Alternative Life Cycle Strategies and Colonization of Young Anurans by <i>Gorgoderina Attenuata</i> in Nebraska

Studies on life cycles and epizootiology of North American frog bladder flukes indicate that adult frogs become infected predominantly by ingesting tadpoles or other frogs that serve as second intermediate hosts for gorgoderid metacercariae. Other studies have indicated that newly metamorphosed frogs are rarely infected with these parasites because they are gapelimited predators that cannot feed on large intermediate hosts such as tadpoles and other frogs. We examined the role of potential intermediate hosts in the recruitment of the frog bladder fluke, Gorgoderina attenuata, to metamorphosed northern leopard frogs, Woodhouse’s toads, and bullfrogs from western Nebraska. We completed the life cycle of G. attenuata in the laboratory in 3 anuran species by experimentally infecting a variety of hosts. In addition, we generated and compared DNA sequence data from life cycle stages collected from a variety of naturally infected hosts. Our field and laboratory data indicate that in Nebraska G. attenuata has a truncated, 2-host life cycle that includes fingernail clams and anurans. Cercariae are ingested directly by tadpoles; unencysted juvenile worms then develop in the kidneys of tadpoles before moving to, and maturing in, the urinary bladder when tadpoles metamorphose. Additionally, G. attenuata can infect metamorphosed leopard frogs, bullfrogs, and toads when metacercariae in damselfly second intermediate hosts are ingested. These worms can also infect adult bullfrogs when they feed on other infected anurans possessing worms in their kidneys. Comparison of our material to published accounts of G. attenuata morphology and life cycles in Massachusetts suggests that previous work may have inadvertently involved 2 different species of gorgoderids. Our comparative approach to life cycle studies in different anuran life stages and multiple species of hosts suggests that tadpoles and metamorphosed anurans have favored alternative life cycle strategies in this trematode. Recently there has been an increased interest in attempts to understand adaptations involved in the evolution of complex life cycles of parasitic organisms (see Poulin and Cribb, 2002; Parker et al., 2003). These studies have derived mathematical models of how complex life cycles might have evolved from simple life cycles or conducted meta-analysis from a phylogenetic perspective on life cycle variation among different genera and families of parasites. The results of these studies suggest that we need additional field and experimental data to better understand the selective pressures that have resulted in the evolution of these highly improbable complex life cycles. A major impediment of this evolutionary understanding comes from the fact that few data exist concerning different life cycle strategies among closely related parasite taxa that are distributed in different combinations of hosts in different parts of the world (see Grabda-Kazubska, 1976; Snyder and Janovy, 1994, 1996; Poulin and Cribb, 2002; Bolek and Janovy, 2007a, 2007b). Bolek and Janovy (2007a) argued that amphibian parasites may be good model systems to address questions of parasite life cycle diversity and evolution. Recent comparative studies on amphibian parasite life cycles, recruitment, and community structure in anuran hosts by Bolek and Coggins (2000, 2001, 2003), Muzzall et al. (2001), Bolek and Janovy (2007a, 2007b, 2008), and Yoder and Coggins (2007) have provided base line data on the distribution, demography, field host specificity, and life history of amphibian parasites. These studies suggest that life cycle strategies of amphibian parasites are adapted to the environment and ecology of their hosts; however, life cycle Received 4 August 2008; revised 26 September 2008; accepted 14 November 2008. * Current address: Department of Zoology, Oklahoma State University, Stillwater, Oklahoma 74078. † Department of Biology, University of Nebraska at Omaha, Omaha, Nebraska 68182. ‡ School of Biological Sciences, University of Nebraska–Lincoln, Lincoln, Nebraska 68588. DOI: 10.1645/GE-1813.1 strategies of congeners and individual species in different life stages of amphibians can vary substantially. This variability may provide insight into the evolution and local adaptation of complex life cycles (Grabda-Kazubska, 1976; Snyder and Janovy, 1994, 1996; Bolek and Janovy, 2007a, 2007b, 2008). Studies on the life cycles and epizootiology of North American amphibian bladder flukes indicate a remarkable plasticity in the use of second intermediate hosts in their life cycles. North