4 0 0 0 0 0 0 0

* SVL � snout-vent length; S � stomach; SI � small intestine; LI � large intestine.

The helminths found in each segment of the gastrointestinal tractwere isolated, counted, fixed in alcohol-formalin-acetic acid (AFA),cleared in Amann’s lactophenol and Faia creosote, and mounted be-tween slides and slide covers with Canada balsam. We analyzed therelationship between parasitism and the presence of plant remains inthe lizards’ stomach contents and tested the hypotheses that the hel-minth intensity (number per parasitized lizard) and the presence of plantremains increase with the lizard’s age. We grouped the parasitized liz-ards by SVL and performed a chi-square test to verify the hypothesisof interdependence between the SVL and intensity of nematodes. Tocompare the prevalence and intensity of parasitism, data were submittedto the Kruskal-Wallis test.

The slides were deposited in the Helminthology Collection of theDepartment of Animal Biology of the Universidade Federal Rural doRio de Janeiro (UFRRJ) under the numbers 789 a–d, 790 a–c, 791 a.The ecological terms employed are those of Bush et al. (1997).

Strongyluris oscari Travassos, 1923 (Ascaridida: Heterakidae), Os-waldocruzia subauricularis (Rudolphi, 1819) (Strongylida: Molineidae),and Aplectana vellardi Travassos, 1926 (Ascaridida: Cosmocercidae)were present in the necropsied lizards. Of the 55 lizards examined (12adult females, 5 immature females, 31 adult males, and 7 immaturemales), 69% (n � 38) were parasitized by nematodes. The differencein the prevalence and mean intensity among the nematode species wasstatistically significant (P � 0.05). The prevalence and mean intensityof S. oscari were significantly higher (58%; 7.6 parasites � 7.3) thanO. subauricularis (29%; 3.2 parasites � 3.1), and A. vellardi (6%; 4.8parasites � 0.5).

The habitat with the highest occupation by the nematodes was thelarge intestine (H � 15.0972; P � 0.0017), where 98% of the S. oscarioccurred. Oswaldocruzia subauricularis was found in the large intestine(25%), small intestine (52%), and stomach (23%). In the case of A.vellardi, 66% of the individuals occurred in the large intestine and 34%in the small intestine. Neither S. oscari nor A. vellardi was found inthe stomach. Prevalence and intensity of infection were higher in malelizards.

Data for abundance and size of the nematode infrapopulations in thelizards as related to site of infection, sex, and SVL are presented inTables I and II. We found greater nematode intensities in lizards withbody lengths between 70 and 80 mm, and a positive relationship be-tween parasite intensity and body size. A chi-square test supported asignificant direct correlation between parasite intensity and body size(SVL).

In our study, vegetable fragments were found in 20% (n � 11) ofthe stomachs examined, of which 100% were from adult lizards. Among

the 12 young lizards examined, with SVL varying from 26.8 to 43.3mm, we found that 75% (n � 9) were not infected by nematodes, andnone contained vegetable fragments in the gastrointestinal content.

The finding of S. oscari and O. subauricularis is the first record ofthese nematodes in E. perditus, and that of A. vellardi is the first reportof this species in reptiles. The presence of O. subauricularis has pre-viously been reported in E. catenatus, and that of S. oscari has beenreported in other lizards, particularly Tropidurus sp. (Vicente et al.,1993) for nematodes of Brazilian reptiles.

Lizards can become infected by nematodes via ingestion of arthro-pods infected with larvae of these parasites (Goldberg and Bursey,1992a; Ribas et al., 1995). According to information provided by An-derson (2000), experimental infection with S. brevicaudata in insectsresulted in encapsulated larvae in the thorax of Culex sp. However, thisneeds to be studied further since the heterakoid nematodes are generallymonoxenous and infect their hosts when the latter ingests eggs withlarvae. The diet of E. perditus consists mainly of ants, woodlice, andarthropod larvae (Sousa, 2000), which, considering the observations ofAnderson (2000) for S. brevicauda, may have contributed to the highintensity of parasitism by S. oscari found in the present work. Oswal-docruzia subauricularis (Molineidae) and A. vellardi (Cosmocercidae)are probably monoxenous, infecting their hosts through ingestion ofeggs or active penetration of larvae (Anderson, 2000).

