1163 J. Parasitol., 90(5), 2004, pp. 1163–1165 q American Society of Parasitologists 2004 Alternative Mechanism of Eimeria bovis Sporozoites to Invade Cells In Vitro by Breaching the Plasma Membrane J. H. Behrendt, W. Clauss, H. Zahner*, and C. Hermosilla* , Institute of Animal Physiology, Justus Liebig University Giessen, Wartweg 95, 35392 Giessen, Germany; *Institute for Parasitology, Justus Liebig University Giessen, Rudolf-Buchheim-Strasse 2, 35392 Giessen, Germany; To whom correspondence should be addressed. e-mail: [email protected]FIGURE 1. Cell wound assay. A–B. ‘‘Positive’’ control. Scratch mark in a BSLEC monolayer, applied in the presence of FITC-dextran. (A, phase contrast picture, arrows indicate dextran positive cells; B, fluorescence picture.). C–D. Pattern after infection of a BSLEC monolayer with Eimeria bovis sporozoites in the presence of FITC-dextran. (C, phase contrast picture, arrows indicate dextran positive cells; D, fluorescence picture.) ABSTRACT: In vitro Eimeria bovis sporozoites invade a wide range of cell types, and in the case of bovine cells, they may develop to first- generation schizonts. Often, however, they subsequently leave their host cell to invade a new one, which seems contrary to the classical way of infecting a cell by forming a parasitophorous vacuole. Using a standard, ‘‘cell wound assay,’’ we show that E. bovis can invade bovine endo- thelial cells by breaching the plasma membrane and may again leave the surviving cell. Eimeria bovis sporozoites also infected VERO and HT29 cells but obviously without damaging the plasma membrane. The same held true when bovine endothelial cells were exposed to tachy- zoites of Toxoplasma gondii and Neospora caninum. According to a literature report dealing with Plasmodium yoelii sporozoites, breaching the membrane of certain host cells may be a common phenomenon for coccidian sporozoites but may not be for merozoites. Eimeria bovis is a common and pathogenic coccidium of cattle, which may cause severe hemorrhagic enteritis (Daugschies et al., 1998). Its endogenous development shows a number of peculiarities. Sporo- zoites are liberated in the host’s gut and must invade endothelial cells of the central lymph capillaries in the villi of the ileum, where they replicate, forming multinucleated macroschizonts, which contain hun- dreds of thousands of first-generation merozoites. Second-generation schizonts and gamonts then develop quickly in epithelial cells of the large intestine (Hammond et al., 1944). It is unclear how the sporozoites reach their destination. If migrating directly, they would need to traverse the ileum epithelial cell layer. Because invasion of cells by coccidia is a complex process involving the release of particular parasite products and an invagination of the host cell plasma membrane to form a par- asitophorous vacuole (PV) around the invading stage (Dubremetz et al., 1998; Entzeroth et al., 1998), one might speculate that sporozoites take a paracellular route on their way to the predetermined host cell. How- ever, while monitoring the invasion of cells by E. bovis sporozoites in vitro, we have often observed parasites entering and leaving cells quick- ly and repeatedly, without harming them severely, as has been described by others (Fayer and Hammond, 1967). A similar observation was re- cently reported by Mota et al. (2001) for sporozoites of Plasmodium yoelii, which similarly traverse hepatocytes in vitro. The latter authors suggested that these sporozoites may use an alternative mode of inva- sion by breaching the host cell membrane without forming a PV. This breaching of the plasma membrane would be followed by a rapid mem- brane repair, i.e., cells usually survive it. To determine whether E. bovis sporozoites also share the ability to simply traverse cells, we performed several experiments using a variety of host cell types and compared the results with those obtained in the course of host cell invasion by Toxoplasma gondii and Neospora can- inum tachyzoites. The E. bovis strain was isolated in 1988 in the field in northern Germany and maintained by passages in calves (Fiege et al., 1992). Sporozoites were isolated from oocysts according to Hermosilla et al. (2002). Toxoplasma gondii (RH strain; Sabin, 1941) tachyzoites were harvested from the peritoneal cavity of BALB/c mice 48 hr after intra- peritoneal injection of the parasites and were washed several times with phosphate-buffered saline (PBS), then centrifuged at 400 g for 10 min. Neospora caninum (strain NC-1; Dubey et al., 1988) was maintained in VERO cells. Tachyzoites were washed off the cultures and prepared as above. The experiments used bovine aortic endothelial cells (BAEC), bovine spleen lymphatic endothelial cells (BSLEC), African green monkey kid- ney cells (VERO), and human colon adenocarcinoma cells (HT29). BAEC were isolated from freshly resected aortas by collagenase diges- tion (20 min in 0.2% collagenase type II [Worthington Biochemical Corp., Lakewood, New Jersey] in Puck’s saline A [PSA, GIBCO, Eg- genstein, Germany] salt solution at 37 C and 5% CO 2 ). They were washed, resuspended in endothelial cell growth medium (ECGM; Prom-
Transcript
1163
J. Parasitol., 90(5), 2004, pp. 1163–1165q American Society of Parasitologists 2004
Alternative Mechanism of Eimeria bovis Sporozoites to Invade Cells In Vitro byBreaching the Plasma Membrane
J. H. Behrendt, W. Clauss, H. Zahner*, and C. Hermosilla*†, Institute of Animal Physiology, Justus Liebig University Giessen, Wartweg 95,35392 Giessen, Germany; *Institute for Parasitology, Justus Liebig University Giessen, Rudolf-Buchheim-Strasse 2, 35392 Giessen, Germany;†To whom correspondence should be addressed. e-mail: [email protected]
FIGURE 1. Cell wound assay. A–B. ‘‘Positive’’ control. Scratch mark in a BSLEC monolayer, applied in the presence of FITC-dextran. (A,phase contrast picture, arrows indicate dextran positive cells; B, fluorescence picture.). C–D. Pattern after infection of a BSLEC monolayer withEimeria bovis sporozoites in the presence of FITC-dextran. (C, phase contrast picture, arrows indicate dextran positive cells; D, fluorescencepicture.)
ABSTRACT: In vitro Eimeria bovis sporozoites invade a wide range ofcell types, and in the case of bovine cells, they may develop to first-generation schizonts. Often, however, they subsequently leave their hostcell to invade a new one, which seems contrary to the classical way ofinfecting a cell by forming a parasitophorous vacuole. Using a standard,‘‘cell wound assay,’’ we show that E. bovis can invade bovine endo-thelial cells by breaching the plasma membrane and may again leavethe surviving cell. Eimeria bovis sporozoites also infected VERO andHT29 cells but obviously without damaging the plasma membrane. Thesame held true when bovine endothelial cells were exposed to tachy-zoites of Toxoplasma gondii and Neospora caninum. According to aliterature report dealing with Plasmodium yoelii sporozoites, breachingthe membrane of certain host cells may be a common phenomenon forcoccidian sporozoites but may not be for merozoites.
Eimeria bovis is a common and pathogenic coccidium of cattle,which may cause severe hemorrhagic enteritis (Daugschies et al., 1998).Its endogenous development shows a number of peculiarities. Sporo-zoites are liberated in the host’s gut and must invade endothelial cellsof the central lymph capillaries in the villi of the ileum, where theyreplicate, forming multinucleated macroschizonts, which contain hun-dreds of thousands of first-generation merozoites. Second-generationschizonts and gamonts then develop quickly in epithelial cells of thelarge intestine (Hammond et al., 1944). It is unclear how the sporozoitesreach their destination. If migrating directly, they would need to traversethe ileum epithelial cell layer. Because invasion of cells by coccidia isa complex process involving the release of particular parasite productsand an invagination of the host cell plasma membrane to form a par-asitophorous vacuole (PV) around the invading stage (Dubremetz et al.,1998; Entzeroth et al., 1998), one might speculate that sporozoites takea paracellular route on their way to the predetermined host cell. How-ever, while monitoring the invasion of cells by E. bovis sporozoites in
vitro, we have often observed parasites entering and leaving cells quick-ly and repeatedly, without harming them severely, as has been describedby others (Fayer and Hammond, 1967). A similar observation was re-cently reported by Mota et al. (2001) for sporozoites of Plasmodiumyoelii, which similarly traverse hepatocytes in vitro. The latter authorssuggested that these sporozoites may use an alternative mode of inva-sion by breaching the host cell membrane without forming a PV. Thisbreaching of the plasma membrane would be followed by a rapid mem-brane repair, i.e., cells usually survive it.
To determine whether E. bovis sporozoites also share the ability tosimply traverse cells, we performed several experiments using a varietyof host cell types and compared the results with those obtained in thecourse of host cell invasion by Toxoplasma gondii and Neospora can-inum tachyzoites.
The E. bovis strain was isolated in 1988 in the field in northernGermany and maintained by passages in calves (Fiege et al., 1992).Sporozoites were isolated from oocysts according to Hermosilla et al.(2002). Toxoplasma gondii (RH strain; Sabin, 1941) tachyzoites wereharvested from the peritoneal cavity of BALB/c mice 48 hr after intra-peritoneal injection of the parasites and were washed several times withphosphate-buffered saline (PBS), then centrifuged at 400 g for 10 min.Neospora caninum (strain NC-1; Dubey et al., 1988) was maintainedin VERO cells. Tachyzoites were washed off the cultures and preparedas above.
The experiments used bovine aortic endothelial cells (BAEC), bovinespleen lymphatic endothelial cells (BSLEC), African green monkey kid-ney cells (VERO), and human colon adenocarcinoma cells (HT29).BAEC were isolated from freshly resected aortas by collagenase diges-tion (20 min in 0.2% collagenase type II [Worthington BiochemicalCorp., Lakewood, New Jersey] in Puck’s saline A [PSA, GIBCO, Eg-genstein, Germany] salt solution at 37 C and 5% CO2). They werewashed, resuspended in endothelial cell growth medium (ECGM; Prom-
1164 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
FIGURE 2. Cell wounding and membrane repair effects in bovineendothelial (BAEC, BSLEC), African green monkey kidney (VERO),and human colon adenocarcinoma cells (HT29) after in vitro infectionwith Eimeria bovis sporozoites (104 sporozoites/10 cm2 cell monolayer).Proportion of dextran-positive (wounded and repaired) cells. Verticalbar: standard error of the mean.
ocell, Heidelberg, Germany) supplemented with L-glutamine (2 mM),penicillin (50 IU/ml), streptomycin (50 mg/ml) and 10% fetal calf serum(FCS), seeded on glass coverslips (15 mm, NeoLab, Heidelberg, Ger-many) and cultured in 4-well culture plates (Nunc, Wiesbaden, Ger-many) at 37 C and 5% CO2. BSLEC were originally prepared frombovine splenic lymphatic vessels and cultured as above. VERO cellsand HT29 cells were purchased from the European collection of cellcultures (ECACC, Salisbury, United Kingdom) and maintained in Ros-well Park Memorial Institute (RPMI) 1640 medium supplemented asabove.
Breaching of cell membrane by parasites and membrane repair weredetermined according to McNeil et al. (1999) by a ‘‘cell wound assay.’’The assay uses a cell-impermeant tracer (fluorescein isothiocyanate[FITC]–labeled dextran, MW 10,000; Molecular Probes, Leiden, Neth-erlands), which penetrates wounded cells and is trapped inside aftermembrane repair. Cell monolayers on glass coverslips were exposed tothe parasites (E. bovis: all cell types; T. gondii and N. caninum: onlyBAEC and BSLEC) in the presence of the tracer (5 mg/ml) diluted inthe corresponding cell culture medium and incubated for 2 hr at 37 Cand 5% CO2. Subsequently, the cells were washed 3 times with PBS toremove exogenous tracer and fixed 15 min in 4% paraformaldehyde(Merck, Darmstadt, Germany). The glass coverslips were transferred tomicroscopic slides, covered with Mowiol mounting medium (Hoechst,Frankfurt am Main, Germany) and incubated at 4 C in darkness over-night. As a positive control, a cell monolayer was incubated as above,but without parasites. After 30 min of incubation, the monolayer wasscratched with the tip of a blood lancet. Using a fluorescence micro-scope (Leitz excitation filter 515–560), the numbers of cells, parasites,and wounded or repaired cells were determined for 10 microscopicfields per coverslip.
The assay gave clear results. Endocytosed FITC–dextran was ob-served as a fine dotted pattern in the cell cytosol and was easily distin-guished from homogenous staining in wounded cells (Fig. 1). Deadcells, which did not reseal, were usually not labeled. In the positivecontrol, dextran-positive cells were detected along the scratch mark(Figs. 1A, B). FITC-dextran did not affect the motility and infectivityof E. bovis sporozoites or T. gondii and N. caninum tachyzoites. Onprinciple, they moved by gliding and flexing on the cell monolayer andgenerally invaded all cell types offered, although in the case of E. bovisthe infection was approximately 5-fold greater in bovine endothelialcells than in VERO or HT29 cells. Eimeria bovis infection of endothe-lial cells resulted in a relatively high proportion of dextran-positivecells. Using an inoculum size of 4 3 104 sporozoites per coverslip, upto 2.5 and 3.5% of BAEC and BSLEC were stained, respectively. Un-infected control cells were only occasionally stained. The number ofdextran-positive cells was a linear function of the number of sporozoitesdetected in the cells. The linear trend was highly significant as evaluatedby the t-test (P , 0.001). Ratios between the numbers of parasites andstained cells were 0.95 6 0.23 and 0.91 6 0.08 in BAEC and BSLEC,respectively. In general, dextran-positive cells did not contain parasites.
These data strongly suggest that E. bovis sporozoites, as well as spo-rozoites of P. yoelii (Mota et al., 2001), possess mechanisms of tra-versing cells before invasion of a final host cell for replication. Althoughon the basis of the performed cell wound assay we could not distinguishbetween breaching of the plasma membrane due to invasion or egress,findings on other apicomplexan parasites, as reviewed by Mota andRodriguez (2001), suggest that a mechanism of entry different from theclassic way, i.e., from forming a PV, is involved. Considering that theability to egress from once-invaded cells has also been reported forsporozoites of other Eimeria species, (Speer et al., 1971; Long andSpeer, 1977; Danforth et al., 1984; Danforth et al., 1992; Chobotar etal., 1993) and T. gondii (Nichols and O’Conner, 1981; Speer et al.,1997) and that sporozoites of coccidia have been frequently found with-out a PV in the cytosol of host cells shortly after infection, both in vivoand in vitro (Aikawa et al., 1984; Danforth et al., 1992; Chobotar etal., 1993), this capability to traverse cells seems common in the Eu-coccidiida. As previously speculated by Mota et al. (2001), the advan-tage could be an ability of sporozoites to reach their final destinationby crossing physical cellular barriers, particularly in cases when thesuitable targets of the parasites are not available at the place of entryinto the host.
On the basis of this assumption, we expected that cells unsuitable forthe further development of the parasite, i.e., VERO or HT29 cells (see
Hermosilla et al., 2002), would be traversed by the sporozoites moreoften than suitable host cells such as those from the bovine endothelium.However, the opposite was observed. Thus, the proportion of dextran-positive cells was considerably low in E. bovis–infected VERO andHT29 cell cultures. In general, it did not exceed the proportion ofstained cells in the uninfected controls (0.026 6 0.016 vs. the controls,which were 0.038 6 0.038 in VERO; and 0.059 6 0.033 vs. the con-trols, which were 0.056 6 0.030 in HT29 cells) (Fig. 2). This meansthat hardly any cell wounding or repair occurred because of E. bovissporozoite penetration. The reason for this cell type-dependent differ-ence is presently not known. It may depend on the cell donor speciesbecause sporozoites did not traverse cells of human or monkey origin,whereas traversal also occurred in other bovine cell types not includedin this study (data not shown). Better understanding of the phenomenonwould require knowledge on the molecular events accompanying theparasites’ breaching of the host cell membrane and the related parasiteegress, which is not available.
However, the ability of coccidian parasites to traverse cells withoutforming a PV seems to be limited to sporozoites. We could not studythe behavior of E. bovis merozoites because there is currently no celltype available, which is regularly invaded by E. bovis merozoites invitro. We repeated the experiments of Mota et al. (2001) with T. gondiimerozoites using endothelial cells instead of hepatocytes and extendedthe study to merozoites of N. caninum. Both parasites invaded BAECand BSLEC rapidly, but obviously only in the common manner byforming a PV (next generation schizonts developed within 2–3 days)because the proportion of dextran-positive cells in infected cell cultureswas almost ‘‘zero.’’ Even using inoculum sizes that resulted in infectionof 20 and 30%, respectively, did not change the pattern (data notshown). This stage-specific capacity of migration through cells mayreflect the different requirements of the different developmental stagesof apicomplexan parasites. In fact, the sporozoite stages are the formsof the parasite that encounter physical barriers to reach their final hostcell and replication site for schizogony. Thus, the ability of E. bovissporozoites to traverse by disrupting the plasma membrane of the hostcell may be an essential requirement to overcome the physical barriersthat separate the sporozoite from their final destination and to ensure asuccessful infection in the mammalian host.
We would like to thank K. Preissner in Giessen for the kind donationof BSLEC.
RESEARCH NOTES 1165
LITERATURE CITED
AIKAWA, M., A. SCHWARTZ, S. UNI, R. NUSSENZWEIG, AND M. HOLLING-DALE. 1984. Ultrastructure of in vitro cultured exo-erythrocyticstage of Plasmodium berghei in a hepatoma cell line. AmericanJournal of Tropical Medicine and Hygiene 33: 792–799.
CHOBOTAR, B., H. D. DANFORTH, AND R. ENTZEROTH. 1993. Ultrastruc-tural observations of host-cell invasion by sporozoites of Eimeriapapillata in vivo. Parasitology Research 79: 15–23.
DANFORTH, H. D., B. CHOBOTAR, AND R. ENTZEROTH. 1984. Cellularpathology in mouse embryonic brain cells following in vitro pen-etration by sporozoites of Eimeria papillata. Zeitschrift fur Paras-itenkunde 70: 165–171.
———, R. ENTZEROTH, AND B. CHOBOTAR. 1992. Scanning and trans-mission electron microscopy of host cell pathology associated withpenetration by Eimeria papillata sporozoites. Parasitology Re-search 78: 570–573.
DAUGSCHIES, A., H. J. BURGER, AND M. AKIMURA. 1998. Apparent di-gestibility of nutrients and nitrogen balance during experimentalinfection of claves with Eimeria bovis. Veterinary Parasitology 77:93–102.
DUBEY, J. P., A. L. HATTEL, D. S. LINDSAY, AND M. J. TOPPER. 1988.Neonatal Neospora caninum infection in dogs: Isolation of thecausative agent and experimental transmission. Journal of theAmerican Veterinary Medical Association 193: 1259–1263.
DUBREMETZ, J. F., N. GARCIA-REQUET, V. CONSEIL, AND M. N. FOUR-MAUX. 1998. Apical organelles and host-cell invasion by Apicom-plexa. International Journal for Parasitology 28: 1007–1013.
ENTZEROTH, R., F. R. MATTIG, AND R. WERNER-MEIER. 1998. Structureand function of the parasitophorous vacuole in Eimeria species.International Journal for Parasitology 28: 1015–1018.
FAYER, R., AND D. M. HAMMOND. 1967. Development of first-generationschizonts of Eimeria bovis in cultured bovine cells. Journal of Pro-tozoology 14: 764–772.
FIEGE, N., D. KLATTE, D. KOLLMANN, H. ZAHNER, AND H. J. BURGER.1992. Eimeria bovis in cattle: Colostral transfer of antibodies andimmune response to experimental infections. Parasitology Research78: 32–38.
HAMMOND, D. M., L. R. DAVIS, AND L. BOWMANN. 1944. Experimentalinfections with Eimeria bovis in calves. American Journal of Vet-erinary Research 5: 303–311.
HERMOSILLA, C., B. BARBISCH, A. HEISE, S. KOWALIK, AND H. ZAHNER.2002. Development of Eimeria bovis in vitro: Suitability of severalbovine, human and porcine endothelial cell lines, bovine fetal gas-trointestinal, Madin-Darby bovine kidney (MDBK) and Africangreen monkey kidney (VERO) cells. Parasitology Research 88:301–307.
LONG, P. L., AND C. A. SPEER. 1977. Invasion of host cells by Coccidia.In Parasite invasion, A. E. R. Tayler and R. Muller (eds.). Black-well Scientific, London, U.K., p. 1–16.
MCNEIL, P. L., M. F. S. CLARKE, AND K. MIYAKE. 1999. Cell woundassays. In Current protocols in cell biology, I. J. S. Bonifacino, M.Dasso, J. B. Harford, J. Lippincott-Schartz, and K. M. Yamada(eds.). Wiley, New York, p. 12.4.1–12.4.15.
MOTA, M. M., G. PRADEL, J. P. VANDERBERG, J. C. R. HAFALLA, U.FREVERT, R. S. NUSSENZWEIG, V. NUSSENZWEIG, AND A. RODRIGUEZ.2001. Migration of Plasmodium sporozoites through cells beforeinfection. Science 291: 141–144.
———, AND A. RODRIGUEZ. 2001. Migration through host cells by ap-icomplexan parasites. Microbes and Infection 3: 1123–1128.
NICHOLS, B. A., AND G. R. O’CONNER. 1981. Penetration of mouse peri-toneal macrophages by the protozoon Toxoplasma gondii. New ev-idence for active invasion and phagocytosis. Laboratory Investi-gation 44: 324–335.
SABIN, A. 1941. Toxoplasmic encephalitis in children. Journal of theAmerican Medical Association 116: 801–814.
SPEER, C. A., L. R. DAVIS, AND D. M. HAMMOND. 1971. Cinemicro-graphic observations of the development of Eimeria larimerensisin cultured bovine cells. Journal of Protozoology 18: 11.
———, J. P. DUBEY, J. A. BLIXT, AND K. PROKOP. 1997. Time lapsevideo microscopy and ultrastructure of penetrating sporozoites,types 1 and 2 parasitophorous vacuoles, and the transformation ofsporozoites to tachyzoites in the VEG strain of Toxoplasma gondii.Journal of Parasitology 83: 565–574.
J. Parasitol., 90(5), 2004, pp. 1165–1169q American Society of Parasitologists 2004
A Focus of Human Infection by Haplorchis taichui (Trematoda: Heterophyidae) in theSouthern Philippines
Vicente Y. Belizario, Jr., Winifreda U. de Leon, Mary Joan J. Bersabe*, Purnomo†, J. Kevin Baird†, and Michael J. Bangs†‡, College ofPublic Health, University of the Philippines, Pedro Gil, Ermita, Manila, Philippines; *Center for Health Development for Southern Mindanao,Department of Health, Davao City, Philippines; †U.S. Naval Medical Research Unit No. 2 (NAMRU-2), Kompleks P2M/PLP, Jl. PercetakanNegara No. 29, Jakarta 10560, Indonesia; ‡To whom correspondence should be addressed. e-mail: [email protected]
ABSTRACT: We report an exceptionally high rate of infection by Hap-lorchis taichui (Nishigori, 1924) in human populations on MindanaoIsland, southern Philippines. This intestinal fluke is seldom encountered,and this is the first report of high prevalence of infection (36%) inhumans by H. taichui in the Philippines. The likely source of haplor-chine infection has been linked to consumption of raw or undercookedfreshwater fish containing infective metacercariae. The most commonclinical symptoms appeared as upper abdominal discomfort or pain andborborygmi. Praziquantel (75 mg/kg divided in 3 doses in 1 day) wasa well-tolerated and effective treatment for infection by H. taichui.
Food-borne trematodiases are an important cause of disease in hu-mans that create an adverse economic impact in endemic tropical areas(Rim et al., 1994; WHO, 1995). Until recently, the prevalence of het-erophyidiasis in humans had been reported as very low in the Philip-pines. For example, less than 1% of more than 30,000 stool specimensexamined in nationwide surveys during 1970–1980 were found to con-tain heterophyid ova (Cross and Basaca-Sevilla, 1984), with site-specific
infection rates not exceeding 3%. Intestinal parasite surveys were con-ducted in 1998 and 1999 in 2 communities in Monkayo, CompostelaValley Province, southern Mindanao in connection with an outbreak ofintestinal capillariasis (Belizario et al., 2000). Survey results indicatedheterophyid infection rates of 17% and 16%, respectively, in the generalpopulation (Belizario et al., 2000, 2001). Infections were reported asheterophyid (Heterophyidae) flukes on the basis of egg size and mor-phology, without definitive species identification. This report examinedadult worms recovered after treatment of patients given praziquanteland provides conclusive identification of Haplorchis taichui (Tremato-da: Heterophyidae) as the parasitic agent. Additional epidemiologicalbackground on the investigations conducted in 2000 from infection foci,including symptomatology and treatment of H. taichui infections, isprovided. All clinical aspects of the investigation were conducted in fullaccordance with Good Clinical Practice Guidelines and provisions ofthe World Medical Association Declaration of Helsinki (amended,1996).
The barangay (village) of San Isidro is in the municipality of Mon-
1166 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
TABLE I. Age and sex distribution of study participants and those in-fected with Haplorchis taichui infection in Barangay San Isidro, Mon-kayo, Compostela Valley, May 2000.
FIGURES 1–2. Haplorchis taichui (Nishigori) 1. Pyriform adult, glycerin whole mount, unstained, showing transverse rows of scale-like spineson tegument. Bar: 200 mm. 2. Carmine-stained anterior portion of adult worm showing subterminal oral sucker and ventrogenital complex (smallarrow) containing ventral sucker armed with prominent spines. Bar: 100 mm.
kayo, Compostela Valley Province, southern Mindanao (Region XI),Philippines, and located inland ;100 km north of coastal Davao City.Farming is the most common occupation, with corn as the major cul-tivated crop, followed by rice, coconut, and banana. Based on RuralHealth Unit (RHU) records, nearly 35% of interviewed households (n5 270) did not have dedicated toilet–latrine facilities. The people ofSan Isidro obtain water from 2 primary sources, i.e., rainwater and theSaug River. Rainwater is preferred for cooking and washing purposes.The river regularly provides freshwater fish, crabs, shrimps, and snailsthat supplement the diet with additional protein. A detailed descriptionof the study site has been provided elsewhere (Belizario et al., 2000).
A cross-sectional, active case detection survey was conducted in May2000 in San Isidro. Health interviews and stool samples were obtainedfrom residents with a history of recent bowel disturbance (abdominaldiscomfort or pain, or diarrhea, or both). The principal complaints,signs, and symptoms were noted, and a fresh stool specimen was ob-tained from each patient. Stool was examined as wet mounts usingeither the Kato thick smear (Kato–Katz) technique or the formalin–ethersedimentation technique (FEST) (Ash and Orihel, 1987; WHO, 1998).Processed stool samples were initially examined under the microscope
RESEARCH NOTES 1167
FIGURES 3–4. Haplorchis taichui 3. Ventral sucker with crescentic group of 14 hollow spines (sclerites). Bar: 25 mm. 4. Operculated eggs(30–35 mm long, 15–18 mm wide) from stool showing prominent ‘‘shoulders’’ at apex and thickened posterior prominence. Bar: 30 mm.
by trained microscopists assigned to the local government health facil-ities, with on-site cross-checking conducted by staff from the DiagnosticParasitology Laboratory, College of Public Health, University of thePhilippines Manila (UP-CPH). A subsequent independent quality con-trol of sample results was conducted by UP-CPH. Using the Kato–Katztechnique, heterophyid eggs in stool were counted along with otherhelminths (Ascaris lumbricoides, Trichuris trichiura, Capillaria philip-pinensis, and hookworm eggs) (Belizario et al., 2001). Heterophyid eggswere counted on a standardized template grid and multiplied by a factorof 24 to approximate the number of eggs present per gram (epg) offeces. A classification of infection intensity (‘burden’) was used to rep-resent light (1–100 epg), moderate (101–1,000 epg), and heavy (.1,000epg) infections (Belizario et al., 2001).
Sixty-seven patients infected with H. taichui were given commer-cially available praziquantel (Distocidet, Shin Poong Pharm. Co., Ltd.,Seoul, Republic of Korea), 75 mg/kg, divided in 3 doses in 1 day. Stoolspecimens were collected from 35 patients between 7 and 14 days aftertreatment and examined using the Kato–Katz technique. The MonkayoRHU microscopist performed the initial examination in the field, andthe 10% formalin-fixed specimens were later forwarded to the UP-CPHDiagnostic Parasitology Laboratory for crosschecking and confirmation.Patient response to drug treatment was categorized as one of the fol-lowing: (1) clinical cure, i.e., absence of eggs on repeat stool exami-nation; (2) clinical improvement, i.e., lower intensity of associatedsymptoms and infection (eggs) on repeat examination; (3) clinical fail-ure, i.e., similar or higher intensity of infection on repeat stool exami-nation; and (4) indeterminate, i.e., no repeat stool examination possible(patient loss to follow-up or noncompliance).
