Relationship of tadpole stage to location of echinostome cercariae encystment and the consequences for tadpole survival

Abstract

The effect of echinostome infections on the survival of Rana pipiens tadpoles was examined in relation to developmental stage of tadpoles. Individual tadpoles of Gosner stages 25, 27, 32-33, and 37-39 were exposed to 1 of 4 levels of cercariae (0, 20, 50, or 100). Only tadpoles at stage 25, the earliest stage infected, died within a 5-day experimental period. This stage-specific mortality rate could be explained, in part, by the stage-specific location of encystment of cercariae, which was documented in a separate experiment. In accordance with kidney development, cercariae predominately encysted in the pronephroi during early stages of tadpole development (stages 25 through 31-32) and only in the mesonephroi and associated ducts at later stages (stages 37 through 46). As the mesonephros develops, renal capacity presumably increases. Thus, tadpoles died only when metacercariae concentrated in the functional portion of the kidney with the most limited renal capacity. As tadpoles aged, they also became less susceptible to infections. On average, 69.5% of cercariae that were exposed to stage 25-26 tadpoles successfully encysted. compared with only 8.4% of cercariae exposed to stage 37-38 tadpoles. Exposures of metamorphic frogs (poststage 46) to cercariae revealed that these individuals can become infected with echinostomes. Collectively, our data highlight the host stage-dependent dynamics of tadpole-echinostome interactions.

Full text

475
J. Parasitol., 89(3), 2003, pp. 475–482
q
American Society of Parasitologists 2003
RELATIONSHIP OF TADPOLE STAGE TO LOCATION OF ECHINOSTOME CERCARIAE
ENCYSTMENT AND THE CONSEQUENCES FOR TADPOLE SURVIVAL
Anna M. Schotthoefer, Rebecca A. Cole*, and Val R. Beasley†
Department of Veterinary Pathobiology, University of Illinois at Urbana-Champaign, Urbana, Illinois 61802.
e-mail: aschotthoefer@cvm.uiuc.edu
ABSTRACT
: The effect of echinostome infections on the survival of Rana pipiens tadpoles was examined in relation to devel-
opmental stage of tadpoles. Individual tadpoles of Gosner stages 25, 27, 32–33, and 37–39 were exposed to 1 of 4 levels of
cercariae (0, 20, 50, or 100). Only tadpoles at stage 25, the earliest stage infected, died within a 5-day experimental period. This
stage-specific mortality rate could be explained, in part, by the stage-specific location of encystment of cercariae, which was
documented in a separate experiment. In accordance with kidney development, cercariae predominately encysted in the pronephroi
during early stages of tadpole development (stages 25 through 31–32) and only in the mesonephroi and associated ducts at later
stages (stages 37 through 46). As the mesonephros develops, renal capacity presumably increases. Thus, tadpoles died only when
metacercariae concentrated in the functional portion of the kidney with the most limited renal capacity. As tadpoles aged, they
also became less susceptible to infections. On average, 69.5% of cercariae that were exposed to stage 25–26 tadpoles successfully
encysted, compared with only 8.4% of cercariae exposed to stage 37–38 tadpoles. Exposures of metamorphic frogs (poststage
46) to cercariae revealed that these individuals can become infected with echinostomes. Collectively, our data highlight the host
stage–dependent dynamics of tadpole–echinostome interactions.
Species of Echinostoma and Echinoparyphium (Trematoda:
Echinostomatidae) are widely distributed and commonly use
frogs as second intermediate hosts (Prudhoe and Bray, 1982;
Kostadinova and Gibson, 2000). For instance, in a 2-yr survey
conducted in Minnesota, the overall mean prevalence (
6
SE) of
echinostome infections across 20 populations of metamorphic
frogs of Rana pipiens was 85.6% (
6
3.77%) (A. Schotthoefer,
unpubl. data). Similarly, echinostome metacercariae were the
most frequently encountered parasites in R. pipiens, R. cates-
beiana, and R. clamitans of New Brunswick, Canada (Mc-
Alpine and Burt, 1998). Infections have also been reported in
Rana sylvatica, Hyla crucifer (Najarian, 1955), Bufo american-
us (Ulmer, 1970), and Acris crepitans (Beasley et al., 2003).