Observation of mineral matter in the stomach contents of E. perditusin the present study suggests that coprophagy (Goldberg and Bursey,1992a) or geophagy (Troyer, 1982) may be an important mechanism ofinfection in lizards, by providing opportunities for the ingestion of eggsand larvae found in the soil and/or penetration of larvae through thebuccal mucosa. Seventeen percent of the lizards studied had molted skinpresent in stomach contents. This may also be considered an opportunityfor infection by helminths, since it is possible that embryonated eggsof oxyurid nematodes such as Aplectana sp. could be retained on moltedskin and be ingested.

In the specimens of E. perditus that contained nematodes and thatwere grouped by SVL, we confirmed the hypothesis that parasite inten-sity increases with age and size. Because of the lack of independencebetween the intensity of parasitism by nematodes and the SVL of E.perditus, we suggest that there is a positive correlation between parasiteintensity and the size (SVL) of E. perditus, a result that agrees withRibas et al. (1995). Higher parasite prevalence and intensity at the levelof nematode infection in males of E. perditus concurs with Vogel andBundy (1987) and Ribas et al. (1995), who made similar observationsin species of Anolis and in C. ocellifer, respectively. Some authors havealso observed that the diets of males and females of some lizard speciesdiffer qualitatively and quantitatively (Shoener, 1967; Fitch, 1981;Floyd and Jenssen, 1983; Barden and Shine, 1994) and that the sexdifferences in the prevalence and intensity of helminths are due to var-iations in body size and prey consumption (Ribas et al., 1995). Sexualdimorphism in males and females is common in some lizard species,and E. perditus males have a longer tail than females (Sousa, 2000),which indicates a greater foraging activity in soil.

Differences in foraging behavior among lizards that actively forage,as seen in Anolis valencieni, compared with other species of Anolis thathave a passive hunting strategy, appear to be reflected in the helminthfauna (Bundy et al., 1987). Thus, for example, active foragers such asC. ocellifer show a high intensity of parasitism by nematodes (Ribas etal., 1995). We may also consider other activities such as basking ormating behaviors between lizards as opportunities for infection by eggingestion of or larvae penetration.

The authors would like to thank Sueli Pontes de Fabio (UFRRJ) for

RESEARCH NOTES 213

identifying the helminths, CAPES for its financial support, and IEF/MGfor logistical assistance. We also thank Juliane Floriano Santos Lopesfor kindly reading the manuscript.

LITERATURE CITED

ANDERSON, R. C. 2000. Nematode parasites of vertebrates: Their de-velopment and transmission. 2nd ed. CABI Publishing, London,U.K., 650 p.

BARDEN, G., AND R. SHINE. 1994. Effects of sex and reproductive modeon dietary composition of the reproductively bimodal scincid liz-ard, Lerista bougainvillii. Australian Journal of Zoology 29: 225–228.

BERON-VERA, B., S. N. PEDRAZA, J. A. RAGA, A. G. PERTIERRA, E. A.CRESPO, M. K. ALONSO, AND R. N. P. GOODALL. 2001. Gastrointes-tinal helminths of Commerson’s dolphins Cephalorhynchus com-mersonii from central Patagonia and Terra del Fuego. Diseases ofAquatic Organisms 47: 201–208.

BROOKS, D. R. 2000. Parasite systematics in the 21st century: Oppor-tunities and obstacles. Memorias do Instituto Oswaldo Cruz 95:99–107.