American species of Gorgodera and Gorgoderina primarily use tadpoles as second intermediate hosts but can also use odonates and molluscs. However, species of Phylodistomum use arthropods as second intermediate hosts, while adult amphibians become infected by ingesting arthropods, snails, tadpoles, or other frogs infected with metacercariae (Krull, 1935; Rankin, 1939; Crawford, 1940; Goodchild, 1943, 1948; Ubelaker and Olsen, 1972). Field studies also indicate that within individual anuran species, newly metamorphosed and juvenile anurans are less commonly infected with bladder flukes than are larger adult frogs because of small gape size, which affects the size of potential intermediate hosts that can be ingested by these frogs (see Bolek and Coggins, 2003; Bolek and Janovy, 2007a). Contrary to these studies, our observations from Nebraska indicate that newly metamorphosed northern leopard frogs, which never feed on tadpoles or other anurans (Bolek and Janvoy, 2007a), are commonly infected with Gorgoderina attenuata, with prevalences reaching 80%. We examined the population structure, recruitment, and route of infection of G. attenuata in young-of-the-year northern leopard frogs (Rana pipiens), Woodhouse’s toads (Bufo woodhousii), and bullfrogs (Rana catesbeiana) from western Nebraska in order to elucidate any differences in this bladder fluke’s life cycle strategy. Additionally, we compare our results to the original life cycle work on G. attenuata from western Massachusetts by Rankin (1939). The conceptual strength of the present study rests with the examination of alternative routes of infections by a single trematode species to 3 different anuran species BOLEK ET AL.—FROG BLADDER FLUKE LIFE CYCLE VARIATION 605 that vary in their life histories and phylogenetic relationships (Lannoo, 2005). MATERIALS AND METHODS Gorgoderina attenuata field studies in tadpoles and metamorphosed leopard frogs During July–September 2001 and July 2002, 125 newly metamorphosed northern leopard froglets were collected from Cedar Creek, Keith County, Nebraska (41 11.194 , 101 21.820 ). All frogs were placed on ice as they were collected and brought into the laboratory. Frogs were killed, the snout vent length (SVL) was measured, and they were examined for bladder fluke metacercariae in the skin and viscera as well as worms in the urinary bladder within 1–6 hr of collection. Additionally, 12 northern leopard frog tadpoles (Gosner stage 39–41) collected from Cedar Creek during July 2002 were brought to the laboratory, maintained through metamorphosis in 45.5-L tanks filled with aged tap water for a period of 3–6 wk, then examined for bladder flukes. To understand at what stage, age, or both newly metamorphosed northern leopard frogs acquired bladder fluke infections, we examined the seasonal recruitment of G. attenuata by collecting 60 northern leopard frog tadpoles (20 individuals every 2–3 wk during May–July 2003) and 20 metamorphosed froglets during July 2003 from Cedar Creek. All tadpoles and froglets were collected with a dip net, placed on ice, and necropsied within 1–6 hr of collection. They were aged and measured according to Gosner (1960) and McDiarmid and Altig (1999) and examined for gorgoderid metacercariae in the skin, musculature, and viscera, as well as for worms in the kidneys and urinary bladder. Kidneys were removed from tadpoles and froglets, pressed between 2 slides, and examined for any juvenile or encysted worms. Worms were fixed in 95% ethanol or AFA, and representative worms were stained and permanent slides prepared. Kidneys of some tadpoles were fixed in Bouin’s fixative, embedded in paraffin, sectioned at 12 m, affixed to slides, stained with hematoxylin and eosin, mounted in Canada balsam, and examined microscopically. Gorgoderina attenuata field studies in 3 sympatric anuran species During June–September 2004, 20 tadpoles each of northern leopard frogs and Woodhouse’s toads were collected from Cedar Creek, 20 bullfrog tadpoles from Breen’s Flyway (a pond adjacent to Cedar Creek) (41 10.914 , 101 21.654 ), along with 20 metamorph or adult northern leopard frogs, 25 metamorph Woodhouse’s toads from Cedar Creek, and 10 metamorph or adult bullfrogs from Cedar Creek and Breen’s Flyway, and examined for gorgoderid infections in their kidneys, urinary bladder, skin, and viscera. A chi-square test for independence was calculated to compare differences in prevalence between sympatric tadpole and metamorphosed anuran species collected during 2004. The KruskalWallis test was calculated to compare differences among mean abundance among sympatric tadpoles and metamorphosed anurans collected during 2004 because variances were