Despite their common occurrence, relationships between
echinostomes and their amphibian hosts have received little at-
tention. However, in recent years, interest in the nature of echi-
nostome–frog interactions has grown because of evidence that
suggests that infections might contribute to the regulation of
amphibian populations. Beasley et al. (2003) found lower ju-
venile recruitment at ponds where tadpoles and juveniles of A.
crepitans had severe echinostome infections, compared with
ponds where infection intensities were low. In addition, infec-
tions have been observed to cause edema, a clinical sign of
disease, and mortality in tadpoles (Beaver, 1937; Fried et al.,
1997), as well as inhibit growth of infected tadpoles in the
laboratory (Fried et al., 1997). Moreover, observations suggest
that human-induced changes to aquatic ecosystems may in-
crease the prevalence of echinostome infections or change the
dynamics of echinostome–tadpole interactions in ways that
harm tadpole populations (Thiemann and Wassersug, 2000a;
Beasley et al., 2003).
In light of worldwide amphibian population decline (see Bar-
inaga, 1990; Wright et al., 2001) and evidence that infection by
another larval trematode, Ribeiroia ondatrae, is responsible for
limb malformations in frog populations (Johnson et al., 1999,
2002), these observations raise concerns about the importance
Received 8 July 2002; revised 9 November 2002; accepted 11 No-
vember 2002.
* USGS, National Wildlife Health Center, Madison, Wisconsin 53711.
† Department of Veterinary Biosciences, University of Illinois at Ur-
bana-Champaign, Urbana, Illinois 61802.
of echinostome infections, and parasitism in general, in deter-
mining the survival of tadpoles and their subsequent recruit-
ment into juvenile frog populations. To better understand the
potential ecological impacts of echinostome infections on tad-
pole populations, we focused on the stage-specific dynamics of
tadpole–echinostome interactions in an attempt to identify the
time period during development when tadpoles are likely to be
negatively affected by echinostome infections. Specifically, we
addressed the effect of infections with respect to kidney devel-
opment.
Echinostome infections in anurans occur primarily in the kid-
neys. In anurans, as in other vertebrates, the kidneys undergo
ontogenetic changes during larval development. Initially, ex-
cretory and osmoregulatory functions are provided by paired
pronephroi. These pronephroi each consist of single convoluted
tubules and glomeruli, located just posterior to the gill cham-
bers (Fig. 1). The tubules lead into pronephric ducts that run
parallel along the dorsal body wall and eventually empty into
the cloaca of amphibian larvae (Fox, 1963). At later stages of
development, mesonephric kidneys differentiate and begin sup-
plementing the functional roles of the pronephroi. The meso-
nephroi, which are located in the mid to posterior portions of
the abdominal cavity and lie against the dorsal body wall, ul-
timately replace the pronephroi as the sole functional kidneys
in adult amphibians (Fig. 1). Nephric ducts connect the pro-
and mesonephroi while both kidneys are functioning. Degen-
eration of the pronephroi and nephric ducts begins in the final
stages of metamorphosis, and by the end of metamorphosis,
they have largely disappeared (Jaffee, 1954a; Viertel and Rich-
ter, 1999).
Because renal capacity is expected to increase with devel-
opment of the mesonephroi, we predicted that tadpoles would
become more tolerant of infections as they reached later stages
of kidney development and, therefore, would experience less
mortality than earlier tadpole stages. Thus, we examined the
effect of infections on tadpole survival when infections were
acquired at different stages of development. We also examined
the stage-specific locations of cercariae encystment to deter-
mine if the location of encystment changed with development
of the mesonephroi. Thiemann and Wassersug (2000b) reported
that echinostome cercariae had a predilection for the pronephroi
in tadpoles of R. clamitans, whereas in tadpoles of R. sylvatica

476 THE JOURNAL OF PARASITOLOGY, VOL. 89, NO. 3, JUNE 2003
F
IGURE
1. Diagrammatic view of kidney development in anurans, as inferred from Jaffee (1954a, 1954b) and Viertel and Richter (1999). A.