BUNDY, D. A. P., P. VOGEL, AND E. A. HARRIS. 1987. Helminth parasitesof Jamaican anoles (Reptilia: Iguanidae): A comparison of the hel-minth fauna of six Anolis species. Journal of Helminthology 61:77–83.

BUSH, A. O., K. D. LAFFERTY, J. M. LOTZ, AND A. W. SHOSTAK. 1997.Parasites meet ecology on its own terms: Margolis et al. revisited.Journal of Parasitology 83: 575–583.

CRISCIONE, C. D., AND F. FONT. 2001. The guest playing host: Coloni-zation of the introduced Mediterranean gecko, Hemidactylus tur-cicus, by helminth parasites in Southeastern Lousiana. Journal ofParasitology 87: 1273–1278.

EZENWA, V. O. 2004. Host social behavior and parasite infection: Amultifactorial approach. Behavioral Ecology 15: 446–454.

FITCH, H. S. 1981. Sexual size differences in reptiles. MiscellaneousPublication, Museum of Natural History, University of Kansas,Lawrence, Kansas 70: 1–72.

FLOYD, H. B., AND T. A. JENSSEN. 1983. Food habits of the Jamaicanlizard Anolis opalinus: Resource partitioning and seasonal effectsexamined. Copeia 1983: 319–331.

FONTES, A. F., J. J. VICENTE, M. C. KIEFER, AND M. VAN SLUYS. 2003.Parasitism by helminths in Eurolophosaurus nanusae (Lacertilia:Tropiduridae) in an area of rocky outcrops in Minas Gerais State,southeastern Brasil. Journal of Herpetology 37: 736–741.

GOLDBERG, S. R., AND C. R. BURSEY. 1992a. Prevalence of the nematodeSpauligodon giganticus (Oxyurida: Pharyngodonidae) in neonatalyarrow’s spiny lizards, Sceloporus jarrovii (Sauria: Iguanidae).Journal of Parasitology 78: 539–541.

———, AND ———. 1992b. Helminths of the bunch grass lizard, Sce-

loporus scalaris slevini (Iguanidae). Journal of the Helmintholog-ical Society of Washington 59: 130–131.

———, ———, AND H. CHEAM. 1995. Helminth parasites of three sym-patric lizards from Grand Cayman Island, Anolis conspersus, Anolissagrei (Polychridae) and Leiocephalus carinatus (Tropiduridae).Caribbean Journal of Science 31: 339–340.

———, ———, AND R. TAWIL. 1993. Gastrointestinal helminths of thecrevice spiny lizard, Sceloporus poinsettii (Phrynosomatidae). Jour-nal of the Helminthological Society of Washington 60: 263–265.

GOMES, D. C., R. P. CRUZ, J. J. VICENTE, AND R. M. PINTO. 2003. Nem-atode parasites of marsupials and small rodents from the BrazilianAtlantic Forest in the State of Rio de Janeiro, Brazil. Revista Bras-ileira de Zoologia 20: 699–707.

KENNEDY, M. J., L. M. KILLICK, AND M. BEVERLEY-BURTON. 1986. Theprevalence of Paradistomum geckonum, Mesocoelium sociale, andPostorchigenes ovatus (Digenea) in lizards (Sauria) from Indonesia.Canadian Journal of Zoology 65: 1292–1294.

MARCOGLIESE, D. J. 2004. Parasites: Small players with crucial roles inthe ecological theater. EcoHealth 1: 151–164.

PFAFFENBERGER, G. S., T. L. BEST, AND D. BRUIN. 1986. Helminths ofcollared lizards (Crotaphytus collaris) from the Pedro Armendarizlava field, New Mexico. Journal of Parasitology 72: 803–806.

RIBAS, S. C., C. F. D. ROCHA, P. F. TEIXEIRA-FILHO, AND J. J. VICENTE.1995. Helminths (Nematoda) of the lizard Cnemidophorus ocellifer(Sauria: Teiidae): Assessing the effect of rainfall, body size and sexin the nematode infection rates. Ciencia e Cultura 47: 88–91.