Use of the Kato–Katz method or FEST revealed that 87 patients hadevidence of heterophyidiasis, i.e., an overall infection rate of 36.0%(Table I). From fresh stool, the Kato–Katz method revealed 31% ofsamples with heterophyid ova compared with 13.6% for FEST. Hap-lorchis taichui infection was seen in all age groups from 19 mo to 73
yr of age (mean 27.2). Infection prevalence was greatest (55.3%) in the15–30-yr-old age group, followed by the 31–45-yr-old cohort, togetherrepresenting 62% of the sampled population found infected among the6 age groups. Infection prevalence was generally higher in age groups5 yr and older, although 16.7% of children less than 5 yr of age had H.taichui infection. Overall, male subjects had a higher, but not statisti-cally significant (chi-square, P , 0.05), infection prevalence comparedwith females (30%). The majority of cases (71%) had moderate to heavyinfection burdens, particularly patients .15 yr of age. Most cases inage groups below 15 yr had light infections. The infection burden (es-timated egg density) ranged from 24 to 26,256 epg, with a mean (geo-metric) egg count of 256 epg.
In San Isidro, 83 patients were interviewed for clinical signs andsymptoms associated with H. taichui infection. A review of their his-tories showed that 38 (45.8%) patients complained of peptic ulcer–likesymptoms, with upper abdominal discomfort or pain reported by 35(42.2%), followed by borborygmi in 20 (24.1%) cases. Other com-plaints included nausea, chronic diarrhea, and weight loss. Fourteen(16.9%) patients reported no complaints and were considered asymp-tomatic. Although this was not a case–control study, most symptomsassociated with H. taichui infection quickly resolved after completingthe 1-day praziquantel treatment.
Thirty-five (52.2%) patients submitted specimens for follow-up stoolexamination after treatment. Using the Kato–Katz technique, 11 patients(31.4%) were found to have at least 1 helminth species infection. Only1 patient (2.9%) had stool containing heterophyid eggs. The arithmeticand geometric means of Haplorchis egg reduction in stool were 99.3%and 98.4%, respectively, with an overall cure of 97%. Similarly, com-plete clinical cure was seen in 34 of 35 (97%) patients, with significantclinical improvement seen in the 1 patient who did not completely re-cover. None of the treated follow-up patients exhibited clinical failure.Clinical outcome was indeterminate in 40 (53.3%) of the treated patientsbecause of loss to follow-up. In agreement with Radomyos et al. (1998),
1168 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
1-day praziquantel administration appears well tolerated and effectiveagainst H. taichui and could be considered for use in a mass treatment–control campaign.
Fresh stool specimens were obtained after treatment with praziquanteland fixed in 10% formalin for later study. For the identification of theheterophyid specimens, formalin-preserved stool was prepared as astandard direct wet mount and stained with iodine for detection of ova.Egg morphology identified a heterophyid-like origin but could not pro-vide definitive species diagnosis. Formalin-preserved adult flukes werewashed in physiological saline solution, mounted whole, either un-stained or stained with carmine red, and mounted in Canada balsam(Ash and Orihel, 1987). The condition of adult worms recovered fromformalin-preserved stool varied from excellent to slight disfigurementof the integument. In most cases, preserved worms expelled after treat-ment with praziquantel could be adequately stained and identified. Pho-tographic documentation was made using a Nikon UFX-DX microscopesystem (Figs. 1–4). All adult worms examined (Figs. 1–2) conformedto taxonomic descriptions for the genus Haplorchis Looss 1899 andspecifically to H. taichui (Pearson and Ow-Yang, 1982). Of particulardifferential diagnostic importance is the ventrogenital complex with adistinctive ventral sucker containing a crescentric group of 12–16 hol-low sclerites, some ending in a laterodorsal notch (Fig. 3). Operculated,thick-shelled eggs (30–35 mm long, 15–18 mm wide) examined fromstool were within the described size range for this species and clearlyshowed the prominent ‘‘shoulders’’ at apex and thickened posteriorprominence (Fig. 4).
In Asia, H. taichui (Nishigori, 1924) has been reported in animalhosts from Iraq, India, Sri Lanka, Taiwan, Thailand, Malaysia, and thePhilippines (Faust and Nishigori, 1926; Pearson and Ow-Yang, 1982;Radomyos et al., 1983, 1998; Waikagul, 1991; Sukontason et al., 2000).In Southeast Asia, human infections of H. taichui have only been de-scribed from Thailand and the Philippines (Cross, 1974; Waikagul,1991). This trematode species has been reported from natural and ex-perimental animal infections in the Philippines (Pearson and Ow-Yang,1982). Natural molluscan (Melania juncea Lea) and picine (Puntius spp.and Ophiocephalus striatus Bloch) species serving as intermediate hostsof this parasite have been described from Luzon and Mindanao (Tub-angui, 1947; Velasquez, 1973a, 1973b). The common source of infec-tion for humans likely involves a variety of freshwater fish that serveas a second intermediate host and harbor infective metacercaria en-cysted in muscle tissue. Humans, domestic dogs, cats, swine, and thebird Bubulcus ibis coromandus (Boddaert) have been reported as defin-itive hosts of H. taichui in the Philippines (Velasquez, 1973a). The fullrange of intermediate hosts and parasite distribution in the Philippinesrequires more investigation.
Haplorchine infections have generally been rare in humans but re-main a concern because of occasional reports of severe pathologicchanges in tissue (e.g., myocarditis) resulting from the minute eggs ofheterophyids penetrating through the intestinal wall and lodging in ex-traintestinal sites (Africa et al., 1935, 1940). High prevalence (63%) ofH. taichui in humans has been reported as the most common intestinalfluke in northern Thailand (Radomyos et al., 1998), indicating that thisparasite can pose a significant health risk in certain endemic foci. In1998, intestinal parasitism was identified as the second leading causeof morbidity in Monkayo. Hookworm was the most common helminthidentified followed by C. philippinensis, heterophyid species, A. lum-bricoides, and T. trichiura (Monkayo RHU Records, 1998). Intestinalparasite surveys conducted after the discovery of capillariasis in thearea in 1998 identified 17 barangays in Mindanao with heterophyidinfections (Belizario et al., 2001), representing the largest cluster ofendemic heterophyidiasis so far detected in the Philippines and outsideThailand. Based on these findings, we suspect many of the infectionswere likely the result of H. taichui, although 5 other members in theHaplorchinae (Haplorchis Looss, Procerovuom Onji and Nishio, andStellantchasmus Onji and Nishio) have also been reported to infect hu-mans in the Philippines (Waikagul, 1991).
Local eating habits almost certainly contributed to the high preva-lence of H. taichui. The Saug River is a major source of fresh food forthe local residents. Most people are fond of consuming a variety offreshwater fish, shrimps, crabs, and frogs. In particular, many freshwaterfish are eaten raw and seasoned only with salt and vinegar, a localpreparation called kinilaw. Among the local varieties of fish eaten arespecies of tilapia (Oreochromis sp.), paitan (Hypophthalmichthys mol-
itrix), dalag (O. striatus), and buriring (Poecilia sp). Other popularmethods of preparing and cooking fish, such as sabaw (boiling for sev-eral minutes) and sugba (grilling over charcoal), may result in incom-plete cooking of flesh and be an insufficient means to kill the encystedmetacecariae. The relatively lower prevalence of infection in sampledgroups below 15 yr of age may reflect a lesser tendency to consumeraw food preparations compared with adults.
The high prevalence of H. taichui found in the southern Philippineswould likely have gone undetected had it not been for investigationson intestinal capillariasis in the area. More recent investigations haveshown the infection to have a wider distribution in Mindanao than pre-viously reported (V. Y. Belizario Jr., pers. comm.). The underreportingof unusual infections by clinicians is not uncommon because of a lackof familiarity with most trematode infections and the inexperience oflaboratory diagnosticians detecting small helminth eggs. A number offactors may have contributed to the high prevalence of H. taichui inMonkayo. It has also been speculated that there may have been anincreased risk of infection with this parasite, in addition to the C. phi-lippinensis outbreak, a nematode associated with the consumption ofraw or undercooked freshwater fish, as an indirect result of the severedrought conditions during the 1997–1998 El Nino Southern Oscillationclimatic event. During that time, food crops were in short supply, whichlikely resulted in an increased reliance on nutrition derived from theriver. It would seem reasonable to conclude that periodic climatic anom-alies and conditions may influence exposure risk to helminthic infec-tions by changing human behavior and food habits. If such associationsexist, the public’s awareness of increased risk during times of environ-mental stress, together with community education on proper hygienichandling and preparation of freshwater fish and the sanitary disposal ofhuman feces, should help diminish the risk of heterophyid infectionsand other food-borne related parasitic zoonoses (WHO, 1995, 1998).Department of Health efforts continue in Mindanao to identify, treat,and educate affected communities to combat this unpleasant and poten-tially dangerous helminthic infection.
We extend special thanks for assistance provided by the Center forHealth Development for Southern Mindanao, Department of Health, thelocal government units of Monkayo, Compostela Valley, and SalcedoEduardo of the University of the Philippines, Los Banos. This studywas supported by the Essential National Health Research, Departmentof Health, RP and the Committee on Research Implementation and De-velopment, College of Medicine, University of the Philippines Manila,with assistance from the U.S. Naval Medical Research and Develop-ment Command, Navy Department. The opinions of the authors do notpurport to reflect the positions of the U.S. Navy or the Department ofHealth of the Republic of the Philippines.
LITERATURE CITED
AFRICA, C. M., W. E. D DE LEON, AND E. Y. GARCIA. 1935. Intestinalheterophyidiasis with cardiac involvement: A contribution to eti-ology of heart failure. Journal of the Philippine Islands MedicalAssociation 15: 358–361.
———, ———, ———. 1940. Visceral complications in intestinal het-erophyidiasis of man. Acta Medicine Philippina Monograph Series1: 1–132.
ASH, L. R., AND T. C. ORIHEL. 1987. Parasites: A guide to laboratoryprocedures and identification. American Society Clinical Patholo-gists Press, Chicago, Illinois, 328 p.
BELIZARIO, V. Y., M. J. BERSABE, W. U. DE LEON, V. Y. HILOMEN, G. V.PALLER, A. D. DE GUZMAN, AND M. G. BUGAYONG. 2001. Intestinalheterophyidiasis: An emerging food-borne parasitic zoonosis insouthern Philippines. Southeast Asian Journal of Tropical Medicineand Public Health 32: 36–42.
———, W. U. DE LEON, D. G. ESPARAR, J. M. GALANG, J. FANTONE,AND C. VERDADERO. 2000. Compostela Valley: A new endemic fo-cus for capillariasis philippinensis. Southeast Asian Journal ofTropical Medicine and Public Health 31: 479–481.
CROSS, J. H. 1974. Diagnostic methods in intestinal fluke infections: Areview. In Diagnostic methods for important helminthiasis and am-oebiasis in Southeast Asia and the Far East, C. Harinsuta and D.C. Reynolds (eds.). SEAMEO-TROPMED Project, Bangkok, Thai-land, p. 87–108.
———, AND V. BASACA-SEVILLA. 1984. Biomedical surveys in the Phil-
RESEARCH NOTES 1169
ippines. Monograph SP-47. United States Naval Medical ResearchUnit 2, Manila, Philippines, 117 p.
FAUST, E. C., AND M. NISHIGORI. 1926. The life cycles of two newspecies of heterophyidae, parasitic in mammals and birds. Journalof Parasitology 13: 91–132.
PEARSON, J. C., AND C. K. OW-YANG. 1982. New species of Haplorchisfrom Southeast Asia together with keys to the Haplorchis-group ofheterophyid trematodes of the region. Southeast Asian Journal ofTropical Medicine and Public Health 13: 35–60.
RADOMYOS, B., D. BUNNAG, AND T. HARINASUTA. 1983. Haplorchis pum-ilio (Looss) infection in man in northeastern Thailand. SoutheastAsian Journal of Tropical Medicine and Public Health 14: 223–227.
———, T. WONGSAROJ, P. WILAIRATANA, P. RADOMYOS, R. PRAEVANICH,V. MEESOMBOON, AND P. JONGSUKSUNTIKUL. 1998. Opisthorchiasisand intestinal fluke infections in northern Thailand. SoutheastAsian Journal of Tropical Medicine and Public Health 22: 123–127.
RIM, H.-J., H. F. FARAG, S. SORNMANI, AND J. H. CROSS. 1994. Food-borne trematodes: Ignored or emerging? Parasitology Today 10:207–209.
SUKONTASON, K. L., K. SUKONTASON, B. KUNTALUE, N. BOONSRIWONG,S. PIANGJAI, U. CHAITHONG, AND P. VANITTANAKOM. 2000. Surfaceultrastructure of excysted metacercariae of Haplorchis taichui(Trematoda: Heterophyidae). Southeast Asian Journal of TropicalMedicine and Public Health 31: 747–754.
TUBANGUI, M. A. 1947. A summary of the parasitic worms reportedfrom the Philippines. Philippine Journal of Science. 76: 225–322.
VELASQUEZ, C. C. 1973a. Observations on some heterophyidae (Trem-atoda: Digenea) encysted in Philippine fishes. Journal of Parasitol-ogy 59: 77–84.
———. 1973b. Intramolluscan stages of Haplorchis taichui (Nishigori)in Melania juncea Lea in the Philippines. Journal of Parasitology59: 281.
WAIKAGUL, J. 1991. Intestinal fluke infections in Southeast Asia. South-east Asian Journal of Tropical Medicine and Public Health 22Suppl: 158–162.
WHO. 1995. Control of food-borne trematode infections. Technical Re-port Series 849. World Health Organization, Geneva, Switzerland,168 p.
WHO. 1998. Guidelines for the evaluation of soil-transmitted helmin-thiasis and schistosomiasis at community level. WHO/CTD/SIP/98.1. World Health Organization, Geneva, Switzerland, 48 p.
J. Parasitol., 90(5), 2004, pp. 1169–1171q American Society of Parasitologists 2004
Observations on Myiasis by the Calliphorids, Bufolucilia silvarum and Bufoluciliaelongata, in Wood Frogs, Rana sylvatica, From Southeastern Wisconsin
Matthew G. Bolek and John Janovy Jr., School of Biological Sciences, University of Nebraska–Lincoln, Lincoln, Nebraska 68588. e-mail:[email protected]
ABSTRACT: Larvae of certain species of blowflies (Calliphoridae) cancause myiasis in frogs and toads, but there are few reports from NorthAmerican amphibians. Of these, most are from toads (bufonids). In thisstudy, we observe primary myiasis in a population of juvenile woodfrogs, Rana sylvatica, collected on 22–23 August 2003, from south-eastern Wisconsin and compare our observations with previous studieson myiasis from toads. Two (5%) of 39 frogs were infected by the blowfly Bufolucilia silvarum, with an intensity of 28 and 31, whereas 1(2.5%) of 39 frogs was infected by the blow fly Bufolucilia elongatawith an intensity of 14. We found that (1) B. silvarum lay eggs onhealthy wood frogs, (2) eggs hatch, with first-instar maggots penetratingunder the skin, (3) maggots develop to mature third instars within 13–16 hr of egg hatching, (4) maggots kill the host within 7–47 hr of egghatching, and (5) maggots consume the entire frog carcass reducing itto bones within 42–59 hr of egg hatching. Our observations on the timeof death and how quickly carcasses of wood frogs were consumed bythese maggots compared with previous studies on toads suggest thatfinding infected juvenile wood frogs may be uncommon. Therefore,myiasis by these flies on wood frogs and other small terrestrial anuransmay be a phenomenon that is much more common than is currentlyobserved. This is the first report of B. silvarum and B. elongata causingmyiasis in wood frogs.
Myiasis in amphibians is caused by larvae of dipterans from Sarco-phagidae, Calliphoridae, and Chloropidae, some of which can causesubstantial mortality in their amphibian hosts (Dasgupta, 1962; Crumpand Pounds, 1985; Schell and Burgin, 2001; Bolek and Coggins, 2002).During 2003, the blowflies, Bufolucilia silvarum and Bufolucilia elon-gate, were found infesting juvenile wood frogs, Rana sylvatica, insoutheastern Wisconsin. Members of this genus have been reported asobligate or facultative parasites of amphibians, particularly species ofBufo, but there are few reports of these flies causing myiasis in NorthAmerican ranid frogs (see Bolek and Coggins, 2002). Our observationson the rapid consumption of wood frog carcasses by these maggotssuggest that myiasis on small terrestrial anurans may be more commonin North America than is currently observed.
In North America, 2 species of Bufolucilia have been reported tocause myiasis in 3 species of amphibians. Bufolucilia elongata causedmyiasis in 1 boreal toad, Bufo boreas boreas, and 6 American toadsfrom Colorado and Wisconsin, respectively, whereas B. silvarum wasreported from 48 bullfrogs, Rana catesbeiana, in California, 1 Americantoad from Nova Scotia (Canada), and 1 American toad from Ontario,Canada (James and Maslin, 1947; Hall, 1948; Anderson and Bennett,1963; Bleakney, 1963; Briggs, 1975). More recently, studies on the lifehistory of B. silvarum from 9 American toads examined by Bolek andCoggins (2002) indicate that these flies deposit eggs on the back andflanks of the amphibian host. The larvae hatch and migrate under theskin where they form a single lesion in the paratoid glands, back, neck,and front or hind legs, where development takes place, and all infectedtoads die within 1 day to 2 wk of infection (Bolek and Coggins, 2002).No other information is available on these flies infecting North Amer-ican amphibians. In their review on the life history of B. silvarum,Bolek and Coggins (2002) hypothesized that juvenile terrestrial anuransthat are diurnal and overlap in their ecology with this fly species maybe more prone to parasitism by B. silvarum than is currently known. Inthis study, we report observations on the life history of B. silvarum andB. elongata in juvenile wood frogs, R. sylvatica, from southeastern Wis-consin and compare these data with infections by these 2 species offlies in American toads examined by Briggs (1975) and Bolek and Cog-gins (2002) from Wisconsin. Our observations indicate that these fliesmay kill small frogs in the genus Rana and consume the carcass morerapidly than toads and therefore may be less commonly observed par-asitizing these hosts in nature. These observations may account for thenumerous reports of myiasis of Bufolucilia spp. and other calliphoridssuch as Phaenicia sericata and Lucilia illustris in toads and rarely inother species of North American frogs (Anderson and Bennett, 1963;Stewart and Foote, 1974; Bolek and Coggins, 2002).
Thirty-nine juvenile wood frogs were collected by hand in the woodsduring the day on 22–23 August 2003 at the University of Wisconsin–Milwaukee field station, Ozaukee County, Wisconsin (438239N,88829W). All frogs were measured (2.97 6 0.27 cm) and examined forexternal lesions, eggs, or maggots. Frogs suspected of being infected
1170 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
FIGURES 1–8. Photographs of successive stages and fate of a singleyoung-of-the-year Rana sylvatica infected with Bufolucilia silvarumfrom southeastern Wisconsin. 1. Juvenile R. sylvatica with recently de-posited B. silvarum eggs attached to the back. Bar 5 1 cm. 2. Highermagnification of eggs. Bar 5 0.5 cm. 3. Frog with a single lesion pre-sent on the right hip, with second instars being present. Bar 5 1 cm.4. Frog with third instars present within 13–16 hr of egg hatching. Bar5 1 cm. 5. Higher magnification of third instars. Bar 5 0.25 cm. 6.Third instars continued to feed on the carcass. Bar 5 1 cm. 7. Thirdinstars quickly reduced the carcass into liquid slurry of tissue and bones.Bar 5 1 cm. 8. Third instars reduced the carcass to bones and migratedout to pupate within 42–59 hr of eggs hatching. Bar 5 1 cm.
were placed in individual 8.45-L tanks lined with moist paper towelsfor observation. Dead frogs were placed in 70-ml plastic jars along withtheir maggots and observed for time of carcass consumption. All otherfrogs were killed and necropsied within 72 hr of collection. Third-stagemaggots from each frog were placed in individual 70-ml plastic jarscontaining moist sand and allowed to pupate. Some were boiled indistilled water and fixed in 95% ethanol and cleared in 10% KOH. Thecephalopharingial skeleton and posterior spiracles of some third instarswere dissected and mounted in glycerin as temporary slides. Adult flieswere fed granulated sugar and banana peels for at least 24 hr beforebeing killed by freezing and were pinned or preserved in 70% ethanol.Adult flies were identified to genus (Shewell, 1987) and to species bykeys in Hall (1948) and Hall and Townsend (1977). Prevalence, inten-sity, and mean intensity are according to Bush et al. (1997). Student’st-test was used to compare differences in snout vent length (SVL), meanintensity of Bufolucilia spp. maggots, and time of survival (in hr) ofAmerican toads collected by Bolek and Coggins (2002) and wood frogsinfected in this study. An approximate t-test was calculated when var-iances were heteroscedastic (Sokal and Rohlf, 1981). Adult and larvalflies were deposited in the Harold W. Manter Laboratory collection(University of Nebraska State Museum, Lincoln, Nebraska; HWML45414, B. silvarum larvae; 45415, B. silvarum adult flies; HWML45416, B. elongata larvae; 45417, B. elongata adult flies).
Only 3 of 39 (7.7%) juvenile wood frogs collected on 22–23 August2003 were infected by green blow flies. Two (5%) of the frogs wereinfected with B. silvarum with intensities of 28 and 31, whereas 1(2.5%) frog was infected with B. elongata with an intensity of 14.
The single frog infected with B. elongata was collected with a singlelesion on the abdomen, with third-instar maggots being clearly visible;this frog died within 24 hr of collection. The other 2 frogs were col-lected with deposited fly eggs. These were white and attached to theamphibians back (Figs. 1, 2). Neither individual appeared to show anydiscomfort and no wound or lesion was observed on the skin of thesefrogs. Eggs hatched on 1 frog collected on 23 August 2003 sometimebetween 11.5 and 19 hr after collection during the night, with secondinstars being present in a single wound on 24 August 2003 (Fig. 3).Eggs hatched within 13.5–15.5 hr of collection on the second frog.Observations indicated that both frogs had single lesions on the righthip (Fig. 3), with third instars being visible on the 2 frogs within 13–16 hr of egg hatching (Figs. 4, 5). These 2 frogs died within 7 and 47hr of egg hatching. Maggots of both species of calliphorids continuedto feed on the carcass (Fig. 6) reducing it to liquid slurry of tissue andbones (Fig. 7). Maggots consumed the entire frog carcass, reducing itto bones within 42–59 hr of egg hatching (Fig. 8). Third-stage maggotsof B. silvarum turned into pupa within 2 days after leaving the carcassremains and emerged as flies within 7–8 days at room temperature.Third-stage maggots of B. elongata turned into pupa within 3 days afterleaving the carcass remains and emerged as flies within 7–8 days atroom temperature.
Comparisons of SVL, mean intensities of Bufolucilia spp., and timeof survival (in hr) of 9 American toads and 3 wood frogs infected withBufolucilia spp. collected by Bolek and Coggins (2002) from WaukeshaCounty, Wisconsin, during 1998 and this study are given in Table I.There were statistically significant differences in SVL, intensities, andtime of survival among these hosts. Although these sample sizes aresmall, the 9 juvenile-infected toads were larger, had significantly lowerintensities of maggots, and lived significantly longer in the laboratory.It is unclear why wood frogs had higher mean intensity of Bufoluciliaspp. than American toads. Observations on egg hatching of B. silvarumin 2 toads by Bolek and Coggins (2002) indicate that toads vigorouslyrub their hind legs over their back as if they were trying to dislodgethe eggs. We observed egg hatching on a single wood frog, which didnot attempt to dislodge any eggs with its hind legs during hatching. Itmay be that American toads can reduce the number of calliphorid mag-gots by dislodging eggs during hatching. More importantly, our obser-vations on the time of death and how quickly carcass of wood frogswere consumed by these maggots suggest that finding infected juvenilewood frogs may be uncommon. Bolek and Coggins (2002) reported thatinfected toads died within 1 day to 2 wk; the maggots never consumedthe entire toad carcass before leaving to pupate. In addition, observa-tions by Briggs (1975) on infections of B. elongata on American toadsin Wisconsin indicated that numerous toads were observed alive anddead at his study site.
These wood frogs were significantly smaller than toads examined byBolek and Coggins (2002). Size probably plays a role in how quicklythey are consumed by maggots of these flies. Adult wood frogs reacha maximum size of only 3–5 cm, and most adults spend long inactiveperiods in the leaf litter or under logs during the summer (Vogt, 1981).The only common wood frogs found during the day in the summer arejuveniles, and we suspect that this age group of frogs overlaps ecolog-ically more commonly with Bufolucilia spp., which are diurnal (Hall,1948; Vogt, 1981). These observations, along with the known time ofconsumption of carcasses of juvenile wood frogs in this study, indicatethat wood frogs and other small anurans that overlap in their ecologywith calliphorids may be more commonly parasitized by these flies butare rarely observed compared with bufonids.
We thank Melissa A. Ewert, Brigham and Women’s Hospital, Boston,
RESEARCH NOTES 1171
TABLE I. Comparisons of SVL, mean intensity (MI), and time until death of juvenile American toads, Bufo americanus, and juvenile wood frogs,Rana sylvatica, infected with maggots of Bufolucilia species from southeastern Wisconsin.
Variable
Host
B. americanus(n 5 9*)
R. sylvatica(n 5 3)
Statistics
t P
SVL (cm 6 1 SD)MI (61 SD)Time until death (hr 6 1 SD)
4.1 6 0.2310.5 6 7.2
114.33 6 100.84
2.6 6 0.2824 6 926 6 20
8.782.862.48
0.00010.0170.035
* Data from Bolek and Coggins (2002).
Massachusetts, for help in collecting frogs and James R. Coggins andGretchen A. Meyer, Department of Biological Sciences, University ofWisconsin–Milwaukee, for allowing us access to collect frogs at theUWM field station. We thank Agustin Jimenez Ruiz, Harold W. ManterLaboratory, University of Nebraska–Lincoln, for reviewing the manu-script and 2 anonymous reviewers for improvements on the manuscript.This work was supported by a grant to M.G.B. from the School ofBiological Sciences Special Funds, University of Nebraska–Lincoln.
LITERATURE CITED
ANDERSON, R. C., AND G. F. BENNETT. 1963. Ophthalmic myiasis inamphibians in Algonquin Park, Ontario, Canada. Canadian Journalof Zoology 41: 1169–1170.
BLEAKNEY, J. S. 1963. First North American record of Bufolucilia sil-varum (Meigen) (Diptera: Calliphoridae) parasitizing Bufo terres-tris americanus Holbrook. The Canadian Entomologist 95: 107.
BOLEK, M. G., AND J. R. COGGINS. 2002. Observations on myiasis bythe calliphorid, Bufolucilia silvarum, in the eastern American toad(Bufo americanus americanus) from southeastern Wisconsin. Jour-nal of Wildlife Diseases 38: 598–603.
BRIGGS, J. L. 1975. A case of Bufolucilia elongata shannon 1924 (Dip-tera: Calliphoridae) myiasis in the American Toad, Bufo american-us Holbrook 1836. Journal of Parasitology 61: 412.
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.
CRUMP, M. L., AND J. A. POUNDS. 1985. Lethal parasitism of an Apo-sematic anuran (Atelopus varius) by Notochaeta bufonivora (Dip-tera: Sarcophagidae). Journal of Parasitology 71: 588–591.
DASGUPTA, B. 1962. On the myiasis of the Indian toad Bufo melanos-tictus. Parasitology 52: 63–66.
HALL, D. G. 1948. The blowflies of North America. The Thomas SayFoundation, Baltimore, Maryland, 477 p.
HALL, R. D., AND L. H. TOWNSEND. 1977. The insects of Virginia No.11. Research division bulletin 123. Virginia Polytechnic Instituteand State University, Blacksburg, Virginia, 48 p.
JAMES, T. M., AND T. P. MASLIN. 1947. Notes on myiasis of the toad,Bufo boreas boreas Baird and Girard. Journal of the WashingtonAcademy of Sciences 37: 366–368.
SCHELL, C. B., AND S. BURGIN. 2001. Batrachomyia strigapes (Diptera)parasitism of Uperoleia laevigata (Anura). Journal of Parasitology87: 1215–1216.
SHEWELL, G. E. 1987. Calliphoridae. In Manual of Nearctic Diptera 2,Monograph 28, J. F. McAlpine (ed.). Research Branch AgricultureCanada, Ottowa, Canada, p. 1133–1145.