At hatching (and at stage 25; Gosner, 1960), the pronephroi are present and functioning. B. In Rana pipiens, the mesonephroi have differentiated
and are functioning by 30 days posthatching (approximately stage 30, Gosner, 1960). C. The pro- and mesonephroi co-function until about stage
40 (Gosner, 1960), when the pronephroi and nephric ducts, connecting the pro- and mesonephroi, begin degenerating. By the end of metamorphosis,
the pronephroi and nephric ducts have largely disappeared, and only the mesonephroi are available for renal function in the frog.
the cercariae encysted more often in the mesonephroi. The R.
clamitans tadpoles used in the study were younger (Gosner
stage 26; see Gosner, 1960) and presumably at an earlier stage
of kidney development than the tadpoles of R. sylvatica (Gosner
stage 34), suggesting that the location of echinostome cercariae
encystment changes with kidney development from the pro-
nephroi to the mesonephroi, however, to our knowledge this
aspect of echinostome–tadpole interactions has not been inves-
tigated. Finally, we exposed metamorphic frogs to cercariae to
determine if frogs remain susceptible to infections after meta-
morphosis (Martin and Conn, 1990). Knowing when during the
frog life cycle individuals are most likely to be affected by
echinostome infections is an important first step in understand-
ing the potential effects of these parasites on wild anuran pop-
ulations.
MATERIALS AND METHODS
Animal husbandry
Egg masses of R. pipiens were purchased from Carolina Biological
Supply (Burlington, North Carolina) and Nasco (Fort Atkinson, Wis-
consin), and placed in 37.8- or 75.7-L glass aquaria with enough aged
or carbon-filtered tap water to submerge the eggs. After tadpoles
hatched and became active, the volume of water in each aquarium was
increased and gently aerated. Tadpoles were fed boiled romaine lettuce
for approximately 2 wk posthatching, after which they were fed com-
mercial tadpole food (Frog Brittle, Nasco), supplemented with ground
algae wafers (Hikari, Kyorin Food Inc. Ltd., Tokyo, Japan). Partial wa-
ter changes were conducted every 2–3 days and sponge filters were
installed 4 wk posthatching to control ammonia and nitrite concentra-
tions. Tadpoles were maintained en masse in aquaria until they reached
the targeted stages of development used in experiments.
A number of tadpoles from each batch of frog eggs were maintained
through the end of metamorphosis (Gosner stage 46). After forelimb
emergence, these tadpoles were transferred to aquaria containing a bed
of dry aquarium gravel sloping into water, which facilitated their emer-
gence from the water as metamorphic frogs. Metamorphic frogs were
fed fruit flies, mealworms, and small crickets before echinostome ex-
posures.
The intermediate snail hosts, Planorbella trivolvis (
5
Helisoma tri-
volvis), were maintained in separate aquaria on a diet of boiled romaine
lettuce and ground algae wafers. Partial water changes were conducted
weekly. All animals were maintained on a 12 hr light: 12 hr dark light
cycle throughout the experiments.
Experimental design
Two separate 2-factor design experiments were conducted to deter-
mine (1) how infections acquired at different stages of development
affect tadpole survival and (2) how the location of cercariae encystment
changes with tadpole stage, i.e., kidney development. The treatment
factors examined in both the experiments were tadpole stage and cer-
cariae exposure dose. The tadpole stages selected were representative
of different stages of kidney development in R. pipiens, as inferred from
Jaffee (1954a, 1954b). For the survival experiment, stages 25, 27, 32–
33, and 37–39 (Gosner, 1960) were used, and stages 25–26, 31–32, 37–
38, and 42 (Gosner, 1960) were used in the location of cercariae en-
cystment experiment. Thus, tadpoles at stages 25 through 27 primarily
depend upon pronephroi for renal function, tadpoles at stages 31
through 39 are expected to have both pro- and mesonephroi function,
and tadpoles at stage 42 exclusively rely on mesonephroi function. In
the survival experiment, individual tadpoles were exposed to 0, 20, 50,
100, or 150 cercariae, whereas in the location of cercariae encystment
experiment, the exposure dose treatments were 25, 50, 100, and 150
cercariae per tadpole. The doses of cercariae used were in the range of
infection intensities observed in wild-caught juveniles of R. pipiens
(data not shown).