SANCHIS, V., J. M. ROIG, M. A. CARRETERO, V. ROCA, AND G. A. LLO-RENTE. 2000. Host-parasite relationships of Zootoca vivipara (Sau-ria: Lacertidae) in the Pyrenees (North Spain). Folia Parasitologica47: 118–122.

SCHOENER, T. W. 1967. The ecological significance of sexual dimor-phism in size in the lizard Anolis conspersus. Science 155: 474–477.

———. 1971. Theory of feeding strategies. Annual Review of Ecologyand Systematics 2: 369–404.

SOUSA, B. M. 2000. Aspectos ecologicos, comportamentais e morfolo-gicos associados a alimentacao de Enyalius perditus Jackson, 1978(Sauria: Polychrotidae). Tese de Doutorado. Universidade FederalRural do Rio de Janeiro, Rio de Janeiro, Brasil, 169 p.

TROYER, K. 1982. Transfer of fermentative microbes between genera-tions in a herbivorous lizard. Science 216: 540–542.

VICENTE, J. J., H. O. RODRIGUES, D. C. GOMES, AND R. M. PINTO. 1993.Nematoides do Brasil. Parte III: Nematoides de Repteis. RevistaBrasileira de Zoologia 10: 19–168.

VOGEL, P., AND D. A. P. BUNDY. 1987. Helminth parasites of Jamaicananoles (Reptilia: Iguanidae): Variation in prevalence and intensitywith host age and sex in a population of Anolis lineatopus. Para-sitology 94: 399–404.

J. Parasitol., 93(1), 2007, pp. 213–215� American Society of Parasitologists 2007

Mitotic Responses to Injected Extracts of Larval and Adult Schistosoma mansoni inBiomphalaria glabrata: Effects of Dose and Colchicine Treatment

John T. Sullivan, Department of Biology, University of San Francisco, San Francisco, California 94117. e-mail: sullivan@usfca.edu

ABSTRACT: Biomphalaria glabrata snails injected with extracts ofSchistosoma mansoni miracidia, mother sporocyst excretory–secretoryproduct, cercariae, and adults, showed increased mitotic activity in his-tological sections of the amebocyte-producing organ (APO) relative towater-injected controls. The mitotic response was generally higher toextracts adjusted to 1.0 mg protein/ml than to a 10-fold lower concen-tration, although in most cases this increase was not statistically signif-icant. Colchicine treatment prior to fixation significantly increased thenumber of mitotic figures in APOs of all groups of extract-injectedsnails, both with respect to water-injected controls and, with 1 excep-

tion, relative to matched colchicine-untreated snails. Extracts of adultworms elicited a pronounced mitotic response, suggesting that adultsmay share a mitogenic molecule with larvae. The high variability incounts of mitotic figures may limit the usefulness of this histologicalmethod.

The hematopoietic amebocyte-producing organ (APO) (Lie et al.,1975) of Biomphalaria glabrata undergoes increased mitotic activityfollowing exposure of snails to miracidia of Schistosoma mansoni, and

214 THE JOURNAL OF PARASITOLOGY, VOL. 93, NO. 1, FEBRUARY 2007

FIGURE 1. Mean standard error of the mean of mitotic figures inhistological sections of the amebocyte-producing organ (APO) of Biom-phalaria glabrata at 48 hr postinjection with artificial pond water(APW), 1 mg/ml bovine serum albumin (BSA), or either 0.1 or 1.0 mgprotein/ml of one of the following extracts of Schistosoma mansoni:miracidia (MIR), excretory-secretory product (ESP), cercariae (CER),or adults (ADU). Shaded bars show counts from colchicine-untreatedsnails, and unshaded bars show counts from colchicine-treated (0.1% 6 hr) snails. Each bar represents counts of 10 APOs, except for 1.0 ESPcolchicine-untreated (n � 5) and 1.0 ESP colchicine-treated (n � 6)snails. Differences between bars marked by the same letter are not sta-tistically significant (P � 0.05, Student–Neuman–Keuls test, separatelyconducted on colchicine-untreated [lowercase letters] and colchicine-treated [uppercase letters] snails). Asterisks indicate a statistically sig-nificant difference between the means of colchicine-untreated and col-chicine-treated snails injected with the same material (P � 0.05, Stu-dent’s 2-tailed t-test).