heteroscedastic (Sokal and Rohlf, 1981). Arthropod survey for gorgoderid metacercariae To examine other potential routs of infection by G. attenuata to metamorphosed anurans, we examined 227 aquatic arthropods for gorgoderid metacercariae from Cedar Creek and Breen’s Flyway. All aquatic arthropods were collected with a dip net during the summer of 2001– 2004. These included anisopteran larvae (Anax junius, N 16; Sympetrum occidiualis, N 4); zygopteran larvae (Amphiagrion abbreviatum, N 14; Ischnura verticalis, N 11); coleopteran adults (Hydrophilidae, N 27); hemipterans (Belostoma sp., N 33); Ephemeroptera larvae (Callibaetis sp., N 20; Caenis sp., N 22); dipteran larvae (Stratiomyidae, N 9); amphipods (Hyalella azteca, N 70); and decapods (Orconectes, sp. N 1). Aquatic arthropods were brought to the laboratory, identified to family, genus, or species using keys in Borror et al. (1989), Merritt and Cummins (1996), Westfall and May (1996), Needham et al. (2000), and Thorp and Covich (2001), and gently teased apart for metacercariae with forceps in insect saline. Gorgoderina attenuata laboratory northern leopard frog tadpole infections Sphaerid pea clams (Pisidium compressum) were collected during June–July 2003 from Cedar Creek by sifting aquatic vegetation and sand in the stream with a 1-mm mesh size strainer. Individual clams were isolated in 1.5-ml well plates filled with aged tap water and observed daily for shedding gorgoderid cercariae. Some cercariae were fixed in AFA and representative cercariae were stained and permanent slides prepared. Northern leopard frog tadpoles (Gosner stage 30–36) were collected from Cedar Creek during May 2003 when no apparent infections were present and maintained in the laboratory in 45.5-L tanks for 2 wk. Tadpoles were divided into 3 groups and assigned to either a time-0 control (N 20), experimental (N 10), or time-T control (N 10) and were isolated in 5-ml well plates filled with aged tap water for 1 hr before exposure. Time-0 controls were dissected at the beginning of the experimental infections, whereas time-T controls were maintained throughout the duration of the experiment and dissected along with the experimental group. Gorgoderid cercariae were isolated from the shedding clams, and 3–10 cercariae were individually provided to each experimental tadpole over a period of 1–3 days. Tadpoles were observed using a dissecting microscope until they ingested an individual cercaria before another cercaria was introduced into the 5-ml well plate containing the tadpole. Additionally, the water in each well containing individual tadpoles was checked for any dead cercariae 12–24 hr after exposure. All infected tadpoles and time-T controls were maintained in groups of 1–4 on a diet of frozen mustard greens and Tetra Min fish food in 45.5-L tanks. To follow infection status, bladder fluke development, and migration, experimental and time-T tadpoles were killed and necropsied 3 days to 4 wk post-exposure (Gosner stage 36–45). Gorgoderina attenuata laboratory damselfly infections Sphaerid pea clams were collected during June–July 2004 from Cedar Creek and processed as described previously. Ischnura verticalis damselfly larvae were collected from Dunwoody Pond, Keith County, Nebraska (41 12.916 , 101 34.704 ). Larvae were divided into 3 groups and starved for 5 days before exposure; they were assigned to either a time-0 control (N 10), experimental (N 10), or time-T control (N 10) and were isolated in 5-ml well plates filled with aged tap water. Time-0 control damselflies were dissected at the beginning of the experimental infections; 1 to 5 cercariae were individually pipetted to each experimental damselfly larva over a period of 1–5 days. Larvae were observed using a dissecting microscope until they ingested an individual cercaria before another cercaria was introduced into the well. All exposed damselflies and time-T controls were dissected 1–6 days postexposure. Gorgoderina attenuata laboratory metamorphosed toad and bullfrog infections Newly metamorphosed Woodhouse’s toads were collected from Beckius Pond, Keith County, Nebraska (41 12.523 , 101 37.266 ) and divided into 3 groups: time-0 control (N 10), experimental (N 1), and time-T control (N 10). Laboratory-infected damselflies were dissected in insect saline (Hoar and Hickman, 1967). Time-0 control toads were dissected at the beginning of the experimental infections; 3 metacercariae were intubated into an experimental toad. The pipette was examined to confirm that no metacercariae remained. The experimental toad and the 10 time-T controls were killed 5 days post-exposure and examined for bladder flukes in the kidneys and urinary