Tadpole survivorship was monitored daily for 5 days postexposure
(PE) in the survival experiment. Because edema can also be an indi-
cation of renal dysfunction (McClure, 1919, 1928), we also recorded

SCHOTTHOEFER ET AL.—ECHINOSTOME INFECTIONS IN TADPOLES 477
T
ABLE
I. Survival of Rana pipiens tadpoles after experimental exposure to various doses of echinostome cercariae at specific Gosner (1960)
stages of development.
Tadpole
stage Trial
Exposure
dose of
cercariae n
Average
(min–max)
infection intensity
% Tadpole
survival
(5-days PE)
25 1 0
20
50
100
3
3
3
5
0
12 (5–17)
29 (24–36)
47.7 (44–51)
100
33.3
0
0
20
20
50
100
5
5
5
5
0
15.2 (10–19)
29.6 (18–38)
36.8 (33–43)
80
60
80
20
27 1 0
20
50
100
150
5
5
5
5
5
0
4.8 (1–10)
12.4 (5–21)
25 (14–44)
45 (33–63)
100
100
100
100
100
20
20
50
150
5
5
2
5
0
6.6 (4–9)
7.8 (3–13)
20 (12–41)
100
100
100
100
32–33 1 0
50
150
4
4
4
0
6 (0–16)
21.7 (8–35)
100
100
100
37–39 1 0
50
4
5
0
0
100
100
its occurrence. The numbers of metacercariae in the pro- and meso-
nephroi of tadpoles were determined at death or at termination of the
experiment by dissection of the pro- and mesonephroi. Whole kidneys
were examined under cover slips under a compound microscope. Be-
tween 48 and 72 hr PE, tadpoles in the location of cercariae encystment
experiment were killed and the distributions of metacercariae deter-
mined similarly. All tadpoles were killed by immersion in 1:1,000 tri-
caine methanesulfonate (MS-222) for 5–10 min or until there was no
response to repeated physical stimuli.
Infection procedures
Anuran larvae: Tadpoles at the appropriate stages of development
were removed from aquaria, weighed and measured (total and snout-to-
vent lengths), and isolated in individual containers filled with aged or
carbon-filtered tap water. Tadpoles at stages 25 through 27 were isolated
in 120-ml containers (estimated water–tadpole volume ratios, 35.6–
11.6:1 cm
3
) and tadpoles of poststage 27 treatments were isolated in
266-ml containers (2.6–0.54:1 cm
3
water–tadpole volume ratios). The
effect that size of tadpole within a stage might have on transmission
success was controlled by ensuring that the distributions of tadpole
weights and lengths were equal among dose treatments. After a 24-hr
acclimation period, tadpoles were exposed to cercariae that had been
shed from snails within an 8-hr period. Each individual tadpole was
exposed to a composite of cercariae from several different snails. All
tadpoles were exposed to cercariae for approximately 24 hr.
Water in containers was changed daily after exposure, until the end
of the experimental period. Because the number of tadpoles that could
be exposed to cercariae at 1 time was limited by the number of tadpoles
that were currently at the targeted stages, or the number of cercariae
that were shed from snails on that day, or both, sample sizes varied
among treatments, and tadpoles at each stage could not be exposed to
all doses. However, in most cases, at least 5 individuals were used in a
treatment for both experiments (Tables I, II).
Metamorphic frogs: To determine whether frogs could serve as sec-
ond intermediate hosts, metamorphic frogs at least 1 wk post-Gosner
stage 46, i.e., after complete tail resorbtion, were exposed to echino-
stome cercariae. Individual frogs were placed in containers (14
3
14
3
7 cm) with sufficient water (
;
150 ml) to cover them but which did not
prohibit them from breathing through their nares.
Individuals within 1 group of frogs (n
5
16) were each exposed to
the total number of cercariae (
.