also in response to injections of freeze–thaw extracts of miracidia andcercariae, although it is quite unresponsive to injections of bacteria andseveral other candidate mitogens (Sullivan et al., 1984, 2004). Mitoticresponses to miracidia and to injected miracidia extracts are signifi-cantly stronger in snails that are genetically resistant to S. mansoni thanin susceptible snails (Sullivan et al., 1984, 2004). Moreover, increasednumbers of mitotic figures occur in this organ when snails are incubatedin 0.1% colchicine from 4 to 24 hr prior to fixation (Sullivan and Castro,2005).

Although extracts of miracidia and cercariae are known to be mito-genic, the activity of other stages has not been investigated. Therefore,mother sporocyst excretory–secretory product (ESP) (Yoshino andLodes, 1988), and extracts of adult worms, were tested. Also, becauseprevious injections of parasite extracts had been normalized for numbersof larvae, the APO-stimulating activity of different life cycle stages,including those tested previously, was compared at a constant proteinconcentration of the extract (1 mg/ml and 0.1 mg/ml). Finally, to as-certain whether colchicine would facilitate measurement of mitotic re-sponses against injected parasite extracts, snails were incubated in col-chicine prior to fixation. Joky et al. (1985) treated B. glabrata withcolchicine to demonstrate mitotic activity of the APO following expo-sure to miracidia of Echinostoma caproni. However, because mitoticfigures were not enumerated, and responses of colchicine-untreatedsnails were not reported, it was not clear that colchicine treatment of-fered any advantage in quantifying responses of the APO to trematodeinfection.

The schistosome-resistant Salvador strain of B. glabrata, measuring9.5–10.5 mm in shell diameter, was used in this study (Paraense andCorrea, 1963). Methods for rearing snails, preparing 15,000 g super-natants of parasite freeze–thaw extracts, injecting 5 �l of extract intothe hemocoel, and histological analysis of APOs were those describedpreviously (Sullivan et al., 2004). Extracts of the following stages wereprepared: miracidia, cercariae, and adult worms. Miracidia were hatchedfrom eggs isolated from livers of mice, cercariae were collected frominfected snails, and adults were perfused from the hepatic portal cir-culation of infected mice. Additionally, ESP from mother sporocyststhat had transformed for 24 hr in vitro from miracidia in Chernin’sbalanced salt solution (CBSS) (Chernin, 1963) was tested. Extracts wereconcentrated with the use of centrifugal filtration devices having a 10kDa nominal molecular weight cutoff. Protein concentration was mea-sured colorimetrically (Bradford, 1976) and adjusted with artificial pondwater (APW) (Sullivan and Castro, 2005) to 1.0 or 0.1 mg/ml. Severaldifferent batches of miracidia extract were used, and yield was variable,with up to an approximately 10,000-fold concentration of the originalmiracidia suspension sometimes required to obtain 1 mg protein/ml,resulting in an estimated 1,100 miracidia equivalents/�l. Preparation ofESP was similar, except that it was not subjected to 5 freeze–thawcycles, and was diluted in CBSS instead of APW. Control injectionsconsisted of 5 �l of APW or 1 mg/ml bovine serum albumin (Bio-Rad,Hercules, California) in APW. Following injections, snails were main-tained at 27 C for 48 hr, and then their pericardial sac was removed,fixed in Bouin’s fluid, and processed for histology. An interval of 48hr postinjection (PI) was chosen because a previous study had shownthat the highest response to miracidia extract occurred at this time (Sul-livan et al., 2004). For colchicine treatment, snails were incubated in 4ml of 0.1% colchicine (Sigma Chemical Company, St. Louis, Missouri)for 6 hr prior to fixation, a procedure that was selected to provide thebest compromise between mitotic arrest and toxicity (Sullivan and Cas-tro, 2005). Total numbers of mitotic figures in the APO of each snailwere counted in hematoxylin and eosin–stained serial sections.