SOKAL, R. R., AND J. F. ROHLF. 1981. Biometry, 2nd ed. W. H. Freemanand Company, New York, 859 p.
STEWART, S., AND R. H. FOOTE. 1974. An unusual infestation by Phaen-icia sericata (Mg.) (Diptera: Calliphoridae). Proceedings of the En-tomological Society of Washington 76: 466.
VOGT, R. C. 1981. Natural history of amphibians and reptiles of Wis-consin. The Milwaukee Public Museum and Friends of the MuseumInc., Milwaukee, Wisconsin, 205 p.
J. Parasitol., 90(5), 2004, pp. 1171–1174q American Society of Parasitologists 2004
Biologic and Molecular Characteristics of Toxoplasma gondii Isolates From StripedSkunk (Mephitis mephitis), Canada Goose (Branta canadensis), Black-Winged Lory (Eoscyanogenia), and Cats (Felis catus)
J. P. Dubey, P. G. Parnell*, C. Sreekumar, M. C. B. Vianna, R. W. De Young†, E. Dahl‡, and T. Lehmann‡, United States Department ofAgriculture, Agricultural Research Service, Animal and Natural Resources Institute, Animal Parasitic Diseases Laboratory, Building 1001,Beltsville, Maryland 20705-2350; *Clemson Veterinary Diagnostic Center, P.O. Box 102406, Columbia, South Carolina 29224-2406; †CaeserKleberg Wildlife Research Institute, MSC 218,700 University Boulevard, Kingsville, Texas 78363; ‡Division of Parasitic Diseases, Centers forDisease Control and Prevention, 4770 Buford Highway, MS: F22, Chamblee, Georgia 30341. e-mail: [email protected]
ABSTRACT: Toxoplasma gondii isolates can be grouped into 3 geneticlineages. Type I isolates are considered virulent to outbred mice, where-as Type II and III isolates are not. In the present report, viable T. gondiiwas isolated for the first time from striped skunk (Mephitis mephitis),Canada goose (Branta canadensis), and black-winged lory (Eos cyan-ogenia). For the isolation of T. gondii, tissues were bioassayed in mice,and genotyping was based on the SAG2 locus. Toxoplasma gondii wasisolated from 3 of 6 skunks, 1 of 4 Canada geese, and 2 of 2 feral cats(Felis catus) from Mississippi. All donor animals were asymptomatic.Viable T. gondii was also isolated from 5 of 5 lories that had died ofacute toxoplasmosis in an aviary in South Carolina. Genotypes of T.
gondii isolates were Type III (all skunks, lories, and the goose) andType II (both cats). All 5 Type III isolates from birds and 2 of the 3isolates from skunks were mouse virulent.
Toxoplasma gondii infections are widely prevalent in human beingsand animals worldwide (Dubey and Beattie, 1988; Tenter et al., 2000).Postnatally, humans become infected by ingesting tissue cysts from un-dercooked meat or consuming food or drink contaminated with oocysts.However, only a small percentage of exposed adult humans developclinical signs after exposure. It is not known whether the severity of
1172 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
TABLE I. Isolation of Toxoplasma gondii from animals.
* Of 5 mice inoculated.† ND, no data.‡ Mice used for genotyping.§ Died of bacterial infection.
toxoplasmosis in individuals is due to the parasite strain, host variabil-ity, or other factors.
Overall, there is low genetic diversity among T. gondii isolates so farexamined. Toxoplasma gondii isolates have been classified in 3 geneticTypes (I, II, and III) based on restriction fragment length polymorphism(Howe and Sibley, 1995; Howe et al., 1997; Grigg, Bonnefoy et al.,2001). It has been suggested that isolates of Type I and II are morelikely to result in clinical toxoplasmosis in humans, but genetic char-acterization has been limited essentially to isolates from patients ill withclinical toxoplasmosis (Howe et al., 1997; Fuentes et al., 2001; Grigg,Gantara et al., 2001; Aspinall et al., 2003). The prevalence of T. gondiitypes in asymptomatic humans is unknown. In 1 study, all 86 isolatesof T. gondii obtained from women (or their fetuses), who acquired in-fection during pregnancy, were genotyped: 73 were Type II, 2 wereType III, 4 were atypical, and 7 were Type I (Ajzenberg et al., 2002).Most reported isolates of T. gondii from domestic animals from theUnited States and Europe were Type II or Type III (Howe and Sibley,1995; Mondragon et al., 1998; Owen and Trees, 1999; Dubey, Gambleet al., 2002; Jungersen et al., 2002; Lehmann et al., 2003), irrespectiveof clinical status. Recently, 70% of isolates of T. gondii obtained fromasymptomatic free-range chickens from Brazil were Type I (Dubey,Graham et al., 2002; Dubey, Graham, Silva et al., 2003; Dubey, Navarroet al., 2003).
Little is known of the prevalence and distribution of genotypes of T.gondii in wildlife species. Recently, we reported genotypes of T. gondiiisolates from white-tailed deer, raccoons, coyotes, black bears, and foxes(Dubey et al., 2004). In this study, we describe the genetic and biologiccharacteristics of isolates of T. gondii from striped skunks (Mephitismephitis), a Canada goose (Branta canadensis), and black-winged lories(Eos cyanogenia). This is the first report of isolation of T. gondii fromskunks, geese, and lories.
The 6 skunks and the 2 feral cats (Table I) were obtained from north-eastern Mississippi in July 2003. They were trapped around humandwellings in Starkville, Mississippi, by the United States Departmentof Agriculture Wildlife Service at the request of owners. The 4 Canadageese were killed by hunters in September 2003. They were from theTennessee–Tombigbee Waterway, in the vicinity of Columbus, Missis-sippi.
Tissues from 5 lories from an aviary in South Carolina were obtainedin 2 batches in October 2001 (Table I). The birds had died of acutetoxoplasmosis (P. G. Parnell, unpubl.). Tissue fluids were obtained fromthese birds for serology because serum was not available. In the firstbatch, livers from 3 lories (nos. 1–3) were sent from South Carolina toMaryland for confirmation of diagnosis. When it was realized that thelivers had been frozen for 6 days at 220 C, the eviscerated carcassesof the animals were retrieved from the cold storage and muscle tissue
was removed for parasite isolation. Fluid for serology assays was ob-tained by compression and partial maceration of liver tissue. In thesecond batch, livers and lung tissue from 2 lories were obtained; fluidwas taken from lung tissue for serology.
Serum samples were tested for antibodies to T. gondii with the mod-ified agglutination test (MAT) as described (Dubey and Desmonts,1987). Sera were diluted 2-fold starting at 1:25 or 1:5 dilutions.
Squash preparations were made from the brains of the skunks andcats and examined microscopically for the presence of tissue cysts.Brains from cats, skunks, and geese were bioassayed in outbred femaleSwiss Webster mice obtained from Taconic Farms, Germantown, NewYork, as reported previously (Dubey, Graham et al., 2002). Each brainwas homogenized, digested in acidic pepsin, washed, and the homog-enate inoculated subcutaneously into 5 mice (Dubey, 1998). Avian tis-sues were homogenized in 0.85% NaCl aqueous solution (saline), andthe homogenates were inoculated into mice without digestion in pepsin.Tissue imprints of mice that died were examined for T. gondii tachy-zoites or tissue cysts. Survivors were bled 5 wk postinoculation (PI),and a 1:25 dilution of serum from each mouse was tested for T. gondiiantibodies with the MAT. Mice were killed 62 days PI, and their brainswere examined for tissue cysts as described (Dubey and Beattie, 1988).Mice were considered infected with T. gondii when tachyzoites or tissuecysts were found in murine tissues.
Samples of lung tissue from dead mice as well as from those killedwere frozen at 270 C for DNA characterization as described (Lehmannet al., 2000). Polymerase chain reaction–restriction fragment lengthpolymorphism genotypes of the SAG2 locus were used to geneticallycharacterize the isolates (Howe et al., 1997).
Toxoplasma gondii was isolated from the brains of 3 of 6 skunks(Table I). Mice inoculated from the brains of 2 skunks died from acutetoxoplasmosis with demonstrable tachyzoites. These isolates of T gondiialso killed subinoculated mice. Tissue cysts of T. gondii were found bydirect microscopic examination of brain smears from skunk no. 3. In-dividual tissue cysts were microisolated from the brain homogenate ofskunk no. 3 as described (Sreekumar et al., 2003) and inoculated sub-cutaneously into 3 mice, 1 tissue cyst per mouse. The mice inoculatedwith individual tissue cysts died from acute toxoplasmosis 13 days PIwith numerous demonstrable tachyzoites in lungs. All isolates of T.gondii from skunks, including clones were Type III.
Antibodies to T gondii were found in 1 feral cat examined, and T.gondii was isolated from the brains of both cats bioassayed; both iso-lates were Type II (Table I). Toxoplasma gondii was isolated from thebrain of 1 of the 4 geese sampled, and this isolate was Type III.
Viable T. gondii was isolated from the tissues of 5 of 5 lories; all 5isolates were Type III and were virulent for mice (Table I). A few miceinoculated with these avian tissues died 1–5 days PI because of bacterial
RESEARCH NOTES 1173
infections and were discarded (Table I). Numerous T. gondii tachyzoitesor tissue cysts were found in tissues of dead birds. No viable T. gondiiwas isolated from frozen livers, confirming earlier findings that freezingkills T. gondii (Dubey and Beattie, 1988). Antibodies to T. gondii werefound in fluids obtained from avian tissues (Table I).
Isolation of viable T. gondii in the present study from the skunk,black-winged lory, and goose is a new host record. Walton and Walls(1964) found T. gondii antibodies in 1 of the 5 skunks from Georgiabut were unable to isolate the parasite from the brains of 4 skunksbioassayed in mice. Diters and Nielsen (1978) histologically diagnosedtoxoplasmosis in a skunk that was immunosuppressed due to concurrentdistemper virus infection. Dubey, Hamir et al. (2002) found a T. gondii–like tissue cyst in tongue of 1 of 4 skunks. However, the diagnosis couldnot be confirmed in either of these 2 previous reports.
Among the avian hosts of T. gondii, passeriform birds are most sus-ceptible for clinical toxoplasmosis (Dubey, 2002). There are a few re-ports of clinical toxoplasmosis in birds of the Psittaciformes. Thesereports include toxoplasmosis in 2 budgerigars (Melopsittacus undula-tus) from Switzerland (Galli-Valerio, 1939), 1 Swanson’s Lorikeet (Tri-choglossus moluccanus) from the Netherlands (Poelma and Zwart,1972), 1 regent parrot (Polytelis anthopeplus), 1 superb parrot (P.swainsonii), and 1 crimson rosella (Platycercus elegans) from Australia(Hartley and Dubey, 1991). There is 1 report of clinical toxoplasmosisin a red lory (E. bornea) from the United States. (Howerh et al., 1991).A T. gondii–like parasite was found histologically in sections of tissuesof this bird. Serum and unfixed tissues were not available for confir-mation in any of the reports in Psittaciformes.
Both antibodies to T. gondii and viable organisms were found inblack-winged lories. A few of the mice died of bacterial infection 1–5days PI with avian tissues. The tissues had been stored in a cooler for7 days before bioassays were performed. The lories had died of T.gondii–associated hepatitis and pneumonia, and therefore muscle tissuewas not collected initially for T. gondii examination.
Before molecular genotyping (Howe and Sibley, 1995), T. gondiiisolates were grouped as mouse virulent or mouse avirulent based oninfectivity in outbred mice. However, mouse virulence has no correla-tion with disease in humans or domestic animals (Dubey and Beattie,1988). For example, the strains of T. gondii isolated from tissues ofaborted ovine fetuses were not mouse virulent. Initial studies by Howeet al. (1997) indicated that all Type I strains of T. gondii were lethalfor mice, irrespective of the dose, whereas Types II and III were rela-tively avirulent for mice. However, data are now accumulating that in-dicate that based on the SAG2 locus the phenotypic and genetic char-acters may not be tightly correlated (Ajzenberg et al., 2002, oDubey,Graham et al., 2002; Lehmann et al., 2004). In the present study, all 5Type III isolates from the birds and 2 of 3 isolates from skunk werevirulent for mice.
LITERATURE CITED
AJZENBERG, D., N. COGNE, L. PARIS, M. H. BESSIERES, P. THULLIEZ, D.FILISETTI, H. PELLOUX, P. MARTY, AND M. L. DARDE. 2002. Geno-type of 86 Toxoplasma gondii isolates associated with human con-genital toxoplasmosis, and correlation with clinical findings. Jour-nal of Infectious Diseases 186: 684–689.
ASPINALL, T. V., E. C. GUY, K. E. ROBERTS, D. H. M. JOYNSON, J. E.HYDE, AND P. F. G. SIMS. 2003. Molecular evidence for multipleToxoplasma gondii infections in individual patients in England andWales: Public health implications. International Journal for Para-sitology 33: 97–103.
DITERS, R. W., AND S. W. NIELSEN. 1978. Toxoplasmosis, distemper, andherpesvirus infection in a skunk (Mephitis mephitis). Journal ofWildlife Diseases 14: 132–136.
DUBEY, J. P. 1998. Refinement of pepsin digestion method for isolationof Toxoplasma gondii from infected tissues. Veterinary Parasitol-ogy 74: 75–77.
———. 2002. A review of toxoplasmosis in wild birds. Veterinary Par-asitology 106: 121–153.
———, AND C. P. BEATTIE. 1988. Toxoplasmosis of animals and man.CRC Press, Boca Raton, Florida, 220 p.
———, AND G. DESMONTS. 1987. Serological responses of equids fedToxoplasma gondii oocysts. Equine Veterinary Journal 19: 337–339.
———, H. R. GAMBLE, D. HILL, C. SREEKUMAR, S. ROMAND, AND P.THULLIEZ. 2002. High prevalence of viable Toxoplasma gondii in-fection in market weight pigs from a farm in Massachusetts. Journalof Parasitology 88: 1234–1238.
———, D. H. GRAHAM, C. R. BLACKSTON, T. LEHMANN, S. M. GENNARI,A. M. A. RAGOZO, S. M. NISHI, S. K. SHEN, O. C. H. KWOK, D. E.HILL, AND P. THULLIEZ. 2002. Biological and genetic characterisa-tion of Toxoplasma gondii isolates from chickens (Gallus domes-ticus) from Sao Paulo, Brazil: Unexpected findings. InternationalJournal for Parasitology 32: 99–105.
———, ———, D. S. SILVA, T. LEHMANN, AND L. M. G. BAHIA-OLIV-EIRA. 2003. Toxoplasma gondii isolates of free-ranging chickensfrom Rio de Janeiro, Brazil: Mouse mortality, genotype, and oocystshedding by cats. Journal of Parasitology 89: 851–853.
———, ———, R. W. DE YOUNG, E. DAHL, M. L. EBERHARD, E. K.NACE, K. WON, H. BISHOP, G. PUNKOSDY, C. SREEKUMAR, M. C. B.VIANNA, S. K. SHEN, O. C. H. KWOK, J. A. SUMNERS, S. DEMARAIS,J. G. HUMPHREYS, AND T. LEHMANN. 2004. Molecular and biologiccharacteristics of isolates of Toxoplasma gondii from wildlife inthe United States. Journal of Parasitology 90: 67–71.
———, A. N. HAMIR, AND M. J. TOPPER. 2002. Sarcocystis mephitisin. sp. (Protozoa: Sarcocystidae), Sarcocystis neurona-like andToxoplasma-like infections in striped skunk (Mephitis mephitis).Journal of Parasitology 88: 113–118.
———, I. T. NAVARRO, D. H. GRAHAM, E. DAHL, R. L. FREIRE, L. B.PRUDENCIO, C. SREEKUMAR, M. C. VIANNA, AND T. LEHMANN. 2003.Characterization of Toxoplasma gondii isolates from free rangechickens from Parana, Brazil. Veterinary Parasitology 117: 229–234.
FUENTES, I., J. M. RUBIO, C. RAMıREZ, AND J. ALVAR. 2001. Genotypiccharacterization of Toxoplasma gondii strains associated with hu-man toxoplasmosis in Spain: Direct analysis from clinical samples.Journal of Clinical Microbiology 39: 1566–1570.
GALLI-VALERIO, B. 1939. Sur une toxoplasmiase du Melopsittacus un-dulatus Shaw. Schweizer Archiv fuer Tierheilkunde 81: 458–460.
GRIGG, M. E., S. BONNEFOY, A. B. HEHL, Y. SUZUKI, AND J. C. BOOTH-ROYD. 2001. Success and virulence in Toxoplasma as the result ofsexual recombination between two distinct ancestries. Science 294:161–165.
———, J. GANATRA, J. C. BOOTHROOYD, AND T. P. MARGOLIS. 2001.Unusual abundance of atypical strains associated with human oc-ular toxoplasmosis. Journal of Infectious Diseases 184: 633–639.
HARTLEY, W. J., AND J. P. DUBEY. 1991. Fatal toxoplasmosis in somenative Australian birds. Journal of Veterinary Diagnostic Investi-gation 3: 167–169.
HOWE, D. K., S. HONORE, F. DEROUIN, AND L. D. SIBLEY. 1997. Deter-mination of genotypes of Toxoplasma gondii strains isolated frompatients with toxoplasmosis. Journal of Clinical Microbiology 35:1411–1414.
———, AND L. D. SIBLEY. 1995. Toxoplasma gondii comprises threeclonal lineages: Correlation of parasite genotype with human dis-ease. Journal of Infectious Diseases 172: 1561–1566.
HOWERTH, E. W., G. RICH, J. P. DUBEY, AND K. YOGASUNDRAM. 1991.Fatal toxoplasmosis in a red lory (Eos bornea). Avian Diseases 35:642–646.
JUNGERSEN, G., L. JENSEN, M. R. RASK, AND P. LIND. 2002. Non-lethalinfection parameters in mice separate sheep Type II Toxoplasmagondii isolates by virulence. Comparative Immunology, Microbi-ology and Infectious Diseases 25: 187–195.
LEHMANN, T., C. R. BLACKSTON, S. F. PARMLEY, J. S. REMINGTON, AND
J. P. DUBEY. 2000. Strain typing of Toxoplasma gondii: Comparisonof antigen-coding and house-keeping genes. Journal of Parasitology86: 960–971.
———, D. H. GRAHAM, E. DAHL, C. SREEKUMAR, F. LAUNER, J. L. CORN,H. R. GAMBLE, AND J. P. DUBEY. 2003. Transmission dynamics ofToxoplasma gondii on a pig farm. Infection, Genetics and Evolu-tion 3: 135–141.
———, ———, ———, L. M. G. BAHIA-OLIVEIRA, S. M. GENNARI,AND J. P. DUBEY. 2004. Variation in the structure of Toxoplasmagondii and the roles of selfing, drift, and epistatic selection in main-taining linkage disequilibria. Infection, Genetics, and Evolution 4:107–114.
MONDRAGON, R., D. K. HOWE, J. P. DUBEY, AND L. D. SIBLEY. 1998.
1174 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
Genotypic analysis of Toxoplasma gondii isolates from pigs. Jour-nal of Parasitology 84: 639–641.
OWEN, M. R., AND A. J. TREES. 1999. Genotyping of Toxoplasma gondiiassociated with abortion in sheep. Journal of Parasitology 85: 382–384.
POELMA, F. G., AND P. ZWART. 1972. Toxoplasmose bij kroonduiven enandere vogels in de Koninklijke Rotterdamse Diergaarde ^Blijdorp&.Acta Zoologica et Pathologica Antverpiensia 55: 29–40.
SREEKUMAR, C., J. R. RAO, A. K. MISHRA, D. RAY, G. C. BANSAL.
(2003). In vitro generation of cloned populations of Toxoplasmagondii. Parasitology Research 90: 489–492.
TENTER, A. M., A. R. HECKEROTH, AND L. M. WEISS. 2000. Toxoplasmagondii: From animals to humans. International Journal for Parasi-tology 30: 1217–1258.
WALTON, B. C., AND K. W. WALLS. 1964. Prevalence of toxoplasmosisin wild animals from Fort Stewart, Georgia, as indicated by sero-logical tests and mouse inoculation. American Journal of TropicalMedicine and Hygiene 13: 530–533.
J. Parasitol., 90(5), 2004, pp. 1174–1176q American Society of Parasitologists 2004
Molecular and Biological Characterization of Hammondia heydorni–Like Oocysts From aDog Fed Hearts From Naturally Infected White-Tailed Deer (Odocoileus virginianus)
J. P. Dubey, C. Sreekumar, K. B. Miska, D. E. Hill, M. C. B. Vianna, and D. S. Lindsay*, Animal Parasitic Diseases Laboratory, Animal andNatural Resources Institute, Agricultural Research Service, United States Department of Agriculture, Beltsville, Maryland 20705; *Center forMolecular Medicine and Infectious Diseases, Department of Biomedical Sciences and Pathobiology, Virginia–Maryland Regional College ofVeterinary Medicine, Virginia Tech, 1410 Prices Fork Road, Blacksburg, Virginia 24061. e-mail: [email protected]
ABSTRACT: Neospora caninum and Hammondia heydorni are morpho-logically and phylogenetically related coccidians that are found in dogs.Although there is serological evidence of N. caninum infection in thewhite-tailed deer (Odocoileus virginianus), the parasite has not been yetisolated from the tissues of this host. In an attempt to isolate N. caninumfrom deer, hearts from 4 deer with antibodies to N. caninum were fedto 2 dogs. One of these dogs shed unsporulated oocysts 12–14 mm indiameter. Sporulated oocysts were not infective to Mongolian gerbils(Meriones ungulatus), and DNA isolated from these oocysts was notamplified using N. caninum–specific primers. However, positive ampli-fication with the H. heydorni–specific first internal transcribed spacer(ITS-1) primers and common toxoplasmatiid ITS-1 primers confirmedthe presence of H. heydorni DNA in the samples. The oocysts wereconsidered to be H. heydorni on the basis of their morphology, biology,and molecular characteristics. This is the first record of a H. heydorni–like parasite in the white-tailed deer.
Neospora caninum is a parasite of livestock and companion animalsand is an important cause of bovine abortion in dairy cattle worldwide(Dubey, 2003). It is transmitted transplacentally, by the ingestion of in-fected tissues and by the ingestion of food and water contaminated withoocysts excreted in the feces of dogs. The domestic dog is the only knowndefinitive host for N. caninum (McAllister et al., 1998). The role of thedog in the epidemiology of N. caninum is currently unclear because ex-perimentally infected dogs usually excrete only a few oocysts, and theparasite has been isolated only a few times from naturally infected dogs(Basso et al., 2001; Gondim et al., 2002; Slapeta, Modry et al., 2002;McGarry et al., 2003). Furthermore, N. caninum oocysts morphologicallyresemble the oocysts of a related coccidian, Hammondia heydorni, andthere is no simple method to distinguish them.
Little is known about the life cycle of H. heydorni or whether addi-tional Hammondia species that use dogs as their definitive host occur(Dubey et al., 2002; Schares et al., 2002; Slapeta, Modry et al., 2002).Hammondia heydorni–like oocysts were found in the feces of dogs thatwere fed naturally infected tissues from cattle (Bos taurus), water buf-falo (Bubalus bubalis), sheep (Ovis aries), goats (Capra hircus), moose(Alces alces), and camels (Camelus dromedarius) (reviewed in Dubeyet al., 2002). In addition, red foxes (Vulpes vulpes) fed tissues fromsheep, cattle, roe deer (Capreolus capreolus), mountain gazelle (Gazellagazella), and reindeer (Rangifer tarandus) shed H. heydorni–like oo-cysts in their feces (Dubey et al., 2002). Until recently, all these oocystsexcreted in feces of dogs and foxes were considered 1 species. However,studies of Schares et al. (2002, 2003) indicated the parasite in foxes isdifferent morphologically and biologically from the parasite in dog fe-ces. Furthermore, molecular studies indicate that there are more than 1genetic variant at present designated as H. heydorni (Sreekumar et al.,
2003). We report isolation of another H. heydorni–like parasite fromthe white-tailed deer (Odocoileus virginianus) for the first time fromthis host.
During studies on the genetic characterization of Toxoplasma gondiiisolates from wildlife (Dubey et al., 2004) in March 2003, tissues from4 white-tailed deer from Mississippi were also examined for antibodiesto N. caninum using the N. caninum–agglutination test (NAT; Romandet al., 1998). The NAT titers were 1:25 in 2 and 1:50 in 2 deer. Hearttissue from these 4 deer were pooled and fed to 2 laboratory-raiseddogs. The dogs had not ingested uncooked meat products before feedingon deer tissues. Feces of these dogs were examined daily for 3 wk forcoccidian oocysts by floatation in sugar solution. Oocysts were collectedfrom feces, sporulated in 2.5% potassium dichromate aqueous solutionat room temperature for 7 days, and then stored at 4 C until used.Sporulated oocysts were washed with water to remove potassium di-chromate by centrifugation, treated with 5.25% sodium hypochloritesolution (Clorox), washed, and divided into aliquots for bioassay, invitro cultivation, and polymerase chain reaction (PCR) studies. Aliquotswere fed to 2 gerbils (Meriones ungulatus) and to 5 interferon gammagene knockout (KO) mice (Dubey and Lindsay, 1998). For in vitrocultivation, 1 aliquot was vortexed for 5 min with 500-mm glass beads(Microbeads, Ferro Corporation, Cleveland, Ohio) and subsequent in-cubation in an excystation medium (sodium taurocholate 250 mg, so-dium deoxycholic acid 400 mg, trypsin [1:250] 25 mg in 100 ml saline,pH 7.5) at 37 C. After excystation, the suspension was washed withgrowth medium and layered over each of the 2 CV1 (African Greenmonkey [Cercopithecus aethiops] kidney cells) and equine dermal cellmonolayers grown over coverslips in multiwell plates. The coverslipswere removed at intervals, fixed with Bouin fixative and stained withGiemsa.
For obtaining DNA, the oocyst suspension was ruptured by 2–3freeze–thaw cycles, followed by grinding of the pellet in small volumes(about 30 ml) in a 0.2-ml microtissue grinder (Wheaton, Fischer Sci-entific, Pittsburgh, Pennsylvania). The DNA was extracted from thehomogenized suspensions using DNAzol (MRC, Cincinnati, Ohio) ac-cording to the manufacturer’s instructions. Five sets of primers, Neos-pora-specific NP6/NP21 (Yamage et al., 1996), H. heydorni–specificfirst internal transcribed spacer (ITS-1) RAPD primers JS4/JS5 (Slapeta,Koudela et al., 2002), H. heydorni–specific HhAP7 and HhAP10 prim-ers, and common toxoplasmatiid ITS-1 primers CT1/CT2 (Sreekumaret al., 2003), were used for PCR amplification of the DNA, accordingto previously described protocols. The PCR products were electropho-resed in a 2% agarose gel, and the gel-cleaned PCR products weredirectly sequenced in both directions using the Big Dye terminator sys-tem, version 3.1 (Applied Biosystems, Foster City, California) using anABI 377 sequencer. The sequence chromatograms were edited usingSequencher 4.1 software (Genecodes Corp., Ann Arbor, Michigan).
RESEARCH NOTES 1175
FIGURE 1. Unsporulated (A) and sporulated (B) oocysts of the Ham-mondia heydorni–like parasite shed in the feces of a dog fed naturallyinfected deer tissues.
TABLE I. Sequence comparison of a 399-bp region of ITS-1 from different isolates of Hammondia heydorni.
Isolatelocation/host(s)
GenBankaccession no.
Polymorphic sites
165 209 293 317 331 356 361 368
United States/Deer, dog*USA?/DogGermany/goat, foxCzech Republic/Dog
* Present study.† Alignment gaps are indicated by dashes.‡ Sequences identical to the deer isolate are indicated with a period.
Searches were performed using BLAST (http://www.ncbi.nlm.nih.gov/BLAST/) to determine whether the sequences were similar to any ofthe previously published sequences.
One dog shed unsporulated oocysts in feces 10–12 days after in-gesting deer tissue. Unsporulated oocysts were 12–14 3 10–12 mm (n5 100) in size, and sporulated oocysts contained 2 sporocysts, 6–8 34–6 mm in size (Fig. 1A, B). There were 4 sporozoites in each sporo-cyst. The gerbils and KO mice fed sporocysts remained asymptomatic,and antibodies to N. caninum were not found in 1:25 dilution of theirserum tested 30 days after inoculation. Protozoans were not found inhistological sections of rodents fed oocysts.