300) shed from a snail in a 6- to 8-hr
period for 24 hr. This number was estimated by counting the numbers
of cercariae in 10, 0.5-ml aliquots of water randomly removed from the
container harboring an infected snail. The average number of cercariae
in 0.5 ml was multiplied by the total volume of water remaining in the
snail’s container; this water was then transferred to a container holding
a frog. Five days PE, individuals were killed in 1:1,000 MS-222 and
fixed whole in 10% neutral buffered formalin. Metacercariae in the
kidneys were enumerated by pressing the fixed kidneys between 2 glass
slides and viewing them under a dissection microscope. In a second
group of frogs (n
5
7), individuals were exposed to doses of 150 cer-
cariae each day for 2 consecutive days, for a total exposure dose of 300
cercariae per frog. These frogs were similarly killed but were examined
for metacercariae on the day after the second exposure.
Source of cercariae: Cercariae used in the study were from naturally
and experimentally infected snails of P. trivolvis. Procedures for estab-
lishing infections in laboratory animals followed those of Fried (1985).
Morphology of cercariae and adults obtained from laboratory hamsters
was consistent with that of Echinostoma trivolvis (Fried et al., 1998;
Kostadinova and Gibson, 2000); however, we cannot rule out the pos-
sibility that other echinostome species were present because cercariae
from naturally infected snails were used. Original source locations from
which snails with echinostome infections were collected varied by ex-
periment. Snails used in the tadpole survival experiment were from
Necedah National Wildlife Refuge, Juneau County, Wisconsin; snails
for the location of cercariae encystment experiment and metamorphic
frog exposures were from privately owned wetlands in Anoka, Meeker,
Wright, and Stearns Counties, Minnesota and Liberty Creek and Albany
State Wildlife Areas, Green County, Wisconsin. Voucher specimens of
adults raised in laboratory hamsters and cercariae were deposited in the
U.S. National Parasite Collection, Beltsville, Maryland as USNPC
092556 and 092554-5, respectively.
Statistical analysis
Mortality and edema were compared among exposure doses for tad-
poles at stage 25 (the only stage at which death was observed) using

478 THE JOURNAL OF PARASITOLOGY, VOL. 89, NO. 3, JUNE 2003
T
ABLE
II. Percentage of individuals infected and average infection intensities acquired by Rana pipiens tadpoles exposed to different doses of
echinostome cercariae at specific Gosner (1960) stages used in the location of cercariae encystment experiment. Predominate kidney functioning
at stages, as inferred by Jaffee (1954a, 1954b), is also indicated.
Tadpole
stage Kidney stage
Exposure
dose of
cercariae n % Infected
Average
(min–max)
infection intensity
25–26 Pronephros 25
50
100
150
5
15
15
5
100
100
100
100
21.6 (18–24)
37.6 (23–50)
56.2 (36–78)
70.2 (47–88)
31–32 Pro- and mesonephros 25
50
100
150
6
6
5
5
100
100
100
100
6.2 (2–16)
14 (7–24)
12.2 (5–20)
34.2 (5–57)
37–38 Mesonephros 25
100
6
6
83.3
83.3
1.5 (0–3)
10.8 (0–37)
42 Mesonephros 100
150
9
8
22.2
12.5
1.5 (1–2)
3
Fisher’s Exact tests. Data from 2 trials involving stage 25 tadpoles in
the survival experiment (Table I) were pooled for this analysis. Analysis
of covariance (ANCOVA), with tadpole weight as a covariate, was used
to examine the influence of tadpole stage and exposure dose on (1)
locations of cercariae encystment within the nephric system, as mea-
sured by the proportion of metacercariae in the mesonephroi, (2) the
right-to-left distributions of metacercariae in tadpoles, using the differ-
ences in the numbers of metacercariae in the right against the left kid-
neys as response variables, and (3) the proportions of cercariae that
successfully encysted in exposed tadpoles, defined here as infection
success. All proportions were arcsine transformed as recommended for
proportions close to 0 and 1 (Zar, 1996). When ANCOVA tests revealed
an overall significant difference, Tukey (honestly significant difference
[HSD]) tests were used to identify differences among groups. To deter-
mine if the differences in right against the left kidneys at each stage
were significantly biased, we used paired t-tests. In these tests, we used
a Bonferroni-adjusted
a
-level, i.e., 0.005, to correct for the increased
experiment-wise type I error associated with doing multiple pairwise
comparisons. The
a
-level in all other analyses was 0.05. Data from stage
42 tadpoles were eliminated from all analyses because the tadpoles at
this stage were exposed to only 1 dose of cercariae (Table II). In ad-
dition, all the metacercariae found in these tadpoles were in the meso-
nephroi and infection success was low (0.3%). Analyses were conducted
using SAS version 8.02 (SAS Institute Inc., Cary, North Carolina). Type
IV sum of squares were used in assessing significance of ANCOVA
tests to account for empty cells at the 50 and 150 cercariae treatments
for the stage 37–38 group (Table II) (Neter et al., 1990).