Data were statistically analyzed both by multiple and by pair-wisecomparisons of means. With separate analyses for colchicine-untreatedand colchicine-treated snails, mean numbers of mitotic figures for eachinjection group were tested by 1-way analysis of variance (ANOVA),and when a statistically significant difference between groups was in-dicated, individual means were compared with the Student–Neuman–Keuls test, using Statmost software (DataMost Corp., Salt Lake City,Utah). Additionally, for each injected substance, mean numbers of mi-totic figures for untreated and colchicine-treated snails were comparedwith the Student’s 2-tailed t-test. For all statistical tests, differences wereconsidered significant at P � 0.05.

Both for colchicine-untreated and colchicine-treated snails, ANOVArevealed statistically significant differences among means (F(9,85) � 3.39,

P � 0.001 and F(9,86) � 4.45, P � 0.00008, respectively). Among col-chicine-untreated snails, injections of all parasite extracts at 1 mg pro-tein/ml yielded a significant increase in mitotic activity relative to APW-injected controls (Fig. 1). At a concentration of 0.1 mg protein/ml, onlyextracts of miracidia and cercariae induced significant responses. Exceptfor the cercaria extract, mitotic responses to 1 mg protein/ml were high-er than to the same extract at 0.1 mg protein/ml. However, these in-creases were not statistically significant.

Among colchicine-treated snails, injections of all parasite extracts,both at 0.1 mg protein/ml and at 1 mg protein/ml, yielded a statisticallysignificant increase in mitotic activity relative to colchicine-treated,APW-injected controls. For all extracts, mitotic responses to 1 mg pro-tein/ml were higher than to the same extract at a concentration of 0.1mg protein/ml. However, these increases were statistically significantonly for extracts of miracidia and adults. Finally, when matched groupsof colchicine-treated and colchicine-untreated snails were compared,numbers of mitotic figures were higher in all cases in the former group,and these increases were statistically significant, except for injectionsof APW, 1 mg/ml BSA, and 0.1 mg protein/ml miracidia extract.

One goal of this study was to ascertain whether colchicine treatmentfacilitated the histological assay of APO mitotic responses. Previouswork had shown that incubation of Salvador B. glabrata in 0.1% col-chicine for 6 hr caused a statistically significant, approximately 3-fold,increase in mitotic figures relative to untreated controls (Sullivan andCastro, 2005). In the present study, exposure to colchicine caused anapproximately 2-fold, although statistically nonsignificant, increase incounts in APW-injected snails. Despite this increased background, col-chicine treatment, presumably by arresting dividing cells in metaphaseand thereby capturing a larger number of mitotic figures over the 6-hrincubation, proved useful in amplifying the response to injected parasiteextracts. For example, whereas injections of 0.1 mg protein/ml ESP and0.1 mg protein/ml adult extract did not stimulate a significant increasein mitotic activity in the absence of colchicine, effects of both treat-

RESEARCH NOTES 215

ments differed significantly from that of APW injection among colchi-cine-treated snails. Thus, although it raised mitotic counts both in con-trols and extract-injected snails, colchicine had a greater effect in thelatter, thereby increasing the sensitivity of the assay.

Mitotic responses to injections of extracts of miracidia and cercariaeamong colchicine-untreated snails were comparable to those reportedpreviously (Sullivan et al., 2004), i.e., both elicited increased mitoticactivity in the APO relative to APW-injected controls. The response tothe miracidia extract at 1 mg protein/ml was nearly the same as thatreported at 48 hr PI in the earlier study, which employed a proteinconcentration of 1.21 mg/ml.