bladder. Bullfrog tadpoles were collected at South Platte River, Paxton, Keith County, Nebraska (41 07.600 , 101 34.611 ), brought to the laboratory, reared through metamorphosis, and divided into 3 groups: time-0 control (N 10), experimental (N 2), and time-T control (N 10). Lab-reared bullfrogs were each fed 4, naturally infected, northern leopard frog tadpoles over a period of 4 days, with 0–10 juvenile worms located in the kidneys (prevalence 55%; mean abundance 1.4 2.2), determined by dissecting 20 tadpoles. Thirteen to 18 days post-exposure, experimental bullfrogs were examined for juvenile and adult bladder flukes in the kidneys and urinary bladder and metacercariae in the tissues. Finally, a single adult northern leopard frog and a single laboratory-reared bullfrog were each fed 7 and 4 newly metamorphosed 606 THE JOURNAL OF PARASITOLOGY, VOL. 95, NO. 3, JUNE 2009 TABLE I. Gorgoderid specimens used in this study, their hosts, geographical origin of specimens, GenBank accession numbers, sequence length, and accession numbers for vouchers of corresponding sequences of gorgoderid specimens. Gorgoderid taxa Stage/location Host species Geographic origin ITS GenBank no. (sequence length) Vouchers G. attenuata RP1WNE Adult, bladder R. pipiens Cedar Creek, Keith County, Nebraska FJ445736 (960) HWML 49000 G. attenuata RP2WNE Adult, bladder R. pipiens Cedar Creek, Keith County, Nebraska FJ445737 (963) HWML 49001 Juvenile Gorgoderina sp. RP3WNE Juvenile, kidney Tadpole of R. pipiens Cedar Creek, Keith County, Nebraska FJ445738 (1,030) HWML 49002 Juvenile Gorgoderina sp. BWEXCC Juvenile, kidney B. woodhousii Cedar Creek, Keith County, Nebraska FJ445739 (918) HWML 49003 G. attenuata BWWNE Adult, bladder B. woodhousii Cedar Creek, Keith County, Nebraska FJ445740 (861) HWML 49004 G. attenuata RCLNY Adult, bladder R. clamitans Queechy, Columbia County, New York FJ445741 (891) HWML 49005 G. simplex RCLWI Adult, bladder R. clamitans Eagle, Waukesha County, Wisconsin FJ445742 (1,060) HWML 49006 G. amplicava RCWI Adult, bladder R. catesbeiana Big Muskego Lake, Waukesha County, Wisconsin FJ445743 (911) HWML 49007 Woodhouse’s toads to see if they could survive after ingesting these potential toxic hosts. Other amphibian field surveys Because Rankin (1939) indicated that G. attenuata formed metacercaria stages in tadpoles and metamorphosed frogs in Massachusetts, we examined different species of adult anurans and/or their tadpoles from other locations for gorgoderid metacercariae stages to see if gorgoderid metacercariae occurred in anurans at other locations than Cedar Creek. During May–September 2000–2006 and March–May 2007 an additional 9 species of adult anurans and/or their tadpoles were collected from 4 locations in Nebraska and Arkansas and examined for gorgoderid metacercariae in the skin, musculature, and viscera as well as juvenile worms in the kidneys and adults in the urinary bladder. These included (1) 69 adult bullfrogs and 2 adult plains leopard frogs (Rana blairi) collected from Nevens Pond, Keith County, Nebraska (41 12.426 , 101 24.510 ); (2) 100 plains leopard frogs, 108 bullfrogs, 100 Woodhouse’s toads, 103 boreal chorus frogs (Pseudacris maculata), 56 cricket frogs (Acris crepitans), and 62 Cope’s gray treefrogs (Hyla chrysoscelis) from Pawnee Lake, Lancaster County, Nebraska (40 51.589 , 96 53.468 ); (3) 53 bullfrog tadpoles and 3 bullfrogs from Elk Creek, Lancaster County, Nebraska (40 53.145 , 96 50.048 ); and (4) 6 bullfrogs, 4 southern leopard frogs, (Rana sphenocephala), 1 green frog (Rana clamitans), 1 pickerel frog (Rana palustris), 1 eastern gray treefrog (Hyla versicolor), and 7 cricket frogs collected from a pond in Washington County, Arkansas (35 46.979 , 94 14.687 ). All adult gorgoderids from experimental life cycle studies and specimens collected from Arkansas, Nebraska, and other locations from North America for molecular work (see below) were identified based on the descriptions and redescriptions provided by Stafford (1902), Cort (1912), Holl (1928), Olsen, (1937), Goodchild (1950), Brooks (1976), and Bolek et al. (2009). Voucher specimens of bladder flukes have been deposited in the H. W. Manter Parasitology Collection, University of Nebraska, Lincoln, Nebraska (accession numbers HWML): 48984, G. attenuata from the urinary bladder of a northern leopard frog from Cedar Creek, Keith County, Nebraska (unless indicated otherwise, all collecting sites are in Keith County); 48985, G. attenuata from the urinary bladder of a bullfrog from Cedar Creek; 48986, G. attenuata from the urinary bladder of Woodhouse’s toad from Cedar Creek; 48987, Gorgoderina sp. from the kidneys of a northern leopard