RESULTS
Mortality was observed only in tadpoles that were exposed
to echinostome cercariae at stage 25. Cumulative incidence of
mortality increased with an increasing exposure dose of cer-
cariae (Table I). Tadpoles exposed to 100 cercariae had sig-
nificantly higher mortality than controls (Fisher’s exact test, P
5
0.01). Generalized edema also was associated with infec-
tions acquired at this early stage (Fisher’s exact test, P
,
0.0001). Of the exposed tadpoles that survived beyond 24 hr
PE (n
5
20), 14 (70%) developed edema; 12 (85.7%) of these
cases developed in tadpoles exposed to 50 or 100 cercariae.
Edema co-occurred with death in 8 cases (57.1%). A positive
association between dose of cercariae and edema also was
detected in the location of cercariae encystment experiment
for tadpoles exposed at stages 25–26 (Fisher’s exact test, P
5
0.012).
Analysis of covariance indicated tadpole stage (F
5
6.14, P
5
0.004) and weight (F
5
9.14, P
5
0.004) significantly influ-
enced the proportions of metacercariae found in the mesoneph-
roi. There also was a significant interaction effect between tad-
pole stage and weight (F
5
9.11, P
5
0.004); the proportions
of metacercariae in the mesonephroi increased with weight
within stages 25–26 and 31–32, as indicated by scatter plots
(data not shown). Tukey (HSD) tests revealed the proportions
of metacercariae in the mesonephroi were significantly different
across all tadpole stages, with these proportions increasing with
tadpole stage (Fig. 2). At stages 37–38, all metacercariae were
found in the mesonephroi and nephric ducts, and at stage 42
all metacercariae were found in the mesonephroi. The propor-
tion of metacercariae in the mesonephroi increased with expo-
sure dose at stages 25–26 (Fig. 2), but overall there was no
significant dose effect in the model (F
5
1.52, P
5
0.217).
Differences between the numbers of metacercariae in the left
compared to the right pronephroi varied significantly by stage
(F
5
3.66, P
5
0.031) but not dose (F
5
1.22, P
5
0.311),
whereas the numbers of metacercariae in the right and left me-
sonephroi did not vary by either stage or dose (F
5
1.21, P
5
0.310). Overall, the differences in the numbers of metacercariae
on the right and left sides of the nephric system were significant
for stage (F
5
3.25, P
5
0.045) but not dose (F
5
2.21, P
5
0.095). More metacercariae were found in the right than the left
pronephros of tadpoles exposed to cercariae at stages 25–26,
although this was not significant (t
5
1.88, P
5
0.067; mean
difference
6
SD, 2.3
6
7.56). In contrast, significantly fewer
metacercariae were found in the right pronephros than the left
of staged 31–32 tadpoles (t
52
3.10, P
5
0.005; mean differ-
ence
6
SD,
2
3.4
6
5.08). When comparing the overall differ-
ences between the numbers of metacercariae on the right and
left sides of the nephric systems of tadpoles at each stage, we
again found that significantly fewer metacercariae were on the
right sides of tadpoles than the left at stages 31–32 (t
52
3.47,
P
5
0.002; mean difference
6
SD,
2
4.0
6
5.46). However,
metacercariae were on average evenly distributed between the
right and left sides of the nephric systems of tadpoles at stages
25–26 (t
5
0.587, P
5
0.561; mean differences
6
SD, 0.975

SCHOTTHOEFER ET AL.—ECHINOSTOME INFECTIONS IN TADPOLES 479
F
IGURE
2. A. Mean (
6
SE) proportions of echinostome metacercar-
iae found in the mesonephroi of Rana pipiens tadpoles exposed to dif-
ferent doses of cercariae at specific Gosner (1960) stages in the location
of cercariae encystment experiment. Tukey (HSD) tests revealed sig-
nificant differences across all stages. B. Mean (
6
SE) proportions of
metacercariae in the pronephroi of the same groups of tadpoles are
displayed for comparative purposes. NA indicates no tadpoles were test-
ed at that treatment.