Mitotic responses to ESP and to miracidia extracts were quite similarin magnitude. ESP of larval S. mansoni contains diverse glycoproteinsthat are known to have a range of inhibitory effects on hemocytes ofB. glabrata (Yoshino and Lodes, 1988; Lodes and Yoshino, 1989, 1990;Connors and Yoshino, 1990; Lodes et al., 1991). Paradoxically, how-ever, despite inhibiting hemocyte function, ESP has a significant mito-genic effect on the APO, the very structure that produces hemocytes.Humphries and Yoshino (2006) recently have shown phosphorylationof p38 mitogen-activated protein kinase in B. glabrata embryonic (Bge)cells, but not hemocytes, following in vitro exposure to ESP. WhetherESP has a similar effect on mitogen-activated protein kinases, particu-larly those involved in cell division, in hemocyte progenitors of theAPO is not known.

An interesting result of this study was the strong response to injec-tions of extracts from adult worms, inasmuch as a snail mitotic responseagainst a stage that occurs only within the vertebrate host would appearto have no adaptive significance. However, based on studies with mono-clonal antibodies, several antigens are known to be conserved amonglife cycle stages of schistosomes. For example, Schmitt et al. (2002)described 2 carbohydrate epitopes that were shared by miracidia, spo-rocysts, and cercariae of S. mansoni, S. haematobium, and S. japonicum,although neither epitope was detected on the surface of adult worms.Similarly, Nayame et al. (2002), described 3 glycan antigens of adultS. mansoni that were also shared by larval stages, although one of these,Lewis x antigen (Gal[Fuc� 1-3]�1-4GlcNAc-R), was not detected onmother or daughter sporocysts. Thus, the response to adult worms maysimply reflect shared mitogenic activity in one or more such conservedmolecules.

The dose–response relationship seen with the 2 tested protein con-centrations was generally weak, i.e., statistically nonsignificant. A sim-ilarly weak effect was found previously with injections of miracidiaextract at 72 hr PI (Sullivan et al, 2004). Detection of subtle changesin response as a function of dose by this method may be problematical,because of the observed high variability in mitotic counts, as discussedbelow.

Mitotic responses to the different extracts at the same protein con-centration were varied, most likely as a result of differences in theamounts of stimulatory substances, mitosis-inhibiting substances, orboth. Because the chemical nature of the active component is notknown, a protein concentration of 1 mg/ml may not result in equivalentactivity in each extract. Although no single extract was significantlymore potent than all others, responses to 1 mg protein/ml adult extractwere pronounced in APOs of some colchicine-treated snails, with 2 of10 having counts exceeding 14,000 mitoses/APO. Therefore, this prep-aration, despite its lack of direct biological relevance to snail-schisto-some interactions, may be a good candidate for further study of mitoticresponses to parasite molecules in the APO.

Even within the same treatment category, variability among mitoticcounts was high, for example ranging from 428 to 15,932 in the caseof colchicine-treated snails injected with 1 mg protein/ml adult extract.Variability for miracidia and cercaria extracts was comparable to thatobserved in the previous study (Sullivan et al., 2004). The degree towhich this variability reflects different effective doses actually deliveredto the APO by the injection method, temporal variations in the numberof mitotic figures within individual snails, or intersnail genetic vari-

ability in responsiveness to the parasite extract is unknown. However,such variability may obscure differences between treatment regimensand may, therefore, limit the usefulness of this histological approachfor identifying the active component or components of the parasite ex-tract.

I am indebted to Fred A. Lewis for infected mouse livers and adultworms, and to Timothy P. Yoshino for samples of ESP. This work wassupported by NIH Grant AI50546, the University of San Francisco Fac-ulty Development Fund, and the Fletcher Jones Foundation.