Sporozoites were found to excyst within 20 min of incubation ofcleaned oocysts in the excystation media. The sporozoites did not infectthe equine dermal cell monolayer. Dividing forms could be seen in theCV1 cells stained on day 14 after inoculation. However, no additionalgrowth was seen, and the parasites gradually disappeared from themonolayers after 4 wk. A H. heydomi–like parasite from the feces of adog was grown in vitro by Speer et al. (1988), but they were unable toinfect new monolayers by passaging, and the parasite died out in cul-ture. Schares et al. (2003) maintained a H. hedorni–like parasite fromthe feces of foxes in vitro and were able to demonstrate oocyst sheddingin fox by feeding the infected culture. However, the parasite that Schareset al. (2003) studied was different from the parasite from dogs; its oo-cysts were larger and it was not transmissible to dogs through tissuesof an intermediate host (sheep). In the present study, only a few oocystswere shed in dog feces, and we were unable to obtain sustainablegrowth in culture.
No amplification was observed with either the N. caninum–specificprimers or the H. heydorni–specific RAPD–derived primers. The H.heydorni–specific ITS-1 primers resulted in the amplification of a prod-uct of the expected size (;270 bp). The sequences (GenBankAY531300) were found to be identical to those of H. heydorni isolatesfrom dogs and fox. The common toxoplasmatiid ITS-1 primers resultedin the amplification of a product in the size range of 399 bp (GenBank
AY530018). The results of the BLAST search of this fragment with thepublic database are shown in Table I. Polymorphism was noticed at 8locations. The sequences of the deer isolate were identical to that of aH. heydorni isolate of dog from the United States. However, the se-quences also had greater similarity to an isolate from fox than othercanine isolates from Germany, Czech Republic, and Australia. Recentevidence has pointed to the presence of molecular differences amongH. heydorni isolates from dogs and foxes. Mohammed et al. (2003)concluded that the isolates of H. heydorni from foxes and dogs consti-tute genetically different populations on the basis of the ribosomal DNAsequences. The deer isolate, which was excreted by a dog, was foundto be closer to the fox isolate than the dog isolates from Germany,Czech Republic, and Australia.
The results of the present study indicate that dogs can excrete H.heydorni–like oocysts in their feces, and these must be distinguishedfrom N. caninum oocysts. The results also point to a greater geneticdiversity among the H. heydorni isolates irrespective of their host.
LITERATURE CITED
BASSO, W., L. VENTURINI, M. C. VENTURINI, D. E. HILL, O. C. H. KWOK,S. K. SHEN, AND J. P. DUBEY. 2001. First isolation of Neosporacaninum from the feces of a naturally infected dog. Journal ofParasitology 87: 612–618.
DUBEY, J. P. 2003. Neosporosis in cattle. Journal of Parasitology89(Suppl.): S42–S46.
———, AND D. S. LINDSAY. 1998. Isolation in immunodeficient miceof Sarcocystis neurona from opossum (Didelphis virginiana) fae-ces, and its differentiation from Sarcocystis falcatula. InternationalJournal for Parasitology 28: 1823–1828.
———, B. C. BARR, J. R. BARTA, I. BJERKAS, C. BJORKMAN, B. L.BLAGBURN, D. D. BOWMAN, D. BUXTON, J. T. ELLIS, B. GOTTSTEIN,A. HEMPHILL, D. E. HILL, D. K. HOWE, M. C. JENKINS, Y. KOBAY-ASHI, B. KOUDELA, A. E. MARSH, J. G. MATTSSON, M. M. MCAL-LISTER, D. MODRY, Y. OMATA, C. A. SPEER, A. J. TREES, A. UGGLA,S. J. UPTON, D J. L. WILLIAMS, AND D. S. LINDSAY. 2002. Rede-scription of Neospora caninum and its differentiation from relatedcoccidia. International Journal for Parasitology 32: 929–946.
———, D. H. GRAHAM, R. W. DE YOUNG, E. DAHL, M. L. EBERHARD,E. K. NACE, K. WON, H. BISHOP, G. PUNKOSDY, C. SREEKUMAR, M.C. B. VIANNA, S. K. SHEN, O. C. H. KWOK, J. A. SUMNERS, S.DEMARAIS, J. G. HUMPHREYS, AND T. LEHMANN. 2004. Molecularand biologic characteristics of isolates of Toxoplasma gondii fromwildlife in the United States. Journal of Parasitology 90: 67–72.
GONDIM, L. F. P., L. GAO, AND M. M. MCALLISTER. 2002. Improvedproduction of Neospora caninum oocysts, cyclical oral transmissionbetween dogs and cattle, and in vitro isolation from oocysts. Jour-nal of Parasitology 88: 1159–1163.
MCALLISTER, M. M., J. P. DUBEY, D. S. LINDSAY, W. R. JOLLEY, R. A.WILLS, AND A. M. MCGUIRE. 1998. Dogs are definitive hosts ofNeospora caninum. International Journal for Parasitology 28:1473–1478.
MCGARRY, J. W., C. M. STOCKTON, D. J. L. WILLIAMS, AND A. J. TREES.
1176 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
2003. Protracted shedding of oocysts of Neospora caninum by anaturally infected foxhound. Journal of Parasitology 89: 628–630.
MOHAMMED, O. B., A. J. DAVIES, H. S. HUSSEIN, P. DASZAK, AND J. T.ELLIS. 2003. Hammondia heydorni from the Arabian mountain ga-zelle and red fox in Saudi Arabia. Journal of Parasitology 89: 535–539.
ROMAND, S., P. THULLIEZ, AND J. P. DUBEY. 1998. Direct agglutinationtest for serologic diagnosis of Neospora caninum infection. Para-sitology Research 84: 50–53.
SCHARES, G., A. O. HEYDORN, A. C. H. MEHLHORN, L. GEUE, M. PETERS,AND F. J. CONRATHS. 2002. In contrast to dogs, red foxes (Vulpesvulpes) did not shed Neospora caninum upon feeding of interme-diate host tissues. Parasitology Research 88: 44–52.
———, J. MEYER, A. BARWALD, F. J. CONRATHS, R. RIEBE, W. BOHNE,K. ROHN, AND M. PETERS. 2003. A Hammondia-like parasite fromthe European fox (Vulpes vulpes) forms biologically viable tissue cystsin cell culture. International Journal for Parasitology 33: 229–234.
SLAPETA, J. R., B. KOUDELA, J. VOTYPKA, D. MODRY, R. HOREJS, AND
J. LUKES. 2002. Coprodiagnosis of Hammondia heydorni in dogs
by PCR based amplification of ITS 1 rRNA: Differentiation frommorphologically indistinguishable oocysts of Neospora caninum.The Veterinary Journal 163: 147–154.
———, D. MODRY, I. KYSELOVA, R. HOREJS, J. LUKES, AND B. KOU-DELA. 2002. Dog shedding oocysts of Neospora caninum: PCR di-agnosis and molecular phylogenetic approach. Veterinary Parasi-tology 109: 157–167.
SPEER, C. A., J. P. DUBEY, J. A. BLIXT, AND B. L. BLAGBURN. 1988.Development of Hammondia heydorni in cultured bovine and ovinecells. Journal of Protozoology 35: 352–356.
SREEKUMAR, C., D. E. HILL, V. M. FOURNET, B. M. ROSENTHAL, D. S.LINDSAY, AND J. P. DUBEY. 2003. Detection of Hammondia hey-dorni-like organisms and their differentiation from Neospora can-inum using random-amplified polymorphic DNA-polymerase chainreaction. Journal of Parasitology 89: 1082–1085.
YAMAGE, M., O. FLECHTNER, AND B. GOTTSTEIN. 1996. Neospora canin-um: Specific oligonucleotide primers for the detection of brain‘‘cyst’’ DNA of experimentally infected nude mice by the polymer-ase chain reaction (PCR). Journal of Parasitology 28: 1473–1478.
J. Parasitol., 90(5), 2004, pp. 1176–1178q American Society of Parasitologists 2004
Ingestion of Cryptosporidium Oocysts by Caenorhabditis elegans
Obed Huamanchay, Linda Genzlinger, Miguel Iglesias, and Ynes R. Ortega*, Center for Food Safety and Department of Food Science andTechnology, University of Georgia, 1109 Experiment Street, Griffin, Georgia 30223; *To whom correspondence should be addressed. e-mail:[email protected]
ABSTRACT: Cryptosporidium parvum has been associated with out-breaks of human illness by consumption of contaminated water, freshfruits, and vegetables. Free-living nematodes may play a role in path-ogen transmission in the environment. Caenorhabditis elegans is a free-living soil nematode that has been extensively studied and serves as agood model to study possible transmission of C. parvum oocysts thatmay come into contact with produce before harvest. The objective ofthis study was to determine whether C. elegans could serve as a poten-tial mechanical vector for transport of infectious C. parvum and Cy-clospora cayetanensis in agricultural settings and whether C. eleganscould ingest, excrete, and protect oocysts from desiccation. Seventy to85% of worms ingested between 0 and 500 oocysts after 1 and 2 hrincubation with oocysts. Most of the nematodes ingested between 101and 200 oocysts after 2 hr. Intact oocysts and empty shells were excretedby nematodes. Infectivity was determined by the neonatal assay withdifferent treatments of worms (intact or homogenized) or oocysts orboth. Adult C. elegans containing C. parvum kept in water were infec-tive for mice. In conclusion, C. elegans adults can ingest and excreteC. parvum oocysts. Caenorhabditis elegans containing C. parvum oo-cysts can infect mice but does not seem to protect oocysts from extremedesiccation at 23 C incubation of a day or longer. Cyclospora oocystswere not ingested by C. elegans. The role of free-living nematodes inproduce contamination needs to be further examined.
Parasites, such as Cryptosporidium parvum, Cyclospora cayetanensis,and Giardia lamblia have been causative agents in outbreaks of humanillness associated with consumption of contaminated water, fresh fruits,and vegetables (Orlandi et al., 2002).
The dissemination of pathogenic parasites onto raw produce has notbeen thoroughly studied (Wasilewska and Webster, 1975) but may beintroduced during pre- and postharvest practices. Potential agriculturalsources of produce contamination include farm workers, animals, irri-gation water, soil, and insects. Flies have been shown to transport in-fectious C. parvum oocysts (Graczyk et al., 1999). If produce comesinto contact with manure, then free-living soil nematode contaminationcould occur. Irrigation water can also be contaminated with feces ofhumans and wild animals, introducing large numbers of parasites intothe environment, particularly in rivers and lakes. Rotifers can ingest C.
parvum oocysts and G. lamblia cysts. The fate of the parasites onceingested by rotifers needs to be studied further (Fayer et al., 2000; Troutet al., 2002).
A well-studied and harmless free-living nematode, Caenorhabditiselegans, may serve as a model to study nematodes acting as carriers ortransport hosts by ingesting harmful bacteria and then protecting suchorganisms from environmental conditions while in the gut (Chang etal., 1960). Salmonella enterica serotype Poona and S. typhimurium canbe ingested by C. elegans, providing protection against inactivation byproduce sanitizers (Labrousse et al., 2000; Caldwell et al., 2003). Theobjective of the present study was to determine whether C. eleganscould serve as a potential mechanical vector for transport of infectiousCryptosporidium in agricultural settings and if the nematode could in-gest, excrete, and protect oocysts from desiccation. Caenorhabditis ele-gans (N2, wild-type strain) was used in all experiments. The wormswere grown on K-agar plates (pH 6.5) (2.36 g potassium chloride, 3 gsodium chloride, 2.5 g Bacto Peptone, and 17 g/L agar in deionizedwater). The agar was autoclaved, cooled, and supplemented with 1 gcholesterol (95%), 11.1 g calcium chloride, and 24.7 g magnesium sul-fate. Medium was distributed in plastic petri plates. Escherichia coliwas cultured in OP50 broth (5 g sodium chloride and 10 g Bacto Pep-tone) for 24 hr, plated in K agar, and incubated at 37 C for 24 hr untilconfluent growth was established. Nematode cultures were fed withconfluent cultures of E. coli OP50.
Cryptosporidium parvum oocysts (Iowa isolate, bovine genotype)were obtained from the Parasitology Laboratory, University of Arizona.Oocysts were labeled with Merifluort Cryptosporidium/Giardia DirectImmunofluorescent Detection Reagent (Meridian Bioscience Inc., Cin-cinnati, Ohio). The oocysts (approximately 3 3 107) were resuspendedin 100 ml K-medium and 40 ml of the Detection Reagent was added.Incubation was at room temperature for 60 min, with mixing every 15min. Detection Reagent was removed by microcentrifugation (8,000rpm for 4 min) and oocysts were washed with fresh K-medium.
Unsporulated C. cayetanensis oocysts were obtained from naturallyinfected individuals with cyclosporiasis. Fecal samples containing oo-cysts were concentrated using the modified ethyl acetate method fol-lowed by a discontinuous sucrose gradient (Ortega et al., 1998).
Between 100 and 200 adult nematodes were placed on K-agar plates
RESEARCH NOTES 1177
FIGURE 1. Cryptosporidium oocysts as they get ingested by Cae-norhabditis elegans. After 1 hr incubation (A). After 2 hr incubationthe anterior portion of the gastrointestinal system of an adult worm isfull of oocysts as observed by 310 magnification (B) and 340 mag-nification (C). Posterior portion of C. elegans contains oocysts (thatwere IFA negative) inside the gut. The preparation was incubated withthe detection reagent specific for Cryptosporidium oocysts and then ob-served by DIC microscopy (D) and fluorescence microscopy (E). Bar5 40 mm.
FIGURE 2. Number of adult Caenorhabditis elegans containing var-ious numbers of Cryptosporidium parvum oocysts in their gastrointes-tinal tract after 2 hr incubation.
with 2 3 106 fluorescein isothiocyanate–tagged C. parvum oocysts. Af-ter specific incubation times, worms were transferred into tubes, washedwith phosphate-buffered saline, and observed by UV and differentialinterference contrast (DIC) microscopy.
About 75–85% of the worms ingested C. parvum oocysts after 1–2hr incubation (Fig. 1). Most of the oocysts were distributed in the an-terior end of the worm after 1 hr incubation (Fig. 1A), and after 2 hr,the oocysts were found throughout the worm (Fig. 1B, C). The numberof oocysts observed per worm was between 1 and 500. Most of thenematodes contained 1–200 oocysts after 1 hr ingestion, increasing to101–200 oocysts after 2 hr of incubation (Fig. 2).
After 12 hr, the oocysts present inside the nematodes did not showfluorescence. By Nomarski (DIC) microscopy, oocysts were observedboth inside and outside the worm (Fig. 1D). Nematodes were washedby centrifugation with K-medium to remove oocysts that could attachto the nematode cuticle. When nematodes were mechanically lysed, theoocysts that were in the gastrointestinal tract (not fluorescent) werereleased. The oocysts fluoresced when the Detection Reagent was added(Merifluor C. parvum Immunofluorescent Assay) (Fig. 1E). Oocystshells and intact oocysts were observed.
Adult worms containing C. parvum oocysts in their gastrointestinaltract, oocysts alone, and worms alone were incubated in water or driedand kept at 23 C for 12 and 24 hr, and 2-, 3-, 4-, 5-, 6-, and 7-dayintervals. Worms were broken open by grinding with a pestle in ho-mogenization tubes. The numbers of oocysts per worm were determinedby UV microscopy (excitation wavelength 490–500, barrier filter 510–530). Nematodes and oocysts alone were used as controls.
Five-day-old neonatal CD-1 mice were fed orally with preparationsof 100 worms containing oocysts (that were broken open by grindingwith pestle in homogenization tubes), 1.5 3 105 oocysts alone, or worms
alone. Each variable was tested in 9 or 10 neonate mice. Five days afterinoculation, mice were killed and 1-cm sections of the ileum were col-lected and mouse feces collected for further examination by acid-faststaining. Tissues were formalin fixed, paraffin embedded, sectioned andstained with hematoxylin–eosin, and examined for the presence of C.parvum parasites.
Infections were observed in mice fed with dried preparations of C.elegans containing C. parvum oocysts for up to 1 day. Dried C. parvumoocysts or dried worms alone did not infect mice. Infection was alsoobserved in mice if oocysts or nematodes containing oocysts were keptin water for as long as 7 days. Nematodes alone did not infect mice.
In the present study, we demonstrated that C. elegans can ingest andexcrete C. parvum oocysts. The antibodies tagging the oocyst cell wallwere no longer present on the oocyst surface after incubation for 12 hr.Enzymatic activity or other conditions in the nematode’s gut (such aspH or ionic strength) may have affected the stability of the antibodiestagging the oocysts. This needs to be studied further. The presence ofsome shells in the nematode’s gut suggested oocyst excystation and theliberation of the sporozoites into the nematode gastrointestinal system.Oocysts that were not affected by this process were still viable andinfectious to mice. Oocysts were dependent on the presence of water.This study demonstrated that C. elegans can harbor C. parvum oocystsand transport the oocyst to discrete distances. Caenorhabditis eleganscan move for distances up to several millimeters in 9–12 min (data notshown). The number of oocysts per worm (approx. 150) is on averagesufficient to contaminate produce and infect susceptible hosts becausethe ID50 of C. parvum in humans varies from 9 to 100 oocysts depend-ing on the isolate (Okhuysen et al., 1999).
Cyclospora cayetanensis oocysts were also fed to C. elegans; how-ever, the nematode did not ingest them (data not shown). This may bebecause of the larger size of the oocysts (8–10 mm) compared withCryptosporidium (4–6 mm). This does not exclude the possibility thatother types of free-living nematodes may ingest C. cayetanensis oo-cysts. The role of nematodes in the transport and distribution of oocystsfrom C. parvum, C. cayetanensis, and other food-borne protozoans fromsoil onto fresh produce needs to be further investigated.
The nematode strain used in this study was provided by the Caenor-habditis Genetics Center, which is funded by the National Institutes ofHealth National Center for Research resources. Funding for this projectwas provided by the USDA Alliance Special Grant 2001-03142.
LITERATURE CITED
CALDWELL, K. N., B. B. ADLER, G. L. ANDERSON, P. L. WILLIAMS, AND
L. R. BEUCHAT. 2003. Ingestion of Salmonella enterica serotypePoona by a free-living nematode, Caenorhabditis elegans, and pro-tection against inactivation by produce sanitizers. Applied and En-vironmental Microbiology 69: 4103–4110.
1178 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
CHANG, S. L., G. BERG, AND N. A. CLARKE. 1960. Survival and protec-tion against chlorination of human enteric pathogens in free-livingnematodes isolated from water supplies. American Journal of Trop-ical Medicine and Hygiene 9: 136–142.
FAYER, R., J. M. TROUT, E. WALSH, AND R. COLE. 2000. Rotifers ingestoocysts of Cryptosporidium parvum. Journal of Eukaryotic Micro-biology 47: 161–163.
GRACZYK, T. K., R. FAYER, M. R. CRANFIELD, B. M. HANGAMI-RUWENDE,R. KNIGHT, J. M. TROUT, AND H. BIXLER. 1999. Filth flies are trans-port hosts of Cryptosporidium parvum. Emerging Infectious Dis-eases 5: 726–727.
LABROUSSE, A., S. CHAUVET, C. COUILLAULT, C. L. KURZ, AND J. J.EWBANK. 2000. Caenorhabditis elegans is a model host for Sal-monella typhimurium. Current Biology 10: 1543–1545.
OKHUYSEN, P. C., C. L. CHAPPELL, J. H. CRABB, C. R. STERLING, AND H.L. DUPONT. 1999. Virulence of three distinct Cryptosporidium par-vum isolates for healthy adults. Journal of Infectious Diseases 180:1275–1281.
ORLANDI, P. A., D. M. T. CHU, J. W. BIER, AND G. J. JACKSON. 2002.Parasites and the food supply. Food Technology 56: 72–81.
ORTEGA, Y. R., C. R. STERLING, AND R. H. GILMAN. 1998. Cyclosporacayetanensis. Advances in Parasitology 40: 399–418.
TROUT, J. M., E. J. WALSH, AND R. FAYER. 2002. Rotifers ingest Giardiacysts. Journal of Parasitology 88: 1038–1040.
WASILEWSKA, L., AND J. M. WEBSTER. 1975. Free-living nematodes asdisease factors of man and his crops. International Journal of En-vironmental Studies 7: 201–204.
J. Parasitol., 90(5), 2004, pp. 1178–1180q American Society of Parasitologists 2004
Protection of Calves Against Cryptosporiosis by Oral Inoculation with Gamma-IrradiatedCryptosporidium parvum Oocysts
Mark Jenkins, James Higgins*, Kali Kniel, James Trout*, and Ron Fayer*, Animal Parasitic Diseases Laboratory, Animal and NaturalResources Institute, Agricultural Research Service, United States Department of Agriculture, Beltsville, Maryland 20705; *EnvironmentalMicrobial Safety Laboratory, Animal and Natural Resources Institute, Agricultural Research Service, United States Department of Agriculture,Beltsville, Maryland 20705. e-mail: [email protected]
ABSTRACT: The purpose of this study was to determine whether gam-ma-irradiated Cryptosporidium parvum oocysts could elicit protectiveimmunity against cryptosporidiosis in dairy calves. Cryptosporidiumparvum Iowa strain oocysts (1 3 106 per inoculation) were exposed tovarious levels of gamma irradiation (350–500 Gy) and inoculated into1-day-old dairy calves. The calves were examined daily for clinicalsigns of cryptosporidiosis, and fecal samples were processed for thepresence of C. parvum oocysts. At 21 days of age, the calves werechallenged by oral inoculation with 1 3 105 C. parvum oocysts andexamined daily for oocyst shedding and clinical cryptosporidiosis.Calves that were inoculated with C. parvum oocysts exposed to 350–375 Gy shed C. parvum oocysts in feces. Higher irradiation doses (450or 500 Gy) prevented oocyst development, but the calves remainedsusceptible to C. parvum challenge infection. Cryptosporidium parvumoocysts exposed to 400 Gy were incapable of any measurable devel-opment but retained the capacity to elicit a protective response againstC. parvum challenge. These findings indicate that it may be possible toprotect calves against cryptosporidiosis by inoculation with C. parvumoocysts exposed to 400-Gy gamma irradiation.
Cryptosporidiosis remains a significant health threat to humans andanimals because of the resiliency of the oocyst stage and the lack ofapproved drugs to prevent or treat infection. It is highly prevalent inpreweaned cattle, constituting a health threat for these young animalsand a significant source of environmental contamination. Although pas-sive immunotherapy with monoclonal antibodies, immune serum, orhyperimmune colostrum specific for Cryptosporidium parvum antigenshas shown some efficacy against cryptosporidiosis, most of these studieshave been conducted in rodent models (for review, see Riggs, 1997,2002; Crabb, 1998). Passive immunotherapy with hyperimmune bovinecolostrum has reduced clinical signs in humans or dairy calves, but C.parvum oocysts continue to be shed in high numbers (Fayer et al., 1989;Ungar et al., 1990; Okhuysen et al., 1998; Perryman et al., 1999). Pre-liminary research has shown that individuals experiencing a patent C.parvum infection, i.e., diarrhea, oocyst shedding, are more resistant tooocyst challenge (Chappel et al., 1999). An ideal vaccine against cryp-tosporidiosis would protect susceptible individuals against C. parvuminfection without causing clinical symptoms during the primary ‘‘im-munization.’’ Exposing C. parvum oocysts to gamma irradiation mayprevent schizont development and thus overt cryptosporidiosis. Gammairradiation of various protozoa has been used to prevent associated dis-eases such as malaria (Clyde et al., 1975; Scheller and Azad, 1995;
Chatterjee et al., 1999), avian coccidiosis (Jenkins, Augustine et al.,1991; Jenkins, Danforth et al., 1991; Jenkins et al., 1993), babesiosis(Phillips, 1971; Purnell et al., 1979), and toxoplasmosis (Dubey et al.,1996). The present study was designed to identify a dose of gammairradiation that inhibits oocyst development and determine whethergamma-irradiated C. parvum oocysts could be used to vaccinate calvesagainst cryptosporidiosis.
Cryptosporidium parvum Iowa strain oocysts were propagated by in-fecting a 1-day-old calf with 1 3 106 oocysts as described (Jenkins etal., 1999). The oocysts were purified by continuous flow centrifugation(Vetterling, 1969), followed by CsCl gradient centrifugation (Kilani andSekla, 1987), suspended in distilled H2O, stored at 4 C before inocu-lation. Cryptosporidium parvum oocysts used in primary inoculationwere 2.1 6 0.9 mo old (range 0.9–3.6 mo). Oocysts used in challengestudies were 3.1 6 0.9 mo old (range 1.6–4.3 mo).
The C. parvum oocysts were suspended at a concentration of 1 3106 oocysts/ml and exposed to various doses of gamma irradiation usinga 137Cs Gammator M radiation source at 10 Gy/min. The excystationrate of both nonirradiated and irradiated oocysts was determined usinga standard in vitro protocol (Gut and Nelson, 1999) and subjected to 1-way analysis of variance using a Tukey–Kramer multiple comparisonstest (GraphPad InStat Version 4.1, GraphPad Software, San Diego, Cal-ifornia). Within 1 hr after irradiation, the oocysts were inoculated peros into newborn (1- to 2-day-old) male Holstein calves obtained fromthe Beltsville Agricultural Research Center Dairy Unit. Calves wereexamined daily for clinical signs of cryptosporidiosis; fecal samples(;40 g) were collected daily from days 1–21 and processed for detect-ing C. parvum oocysts as described (Fayer et al., 2000). In brief, fecalsamples were homogenized by vortexing, a 10-g subsample was mixedwith water and sieved, centrifuged, and the pellet was suspended in 1.4g/ml aqueous cesium chloride, followed by centrifugation at 16,000 g.After centrifugation, the supernatant was aspirated, and the pellet waswashed with deionized water, stained with Merifluor reagent (MeridianDiagnostics, Cincinnati, Ohio), and examined by fluorescence micros-copy for the presence of oocysts. For RNA extraction, purified oocystswere subjected to 3 freeze–thaw cycles of 280 and 55 C, after whichthe oocyst lysate was processed using the Qiagen Viral RNA kit (Va-lencia, California). Purified RNA (10 ml) was used as template for re-verse transcriptase–polymerase chain reaction (RT-PCR) using theCPOP716F/CPOP992R primer pair, which amplifies a 315-bp fragmentof the gene coding for an uncharacterized protein of the C. parvumdouble-stranded RNA viral symbiont (‘‘KSU-1’’ strain, GenBank
RESEARCH NOTES 1179
TABLE I. Efficacy of Cryptosporidium parvum oocysts exposed to var-ious doses of gamma irradiation to confer protection in each of 8 dairycalves against clinical cryptosporidiosis and C. parvum oocyst shed-ding.
Calf no.Irradiationdose (Gy)
Primary infection*
Clinicalsigns‡
Oocystshedding§
Secondary infection†
Clinicalsigns
Oocystshedding
1234
0350375400
11222
11111112
ND¶NDND
2
NDNDND2
5678
400400450500
2222
2222
2211
21
1111
* Calves were infected at 1–2 days of age with 1 3 106 C. parvum Iowa strainoocysts irradiated at level shown.
† Calves were challenged at 21 days of age with 1 3 105 C. parvum Iowa strainoocysts.
‡ Clinical signs: 2, indicates no diarrhea; 1, loose stools; 11, watery diarrheafor at least 3 consecutive days.
§ Oocysts detected by immunofluorescence staining, identified as C. parvum Iowastrain by RT-PCR, purified by CsCl density gradient centrifugation, and enu-merated by hemacytometer counting. 2, indicates no oocysts detected; 1, ,103
oocysts/g; 11, 103–104 oocysts/g; 111, .104 oocysts/g.¶ ND, not done.
TABLE II. Percent excystation of Cryptosporidium parvum oocysts ex-posed to various doses of gamma irradiation.