frog tadpole from Cedar Creek; 48988, Gorgoderina sp. from the kidneys of a Woodhouse’s toad tadpole from Cedar Creek; 48989, Gorgoderina sp. from the kidneys of a bullfrog from Breen’s Flyway; 48990, G. attenuata from the urinary bladder of an experimentally infected northern leopard frog tadpole; 48991, Gorgoderina sp. from the kidneys of an experimentally infected northern leopard frog tadpole; 48992, G. attenuata from the urinary bladder of an experimentally infected bullfrog; 48993, G. attenuata from the urinary bladder of a bullfrog from Nevens Pond; 48994, G. attenuata from the urinary bladder of a bullfrog from Pawnee Lake, Lancaster County, Nebraska; 48995, G. attenuata from the urinary bladder of a plains leopard frog from Pawnee Lake; 48996, G. attenuata from the urinary bladder of a bullfrog from Elk Creek, Lancaster County, Nebraska; 48997, Gorgoderina sp. from the kidney of a bullfrog tadpole from Elk Creek; 48998, G. attenuata from the urinary bladder of a southern leopard frog from a pond in Washington County, Arkansas; and 48999, gorgoderid cercaria from P. compressum from Cedar Creek. Specimen collection for molecular characterization Because the original 3-host life cycle of G. attenuata was elucidated from green frogs and other amphibian species from western Massachusetts (specific location not given) by Rankin (1939), we collected a single green frog infected with G. attenuata near the border of New York and Massachusetts (Queechy, Columbia County, New York; 42 24.269 , 73 25.627 ) to compare the complete internal transcribed spacer region (ITS) of the ribosomal DNA (ITS 1 5.8S ITS 2) of worms from the general location of the original life cycle study and worms from northern leopard frogs and Woodhouse’s toads from Cedar Creek, Nebraska. Additionally, Gorgoderina simplex and Gorgodera amplicava were collected from green frogs from Eagle, Waukesha County, Wisconsin (42 53.225 , 88 29.545 ) and bullfrogs from Big Muskego Lake, (42 51.241 , 88 7.456 ), respectfully. These worms were collected during March–September 2003–2004 to compare the ITS sequence of G. attenuata to other species of amphibian gorgoderids. Living worms recovered from amphibians were allowed to release eggs in water, then thoroughly rinsed in water, identified, and fixed in 95% ethanol. A 2to 5-mm piece of the left or right side of each single alcohol-preserved worm was then removed for DNA extraction. Once DNA was obtained, the remaining adult worm was stained and identified based on the original descriptions. Stained voucher specimens of all species were preserved. Additionally, live juvenile worms identified as Gorgoderina sp. from the kidneys of naturally and experimentally infected tadpoles or metamorphosed naturally and experimentally infected Woodhouse’s toads were fixed in 95% ethanol. Because juvenile worms from kidneys of naturally and experimentally infected tadpoles and metamorphosed anurans contained small amounts of tissue, we used the entire specimen of these individuals for DNA extraction. Species used in the analysis, hosts, collection location, sequence length, and the voucher accession number of species are listed in Table I. DNA extraction, amplification, and sequencing Genomic DNA was extracted from single adult worm pieces and entire juvenile worms from the kidneys of tadpoles, frogs, and toads using DNeasy tissue kits (Qiagen, Valencia, California). With the exception of a few bases at the 5 and 3 ends, the entire ITS rDNA (ITS 1 5.8S ITS 2) was amplified by polymerase chain reaction (PCR) using a forward primer in the 18S region, Br1 (5 -GTA GGT GAA CCT GCA GAA GG), a digenean-specific reverse primer in the 28S region, DigL1 (5 -GTG ATA TGC TTA AGT TCA GC), and a 58S2 (5 -TAA GCC GAC CCT CGG ACA GG) digenean-specific internal reverse primer. Reactions were performed in a 25 l total volume acBOLEK ET AL.—FROG BLADDER FLUKE LIFE CYCLE VARIATION 607 cording to instructions accompanying the FildeliTaq PCR master mix kit (USB Corporation, Cleveland, Ohio). Reactions were run on a Biometra UNO under the following cycling conditions: 94 C for 4 min followed by 40 cycles of 94 C for 30 sec, 50–56 C for 30 sec, and 72 C for 2 min, followed by 1 cycle of 72 C for 5 min for final elongation. Unincorporated PCR primers and nucleotides were removed from PCR products using High Pure PCR Purification Kit (Roche Diagnostics, Mannheim, Germany). Sequences were determined directly from PCR templates by cycle sequencing using Big Dye fluorescent dye terminators and protocols and an ABI 377 automated sequencer (Perkin-Elmer, Foster City, California). Primers used for PCR amplification were also used in sequencing reactions.