LITERATURE CITEDBRADFORD, M. M. 1976. A rapid and sensitive method for the quanti-

tation of microgram quantities of protein utilizing the principle ofprotein-dye binding. Analytical Biochemistry 72: 248–254.

CHERNIN, E. 1963. Observations on hearts explanted in vitro from thesnail Australorbis glabratus. Journal of Parasitology 49: 353–364.

CONNORS, V. A., AND T. P. YOSHINO. 1990. In vitro effect of larvalSchistosoma mansoni excretory-secretory products on phagocyto-sis-stimulated superoxide production in hemocytes from Biom-phalaria glabrata. Journal of Parasitology 76: 895–902.

HUMPHRIES, J. E., AND T. P. YOSHINO. 2006. Schistosoma mansoni ex-cretory-secretory products stimulate a p38 signalling pathway inBiomphalaria glabrata embryonic cells. International Journal forParasitology 36: 37–46.

JOKY, A., M. MATRICON-GONDRAN, AND J. BENEX. 1985. Response tothe amoebocyte-producing organ of sensitized Biomphalaria gla-brata after exposure to Echinostoma caproni miracidia. Journal ofInvertebrate Pathology 45: 28–33.

LIE, K. J., D. HEYNEMAN, AND P. YAU. 1975. The origin of amebocytesin Biomphalaria glabrata. Journal of Parasitology 63: 574–576.

LODES, M. J., V. A. CONNORS, AND T. P. YOSHINO. 1991. Isolation andfunctional characterization of snail hemocyte-modulating polypep-tide from primary sporocysts of Schistosoma mansoni. Molecularand Biochemical Parasitology 49: 1–10.

———, AND T. P. YOSHINO. 1989. Characterization of excretory-secre-tory proteins synthesized in vitro by Schistosoma mansoni primarysporocysts. Journal of Parasitology 75: 853–862.

———, AND ———. 1990. The effect of schistosome excretory-secre-tory products on Biomphalaria glabrata hemocyte motility. Journalof Invertebrate Pathology 56: 75–85.

NAYAME, A. K., T. P. YOSHINO, AND R. D. CUMMINGS. 2002. Differentialexpression of LacdiNAc, fucosylated LacdiNAc, and Lewis x gly-can antigens in intramolluscan stages of Schistosoma mansoni.Journal of Parasitology 88: 890–897.

PARAENSE, W. L., AND L. R. CORREA. 1963. Variation in susceptibilityof populations of Australorbis glabratus to a strain of Schistosomamansoni. Revista do Instituto de Medicina Tropical de Sao Paulo5: 15–22.

SCHMITT, J., M. WUHRER, J. HAMBURGER, J. JOURDANE, R. M. R. RAMZY,R. GEYER, AND A. RUPPEL. 2002. Schistosoma mansoni and Schis-tosoma haematobium: Identification and characterization of gly-coconjugate antigens in the hemolymph of infected vector snails.Journal of Parasitology 88: 505–513.

SULLIVAN, J. T., AND L. CASTRO. 2005. Mitotic arrest and toxicity inBiomphalaria glabrata (Mollusca: Pulmonata) exposed to colchi-cine. Journal of Invertebrate Pathology 90: 32–38.

———, T. C. CHENG, AND K. H. HOWLAND. 1984. Mitotic responses ofthe anterior pericardial wall of Biomphalaria glabrata (Mollusca)subjected to challenge. Journal of Invertebrate Pathology 44: 114–116.

———, S. S. PIKIOS, AND A. Q. ALONZO. 2004. Mitotic responses toextracts of miracidia and cercariae of Schistosoma mansoni in theamebocyte-producing organ of the snail intermediate host, Biom-phalaria glabrata. Journal of Parasitology 90: 92–96.

YOSHINO, T. P., AND M. J. LODES. 1988. Secretory protein biosynthesisin snail hemocytes: In vitro modulation by larval schistosome ex-cretory-secretory products. Journal of Parasitology 74: 538–547.