Irradiationdose (Gy) Percent oocyst excystation
0350375400450500
94 6 584 6 971 6 870 6 866 6 460 6 8
CPU95996; Khramtsov et al., 1997; Kozwich et al., 2000). RT-PCRwas carried out using the Superscripty 1-step reagent (Invitrogen,Carlsbad, California), with the reverse transcription performed at 55 Cfor 30 min, followed by 4 min at 95 C, and 40 cycles of 15 sec at 95C, 45 sec at 55 C, and 1 min at 72 C. The sequence of the forwardprimer is 59-CCGGAAGCAGTGCAATCTGTTAGTCTCACCTTCT-ACTCAT-39, and the sequence of the reverse primer is 59-GA-CCTAATCTCATTGTATATCGCGCGCACGTATATCGGTA-39. RT-PCR products were cycle-sequenced using the Big Dyet 3.1 reagent(Applied Biosystems, Foster City, California) and an ABI Model 3100automated fluorescence sequencing instrument (Applied Biosystems).Sequence data were analyzed using MacVectory 6.5 software (OxfordMolecular, Madison, Wisconsin). The purpose of this assay was to en-sure that oocysts shed by calves had originated from the C. parvumIowa strain oocysts used in the inoculation rather than from contami-nation with local C. parvum strains, e.g., Beltsville-1. For purposes oftesting vaccine efficacy of irradiated oocysts, the calves were then chal-lenged at 21 days of age by oral inoculation with 1 3 105 C. parvumIowa strain oocysts. This challenge dose was based on preliminary stud-ies wherein reproducible clinical signs and shedding of high numbersof C. parvum oocysts were achieved. Calves were challenged at 3 wkof age to allow sufficient time for development of immunity. Inoculationof naive 21-day-old calves with 1 3 105 C. parvum Iowa strain oocystsproduced an average of 1.9 3 104 6 0.6 3 104 oocysts/g of feces (M.Jenkins, unpubl. obs.). Fecal samples were collected twice daily for 3–4 wk after challenge, and fecal smears were examined by immunoflu-orescence staining for the presence of C. parvum oocysts following themanufacturer’s instructions (Meridian Diagnostics). All oocyst-positivefecal samples were pooled, weighed, and processed for total oocystcounts by sieving, sucrose flotation, and CsCl gradient centrifugation.Similar to primary infection, the calves challenged at 21 days of agewere examined daily for clinical signs, and samples of feces were col-lected for oocyst enumeration. Calves that shed oocysts after a primaryinfection with nonirradiated or irradiated C. parvum Iowa strain oocystswere not challenged at 21 days of age because the goal of this studywas to develop an immunization regimen that did not entail clinicalsigns or oocyst shedding.
Although calves that were infected at 1–2 days of age with 350- or375-Gy-irradiated C. parvum oocysts exhibited reduced clinical signs,high numbers of oocysts were found in feces from these calves (TableI). Increasing the irradiation dose to 400 Gy eliminated detectable oo-
cyst shedding as well as clinical cryptosporidiosis as did higher doses(450 or 500 Gy) (Table I). Although the prepatent period for 350- or375-Gy-exposed C. parvum oocysts was similar to that of nonirradiatedoocysts, the duration of oocyst shedding arising from irradiated oocystswas shorter (M. Jenkins unpubl. obs.). Also, calves inoculated with 450-or 500-Gy-treated C. parvum oocysts showed prepatency and patencyperiods after challenge infection similar to those of naive animals (M.Jenkins unpubl. obs.).
Irradiation appeared to have an effect on oocyst excystation. A sig-nificant decrease (P , 0.01) in excystation rate was observed betweennonirradiated oocysts and oocysts that were exposed to greater than350-Gy gamma irradiation. For instance, in C. parvum oocysts exposedto 400-Gy irradiation, excystation was 20% less than in nonirradiatedcontrol oocysts (Table II). However, the absence of oocyst shedding incalves infected with C. parvum oocysts irradiated at $400 Gy wouldnot be solely the result of the decrease in excystation because calvesinfected with 1 3 106 nonirradiated oocysts produce greater than 3 3109 oocysts (M. Jenkins, unpubl. obs.). Even if 50% of the oocysts weredestroyed by irradiation, the number of oocysts arising from the re-mainder would easily be detectable in feces.
In a subsequent challenge, calves that received a primary inoculationof C. parvum oocysts irradiated at 400 Gy were resistant to oocystchallenge infection at 21 days of age. Clinical symptoms were absentafter challenge in all 3 calves (calves 4–6) that had been inoculated at1–2 days of age with 400-Gy-irradiated oocysts, and only calf 6 shedoocysts after challenge (Table I). Oocyst production (5 3 102 oocysts/g) in this calf was 2.5% of that observed (;2 3 104 oocysts/g) in naivecontrol calves that had been challenged at 3 wk of age with C. parvumIowa oocysts. Although it is possible that oocysts may have been shedby calves 4 and 5 in this group, the number of oocysts would havebeen extremely low based on the sensitivity of this assay (Fayer et al.,2000). Higher irradiation doses (450 and 500 Gy) not only inhibitedparasite development in the primary ‘‘infection’’ but also ablated anyprotective effects. These results indicate that there is a narrow range ofoptimum irradiation dose that can inhibit parasite development but stillmaintain the capacity of the radiation-attenuated parasites to elicit aprotective immune response.
These findings also indicate that exposure of C. parvum oocysts to400-Gy gamma irradiation does not affect the capacity of oocysts toinduce a protective response in calves but does prevent the parasite fromdeveloping to the oocyst stage. Determining at what point developmentis blocked in the C. parvum life cycle would be difficult in calves andprobably not relevant to test in mice. Studies in related protozoans in-dicate that development is blocked after sporozoite penetration of hostcells (Jenkins, Augustine et al., 1991; Jenkins, Danforth et al., 1991;Silvie et al., 2002). Presumably, irradiation doses above the optimumlevel prevent sporozoite invasion or intracellular metabolism (or both),which are necessary for expression of parasite antigens on host cellsand the induction of immunity.
Comparing the present study with findings in mouse models of cryp-tosporidiosis indicates that results in mice may not be predictive offindings in calves (Jenkins et al., 1998; Yu and Park, 2003). In neonatalBALB/C mice, C. parvum development was blocked by exposure ofoocysts to 250-Gy irradiation (Jenkins et al., 1998). Preliminary dosetitration studies revealed no apparent diminution in oocyst productionin calves inoculated with C. parvum oocysts exposed to less than 300-Gy irradiation (M. Jenkins, unpubl. obs.). Somewhat conflicting with
1180 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
the current study in calves and our previous studies in mice is a recentreport that 25,000 Gy was necessary to prevent C. parvum oocyst shed-ding in weaned C57/Bl6 mice (Yu and Park, 2003). The observed dis-crepancy may be due to differences in age and strain of mice used.These authors also found that oocysts were shed for an extended periodof time, and that peak oocyst shedding was delayed in mice receivingirradiated oocysts. In the present study, calves were examined for atleast 3 wk after the primary and challenge infections to ensure thatabsence of oocyst secretion was due to the effect of gamma irradiationor immunity rather than delayed development. This precaution was tak-en because C. parvum is unique among protozoans in that an autoin-fective oocyst stage is produced during the life cycle, and thus smallnumbers of viable oocysts may eventually lead to excretion of highnumbers of oocysts, albeit with some time delay.
The present study suggests that irradiated C. parvum oocysts can beused to prevent clinical signs and oocyst shedding in dairy calves. Also,irradiation doses greater than 450 Gy inhibit C. parvum from producingpatent infection, which is considerably less than the irradiation levelrequired to destroy bacteria on the surface of foods. The stability of400-Gy-irradiated C. parvum oocysts is unknown and will require ad-ditional studies before this approach is considered useful in preventingcryptosporidiosis in dairy calves.
The authors acknowledge the excellent technical assistance of Chris-tina Hohn, Kristi Ludwig, Robert Palmer, and Marisol Ponte. The au-thors thank James Harp for providing the original C. parvum Iowa strainoocysts used in this study.
LITERATURE CITED
CHAPPELL, C. L., P. C. OKHUYSEN, C. R. STERLING, C. WANG, W. JAK-UBOWSKI, AND H. L. DUPONT. 1999. Infectivity of Cryptosporidiumparvum in healthy adults with pre-existing anti-C. parvum serumimmunoglobulin G. American Journal of Tropical Medicine andHygiene 60: 157–164.
CHATTERJEE, S., P. DRUILHE, AND M. WERY. 1999. Irradiated sporozoitesprime mice to produce high antibody titres upon viable Plasmo-dium berghei sporozoite challenge, which act upon liver-stage de-velopment. Parasitology 118: 219–225.
CLYDE, D. F. 1975. Immunization of man against falciparum and vivaxmalaria by use of attenuated sporozoites. American Journal ofTropical Medicine and Hygiene 24: 397–401.
CRABB, J. H. 1998. Antibody-based immunotherapy of cryptosporidio-sis. Advances in Parasitology 40: 121–149.
DUBEY, J. P., M. C. JENKINS, D. W. THAYER, O. C. KWOK, AND S. K.SHEN. 1996. Killing of Toxoplasma gondii oocysts by irradiationand protective immunity induced by vaccination with irradiatedoocysts. Journal of Parasitology 82: 724–727.
FAYER, R., C. ANDREWS, B. L. P. UNGAR, AND B. BLAGBURN. 1989.Efficacy of hyperimmune bovine colostrum for prophylaxis ofcryptosporidiosis in neonatal calves. Journal of Parasitology 75:393–397.
———, J. M. TROUT, T. K. GRACZYK, AND E. J. LEWIS. 2000. Prevalenceof Cryptosporidium, Giardia, and Eimeria infections in postweaned and adult cattle on three Maryland farms. Veterinary Par-asitology 93: 103–112.
GUT, J., AND R. G. NELSON. 1999. Cryptosporidium parvum: Synchro-nized excystation in vitro and evaluation of sporozoite infectivitywith a new lectin-based assay. Journal of Eukaryotic Microbiology46: 56S–57S.
JENKINS, M. C., P. C. AUGUSTINE, J. R. BARTA, M. D. CASTLE, AND H.D. DANFORTH. 1991. Development of resistance to coccidiosis inthe absence of merogonic development using X-irradiated Eimeriaacervulina oocysts. Experimental Parasitology 72: 285–293.
———, H. D. DANFORTH, AND P. C. AUGUSTINE. 1991. X-irradiation ofEimeria tenella oocysts provides direct evidence that sporozoite
invasion and early schizont development induce a protective im-mune response(s). Infection and Immunity 59: 4042–4048.
———, P. G. SEFERIAN, P. C. AUGUSTINE, AND H. D. DANFORTH. 1993.Protective immunity against coccidiosis elicited by radiation-atten-uated Eimeria maxima sporozoites that are incapable of asexualdevelopment. Avian Diseases 37: 74–82.
———, J. TROUT, AND R. FAYER. 1998. Development and applicationof an improved semiquantitative technique for detecting low-levelCryptosporidium parvum infections in mouse tissue using poly-merase chain reaction. Journal of Parasitology 84: 182–186.
———, ———, C. MURPHY, J. A. HARP, J. HIGGINS, W. WERGIN, AND
R. FAYER. 1999. Cloning and expression of a DNA sequence en-coding a 41-kilodalton cryptosporidium parvum oocyst wall pro-tein. Clinical and Diagnostic Laboratory Immunology 6: 912–920.
KHRAMTSOV, N. V., K. M. WOODS, M. V. NESTERENKO, C. C. DYKSTRA,AND S. J. UPTON. 1997. Virus-like, double-stranded RNAs in theparasitic protozoan Cryptosporidium parvum. Molecular Microbi-ology 26: 289–300.
KILANI, R. T., AND L. SEKLA. 1987. Purification of Cryptosporidiumoocysts and sporozoites by cesium chloride and Percoll gradients.American Journal of Tropical Medicine and Hygiene 36: 505–508.
KOZWICH, D., K. A. JOHANSEN, K. LANDAU, C. A. ROEHL, S. WORONOFF,AND P. A. ROEHL. 2000. Development of a novel, rapid integratedCryptosporidium parvum detection assay. Applied and Environ-mental Microbiology 66: 2711–2717.
OKHUYSEN, P. C., C. L. CHAPPELL, J. CRABB, L. M. VALDEZ, E. T. DOUG-LASS, AND H. L. DUPONT. 1998. Prophylactic effect of bovine anti-Cryptosporidium hyperimmune colostrum immunoglobulin inhealthy volunteers challenged with Cryptosporidium parvum. Clin-ical and Infectious Diseases 26: 1324–1329.
PERRYMAN, L. E., S. J. KAPIL, M. L. JONES, AND E. L. HUNT. 1999.Protection of calves against cryptosporidiosis with immune bovinecolostrum induced by a Cryptosporidium parvum recombinant pro-tein. Vaccine 17: 2142–2149.
PHILLIPS, R. S. 1971. Immunity of rats and mice following injectionwith 60Co irradiated Babesia rodhaini infected red cells. Parasitol-ogy 62: 221–231.
PURNELL, R. E., D. LEWIS, AND D. W. BROCKLESBY. 1979. Babesia ma-jor: Protection of intact calves against homologous challenge bythe injection of irradiated piroplasms. International Journal for Par-asitology 9: 69–71.
RIGGS, M. W. 1997. Immunology: Host response and development ofimmunotherapy and vaccines. In Cryptosporidium and cryptospo-ridiosis, R. Fayer (ed.). CRC Press, Boca Raton, Florida, p. 129–161.
———. 2002. Recent advances in cryptosporidiosis: The immune re-sponse. Microbes and Infection 4: 1067–1080.
SCHELLER, L. F., AND A. F. AZAD. 1995. Maintenance of protective im-munity against malaria by persistent hepatic parasites derived fromirradiated sporozoites. Proceedings of the National Academy ofSciences of the United States of America 92: 4066–4088.
SILVIE, O., J. P. SEMBLAT, J. F. FRANETICH, L. HANNOUN, W. ELING, AND
D. MAZIER. 2002. Effects of irradiation on Plasmodium falciparumsporozoite hepatic development: Implications for the design of pre-erythrocytic malaria vaccines. Parasite Immunology 24: 221–223.
UNGAR, B. L. P., D. J. WARD, R. FAYER, AND C. A. QUINN. 1990. Ces-sation of Cryptosporidium-associated diarrhea in an acquired im-munodeficiency syndrome patient after treatment with hyperim-mune bovine colostrum. Gastroenterology 98: 486–489.
VETTERLING, J. M. 1969. Continuous-flow differential density flotationof coccidial oocysts and a comparison with other methods. Journalof Parasitology 55: 412–417.
YU, J. R., AND W. Y. PARK. 2003. The effect of gamma-irradiation onthe viability of Cryptosporidium parvum. Journal of Parasitology89: 639–642.
RESEARCH NOTES 1181
J. Parasitol., 90(5), 2004, pp. 1181–1183q American Society of Parasitologists 2004
Evaluation of the Presence of a Thapsigargin-Sensitive Calcium Store inTrypanosomatids Using Trypanosoma evansi as a Model
M. Mendoza*, A. Mijares*, H. Rojas*, C. Colina†‡, V. Cervino†, R. DiPolo*, and G. Benaim†‡, Universidad Nacional Experimental SimonRodrıguez, Instituto de Estudios Cientıficos y Tecnologicos, Centro de Estudios Biomedicos y Veterinarios, Apartado Postal 47925, Caracas1041A, Venezuela; *Laboratorio de Permeabilidad Ionica, Centro de Biofısica y Bioquımica, Instituto Venezolano de Investigaciones Cientıficas,Apartado Postal 21827, Caracas 1020 A, Venezuela; †Instituto de Biologıa Experimental, Escuela de Biologıa, Facultad de Ciencias,Universidad Central de Venezuela, Apartado Postal 47114, Caracas 1041 A, Venezuela; ‡Instituto de Estudios Avanzados, Apartado Postal17606, Caracas 1015 A, Venezuela. e-mail: [email protected]
ABSTRACT: Ca21 plays an important role in the regulation of severalimportant activities in different trypanosomatids. These parasites pos-sess a Ca21 transport system in the endoplasmic reticulum (ER) involvedin Ca21 homeostasis, which has been reported to be insensitive to thap-sigargin, a classical inhibitor of the sarcoplasmic–ER Ca21 adenosinetriphosphatase (ATPase) (SERCA) in most eukaryotic cells. However,currently there is a controversy regarding the existence of a thapsigar-gin-sensitive ER Ca21 store in these parasites. Therefore, we decided toexplore the effect of this inhibitor using different methodological ap-proaches. First, we selected Trypanosoma evansi as a parasite model towarrant the homogeneity of the population because this parasite hasonly a single life cycle, i.e., bloodstream-form trypomastigotes. Second,we compared the thapsigargin effect on Ca21 homeostasis by spectro-photometrical Ca21 measurements using 3 different approaches: whole-cell populations, cells that have been permeabilized by treatment withdigitonin, and intact single cells. Our results demonstrate that a lowconcentration of thapsigargin induces Ca21 release from intracellularCa21 stores in this parasite, which can be observed independently of themethod used. Furthermore, the addition of thapsigargin before or afternigericin did not abolish its effect, showing that thapsigargin acts spe-cifically on the ER. In conclusion, our results indicate the presence ofa nonmitochondrial thapsigargin-sensitive Ca21 store in T. evansi.
Calcium plays an important role in the regulation of multiple cellularactivities in different trypanosomatids (Docampo and Moreno, 1996;Stodjl and Clarke, 1996; Ruiz et al., 1998). The presence of differentCa21 transport systems in these parasites warrants an efficient regulationof the intracellular calcium concentration ([Ca21]i), which can reachapproximately 4 orders of magnitude with respect to the extracellularmilieu (Benaim, 1996). These mechanisms are associated with intra-cellular specialized organelles, as well as in the plasma membrane (Rub-en and Hutchinson, 1991; Docampo, 1993; Benaim, 1996; Docampoand Moreno, 2001). As in other eukaryotic cells, a Ca21 transport sys-tem has been localized at the endoplasmic reticulum (ER), which isinvolved in the intracellular Ca21 homeostasis. This system is charac-terized by adenosine triphosphate (ATP) stimulation and sodium or-thovanadate inhibition (Vercesi et al., 1991; Moreno, Docampo et al.,1992). Thapsigargin is a potent and classical inhibitor of the sarcoplas-mic–ER Ca21 adenosine triphosphatase (ATPase) (SERCA) (Thastrupet al., 1990), causing the release of Ca21 from the inositol-1,4,5-tris-phosphate–sensitive store in different cells. However, there is a contro-versy regarding the effect of thapsigargin on the ER Ca21 store in try-panosomatids. Thus, although some groups have detected the presenceof a nonmitochondrial Ca21 store sensitive to thapsigargin in Trypano-soma brucei at low concentration (1 mM) (Ruben and Akins, 1992),other groups failed to detect any effect of thapsigargin, even at higherconcentration (8 mM), in both T. brucei (Vercesi et al., 1993) and Try-panosoma cruzi (Moreno, Vercesi et al., 1992), using whole cells orcells that have been exposed to treatment with digitonin. In contrast,other studies in trypanosomatids aimed at correlating diverse cellularevents with [Ca21]i such as cell cycle control or invasion of mammaliancells and have shown that these processes are sensitive to low concen-tration of thapsigargin (1 mM) (Stojdl and Clarke, 1996; Yoshida et al.,2000; Neira et al., 2002).
Although species or cellular stage differences may account for thisdiscrepancy, another plausible explanation might be related to the dif-ferent methodological approaches used in these studies. Therefore, we
decided to explore the effect of thapsigargin on the Ca21 homeostasisin Trypanosoma evansi by comparing spectrophotometric Ca21 mea-surements in cell populations (whole or digitonin-treated cells) and in-tact single parasites. The latter approach, recently reported by our group,allows for the measurement of free cytosolic calcium concentration withhigher time resolution (Mendoza et al., 2001, 2002). Furthermore, theuse of T. evansi as a model is advantageous because it is a parasite inwhich the only cellular stage is the bloodstream form (trypomastigote),thus avoiding the possible problems related to heterogeneity of the pop-ulations used for studies of Ca21 homeostasis in other trypanosomatids.Finally, there are only few reports with respect to the way in which T.evansi regulates its intracellular calcium concentration (Mendoza et al.,2001, 2002) and there is no report on the effect of thapsigargin in thisparasite.
In this study, first we measured Ca21 transport by intracellular organ-elles in situ, using arsenazo III in digitonin-treated cells of T. evansi asdescribed previously (Benaim et al., 1990). Figure 1 shows the effectof a low concentration of thapsigargin (1 mM) on Ca21 transport in apermeabilized cell of T. evansi cells in the presence of oligomycin (aF1–F0 mitochondrial ATPase inhibitor) and antimycin A (a respiratorychain inhibitor). As observed (Fig. 1), thapsigargin produced a Ca21
release under this condition. We included antimycin A in the assaybecause little is known about mitochondrial function and the possibleeffect of this drug in T. evansi mitochondrion. In addition, this drughas been used by other authors in the experiments used to evaluate thethapsigargin effect in T. brucei procyclic trypomastigotes (Vercesi etal., 1993) and in T. cruzi (Docampo et al., 1993). However, identicalresults were obtained when antimycin A was omitted from the medium(results not shown). These results are in agreement with the fact that T.evansi belongs to T. brucei clade trypanosome (Stevens and Gibson,1999) and only has bloodstream forms that possessed an undevelopedmitochondrion depending only on glycolysis to generate ATP. In thesetrypanosomes, Ca21 uptake by mitochondria depends on the electro-chemical proton gradient membrane generated by an oligomycyn-sen-sitive ATPase. Therefore, the addition of oligomycin prevents Ca21 up-take by the mitochondria in these assays. Indeed, we have previouslyshown that oligomicin produced a release of Ca21 from the mitochon-drial pool in T. evansi (Mendoza et al., 2002). It can be seen that inthe absence of ATP, Ca21 uptake is low (Fig. 1Aa, Ba) and thapsigargininduces a low Ca21 release (Fig. 1Aa) when compared with the effectproduced in the presence of ATP (Fig. 1Ab). Basically, most of theCa21 uptake is due to an ATP-dependent Ca21 sequestration. These re-sults suggest the presence of a nonmitochondrial Ca21 store that dependson a thapsigargin-sensitive Ca21 ATPase. These results contrast withthose reported in other trypanosomatids, where thapsigargin (1–8 mM)did not have any effect on intracellular Ca21 release in digitonin-treatedcells of T. cruzi, T. brucei (Moreno, Vercesi et al., 1992; Vercesi et al.,1993), or Herpetomonas sp. (Sodre et al., 2000). Figure 1Ab shows thatthe addition of the Ca21 ionophore A23187 produces a large Ca21 re-lease, probably from other compartments, such as acidocalcisomes, inwhich Ca21 uptake is also an ATP-dependent process (Docampo andMoreno, 2001). In addition, as seen in Figure 1Ac, when digitonizedcells were pretreated with vanadate, a significant decrease in thapsigar-gin-induced Ca21 release is observed, in comparison with the effect ofthapsigargin alone (Fig. 1Ab). This inhibition suggests that a ‘‘P’’-typeCa21 ATPase, probably a SERCA Ca21 pump, is responsible for thiseffect as reported previously in other trypanosomatids (Vercesi et al.,
1182 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
FIGURE 1. Effect of thapsigargin on Ca21 uptake by digitonin-per-meabilized cells of Trypanosoma evansi. The reaction medium contains125 mM sucrose, 65 mM KCl, 10 mM Tris–HCl, 2.5 mM KH2PO4, 1mM MgCl2, pH 7.4, 40 mM Arsenazo III, 30 mM digitonin, 5 mM CaCl2,2 mg/ml oligomycin, and 2 mg/ml antimycin A in the presence or ab-sence of 1 mM ATP, at a final density of 106 cells/ml. Calcium move-ments were followed by measuring the changes in the absorbance spec-trum of the metallochromic indicator using an Aminco DW-2a dualwavelength (675–685 nm) spectrophotometer at 30 C. The arrows in-dicate the additions of thapsigargin (1 mM) (Tg), calcium ionophoreA23187 (10 mM), sodium vanadate (400 mM) (V), and nigericin (4 mM)(Nig). Traces shown are representative of at least 3 independent exper-iments conducted on separate preparations.
FIGURE 2. Effect of the thapsigargin in [Ca21]i in single (A) andpopulation (B) cells loaded with Fura-2 (8 mM) of Trypanosoma evansiincubated in the absence of extracellular Ca21 (100 mM ethylenegly-coltetraacetic acid). The medium contains 145 mM NaCl, 4 mM KCl,2 mM MgCl2, 11 mM glucose, and 10 mM N-2-hydroxyethylpiperazine-N9-2-ethanesulfonic acid–NaOH (pH 7.3). In cell populations, the ex-periments were carried out in a Hitachi F-2000 spectrofluorometer. Insingle cells, a fluorescence-imaging apparatus (Ion Optix Co., Milton,Massachusetts) was used to obtain fluorescent recordings of Fura-2–loaded parasites, and the solutions were delivered to the parasite usinga DAD-12 perfusion system (Adam & List Associates, Westbury, NewYork). The arrows indicate the additions of thapsigargin (2 mM) andnigericin (8 mM) in the medium. In the dot trace thapsigargin was ap-plied before nigericin, and in the solid trace thapsigargin was appliedafter nigericin in both panels (A and B). All the experiments werecarried out at 30 C. Traces shown are representative of at least 3 in-dependent experiments conducted on separate preparations.
1991; Moreno, Docampo et al., 1992). Second, to evaluate the effect ofthapsigargin on [Ca21]i in whole T. evansi cells loaded with Fura-2 asCa21 indicator, we followed [Ca21]i in single cells or cell populationsbased on the methods described by Mendoza et al. (2001) and Morenoand Zhong (1996), respectively. Our experiments were performed inethyleneglycoltetraacetic acid–containing buffers; therefore, the increasein [Ca21]i reflected its release from intracellular stores. As shown inFigure 2 (single cell [A] and cell populations [B]), addition of thapsi-gargin (2 mM) produces a small increase in [Ca21]i of about 35 and 50nM, respectively. Our results are consistent with those reported in T.brucei by Ruben and Akins (1992), showing the existence of a thapsi-gargin-sensitive Ca21 pool in whole-cell populations. These authors re-ported that the addition of 1 mM thapsigargin elevated [Ca21]i by ap-proximately 50–75 nM, thus indicating that a thapsigargin-sensitivestore has a low calcium capacity, which is characteristic of the ER. Ourexperiments (Fig. 2A) using [Ca21]i measurements in single cells strong-ly support their conclusion. In contrast, other researchers have not de-tected a thapsigargin-induced Ca21 release from intracellular stores atlow concentrations (1–2 mM) using the whole-cell population approacheither in T. brucei (Verccesi et al., 1993) or in T. cruzi (Moreno, Vercesiet al., 1992; Docampo et al., 1993). The latter authors (Docampo et al.,1993) reported an increase in [Ca21]i in the absence of extracellularcalcium only when they used high concentrations of thapsigargin (20mM). Our results demonstrate that independent of the different meth-odological approaches used, thapsigargin at low concentration producesa clear effect on Ca21 homeostasis in T. evansi.
It has been postulated that thapsigargin could have unspecific effectsin trypanosomatids and other cells. Thus, thapsigargin at high concen-trations (4–20 mM) was able to release Ca21 from mitochondria in T.cruzi and T. brucei, probably by collapsing the mitochondrial membrane
potential and not by the classical SERCA pump inhibition (Docampoet al., 1993; Vercesi et al., 1993). Therefore, we decided to determineif the Ca21 mobilization induced by low concentrations of thapsigargin(1–2 mM) was indeed from the ER or from other sources, including themitochondria and the acidocalcisomes. For this purpose, we evaluatedthe effect of thapsigargin in combination with nigericin (a K1–H1 ex-changer), which is known to produce Ca21 release from intracellularstores (acidocalcisomes and mitochondria) as a consequence of the col-lapse of the proton gradient present in the membrane of these organelles(Ruben et al., 1991; Vercesi et al., 1993; Docampo et al., 1995; Men-doza et al., 2002). Interestingly, Figures 1B, 2A, B (whole and digito-nin-treated cells, respectively) show that thapsigargin, at low concen-trations, produces Ca21 release before and after (Figs. 1Bb, Bc, 2A, B)nigericin treatment. The effects of thapsigargin and nigericin on intra-cellular calcium concentration and Ca21 transport were additive. Evenmore, regardless of the order in which the treatment was given, any ofthese drugs elevated [Ca21]i independently. These data support the no-tion that these drugs act on different intracellular Ca21 stores in T. evan-si, i.e., one small thapsigargin-sensitive store that resides within the ERand another larger nigericin-sensitive store that resides in organelleswhose calcium accumulation is dependent on the proton gradient pre-sent in their membrane. These results, together with those describedabove, where thapsigargin produces a Ca21 release even in the presenceof oligomycin, rule out the participation of a nonmitocondrial compo-nent in the effect of thapsigargin at low concentrations in T. evansi(Fig. 1). These results are in agreement with those reported for T. bru-cei, where a thapsigargin-sensitive Ca21 pool apparently resides in theER (Ruben and Akins, 1992).
In agreement with our findings, the presence of a SERCA-type Ca21
ATPase in these parasites has been supported by cloning and sequencingin T. brucei (Nolan et al., 1994), T. cruzi (Furuya et al., 2001), andLeishmania mexicana amazonensis (Lu et al., 1997). In T. brucei, thisCa21 ATPase activity in vitro possesses a high affinity for Ca21, whichis sensitive to vanadate and to low concentrations of thapsigargin (No-lan et al., 1994). However, in T. cruzi, this pump was insensitive to lowconcentrations (0.3–1 mM) of thapsigargin (Furuya et al., 2001). In thesame study, the authors pointed out differences between the sequencesof the M3 and S3 fragments in these trypanosomatids, which are re-sponsible for the sensitivity of SERCA pumps to thapsigargin (Furuyaet al., 2001). They concluded that differences in the transmembranesequences of the thapsigargin-binding site could account for the ob-served differences in thapsigargin sensitivity of these pumps in thesetrypanosomatids (Norregaard et al., 1994). In this regard, it is interesting
RESEARCH NOTES 1183
to note that T. evansi belongs to the T. brucei clade of trypanosomes(Stevens and Gibson, 1999). Thus, it is also possible that the effect ofthapsigargin observed in T. evansi is a consequence of a closer phylo-genetic relationship between this parasite and T. brucei, when comparedwith other trypanosomatids.
The thapsigargin-sensitive Ca21 store in T. evansi is presumed toreside in the ER based on (1) the low capacity of this store, (2) intra-cellular localization because it occurs in the absence of extracellularcalcium, (3) ATP dependency and sensitivity to vanadate, (4) specificCa21 release by low concentration of thapsigargin, and (5) the fact thatnigericin and thapsigargin effects were additive and have different sizes.In conclusion, the results obtained in this work indicate that the ER isthe intracellular Ca21 store targeted by low concentrations of thapsigar-gin for Ca21 release in T. evansi, as reported for T. brucei (Ruben andAkins, 1992). These results also suggest that the Ca21 pump present inthis organelle has strong similarities with the SERCA pump that hasbeen described in higher eukaryotic cells.
This work was supported by grants from the Fondo Nacional de In-vestigaciones Cientıficas y Tecnologicas (FONACIT-Venezuela, GrantsS1-99000946, S1-99000058, S1-2000000613, G-2000001152, and G-2001000637), CDCH–UCV PI 03-33-4798-00, and Fundacion Polar-Venezuela.
LITERATURE CITED
BENAIM, G. 1996. Intracellular calcium signal and regulation in Leish-mania. In Molecular and immune mechanism in pathogenesis ofcutaneous leishmaniasis, F. Tapia, G. Caceres-Dittmar, and M. A.Sanchez (eds.). R. G. Landes Co., Biomedical Publishers, Austin,Texas, p. 89–106.
———, R. BERMUDEZ, AND J. A. URBINA. 1990. Ca21 transport in iso-lated mitochondrial vesicles from Leishmania braziliensis promas-tigotes. Molecular and Biochemical Parasitology 39: 61–68.
DOCAMPO, R. 1993. Calcium homeostasis in Trypanosoma cruzi. Bio-logical Research 26: 189–196.
———, AND S. N. J. MORENO. 1996. The role of Ca21 in the processof cell invasion by intracellular parasites. Parasitology Today 12:61–65.
———, AND ———. 2001. The acidocalcisome. Molecular and Bio-chemical Parasitology 114: 151–159.
———, S. N. J. MORENO, AND A. E. VERCESI. 1993. Effect of thapsi-gargin on calcium homeostasis in Trypanosoma cruzi trypomasti-gotes and epimastigotes. Molecular and Biochemical Parasitology59: 305–313.
———, D. S. SCOTT, A. E. VERCESI, AND S. N. J. MORENO. 1995. In-tracellular calcium storage in acidocalcisomes of Trypanosoma cru-zi. Biochemical Journal 310: 1005–1012.
FURUYA, T., M. OKURA, F. A. RUIZ, D. A. SCOTT, AND R. DOCAMPO.2001. TcSCA complements yeast mutant defective in Ca21 pumpsand encodes a Ca21-ATPase that localizes to the endoplasmic re-ticulum of Trypanosoma cruzi. Journal of Biological Chemistry276: 32437–32445.
LU, H. G., L. ZHONG, K. P. CHANG, AND R. DOCAMPO. 1997. IntracellularCa21 pool content and signaling, and expression of a calcium pumpare linked to virulence in Leishmania mexicana amazonensis. Jour-nal of Biological Chemistry 272: 9464–9473.
MENDOZA, M., A. MIJARES, H. ROJAS, M. RAMOS, AND R. DIPOLO. 2001.Trypanosoma evansi: A convenient model for studying intracellularCa21 homeostasis using fluorometric ratio imaging from single par-asites. Experimental Parasitology 99: 213–219.
———, ———, ———, J. P. RODRıGUEZ, J. A. URBINA, AND R. DI-POLO. 2002. Physiological and morphological evidences for thepresence acidocalcisomes in Trypanosoma evansi: Single cell fluo-rescence and 31P NMR studies. Molecular and Biochemical Para-sitology 125: 23–33.
MORENO, S. N. J., R. DOCAMPO, AND A. E. VERCESI. 1992. Calciumhomeostasis in procyclic and bloodstream forms of Trypanosomabrucei. Journal of Biological Chemistry 267: 6020–6026.
———, A. E. VERCESI, O. P. PIGNATARO, AND R. DOCAMPO. 1992. Cal-cium homeostasis in Trypanosoma cruzi amastigotes: Presence ofinositol phosphates and lack of an inositol 1,4,5-trisphosphate-sen-sitive calcium pool. Molecular and Biochemical Parasitology 52:251–261.
———, AND L. ZHONG. 1996. Acidocalcisomes in Toxoplasma gondiitachyzoites. Biochemical Journal 313: 655–659.
NEIRA, I., A. T. FERREIRA, AND N. YOSHIDA. 2002. Activation of distinctsignal transduction pathways in Trypanosoma cruzi isolates withdifferential capacity to invade host cells. International Journal forParasitology 32: 405–414.
NOLAN, D. P., P. REVELARD, AND E. PAYS. 1994. Overexpression andcharacterization of a gene for a Ca21-ATPase of the endoplasmicreticulum in Trypanosoma brucei. Journal of Biological Chemistry269: 26045–26051.
NORREGAARD, A., B. VILSEN, AND J. P. ANDERSEN. 1994. Transmembranesegment M3 is essential to thapsigargin sensitivity of sarcoplasmicreticulum Ca21-ATPase. Journal of Biological Chemistry 269:26598–26601.
RUBEN, L., AND C. D. AKINS. 1992. Trypanosoma brucei: The tumorpromoter thapsigargin stimulates calcium release from an intracel-lular compartment in slender bloodstream forms. Experimental Par-asitology 74: 332–339.
———, A. HUTCHINSON, AND J. MOEHLMAN. 1991. Calcium homeostasisin Trypanosoma brucei. Journal of Biological Chemistry 266:24351–24358.
RUIZ, R. C., S. FAVORETO, M. L. DORTA, M. E. M. OSHIRO, A. T. FER-REIRA, P. M. MANQUE, AND N. YOSHIDA. 1998. Infectivity of Try-panosoma cruzi strains is associated with differential expression ofsurface glycoproteins with differential Ca21 signalling activity. Bio-chemical Journal 330: 505–511.
SODRE, C. L., B. L. M. MOREIRA, F. B. NOBREGA, F. R. GADELHA, J. R.MEYER-FERNANDES, P. M. DUTRA, A. E. VERCESI, A. H. C. S. LOPES,H. M. SCOFANO, AND H. BARRABIN. 2000. Characterization of theintracellular Ca21 pools involved in the calcium homeostasis inHerpetomonas sp. promastigotes. Archives of Biochemistry andBiophysics 380: 53–91.
STEVENS, J. R., AND W. C. GIBSON. 1999. The evolution of pathogenictrypanosomes. Cadernos de Saude Publica 15: 673–684.
STOJDL, D. F., AND M. W. CLARKE. 1996. Trypanosoma brucei: Analysisof cytoplasmic Ca21 during differentiation of bloodstream stages invitro. Experimental Parasitology 83: 134–146.
THASTRUP, O., P. J. CULLEN, B. K. DROBAK, M. R. HANLEY, AND A. P.DAWSON. 1990. Thapsigargin, a tumor promoter, discharges intra-cellular Ca21 stores by specific inhibition of the endoplasmic retic-ulum Ca21-ATPase. Proceeding of the National Academy of Sci-ences of the United States of America 87: 2466–2470.
VERCESI, A. E., M. E. HOFFMAN, C. F. BERNARDES, AND R. DOCAMPO.1991. Regulation of intracellular calcium homeostasis in Trypano-soma cruzi. Effects of calmidazolium and trifluoperazine. Cell Cal-cium 12: 361–369.
———, S. N. J. MORENO, C. F. BERNARDES, A. R. MEINICKE, E. C.FERNANDES, AND R. DOCAMPO. 1993. Thapsigargin causes Ca21 re-lease and collapse of the membrane potential of Trypanosoma bru-cei mitochondria in situ and isolated rat liver mitochondria. Journalof Biological Chemistry 25: 8564–8568.
YOSHIDA, N., S. FAVORETO JR., A. T. FERREIRA, AND P. M. MANQUE.2000. Signal transduction induced in Trypanosoma cruzi metacy-clic trypomastigotes during the invasion of mammalian cells. Bra-zilian Journal of Medical Biological Research 33: 269–278.
1184 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
J. Parasitol., 90(5), 2004, pp. 1184–1185q American Society of Parasitologists 2004
Exogenous Nucleosides are Required for the Morphogenesis of the Human FilarialParasite Brugia malayi
T. V. Rajan, Department of Pathology, University of Connecticut Health Center, Farmington, Connecticut. e-mail: [email protected]
FIGURE 2. All wells contained 200 ml. a-MEM 12561. In the pres-ence of serum (MBS #2), molting efficiency is seen to be high. In theabsence of serum (MEM #2), molting does not take place. On supple-mentation of this medium with 10–50 mM nucleosides (all 4), moltingis restored to the level seen in the presence of serum.
FIGURE 1. L3 were cultured at 5 larvae/well in 96-well, flat-bottomcluster dishes. All wells received vitamin C supplementation to a finalconcentration of 75 mM on day 5 of culture. Culture conditions usedin this experiment were: a-MEM supplemented with 10% fetal bovineserum (MBS #1 in Figure); a-MEM 12571 without fetal bovine serumsupplementation (MEM #1); a-MEM supplemented with 10% fetal bo-vine serum (MBS #2); a-MEM without 10% fetal bovine serum sup-plementation (MEM #2). It is noted that the molting efficiency is highin either formulation of a-MEM as long as the medium is supplementedwith 10% fetal bovine serum. However, in the absence of serum, molt-ing is still efficient in MEM 12571 but does not take place in MEM12561.
FIGURE 3. Columns labeled MBS #2 and MEM #2 as in Figures 1and 2. Other culture conditions included those in which the wells weresupplemented with 3 nucleotides in all 4 combinations. It is seen thatthe omission of any single ribonucleotide does not significantly reducemolting efficiency in comparison with that seen in the presence of se-rum.
ABSTRACT: The nematode parasites Wuchereria bancrofti, Brugia ma-layi, and B. timori cause a human disease known as lymphatic filariasis,which afflicts approximately 120 million people worldwide. The para-sites enter the human host from the mosquito as L3 or infective larvaeand subsequently differentiate through 2 molts. In this communication,I report that B. malayi and B. pahangi depend on an exogenous sourceof at least 1 purine and 1 pyrimidine nucleoside to complete the L3 toL4 molt. The requirement for exogenous nucleosides opens the door forpossible chemotherapeutic intervention.
The filarial worms are obligate nematode parasites. They require 2hosts (a mammal and an insect) to complete their life cycles (Rajan andGundlapalli, 1997). They enter the mammal as an L3 and develop intothe L4 in approximately 8 days. This transformation requires massivereorganization of the structure of the organisms. The L3 is 760–900mm in length, has a sealed buccal cavity, and lacks an open gut andgonads. The L4 is greater than 2 mm in length, has an open buccalcavity, and a fully formed gut and gonads. Development from L3 to L4is initiated by a shift to 37 C. During this period, the organism synthe-sizes a new cuticle underneath the old one, after which the old cuticleis cast off.
I have been attempting to determine the minimal culture conditionsthat will support their growth and development from the infective (L3)to the L4 stage. I have recently shown that Brugia malayi and B. pa-hangi L3s require vitamin C to molt successfully to L4 (Rajan et al.,2003) and that this compound is optimally effective if added on day 5of culture. Even when vitamin C is supplemented, however, I noted thatthe molting efficiency was significantly higher in culture medium basedon a-minimum essential medium (MEM) than Roswell Park MemorialInstitute (RPMI) 1640. In a typical experiment, molting in a-MEM was43.33 6 6.8% in contrast to 0 6 0% in RPMI. This led me to comparethe constituents of these 2 formulations. I found that in addition tomultiple quantitative differences in the precise concentrations of several
shared components, there were some qualitative differences. a-MEMcontains pyruvate (1 mM) and all 4 ribonucleosides and deoxyribonu-cleosides (at ;40 mm). These multiple differences made it difficult toapproach the puzzle of the differential ability of these 2 serum-freemedia to support the molt.
Fortunately, 2 different formulations of a-MEM are marketed byGIBCO (Invitrogen, Carlsbad, California). These 2 liquid a-MEM prep-arations (catalogs 12561 and 12571) differ from each other solely inthe presence of all 4 nucleosides and deoxynucleosides in the latter. The2 formulations are otherwise identical, unlike the more extensive quan-titative and qualitative differences between RPMI and a-MEM. Thispermitted me to investigate whether the 1 qualitative difference betweena-MEM 12561 and 12571 could be responsible for the observation.
In a head-to-head comparison of the 2 a-MEM preparations, I noted
RESEARCH NOTES 1185
FIGURE 4. Columns denoted MBS #1, MEM #1, MBS #2, and MEM#2 are as in Figure 1. It is noted that the addition of all 4 nucleotides(ACGU) at 40 mM to a-MEM #2 increases molting efficiency. In thisparticular experiment, the molting efficiency in the presence of all 4nucleotides is not as high as that seen in the presence of 10% serumsupplementation. In the presence of 2 nucleotides at a time, molting isstill high as long as 1 nucleotide is a purine and the other a pyrimidine.
FIGURE 5. Columns refer to the mean percent molting for each con-dition, with the standard error of the mean. Larvae molt well in thepresence of 10% FBS but not in serum-free medium. Addition of asingle ribonucleoside does not result in a significant increase in molting.In some experiments, minimal molting was seen in the presence ofeither C alone or U alone or both but this increase was not statisticallysignificant.that molting was significantly more efficient in serum-free a-MEM (cat-
alog 12571) than in serum-free a-MEM (catalog 12561) (Fig. 1). Thisled me to conclude that B. malayi L3 requires an exogenous source ofribo- or deoxyribonucleosides, or both.
To demonstrate whether either or both ribo- and deoxyribonucleo-sides are required, I supplemented a-MEM 12561 with all 4 ribonucle-osides at varying concentrations. The data shown in Figure 2 demon-strate that supplementation of a-MEM 12561 with all 4 ribonucleosides(without any deoxyribonucleosides) at concentrations ranging from 20to 40 mm results in molting efficiencies that are comparable with thoseseen in serum-supplemented medium. At higher concentrations (.60mm), there was some inhibition of molting.
To determine whether all 4 ribonucleosides were required, I comparedthe molting efficiency of L3s in RPMI supplemented with 3 nucleotidesat a time (ACG, AGU, ACU, CGU). These data (Fig. 3) demonstratethat no one nucleoside appeared to be absolutely limiting because molt-ing did take place in all combinations.
To determine the minimal requirement of nucleosides, I next incu-bated worms under our standard culture conditions (a-MEM, catalog12561 without added fetal bovine serum), supplemented with combi-nations of 2 nucleosides at a time. These data (Fig. 4) indicate thatwhen the combination of 2 nucleosides includes 1 purine and 1 pyrim-idine (A 1 C or U or G 1 C or U), molting approaches rates that areseen in the presence of all 4 nucleosides. However, if both the nucle-osides are purines (adenosine 1 guanosine) or pyrimidines (cytidine 1uridine), molting frequency is dramatically reduced. These data wouldimply that the worms require both a purine and a pyrimidine nucleosideto generate the full complement of precursors for RNA and DNA syn-theses. Consistent with this observation, when I reduced the number ofnucleosides to just 1, molting is extremely inefficient (Fig. 5). Nucleicacid bases themselves, nucleotides, or their di- or triphosphates do notsubstitute for the nucleosides (data not shown).
The data demonstrate that in addition to vitamin C, B. malayi L3requires exogenous nucleosides to progress to L4. These data extendthe observations of Chen and Howells (1981) who showed that radio-actively labeled adenosine (a ribonucleoside) from the culture mediumis incorporated into the tissues but radioactively labeled thymidine (adeoxyribonucleoside) is not.
The absolute requirements for nucleosides for the molting of B. ma-layi raise the possibility that it should be possible to block molting andfurther development of L3 from natural infections by nucleoside analogsthat are toxic after uptake and incorporation. Howells et al. (1981) tested2 compounds (5-fluorouracil and 5-fluorocytosine) for micro- and ma-crofilaricidal efficacy and found that both could block development ofthe parasites, albeit temporarily. I am currently extending these obser-vations and exploring the effects of other analogs.
In conclusion, I report a completely defined culture condition thatsupports the L3 to L4 molt of B. malayi. The optimal condition is theuse of a-MEM (catalog 12561), supplemented at the beginning of cul-ture with 10–40 mM nucleosides and supplemented further with 75 mMvitamin C on day 5 of culture.
This work was made possible by grants from the NIH to T.V.R. (AI-042362, AI-050228, and AI-039075). I thank Carol McGuiness for ex-cellent technical assistance and Jennifer Wegh for typing the manu-script.
LITERATURE CITED
CHEN, S. N., AND R. E. HOWELLS. 1981. Brugia pahangi: Uptake andincorporation of nuclei acid precursors by microfilariae and macro-filariae in vitro. Experimental Parasitology 51: 296–306.
HOWELLS, R. E., J. TINSLEY, E. DEVANEY, AND G. SMITH. 1981. Theeffect of 5-fluorouracil and 5-fluorocytosine on the development ofthe filarial nematodes Brugia pahangi and Dirofilaria immitis. ActaTropica 38: 289–304.
RAJAN, T. V., AND A. V. GUNDLAPALLI. 1997. Lymphatic filariasis. Chem-ical Immunology 66: 125–158.
———, N. PACIORKOWSKI, I. KALAJZIC, AND C. MCGUINESS. 2003. As-corbic acid is a requirement for the morphogenesis of the humanfilarial parasite, Brugia malayi. Journal of Parasitology 89: 868–870.
1186 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
J. Parasitol., 90(5), 2004, pp. 1186–1188q American Society of Parasitologists 2004
Observations on the Leech Placobdella ornata Feeding From Bony Tissues of Turtles
Mark E. Siddall and Eugene S. Gaffney*, Division of Invertebrate Zoology, American Museum of Natural History, Central Park West at 79thStreet, New York 10024; *Division of Paleontology, American Museum of Natural History, Central Park West at 79th Street, New York 10024.e-mail: [email protected]
FIGURE 1. (A) Four Placobdella ornata (arrows) feeding from the xiphoplastron bone of an adult Chelydra serpentina. (B) Right bridge of asnapping turtle with a large leech attached at the hypoplastron margin exhibiting pitting (circle) immediately after the removal of a leech. (C)Blood upwelling near the hyoplastron–hypoplastron suture shortly thereafter. (D) Three P. ornata feeding (arrows) from the sulcus of the poste-riormost right costal scales on the carapace of an adult Chrysemys picta and (E) 1 of those 3 demonstrating blood filled ceca (arrow) after removal.
ABSTRACT: The leech Placobdella ornata was observed feeding fromthe blood sinuses of the plastron and carapace bones of Chelydra ser-pentina and Chrysemys picta. Evidence of successful feeding includedblood upwelling from the point of attachment and gastric ceca of theleeches freshly filled with blood after removal. There was an apparentpreference for the sulci between scales of the shell.
Of the 8 species of Placobdella in North America, 3 in particular arecommon ectoparasites of freshwater chelonians: Placobdella papillifera(Verrill, 1872), Placobdella parasitica (Say, 1824), and Placobdella or-nata (Verrill, 1872), the last 2 being most common in the vicinity ofthe Great Lakes (Davies, 1971; Klemm, 1982). The ecological inter-actions between these leeches and their turtle hosts have been of interestin terms of studies of seasonal prevalence (e.g., Ernst, 1971; Koffler etal., 1978; Brooks et al., 1990; Saumure and Livingston, 1994; Grahamet al., 1997), relationship to blood parasitology (Siddall and Desser,1990, 1991, 1992), and also regarding cleaning symbioses exhibited
between grackles and turtles (Vogt, 1979) and even between differentturtle species (Krawchuck et al., 1997). Species of Placobdella exhibitseasonal variation in prevalence and intensity on a variety of turtlehosts, with P. parasitica being the species most commonly found onthe surface of wild-caught turtles (Ernst, 1971; Koffler et al., 1978;MacCulloch, 1981; Brooks et al., 1990; Saumure and Livingston, 1994;Graham et al., 1997).
Placobdellid leeches are most often found feeding from accessiblesoft body regions. For example, P. multilineata, which feeds exclusivelyon the American alligator, usually is found inside the mouth in the softerportions of the palate corresponding to tooth sockets (Cherry and Ager,1982; Davies and Wilkialis, 1982; Yang and Davies, 1985). Thoseleeches specializing on turtles usually are found on the posterior legsand inguinal region and secondarily on the anterior legs and axillaryregion (Ernst, 1971; MacCulloch, 1981; Brooks et al., 1990). Occa-sionally, leeches have been reported as being present, but rare, on theplastron (Brooks et al., 1990) or on the carapace (MacCulloch, 1981),
RESEARCH NOTES 1187
though Koffler et al. (1978) found this to be a common occurrence onturtles in Saskatchewan. Because P. parasitica is known to remain onits host between blood feedings (Brooks et al., 1990; Graham et al.,1997), those instances of leeches attached to bony surfaces have beenthought to be phoretic only.
On 2 occasions (July 1990 and May 2001), leeches on the plastronor carapace of turtles in Algonquin Park, Ontario, Canada, were deter-mined to have been blood feeding from bony tissues. In connectionwith transmission studies of Haemogregarina balli (Siddall and Desser,1991), approximately 30 unfed P. ornata were placed in a holding tankwith an adult female snapping turtle (Chelydra serpentina). After 3 hr,the turtle was removed from the tank so that the blood-fed leeches couldbe retrieved. Most of the leeches remaining on the snapping turtle wereattached to the inguinal region and hind legs; however, 7 individualswere firmly attached to the plastron by the oral sucker (Fig. 1A–C).One leech was attached at the sulcus of the left anal and femoral scales(Fig. 1A) and 3 near the margin of the right anal scale (Fig. 1A), all ofwhich overlie the xiphoplastron bone. One was on the left abdominalscale (not shown) and 2 on the left bridge overlying the hyoplastronand hypoplastron bones. To determine whether these leeches were mere-ly resting in these positions or were feeding from the blood sinuses ofthe underlying bones, they were gently removed and the attachmentpoints examined. Figure 1B shows 1 such attachment point immediatelyafter the removal of a leech from the left bridge, and Figure 1C illus-trates blood upwelling from the attachment point a few seconds later.In May 2001, an additional 12 P. ornata were permitted to feed on anadult painted turtle (Chrysemys picta) by placing them on the carapaceof the host. Before attachment, the gastric ceca of each leech weredetermined to be free of blood. Most of the leeches crawled off of thecarapace and onto the soft tissues in the hind leg sockets. However, 3remained on the carapace attached at the sulcus of the 2 right posteriorcostal scales (Fig. 1D). Although blood was not seen welling up fromthe attachment site after removal of the leeches, each exhibited blood-filled gastric tissues, indicating that they had acquired a blood mealfrom the attachment point (Fig. 1E).
It is clear from these observations that P. ornata is capable of pen-etrating the bony tissues of turtle shells with its proboscis to obtain ablood meal. These observations may explain pitting observed in fossilturtles from the Cenozoic (Hutchison and Frye, 2001). The specificattachment sites observed here suggest that in those circumstances inwhich leeches eschew the softer skin areas, there is a preference for thesulci between scales overlying the bones. However, these sulci do notcorrespond to sutures between bony plates of the plastron or the cara-pace (Zangerl, 1969). The salivary structure and histochemical charac-teristics of placobdellid leeches indicate a complex mixture of salivarysecretions originating in at least 4 cell types (Siddall, 1991; Moser andDesser, 1995). The precise enzymatic activities of these secretions arenot known; however, one might expect both a mechanism for decalci-fication and for digestion of connective tissue matrix proteins to permitpenetration of the blood sinuses of bony tissue observed here. The mostextensively studied salivary enzyme from a glossiphoniid leech is he-mentin, a neutral mettaloprotease with a high affinity for cleaving fi-brinogen (Malinconico et al., 1984), which in turn is strongly regulatedby free calcium ions (Budzynski, 1991). As well, there is considerableevidence for salivary components in placobdellids that are not antihe-mostatic like the hyaluronoglucoronidase activity known from virtuallyevery family of leeches (Hovingh and Linker, 1999). The only salivaryproteins characterized so far from P. ornata are the ornatins (Mazur etal., 1991), which, though they are involved in platelet disaggregation,also have potent disintegrin activity (McLane et al., 1995).
At present, only P. ornata has been observed feeding from the bloodsinuses of turtle bone notwithstanding the fact that P. parasitica is moreabundant on turtles (Brooks et al., 1990). This difference, however, ismore a manifestation of these 2 species having markedly different feed-ing behaviors. Specifically, between feedings P. parasitica is known toremain on its turtle host, digesting until ready for a subsequent bloodmeal, whereas P. ornata immediately leaves its host on completion offeeding (Siddall and Desser, 1992; Saumure and Livingston, 1994; Gra-ham et al., 1997). Because P. parasitica can occupy large portions ofthe available soft body regions in the inguinal and axial areas (Brookset al., 1990), it is possible that feeding through bone is an adaptationacquired by P. ornata, permitting acquisition of a blood meal fromsurfaces not occupied by its congener.
The authors are grateful for support from the National Science Foun-dation (BIO-DEB 0119329 and BIO-DEB 0108163).
LITERATURE CITED
BROOKS, R. J., D. A. GALBRAITH, AND J. A. LAYFIELD. 1990. Occurrence ofPlacobdella parasitica (Hirudinea) on snapping turtles, Chelydra ser-pentina, in Southeastern Ontario. Journal of Parasitology 76: 190–195.
BUDZYNSKI, A. Z. 1991. Interaction of hementin with fibrinogen andfibrin. Blood Coagulation and Fibrinolysis 2: 149–152.
CHERRY, R. H., AND A. L. AGER JR. 1982. Parasites of American alli-gators (Alligator mississipiensis) in South Florida. Journal of Par-asitology 68: 509–510.
DAVIES, R. W. 1971. A key to the freshwater Hirudinoidea of Canada.Journal of the Fisheries Research Board of Canada 28: 543–552.
DAVIES, R. W., AND J. WILKIALIS. 1982. Observations on the ecologyand morphology of Placobdella papillifera (Verril) (Hirudinoidea:Glossiphoniidae) in Alberta, Canada. American Midland Naturalist107: 316–324.
ERNST, C. H. 1971. Seasonal incidence of leech infestation on the paint-ed turtle, Chrysemys picta. Journal of Parasitology 57: 32.
GRAHAM, T. E., R. A. SAUMURE, AND B. ERICSON. 1997. Map turtlewinter leech loads. Journal of Parasitology 83: 1185–1186.
HOVINGH, P., AND A. LINKER. 1999. Hyaluronidase activity in leeches(Hirudinea). Comparative Biochemistry and Physiology, Series BBiochemistry and Molecular Biology 124: 319–326.
HUTCHISON, H., AND F. L. FRYE. 2001. Evidence of pathology in earlyCenozoic turtles. Paleobios 21: 12–19.
KLEMM, D. J. 1982. The leeches (Annelida: Hirudinea) of North Amer-ica. Aquatic Biology Section, Environmental Monitoring and Sup-port Laboratory, Office of Research and Development, U.S. Envi-ronmental Protection Agency, Cincinnati, Ohio, 177 p.
KOFFLER, B. R., R. A. SEIGEL, AND M. T. MENDONCA. 1978. The seasonaloccurrence of leeches on the wood turtle, Clemmys insculpta (Rep-tilia, Testudines, Emydidae). Journal of Herpetology 12: 571–572.
KRAWCHUCK, M. A., N. KOPER, AND R. J. BROOKS. 1997. Observationsof a possible cleaning symbiosis between painted turtles, Chryse-mys picta, and snapping turtles, Chelydra serpentina, in CentralOntario. Canadian Field-Naturalist 111: 315–317.
MACCULLOCH, R. D. 1981. Leech parasitism on the western paintedturtle, Chrysemys picta belli, in Saskatchewan. Journal of Parasi-tology 67: 128–129.
MALINCONICO, S. M., J. B. KATZ, AND A. Z. BUDZYNSKI. 1984. Hemen-tin: Anticoagulant protease from the salivary gland of the leechHaementeria ghilianii. Journal of Laboratory and Clinical Medi-cine 103: 44–58.
MAZUR, P., W. J. HENZEL, J. L. SEYMOUR, AND R. A. LAZARUS. 1991.Ornatins: Potent glycoprotein IIb-IIIa antagonists and platelet ag-gregation inhibitors from the leech Placobdella ornata. EuropeanJournal of Biochemistry 202: 1073–1082.
MCLANE, M. A., J. GABBETA, A. K. RAO, L. BEVIGLIA, R. A. LAZARUS,AND S. NIEWIAROWSKI. 1995. A comparison of the effect of decorsinand two disintegrins, albolabrin and eristostatin, on platelet func-tion. Thrombosis and Haemostasis 74: 1316–1322.
MOSER, W. E., AND S. S. DESSER. 1995. Morphological, histochemical,and ultrastructural characterization of the salivary glands and pro-boscises of three species of glossiphoniid leeches (Hirudinea:Rhynchobdellida). Journal of Morphology 225: 1–18.
SAUMURE, R. A., AND P. J. LIVINGSTON. 1994. Graptemys geographica(common map turtle). Parasites. Herpetological Review 25: 121.
SIDDALL, M. E. 1991. The life history, ultrastructure, and epizootiologyof Haemogregarina balli (Apicomplexa: Adeleina: Haemogregar-inidae) and phylogenetic implications. M.S. Thesis. University ofToronto, Toronto, Canada, 117 p.
———, AND S. S. DESSER. 1990. Gametogenesis and sporogonic de-velopment of Haemogregarina balli (Apicomplexa: Adeleina: Hae-mogregarinidae) in the leech Placobdella ornata. Journal of Pro-tozoology 37: 511–520.
———, AND ———. 1991. Merogonic development of Haemogrega-rina balli (Apicomplexa: Adeleina: Haemogregarinidae) in theleech Placobdella ornata (Glossiphoniidae), its transmission to achelonian intermediate host and phylogenetic implications. Journalof Parasitology 77: 426–436.
1188 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
———, AND ———. 1992. Prevalence and intensity of Haemogrega-rina balli (Apicomplexa: Adeleina: Haemogregarinidae) in threeturtle species from Ontario, with observations on intraerythrocyticdevelopment. Canadian Journal of Zoology 70: 123–128.
VOGT, R. C. 1979. Cleaning/feeding symbiosis between grackles (Quis-calus: Icteridae) and map turtles (Graptemys: Emydidae). Auk 96:608–609.
YANG, T., AND R. W. DAVIES. 1985. The morphology of Placobdellamultilineata (Hirudinoidea: Glossiphoniidae), a parasite of Crocod-ilia. Canadian Journal of Zoology 63: 550–551.
ZANGERL, R. 1969. The turtle shell. In Biology of the Reptilia, vol. 1,C. Gans, A. d’A. Bellairs, and T. Parsons (eds.). Academic Press,New York, p. 311–339.
J. Parasitol., 90(5), 2004, pp. 1188–1190q American Society of Parasitologists 2004
Trypanosoma brucei: Unexpected Azide Sensitivity of Bloodstream Forms
Dietmar Steverding*† and Stefan Scory†‡, *School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG, U.K.;†Abteilung Parasitologie, Hygiene-Institut der Ruprecht-Karls Universitat, Im Neuenheimer Feld 324, 69120 Heidelberg, Germany; ‡Presentaddress: BD Biosciences, Tullastrasse 8-12, 69126 Heidelberg, Germany. e-mail: [email protected]
FIGURE 1. Effect of azide on the growth of bloodstream forms ofTrypanosoma brucei. Culture-adapted TC221 trypanosomes (Hirumi etal., 1980) were incubated at an initial cell density of 104 parasites permilliliter with varying concentrations of sodium azide (closed squares)or with 1 mM sodium cyanide (open square) in Baltz medium supple-mented with 16.7% heat-inactivated fetal bovine serum (Baltz et al.,1985). After 48 hr of culture at 37 C in a humidified atmosphere con-taining 5% CO2, living cells were counted using a hemocytometer. Theexperiment was repeated 3 times, and mean values 6 SD are shown.The numbers of cells (%) refer to the control in the absence of inhibitor.
ABSTRACT: Bloodstream forms of Trypanosoma brucei lack cyto-chromes and are, therefore, insensitive to cyanide. Azide is a toxic anionthat bears chemical and biological properties in common with cyanideand may act in a similar way by inhibition of cytochrome c oxidase. Itwas, therefore, surprising to find that bloodstream forms of T. bruceiare sensitive to azide; growth is reduced by 50% with 0.1 mM azide.So far, the only enzyme known in bloodstream forms of T. brucei tobe sensitive to azide is the iron-containing superoxide dismutase. How-ever, because the activity of the superoxide dismutase was not affectedin parasites incubated for 16 hr with 0.5 mM azide (a concentration atwhich no cell proliferates), the toxic action of azide cannot be due toinhibition of this enzyme. These results indicate that the general toxicityof azide is different from that of cyanide.
Bloodstream forms of Trypanosoma brucei have a poorly developedmitochondrion lacking a functional Krebs cycle and cytochromes (Bow-man and Flynn, 1976). As a result, the mammalian life cycle stage ofthe parasite relies entirely on glycolysis for energy production. In try-panosomes, the first 7 glycolytic enzymes are compartmentalized in aspecial organelle, the glycosome (Opperdoes, 1987; Clayton and Mich-els, 1996). Within the glycosome, reduced form of nicotinamide adeninedinucleotide is reoxidized during the reduction of dihydroxyacetonephosphate to glycerol-3-phosphate. Glycerol-3-phosphate is reconvertedto dihydroxyacetone phosphate by a mitochondrial alternative oxidasethat is insensitive to cyanide (Clarkson et al., 1989; Fukai et al., 2003).For these reasons, bloodstream forms of T. brucei are totally insensitiveto cyanide (Bowman and Flynn, 1976).
Azide is a toxic agent exhibiting some chemical properties and bio-logical effects common to cyanide. Thus, the toxic effect of azide maybe like that of cyanide, i.e., due to inhibition of cytochrome c oxidase(Wilson and Erecinska, 1978; Bennett et al., 1996). Because blood-stream forms of T. brucei lack cytochrome c oxidase, they should beinsensitive to azide. Surprisingly, incubation of T. brucei bloodstreamforms with sodium azide inhibited the growth of the parasites in a dose-dependent manner, with a 50% effective dose (ED50) value of 0.1 mM(Fig. 1). The lowest concentration of sodium azide at which no cellproliferates was 0.25–0.5 mM. These toxicity levels of sodium azidefor bloodstream forms of T. brucei are in the range of those for othereukaryotes, e.g., human EUE fibroblast: ED50 after 48 hr 5 0.24 mM;cytostatic concentration after 48 hr 5 1.5 mM (Slamenova and Gabe-lova, 1980). In contrast, 1 mM sodium cyanide did not affect the growthof the cells at all (Fig. 1).
A possible target for the toxic action of azide is the parasite’s iron-containing superoxide dismutase, the only enzyme in bloodstream formsof T. brucei that is sensitive to azide (Le Trant et al., 1983; Kabiri andSteverding, 2001). However, the concentration necessary to inhibit theactivity of the enzyme in crude cell extracts by 50% is 5 mM (Le Trantet al., 1983) and thus 50 times higher than the ED50 value of 0.1 mMfor the growth inhibition of the bloodstream-form trypanosomes. Onthe other hand, because the superoxide dismutase has a long half-life(Breidbach et al., 2002), exposure of T. brucei bloodstream forms toazide for any length of time may result in a slow, but progressive,
inhibition of the enzyme. Therefore, the activity of superoxide dismu-tase was determined indirectly by the inhibition of pyrogallol autoxi-dation (S. Marklund and G. Marklund, 1974) in extracts of T. bruceibloodstream forms that have been incubated for 16 hr with, or without,0.5 mM sodium azide, the concentration at which the proliferation ofthe cells was almost abolished. No difference in activity of superoxidedismutase was observed in extracts of bloodstream-form trypanosomestreated with and without sodium azide. Both extracts prevented the au-toxidation of pyrogallol to the same extent (Fig. 2). Likewise, deter-mining the activity of superoxide dismutase by a negative staining pro-cedure after native polyacrylamide gel electrophoresis (Beauchamp andFridovich, 1971; Breidbach et al., 2002) gave the same result; the in-tensities of the achromatic superoxide dismutase bands of extracts fromT. brucei bloodstream forms treated with and without azide were indis-tinguishable (data not shown). From these results, it can be excluded
RESEARCH NOTES 1189
FIGURE 2. Effect of azide on the activity of superoxide dismutasein bloodstream forms of Trypanosoma brucei. Culture-adapted TC221trypanosomes (Hirumi et al., 1980) were grown in Baltz medium sup-plemented with 16.7% heat-inactivated fetal bovine serum (Baltz et al.,1985) in presence or absence of 0.5 mM azide. After 16 hr, the cellswere harvested and lysed in 5 mM Tris, pH 7.8, 0.1 mM ethylenedi-aminetetraacetic acid, and 0.4 mM PMSF (Breidbach et al., 2002). Thesuspension was centrifuged, and the activity of superoxide dismutase inthe supernatant was determined indirectly by the inhibition of pyrogallolautoxidation (S. Marklund and G. Marklund, 1974). Aliquots corre-sponding to 1.1 3 108 cell equivalents were added to 0.2 mM pyrogallolin 50 mM Tris, pH 8.0, 1 mM DTPA, and the increase in absorbanceat 420 nm was followed photometrically. Extracts from both azide-treated (closed squares) and control (closed triangles) trypanosomes pre-vented the autoxidation of pyrogallol, whereas it was not inhibited inthe absence of cell extract (open circles). In addition, the presence of0.5 mM hydrogen peroxide (open diamonds), an inhibitor of iron-con-taining superoxide dismutase (S. Marklund and G. Marklund, 1974;Kabiri and Steverding, 2001), in the buffer did not prevent the autoxi-dation of pyrogallol, demonstrating that the inhibition of the autoxida-tion of pyrogallol by trypanosome cell extracts was specifically due tosuperoxide dismutase. A representative result from 2 independent ex-periments is shown.
that the toxic effect of azide to bloodstream forms of T. brucei is dueto inhibition of the iron-containing superoxide dismutase.
Azide can be oxidized by catalase in the presence of hydrogen per-oxide, resulting in the generation of nitric oxide (Keilin and Hartree,1954; Nicholls, 1964), which has been shown to exert an antimicrobialeffect on bloodstream forms of T. brucei (Vincendeau et al., 1992).However, such a mechanism for the toxic action of azide can also beexcluded because trypanosomes lack catalase (Fulton and Spooner,1956; Boveris et al., 1980).
Recently, it has been demonstrated that azide catalytically inactivatestopoisomerase II (Ju et al., 2001). Because T. brucei has a topoisom-erase II encoded by a single-copy gene (Strauss and Wang, 1990), theantitrypanosomal activity of azide may be due to the inactivation ofthis enzyme. However, topoisomerase II is inactivated by 20 mM so-dium azide (Ju et al., 2001), a concentration 40 times higher than thelowest concentration of azide (0.5 mM) at which no bloodstream formproliferates. In addition, using inducible RNA interference it was shownthat the growth of topoisomerase II RNAi-induced T. brucei blood-stream forms was only slightly reduced over a period of 11 days com-pared with uninduced cells (Timms et al., 2002). Thus, it appears thattopoisomerase II is not essential for bloodstream forms of T. brucei,and therefore, that the trypanocidal activity of azide cannot be attributedto inactivation of this enzyme.
The real mechanism of the toxic action of sodium azide is still puz-
zling. Although azide has the ability to inhibit cytochrome c oxidase,it is unlikely that this accounts for its toxicity (Smith and Wilcox, 1994).For instance, treatment of rats by subcutaneous infusion of 1 mg kg21
hr21 sodium azide induced only partial inhibition of cytochrome c ox-idase after 7 days (Bennett et al., 1996). Azide-induced death in labo-ratory animals and in humans appears to be the result of nonasphyxialconvulsions and cardiovascular collapse, respectively (Smith and Wil-cox, 1994). On the other hand, there is no doubt that azide can beconverted in vivo and in isolated tissue and cells into nitric oxide, whichexplains its vasodilatory activity (Kaplita et al., 1984; Shahidullah etal., 2002). Because only small amounts of azide are oxidized to nitricoxide by the action of catalase (Keilin and Hartree, 1954), it is unclearwhether the toxic effect of azide is due to nitric oxide or to the parentanion. The finding of the present study, that bloodstream forms of T.brucei are unexpectedly sensitive to azide, may suggest that indeed theanion itself exerts the toxic effect. Because the mammalian life cyclestage of T. brucei lacks both cytochrome c oxidase and catalase, blood-stream-form trypanosomes may be an ideal model for studying themechanism of the toxic action of sodium azide, a chemical of rapidlygrowing commercial importance (Smith and Wilcox, 1994).
The authors thank Mark Viney for critical reading of the manuscript.D.S. was supported by The Wellcome Trust (Grant Ref. 064436).
LITERATURE CITED
BALTZ, T., D. BALTZ, C. GIROUD, AND J. CROCKETT. 1985. Cultivationin a semi-defined medium of animal infective forms of Trypano-soma brucei, T. equiperdum, T. evansi, T. rhodesiense and T. gam-biense. EMBO Journal 4: 1273–1277.
BEAUCHAMP, C., AND I. FRIDOVICH. 1971. Superoxide dismutase: Im-proved assays and an assay applicable to acrylamide gels. Analyt-ical Biochemistry 44: 276–287.
BENNETT, M. C., G. W. MLADY, Y.-H. KWON, AND G. M. ROSE. 1996.Chronic in vivo sodium azide infusion induces selective and stableinhibition of cytochrome c oxidase. Journal of Neurochemistry 66:2606–2611.
BOVERIS, A., H. SIES, E. E. MARTINO, R. DOCAMPO, J. F. TURRENS, AND
A. O. STOPPANI. 1980. Deficient metabolic utilization of hydrogenperoxide in Trypanosoma cruzi. Biochemical Journal 188: 643–648.
BOWMAN, J. B. R., AND I. W. FLYNN. 1976. Oxidative metabolism oftrypanosomes. In Biology of the Kinetoplastida, vol. 1, W. H. R.Lumsden and D. A. Evans (eds.). Academic Press, London, U.K.,p. 435–476.
BREIDBACH, T., S. SCORY, R. L. KRAUTH-SIEGEL, AND D. STEVERDING.2002. Growth inhibition of bloodstream forms of Trypanosomabrucei by the iron chelator deferoxamine. International Journal forParasitology 32: 473–479.
CLARKSON, A. B., JR., E. J. Bienen, G. Pollakis, and R. W. Grady. 1989.Respiration of bloodstream forms of the parasite Trypanosoma bru-cei brucei is dependent on a plant-like alternative oxidase. Journalof Biological Chemistry 264: 17770–17776.
CLAYTON, C. E., AND P. MICHELS. 1996. Metabolic compartmentation inAfrican trypanosomes. Parasitology Today 12: 465–471.
FUKAI, Y., C. NIHEI, K. KAWAI, Y. YABU, T. SUZUKI, N. OHTA, N. MIN-AGAWA, K. NAGAI, AND K. KITA. 2003. Overproduction of highlyactive trypanosome alternative oxidase in Escherichia coli heme-deficient mutant. Parasitology International 52: 237–241.
FULTON, J. D., AND D. F. SPOONER. 1956. Inhibition of the respirationof Trypanosoma rhodesiense by thiols. Biochemical Journal 63:475–481.
HIRUMI, H., K. HIRUMI, J. J. DOYLE, AND G. A. M. CROSS. 1980. In vitrocloning of animal-infective bloodstream forms of Trypanosomabrucei. Parasitology 80: 371–382.
JU, R., Y. MAO, M. J. GLICK, M. T. MULLER, AND R. D. SNYDER. 2001.Catalytic inhibition of DNA topoisomerase IIa by sodium azide.Toxicological Letters 121: 119–126.
KABIRI, M., AND D. STEVERDING. 2001. Identification of a developmen-tally regulated iron superoxide dismutase of Trypanosoma brucei.Biochemical Journal 360: 173–177.
KAPLITA, P. V., H. L. BORISON, L. E. MCCARTHY, AND R. P. SMITH. 1984.Peripheral and central actions of sodium azide on circulatory and
1190 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
KEILIN, D., AND E. F. HARTREE. 1954. Reactions of methaemoglobin andcatalase with peroxides and hydrogen donors. Nature 173: 720–723.
LE TRANT, N., S. R. MESHNICK, K. KITCHENER, J. W. EATON, AND A.CERAMI. 1983. Iron-containing superoxide dismutase from Crithid-ia fasciculata. Purification, characterization, and similarity to leish-manial and trypanosomal enzymes. Journal of Biological Chemis-try 258: 123–130.
MARKLUND, S., AND G. MARKLUND. 1974. Involvement of the superoxideanion radical in the autoxidation of pyrogallol and a convenientassay for superoxide dismutase. European Journal of Biochemistry47: 469–474.
NICHOLLS, P. 1964. The reactions of azide with catalase and their sig-nificance. Biochemical Journal 90: 331–343.
OPPERDOES, F. R. 1987. Compartmentation of carbohydrate metabolismin trypanosomes. Annual Review of Microbiology 41: 127–151.
SHAHIDULLAH, M., A. DUNCAN, P. D. STRACHAN, K. M. RAFIQUE, S. L.BALL, M. J. W. MCPATE, S. NELLI, AND W. MARTIN. 2002. Role ofcatalase in the smooth muscle relaxant actions of sodium azide andcyanamide. European Journal of Pharmacology 435: 93–101.
SLAMENOVA, D., AND A. GABELOVA. 1980. The effects of sodium azideon mammalian cells cultivated in vitro. Mutation Research 71:253–261.
SMITH, R. P., AND D. E. WILCOX. 1994. Toxicology of selected nitricoxide-donating xenobiotics, with particular reference to azide. Crit-ical Reviews in Toxicology 24: 355–377.
STRAUSS, P. R., AND J. C. WANG. 1990. The TOP2 gene of Trypanosomabrucei: A single copy gene that shares extensive homology withother TOP2 genes encoding eukaryotic DNA topoisomerase II. Mo-lecular and Biochemical Parasitology 38: 141–150.
TIMMS, M. W., F. J. VAN DEURSEN, E. F. HENDRIKS, AND K. R. MATTHEWS.2002. Mitochondrial development during life cycle differentiationof African trypanosomes: Evidence for a kinetoplast-dependent dif-ferentiation control point. Molecular Biology of the Cell 13: 3747–3759.
VINCENDEAU, P., S. DAULOUEDE, B. VEYRET, M. L. DARDE, B. BOUTEILLE,AND J. L. LEMESRE. 1992. Nitric oxide-mediated cytostatic activityon Trypanosoma brucei gambiense and Trypanosoma brucei bru-cei. Experimental Parasitology 75: 353–360.
WILSON, D. F., AND M. ERECINSKA. 1978. Ligands of cytochrome c ox-idase. Methods in Enzymology 53: 191–201.
J. Parasitol., 90(5), 2004, pp. 1190–1193q American Society of Parasitologists 2004
Identification of the Copepod Intermediate Host of the Introduced Broad Fish Tapeworm,Diphyllobothrium latum, in Southern Chile
P. Torres, L. Villalobos*, S. Woelfl*, and S. Puga, Instituto de Parasitologıa, Facultad de Medicina, Campus Isla Teja, Universidad Austral deChile, Valdivia, Chile; *Instituto de Zoologıa, Facultad de Ciencias, Campus Isla Teja, Universidad Austral de Chile, Valdivia, Chile. e-mail:[email protected]
TABLE I. Prevalence and intensity of experimental infection with pro-cercoids of Diphyllobothrium latum in Diaptomus diabolicus at 20 6 1C in different sampling periods.
Samplingperiods(days) Prevalence (%)
Mean intensity(range)
2369
10
6/11 (55)6/8 (75)2/2 (100)3/3 (100)3/3 (100)
3.2 (1–6)2 (1–4)1.5 (1–2)2 (1–3)2 (2)
1316202331
7/9 (78)5/5 (100)2/2 (100)5/6 (83)2/7 (29)
3.3 (2–6)2.9 (1–6)4.0 (3–5)3.0 (1–6)4.0 (1–7)
Total 41/56 (73) 2.8 (1–7)
ABSTRACT: The broad fish tapeworm, Diphyllobothrium latum, is anexotic species in both Chile and Argentina, and until now, its copepodhost has remained unknown in South American waters. The objectiveof this study was to identify calanoid copepod species that may beintermediate hosts for D. latum in Lake Panguipulli, Chile. In this lake,the highest levels of infection by this tapeworm occur in the introducedrainbow trout, Oncorhynchus mykiss. Of the 2 calanoid copepods foundin Lake Panguipulli, Diaptomus diabolicus and Boeckella gracilipes,only D. diabolicus became infected on experimental exposure to cora-cidia. Prevalence (mean intensity) of experimental infection in adultcopepods was 73.2% (2.8 procercoids per host). Diaptomus diabolicushas been demonstrated to be a new intermediate host; this is the firstrecord of a copepod host for D. latum in South America.
Human infections produced by the broad fish tapeworm, Diphyllo-bothrium latum, have occurred historically throughout Europe, althoughin recent times the infection occurs mainly in the Baltic countries andScandinavia (Marty and Neafie, 2000). A recent resurgence has beendocumented in Italy, Switzerland, and France (Golay and Mariaux,1995; Peduzzi and Boucher-Rodoni, 2001).
Diphyllobothrium latum has been introduced to North America(Grove, 1990) and to South America in Chile (Neghme et al., 1950)and Argentina (Revenga, 1993). The prevalence of this tapeworm variesbetween 0.2 and 3.4% in human lakeshore populations in southern Chile(Neghme and Bertin, 1951; Torres, Franjola et al., 1989; Torres et al.,1991, 1998). The principal source of the infection is the consumptionof insufficiently cooked, raw (as cebiche, sushi, and sashimi), andsmoked fish. The latter includes rainbow trout, Oncorhynchus mykiss;brown trout, Salmo trutta; and native fishes such as perch, Percichthystrucha; smelt, Basilichthys australis; cauque, Odonthestes (Cauque)mauleanum; and puyes, Galaxias maculatus (Torres et al., 1998).
Diphyllobothrium latum infections in fishes have been described be-tween 398S and 418S (Torres et al., 1991, 1998) in the north Patagonianlake district (Thomasson, 1963). The highest prevalence and intensityof infection by D. latum was in rainbow trout collected from Lake
Panguipulli (398419S, 728139W) by Torres et al. (1991). The adult stageof D. latum develops in humans, dogs, and cats; their eggs being elim-inated in the feces of the host. The elimination of raw sewage into thelakes and rivers spreads the eggs in the aquatic ecosystems in southernChile. In the water, these eggs develop into coracidia. In the NorthernHemisphere, coracidia of D. latum are ingested by copepod microcrus-taceans (first intermediate hosts) and form procercoid larvae. Once aplanktivorous fish ingests an infected copepod, the procercoids developinto plerocercoids (Von Bonsdorff and Bylund, 1982).
Copepod species in Chile that act as first intermediate hosts for D.latum are not known. Von Bonsdorff (1977) listed 17 species of cal-anoid copepods as hosts for D. latum, including 7 species of Diaptomus
RESEARCH NOTES 1191
FIGURES 1–5. Experimental development of Diphyllobothrium latum in the copepod Diaptomus diabolicus at 20 C. 1. Initial stage in the bodycavity showing their embryonic hooks (arrows) on day 3 after infection. Bar 5 39 mm. 2. Undeveloped procercoid showing their embryonichooks (arrow) on day 9 after infection. Bar 5 25 mm. 3. Developed procercoid showing calcareous corpuscles (arrow) and cercomer (head ofarrow) on day 20 after infection. Bar 5 147 mm. 4. Undeveloped procercoid (arrows) with embryonic hooks (arrowhead), anterior to the eye(broad arrow) of copepod on day 10 after infection. Bar 5 25 mm. 5. Developed procercoids (arrows) in dorsal and ventral regions of body cavityon day 23 after infection. Bar 5 126 mm.
(in the Northern Hemisphere) and 1 of Boeckella (in Australia), whichwere highly susceptible to infection. Of 14 species of cyclopoid cope-pods, only 3 species of Cyclops were highly susceptible. Cyclops spp.are not present in Chile. Seven species of Cyclopoida and 3 species ofCalanoida copepods have been recorded in the north Patagonian district(Araya and Zuniga, 1985; Villalobos and Zuniga, 1991; Villalobos,1994). We hypothesized that 2 species of calanoid copepods observedin Lake Panguipulli, Diaptomus diabolicus and Boeckella gracilipes(Soto and Zuniga, 1991), might be the first intermediate hosts of D.latum because of their potentially high susceptibility as suggested byVon Bonsdorff (1977). This study provides the first results on the sus-ceptibility, prevalence, and intensity of experimental infection in thesecopepods.
The experimental protocol included infections of dogs with plerocer-coids of D. latum, culture of eggs for the development of coracidia, andexperimental exposures of the copepods to coracidia.
In March 2002, 21 rainbow trout were collected by gillnetting in Lake
Panguipulli, placed whole on ice, and transported to the laboratory with-in 12 hr, where their viscera and musculature were examined for ple-rocercoids. Identification of the plerocercoids was performed accordingto Andersen et al. (1987), P. Torres, J. Torres et al. (1989), and Revengaet al. (1995). Fifteen plerocercoids suspended in 0.9% NaCl were ad-ministered orally to each of 2 mixed-breed dogs aged 2 and 5 yr. Thedogs were housed and maintained individually according to the guide-lines of the Small Animal Clinic of the Veterinary Medicine Faculty,Universidad Austral de Chile. They received professional care duringthe study; they were fed standard dry dog chow and water ad libitum.Before administration of the plerocercoids, both dogs were tested fornatural infections of Diphyllobothrium spp. by fecal exam using phe-nol–alcohol–formalin sedimentation (PAFS) method (Burrows, 1967);natural infections of Toxocara canis and Dipylidium caninum weretreated with 50 mg praziquantel, 5 mg pamoate of pirantel, and 15 mgof Febantel (Galgovett, Laboratorio, Santiago, Chile) per kilogram ofbody weight. Fecal examinations were done 5 days after treatment to
1192 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
verify elimination of natural infections. After each dog was fed the 15plerocercoids, their feces were examined daily using PAFS.
Feces recovered between 20 and 60 days after the initiation of thepatent period (21–24 days) were homogenized in artificial pond water(APW) (Ulmer, 1970) and passed through 5 brass screens of consecutivemesh sizes (120, 90, 75, 53, and 38 mm) for separation and suspensionof D. latum eggs in APW at 4 6 1 C. Eggs were incubated for 20–30days at 20 6 1 C in aerated, distilled water, pH 7.2, in 1-L Erlenmeyerflasks wrapped in black plastic to exclude light. The dogs were killed3 mo after infection; 2 and 5 adults of D. latum were recovered fromthem.
Copepods were collected from Lake Panguipulli by making verticalhauls in the pelagic zone of the lake, using a Ruttner nett (Hydrobios,Kiel, Germany) with a mouth opening of 10.4-cm diameter and a 90-mm mesh size. Specimens of D. diabolicus and B. gracilipes were sep-arated using a stereoscopic microscope and deposited into 1-L Erlen-meyer flasks containing filtered and aerated Panguipulli Lake water(LW) at 20 6 1 C with a 12:12–hr light–dark period. The water waschanged every 3 days. The copepods were fed daily with a mixture ofchow pellets, dried yeast, and alfalfa (Anonymous, 1992). For infectionof the adult copepods, 100 individuals of each species, including bothsexes, were taken from their holding flasks and suspended separatelyby species in 20 ml LW in 10-cm-diameter petri dishes at 20 6 1 C.Coracidia were obtained from their incubation flasks by manually de-canting off about two-thirds of the supernatant water and centrifugingthe remainder of the flask contents at 1,000 rpm for 1 min, followedby decanting half this supernatant and resuspending the sedimented pel-let in the remaining supernatant water. About 5,000 coracidia were de-posited into the petri dishes containing the copepods for 4 hr, afterwhich the copepods of each species were separately trapped on a 100-mm-mesh screen and deposited into 1-L of LW in Erlenmeyer flasksand maintained in the same conditions as described above.
Samples of copepods (2–11 individuals) of each species were mea-sured (head to furcal rami) with an ocular micrometer and sexed at 2,3, 6, 9, 10, 13, 16, 20, 23, and 31 days after infection on slides undercover glasses. The samples did not include the copepods that died dur-ing the experiment. The parasites were removed from copepods by gent-ly pressing the cover glass; their development and presence of cercomerand calcareous corpuscles were determined.
Only D. diabolicus was susceptible to experimental infection by D.latum (Table I). Of the 100 specimens of D. diabolicus maintained at20 6 1 C, 44 copepods died throughout the experiment; these copepodswere not examined for procercoids. Living copepods (35 females and21 males) ranged from 0.9 to 1.8 mm in length. Prevalence and meanintensity of infection in different sampling periods ranged from 29 to100% (with mean of 73%) and 1.5 to 4.0 procercoids per host (withmean of 2.8 procercoids per host), respectively (Table I). The extent ofdevelopment of D. latum that occurred in the body cavity of D. dia-bolicus varied (Figs. 1–3). In this study, cercomers and calcareous cor-puscles appeared, beginning at 13 days of infection. Four (7%) cope-pods showed procercoids in the ocular region, in front of the eyes (Fig.4). The majority of procercoids were in the dorsal region, i.e., the ceph-alothorax, in the 73% of copepods. In heavier infections, the ventralregion of the body also harbored procercoids, as in 20% of the copepods(Fig. 5). Of the 100 specimens of B. gracilipes maintained at 20 6 1C, 33 died during the experiment and none of the 67 individuals ex-amined (52 females and 15 males, ranged from 0.7 to 1.1 mm in length)was infected. Twenty adult individuals of each copepod species wereexamined before experimental use, and none of them was found to beinfected.
The high prevalence and intensity of experimental infection by D.latum in D. diabolicus, in general, agrees with results for other speciesof Calanoida in other countries, i.e., Diaptomus sanguineus, Diaptomuspiscinae, and Diaptomus mississippiensis in the United States (Humes,1950); Boeckella minuta in Australia (Bearup, 1957); Eudiaptomusgracilis in Switzerland (Michajlow, 1963); Eudiaptomus coeruleus, Eu-diaptomus gracilis, and Eudiaptomus zachariasi in Finland; and E.gracilis and Acanthodiaptomus denticornis in Norway (Guttowa, 1961).In these studies, prevalences ranged from 15 to 100% with 1–22 pro-cercoids per host. The occurrence of procercoids in front of the eyesmay alter the swimming behavior of infected copepods and make themmore susceptible to predation by fish intermediate hosts.
Diaptomus diabolicus is a new experimental intermediate host for D.
latum, and this article records the first copepod host for Diphylloboth-rium sp. in South America. Diaptomus diabolicus is herbivorous, en-demic to Chile, and has been reported from lakes Panguipulli, Rinihue,Ranco, Rupanco, and Puyehue (Araya and Zuniga, 1985), where D.latum has also been reported in fishes (Neghme and Bertin, 1951; Wet-zlar, 1979; Torres et al., 1991, 1998). This copepod has not been re-covered from lakes Caburgua, Llanquihue, and Todos los Santos (Arayaand Zuniga, 1985), from which D. latum also has not been reported(Torres et al., 1983, 1991). However, this copepod has been found inLake Huillinco on Chiloe Island, where D. latum has not been foundin preliminary surveys (Torres et al., 1990). Lake Huillinco could be-come another Chilean lake where D. latum could be maintained. Thisis because of the physicochemical conditions of the water, the presenceof susceptible fish to D. latum, and the consumption of these fishes byhumans or domestic animals (Torres et al., 1990). Recently, infectionby Diphyllobothrium sp. was recorded in a rainbow trout from a marineculture center from southern Chile (Torres et al., 2002).
Although this experiment suggested that adult B. gracilipes is notsusceptible to infection by D. latum, it is still necessary to assay co-pepodids of this species because susceptibility to infection in the North-ern Hemisphere was shown to depend on the age of some host species(Guttowa, 1961). Future directions for this work will be collecting D.diabolicus naturally infected with D. latum, examining other copepodspecies to clarify that D. diabolicus is the most likely intermediate hostfor the D. latum in Chilean lakes, demonstrating interspecific differenc-es in susceptibility, and determining whether the swimming behavior ofthe infected copepod is altered.
We thank Cesar Cuevas, Jose Nunez, and Mariano Grandjean for theirtechnical assistance in the field and Louis Di Salvo for English revisionof the manuscript. This work was supported by grant 200224 from theDireccion de Investigacion y Desarrollo de la Universidad Austral deChile.
LITERATURE CITED
ANDERSEN, K., H. CHING, AND R. VIK. 1987. A review of freshwaterspecies of Diphyllobothrium with redescriptions and distribution ofD. dendriticum (Nitzsch, 1824) and D. ditremum (Creplin, 1825)from North America. Canadian Journal of Zoology 65: 2216–2228.
ANONYMOUS. 1992. Standard methods for the examination of waste andwastewater, 18th ed. American Public Health Association, Wash-ington, D.C., 1541 p.
ARAYA, J. M., AND L. ZUNIGA. 1985. Manual taxonomico del zooplanc-ton lacustre de Chile. Boletın Informativo Limnologico 8: 1–110.
BEARUP, A. J. 1957. Experimental vectors of the first larval stage ofDibothriocephalus latus (Cestoda) in Australia. Australian Journalof Experimental Biology 35: 187–192.
BURROWS, R. 1967. A new fixative and technics for the diagnosis ofintestinal parasites. American Journal of Clinical Pathology 48:342–346.
GOLAY, M., AND J. MARIAUX. 1995. Situation de Diphyllobothriumlatum L. 1758 (Cestoda: Pseudophyllidea) dans quatre lacs du pla-teau Suisse. Bulletin de la Societe Neuchateloise des Sciences Na-turelles 118: 79–86.
GROVE, D. I. 1990. A history of human helminthology. C.A.B. Inter-national, Oxon, U.K., 848 p.
GUTTOWA, A. 1961. Experimental investigations on the systems procer-coids of Diphyllobothrium latum (L.)—Copepoda. Acta Parasito-logica Polonica 9: 371–408.
HUMES, A. G. 1950. Experimental copepod hosts of the broad tapewormof man. Dibothriocephalus latus (L.). Journal of Parasitology 36:541–547.
MARTY, A. M., AND R. C. NEAFIE. 2000. Pathology of infection, volume1, helminthiases. In Diphyllobothriasis and sparganosis, W. M.Meyers (ed.). Armed Forces Institute of Pathology, Washington,D.C., p. 165–183.
MICHAJLOW, W. 1963. Results of experimental infecting of copepodafrom Hausersee (Switzerland) with the larvae of Diphyllobothriumlatum (L.) (Cestoda). Bulletin de l’Academie Polonaise des Sci-ences 11: 347–351.
NEGHME, A., AND V. BERTIN. 1951. Diphyllobothrium latum en Chile,IV. Estado actual de las investigaciones epidemiologicas. RevistaChilena de Higiene y Medicina Preventiva 13: 8–11.
RESEARCH NOTES 1193
———, R. DONCKASTER, AND R. SILVA. 1950. Diphyllobothrium latumen Chile. Revista Medica de Chile 78: 410–411.
PEDUZZI, R., AND R. BOUCHER-RODONI. 2001. Resurgence of humanbothriocephalosis (Diphyllobothrium latum) in the subalpine lakeregion. Journal of Limnology 60: 41–44.
REVENGA, J. 1993. Diphyllobothrium dendriticum and Diphyllobothriumlatum in fishes from southern Argentina: Association, abundance,distribution, pathological effects, and risk of human infection. Jour-nal of Parasitology 79: 379–383.
———, C. J. PERFUMO, C. A. UBEDA, AND L. G. SEMENAS. 1995. Difi-lobotriasis en salmonidos introducidos en el Parque y Reserva Na-cional Nahuel Huapi, Argentina: Patologıa de las lesiones produ-cidas por Diphyllobothrium spp. Archivos de Medicina Veterinaria27: 115–122.
SOTO, D., AND L. ZUNIGA. 1991. Zooplancton assemblages of Chileantemperate lakes: A comparison with North American counterparts.Revista Chilena de Historia Natural 64: 569–581.
THOMASSON, K. 1963. Araucarian lakes. Acta Phytogeographica Suecica47: 1–139.
TORRES, P., V. CUBILLOS, W. GESCHE, C. REBOLLEDO, A. MONTEFUSCO,J. C. MIRANDA, J. ARENAS, A. MIRA, M. NILO, AND C. ABELLO.1991. Difilobotriasis en salmonidos introducidos en lagos del surde Chile: Aspectos patologicos, relacion con infeccion humana,animales domesticos y aves piscıvoras. Archivos de Medicina Ve-terinaria 23: 165–183.
———, L. FIGUEROA, AND R. FRANJOLA. 1983. Researches on Pseudo-phyllidea (Carus, 1813) in the south of Chile. IX. Types of plero-cercoids in trouts from five lakes and new cases of Diphylloboth-rium latum in man and D. pacificum in a dog. International Journalfor Zoonoses 10: 15–20.
———, R. FRANJOLA, J. PEREZ, S. AUAD, F. UHEREK, J. C. MIRANDA, L.FLORES, J. RIQUELME, S. SALAZAR, C. HERMOSILLA, AND R. ROJO.1989. Epidemiologıa de la difilobotriasis en la cuenca del rıo Val-divia, Chile. Revista de Saude Publica, Sao Paulo 23: 45–57.
———, W. GESCHE, A. MONTEFUSCO, J. C. MIRANDA, P. DIETZ, AND R.HUIJSE. 1998. Diphyllobothriosis humana y en peces del lago Ri-nihue, Chile: Efecto de la actividad educativa, distribucion esta-
cional y relacion con sexo, talla y dieta de los peces. Archivos deMedicina Veterinaria 30: 31–45.
———, J. C. LOPEZ, V. CUBILLOS, C. LOBOS, AND R. SILVA. 2002. Vis-ceral diphyllobothriosis in a cultured rainbow trout, Oncorhynchusmykiss (Walbaum), in Chile. Journal of Fish Diseases 25: 375–379.
———, E. RUIZ, C. REBOLLEDO, A. MIRA, V. CUBILLOS, N. NAVARRETE,W. GESCHE, A. MONTEFUSCO, L. VALDES, AND A. ALBERDI. 1990.Parasitismo en peces y comunidades humanas riberenas de los la-gos Huillinco y Natri (Isla de Chiloe), Chile. Boletın Chileno deParasitologıa 45: 47–55.
———, J. TORRES, O. GARRIDO, AND J. THIBAUT. 1989. Investigacionessobre Pseudophyllidea (Carus, 1813) en el sur de Chile. X. Obser-vaciones experimentales sobre la coexistencia de plerocercoides deDiphyllobothrium latum (L.) y D. dendriticum (Nitzsch) en sal-monidos de la cuenca del rıo Valdivia. Archivos de Medicina Ve-terinaria 21: 51–57.
ULMER, M. J. 1970. Notes on rearing of snails in the laboratory. InExperiments and techniques in parasitology, A. J. MacInnis and M.Voge (eds.). W. W. Freeman and Company, San Francisco, Cali-fornia, p. 143–144.
VILLALOBOS, L. 1994. Zooplancton of Andine temperate lakes in South-America with special emphasis on the Daphnia species: Taxonomy,geographical distribution, ecology, and functional morphology ofthe filtering apparatus. Ph.D. Dissertation. University of Konstanz,Konstanz, Germany, 197 p.
———, AND L. ZUNIGA. 1991. Latitudinal gradient and morphologicalvariability of copepods in Chile: Boeckella gracilipes Daday. Ver-handlungen Internationale Vereinigung Limnnologie 24: 2834–2838.
VON BONSDORFF, B. 1977. Diphyllobothriasis in man. Academic Press,New York, New York, 189 p.
———, AND G. BYLUND. 1982. The ecology of Diphyllobothrium latum.Ecology Diseases 1: 21–26.
WETZLAR, H. 1979. Beitrage zur biologie und Bewirtschaftung von For-ellen (Salmo gairdneri und S. trutta) in Chile. Ph.D. Dissertation.Albert Ludwigs Universitat, Freiburg, Germany, 264 p.
J. Parasitol., 90(5), 2004, pp. 1193–1196q American Society of Parasitologists 2004
Alteration of Antibodies Against the Fifth-Stage Larvae and Changes in Brain MagneticResonance Images in Experimentally Infected Rabbits with Angiostrongylus cantonensis
Lian-Chen Wang and Yung-Liang Wan*†, Department of Parasitology, School of Medicine, Chang-Gung University, Kueisan, Taoyuan, Taiwan(333); *Department of Diagnostic Radiology, Chang-Gung Memorial Hospital at Linkou, School of Medicine and Medical Technology, ChangGung University, Kueisan, Taoyuan, Taiwan (333); †To whom correspondence should be addressed. e-mail: [email protected]
ABSTRACT: Magnetic resonance (MR) imaging has been suggested tobe helpful in delineating the lesions during the acute phase of angios-trongyliasis caused by Angiostrongylus cantonensis. In this study, an-tibody titers in serum samples of 3 rabbits were determined by enzyme-linked immunosorbent assay, and brain MR images were obtained from6 rabbits. The antibody titer elevated rapidly in the first 4 wk postin-fection (PI) before reaching a plateau. However, suspicious changes inbrain MR images near the left lateral ventricle and hippocampus werefound only in 1 rabbit on day 28 PI. These findings indicate that im-munologic responses in the central nervous system at the early stage ofangiostrongyliasis are not sufficient to be observed by image studies.
Angiostrongylus cantonensis is a parasitic nematode of wild rats. Af-ter acquiring the infection through the ingestion of infected snails, i.e.,Achatina spp. and Pila spp., slugs, planaria, or freshwater prawns, thethird-stage larvae reach the central nervous system before migrating tothe lung of the definitive host to complete their life cycles (Mackerrasand Sandars, 1955; Bhaibulaya, 1975). Humans may also become in-fected by accidental ingestion of the third-stage larvae. Although these
infective larvae develop into a fifth stage, most larvae remain in thebrain and cause eosinophilic meningitis or eosinophilic meningoen-cephalitis (Alicata, 1965).
History taking and clinical observations may be helpful in the diag-nosis of angiostrongyliasis caused by A. cantonensis. However, moreconclusive diagnosis is based on immunologic findings (Weller, 1993).Recently, magnetic resonance (MR) imaging has been suggested to bea favorable noninvasive technique in delineating the lesions during theacute phase (Clouston et al., 1990). The coexistence of immunologicand imaging changes suggests that there may be some correlation be-tween the 2 phenomena.
A life cycle of A. cantonensis has been maintained in our laboratorysince 1980. It has been cycled through Biomphalaria glabrata snailsand Sprague–Dawley rats (Wang et al., 1989). Male New Zealand whiterabbits (1.5 kg) were purchased from the Laboratory Animal Center,College of Medicine, National Taiwan University, Taipei, Taiwan, andNational Veterinary Research Institute, Chunan, Hsinchu, Taiwan.These animals were kept at the Laboratory Animal Center of Chang-Gung University until they attained a body weight of 2.5–3.0 kg. They
1194 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
FIGURE 1. Antibody levels after experimental infection with 200 L3Angiostrongylus cantonensis over time, in rabbits (n 5 3).
were reared in metal cages and provided with food and drinking waterad lib.
Third-stage larvae of A. cantonensis were isolated from B. glabratasnails. After anesthetizing with 0.6-ml ketamine (50 mg/ml) in 2%Rompum solution (0.6 ml), each rabbit was experimentally infectedwith 200 larvae by stomach intubation. The infected animals were sep-arately reared. Changes in behavior and activities of these animals wererecorded.
Serum samples were prepared from the whole blood of 3 rabbits ondays 7, 14, 21, 28, 35, 42, and 49 postinfection (PI). These sampleswere analyzed by enzyme-linked immunosorbent assay using solubleextracts from homogenized fifth-stage larvae of A. cantonensis as anantigen (Wang et al., 1989).
Brain MR images were obtained from 6 rabbits on days 5, 7, 10, 14,21, and 28. MR scannings were performed on a 1.5-T whole-body scan-ner (Magnetom Vision, Siemens Medical Systems, Erlangen, Germany)with a field gradient strength of 25 mT/m using a standard quadraturehead coil. The rabbits were anesthetized with ketamine (1.5 ml) in 2%Rompum solution (1.5 ml). Coronal T1-weighted images of the entirebrain were obtained using a spin-echo sequence with a TR/TE of 464/3.5 ms, a slice thickness of 1.5 mm, distance factor of 0.1, 80-mm fieldof view, and a 256 3 256 image matrix. Coronal T2-weighted imagesof brain were obtained using a fast spin-echo sequence with a TR/TEof 2,136/119 ms, a slice thickness of 2.0 mm, distance factor of 0.1,90-mm field of view, and a 240 3 256 pixel matrix. Ten radio frequency(RF) excitations for T1- and 15 RF excitations for T2-weighted imageswere used and summed for signal averaging to increase the signal-to-noise ratio. Acquisition time was 19 min 51 sec for T1-weighted imagesand 8 min 34 sec for T2-weighted images. Ten to 11 excitations weresummed for signal averaging. Labeling of the different structures onMR images was done according to Shek et al. (1986). After obtainingT2- and T1-weighted MR images, enhanced T1-weighted MR imageswere obtained after intravenous administration of gadolinium–diethy-lenetriaminepentaacetic acid (Gd-DTPA; Magnevist, Schering AG,Pharmaceutical Division, Berlin, Germany) in a dosage of 0.1 mmol/kg.
The antibody titer elevated rapidly in the first 4 wk PI and thenreached a plateau (Fig. 1). Although hallmarks of immunologic respons-es may occur during the early stage of infection, changes in brain MRimages were suspected only in 1 rabbit on day 28. The preenhancedT1-weighted MR images (Fig. 2A–C) revealed hypointensities near theleft lateral ventricle and left hippocampus, and T2-weighted MR images(Fig. 2G–I) did not show any abnormality. Postenhanced MR imagesafter intravenous administration of gadolinium show hyperintensitiesnear the left lateral ventricle and left hippocampus (Fig. 2D–F).
The rabbit is not a natural definitive host of A. cantonensis. Experi-mental infections with 100, 200, or 500 third-stage larvae to rabbits donot cause apparent neurological symptoms, whereas all rabbits infected
with 1,000 larvae died within 3 days (Alicata, 1965). Clinical manifes-tations have been found to be associated with the number of third-stagelarvae inoculated (Jindrak and Magnusson, 1981). In our preliminarystudy, we found that all rabbits inoculated with 500 or 1,000 larvaedied within 72 hr. These rabbits developed paralysis of the hindquarters24 hr PI and even required artificial feeding. Among those infected with200, 300, or 400 larvae, no apparent neurological disorder was observed(data not shown). Thus, rabbits infected with 200 third-stage larvae ofA. cantonensis should be suitable models for the immunological andradiological investigations.
Detection of antibody titer in experimentally infected rats shows sig-nificant variations among different examination techniques. By indirecthemagglutination antibody test, antibodies may be detected as early asday 35 PI, with a peak on day 50 that persisted up to 145 days (Kamiyaand Tanaka, 1969; Kamiya, 1970; Chen and Suzuki, 1974; Yong andDobson, 1982). However, precipitating antibodies may be detected 1wk PI (Chen, 1974; Dharmkrong-At et al., 1978) and reaginic antibodiesin week 2 PI (Yoshimura and Yamagishi, 1976). In experimentally in-fected rabbits, the titer of antibodies against the fifth-stage larvae of A.cantonensis elevates rapidly in the first 4 wk before reaching a plateau.These changes are different from those in rats. In the postmortem ex-amination of rabbits infected with 500 or 1,000 third-stage larvae, wedid not discover any larvae from the brain of these animals (data notshown). These findings indicate that the rabbit may have strong im-munity to A. cantonensis larvae and that larvae may be destroyed im-mediately after passing through the blood–brain barrier. Histologicchanges in the brain may be induced by the fragments of the larvae.However, we did not observe any change in the brain MR images of 5rabbits examined from days 5 to 21 PI. The incubation period of theparasite leading to eosinophilic meningitis in humans ranges from 2 to45 days, with an average of approximately 13–16 days (Hwang andChen, 1991). The negative findings of MR imaging indicate that theimmunologic responses in the central nervous system at the early stageof infection is not sufficient to be observed by image studies.
Case reports on brain MR findings of human angiostrongyliasiscaused by A. cantonensis are not unusual. For example, radicular andcerebral parenchymal involvements have been observed in the brain MRimages of a patient with eosinophilic meningitis (Clouston et al., 1990).A worm was demonstrated on computed tomographic myelography andMR imaging of the spinal cord in a 45-yr-old man with a radiculomye-lopathy associated with an eosinophilic pleocytosis and cerebrospinalfluid antibodies to A. cantonensis but without signs or symptoms ofmeningitis (Wood et al., 1991). The brain MR images of a 17-yr-oldgirl with eosinophilic meningoencephalitis and positive in immunolog-ical tests showed multiple, small, high intensity areas on Gd-DTPA–enhanced T1-weighted images (Ogawa et al., 1998). In 6 patients witheosinophilic meningoencephalitis, MR images displayed the promi-nence of the Virchow–Robin spaces, subcortical enhancing lesions, and
RESEARCH NOTES 1195
FIGURE 2. Coronal MR images of a rabbit’s brain at different levels on day 28 after experimental infection with Angiostrongylus cantonensis.(A–C) T1-weighted MR images before gadolinium enhancement show equivocal hypointensities near the left lateral ventricles and hippocampus(B). (D–F) T1-weighted MR images after gadolinium enhancement show hyperintensities near the left lateral ventricles and left hippocampus(arrows). (G–I) T2-weighted MR images show unremarkable findings.
abnormal high T2 signal lesions in the periventricular regions (Kanpit-taya et al., 2000). On the basis of the brain MR images of 13 patients,high signal intensities were observed over the globus pallidus and ce-rebral peduncle on T1-weighted imaging, leptomeningeal enhancement,ventriculomegaly, and punctate areas of abnormal enhancement withinthe cerebral and cerebellar hemisphere on postenhanced T1-weightedimages, and a hyperintense signal on T2-weighted images (Tsai et al.,2003).
In this study, suspicious abnormal findings were observed near theleft lateral ventricle and hippocampus in a rabbit on day 28 PI. Thesefindings suggest tissue reactions to dead or dying worms or an impairedblood–brain barrier. Similar findings have been demonstrated in humanangiostrongyliasis (Clouston et al., 1990; Ogawa et al., 1998; Kanpit-taya et al., 2000; Tsai et al., 2003). These characteristic changes in brainMR images are readily distinguishable from those of cerebral cysticer-cosis, toxoplasmosis, paragonimiasis, and sparganosis (Chang and Han,1998). However, further investigations should be conducted to clarifythe correlation between histopathologic changes and MR findings.
This study was supported in part by a grant (NHRI-EX91-8908EL)from the National Health Research Institute, Republic of China, andalso by the Chang-Gung Medical Research Program (CMRP970). Wealso thank Shih-Ming Jung, Chien-Chuan Chen, and Shao-Wai Lee,Department of Diagnostic Radiology, Chang-Gung Memorial Hospital,for their help in acquiring and analyzing the magnetic resonance imagesand Chia-Lin Yang, Jan-Mei Lo, Chia-Chen Ko, Shin-Yi Huang, andMu-Jung Chen for their technical assistance.
LITERATURE CITED
ALICATA, J. E. 1965. Biology and distribution of the rat lung worm,Angiostrongylus cantonensis, and its relationship to eosinophilicmeningoencephalitis and other neurological disorders of man andanimals. Advances in Parasitology 3: 223–248.
BHAIBULAYA, M. 1975. Comparative studies on the life history of An-
giostrongylus mackerrasae Bhaibulaya, 1968 and Angiostrongyluscantonensis Chen, 1935. International Journal for Parasitology 5:7–20.
CHANG, K. H., AND M. H. HAN. 1998. MRI of CNS parasitic diseases.Journal of Magnetic Resonance Imaging 8: 297–307.
CHEN, S. N. 1974. Studies on immunodiagnosis of angiostrongyliasis.5. Double immunodifusion and counterelectropheresis for detectionof antibody in infected rats and human sera. Chinese Journal ofMicrobiology 7: 114–118.
———, AND T. SUZUKI. 1974. Fluorescent antibody and indirect hem-agglutination test for Angiostrongylus cantonensis infections in ratsand rabbits. Journal of the Formosan Medical Association 73: 393–400.
CLOUSTON, P. D., A. J. CORBETT, D. S. PRYOR, AND R. GARRICK. 1990.Eosinophilic meningitis: Cause of a chronic pain syndrome. Journalof Neurology, Neurosurgery and Psychiatry 53: 778–781.
DHARMKRONG-AT, A., V. UAHKOWITHCHAI, AND S. SIRISINHA. 1978. Thehumoral and cell-mediated immune responses to somatic and met-abolic antigens in rats infected with Angiostrongylus cantonensis.Southeast Asian Journal of Tropical and Public Health 9: 330–337.
HWANG, K. P., AND E. R. CHEN. 1991. Clinical studies on angiostron-gyliasis cantonensis among children in Taiwan. Southeast AsianJournal of Tropical Medicine and Public Health 22: 194–199.
JINDRAK, K., AND M. S. MAGNUSSON. 1981. Polymyositis and polyneu-ritis in experimental Angiostrongylus cantonensis infection of rab-bits. Annals of Tropical Medicine and Parasitology 75: 205–210.
KAMIYA, M. 1970. Haemagglutination test in rats infected with Angios-trongylus cantonensis. Southeast Asian Journal of Tropical Medi-cine and Public Health 1: 570–571.
———, AND H. TANAKA. 1969. Hemagglutination test in rats infectedwith Angiostrongylus cantonensis. Japanese Journal of Parasitology39: 593–599.
KANPITTAYA, J., S. JITPIMOLMARD, S. TIAMKAO, AND E. MAIRIANG. 2000.
1196 THE JOURNAL OF PARASITOLOGY, VOL. 90, NO. 5, OCTOBER 2004
MR findings of eosinophilic meningoencephalitis attributed to An-giostrongylus cantonensis. American Journal of Neuroradiology21: 1090–1094.
MACKERRAS, M. J., AND D. F. SANDARS. 1955. The life-history of the ratlung worm, Angiostrongylus cantonensis (Chen) (Nematoda: Meta-strongylidae). Australian Journal of Zoology 3: 1–21.
OGAWA, K., M. KISHI, T. OGAWA, N. WAKATA, AND M. KINOSHITA. 1998.A case of eosinophilic meningoencephalitis caused by Angiostron-gylus cantonensis with unique brain MRI findings. Rinsho Shink-eigaku 38: 22–26.
SHEK, J. W., G. Y. WEN, AND H. M. WISNIEWSKI. 1986. Atlas of therabbit brain and spinal cord. S. Karger A. G., Basel, Switzerland,139 p.
TSAI, H. C., Y. C. LIU, C. M. KUNIN, P. H. LAI, S. S. LEE, Y. S. CHEN,S. R. WANN, W. R. LIN, C. K. HUANG, L. P. GER, H. H. LIN, AND
M. Y. YEN. 2003. Eosinophilic meningitis caused by Angiostron-gylus cantonensis associated with eating raw snails: Correlation of
brain magnetic resonance imaging scans with clinical findings.American Journal of Tropical Medicine and Hygiene 68: 281–285.
WANG, L.-C., D. CHAO, AND E.-R. CHEN. 1989. Acquired immunity inrats against Angiostrongylus cantonensis infection. InternationalJournal for Parasitology 19: 617–620.
WELLER, P. F. 1993. Eosinophilic meningitis. American Journal of Medicine 95: 250–253.
WOOD, G., S. DELAMONT, M. WHITBY, AND R. BOYLE. 1991. Spinalsensory radiculopathy due to Angiostrongylus cantonensis infec-tion. Postgraduate Medical Journal 67: 70–72.
YONG, W. K., AND C. DOBSON. 1982. Antibody responses in rats infectedwith Angiostrongylus cantonensis and the passive transfer of pro-tective immunity with immune serum. Zeitschrift fur Parasitenkun-de 67: 329–336.
YOSHIMURA, K., AND Y. YAMAGISHI. 1976. Reaginic antibody productionin rabbit and rats infected with Angiostrongylus cantonensis. Jap-anese Journal of Veterinary Science 38: 33–40.