F
IGURE
3. Mean (
6
SE) proportions of echinostome cercariae that
successfully encysted in Rana pipiens tadpoles exposed to different dos-
es of cercariae at specific Gosner (1960) developmental stages in the
location of the cercariae encystment experiment. Values with different
letters were significantly different according to Tukey (HSD) multiple
comparisons tests. NA indicates no tadpoles were tested.
6
10.502) and 37–38 (t
5
0.953, P
5
0.361; mean differences
6
SD, 0.8
6
2.658).
Infection success was significantly influenced by tadpole
stage (F
5
3.26, P
5
0.045), exposure dose (F
5
4.34, P
5
0.008), and the interaction between tadpole stage and exposure
dose (df
5
1, 64, F
5
4.02, P
5
0.049). In general, infection
success decreased as tadpole stage increased. For tadpoles ex-
posed to cercariae at stages 25–26, infection success also de-
clined with increasing exposure dose (Fig. 3; Table II).
Ten of 23 metamorphic frogs (43.5%) that were exposed to
echinostome cercariae became infected. About 85% of the in-
dividuals exposed to 2 daily doses of 150 cercariae each be-
came infected, compared with only 25% of the individuals ex-
posed to a single dose of more than 300 cercariae. In addition,
more cercariae successfully encysted in frogs when adminis-
tered by the 2-day, 150-dose regimen, than the single, high-
dose method. A total of 2,100 cercariae was exposed to 7 frogs
using the 2-day, 150-dose method; 108 of these cercariae
(5.1%) became encysted. In contrast, only 37 of the estimated
18,170 cercariae (0.2%) exposed to 16 frogs at single doses
ranging between 360 and 2,220 (average, 1,136) cercariae suc-
cessfully infected frogs. All metacercariae in the metamorphic
frogs were found in the mesonephroi.
DISCUSSION
In the present study, only tadpoles at stage 25, the earliest
staged tadpoles exposed to echinostome cercariae in the exper-
iments, experienced mortality associated with infections. The
degree to which tadpoles at this stage were affected, further-
more, was a function of exposure dose (Table I), which is con-
sistent with the previous report by Fried et al. (1997) that E.
trivolvis infections caused intensity-dependent mortality in ear-
ly-staged R. pipiens (e.g., stages 24–25; Shumway, 1940). This
stage-specific mortality can be explained to some extent by the
relationship that was observed between the location of cercariae
encystment and stage of kidney development. Cercariae pre-
dominately encysted in the pronephroi when tadpoles were at
early stages of development and in the mesonephroi when tad-
poles were approaching the end of metamorphosis (Fig. 2). Lo-
cation of cercariae encystment, therefore, corresponded to the
progression of kidney development in R. pipiens. Because tad-
poles at early stages of development rely primarily on the pro-
nephros for osmoregulatory and excretory functions (Jaffee,
1954a, 1954b), the mortality observed at stage 25 was probably
related to the inability of tadpoles at this early stage of devel-
opment to tolerate disruption of renal function associated with
the concentration of metacercariae in this organ. Tadpoles
quickly became tolerant of infections, however, as tadpoles at
stage 27, which are about 2 wk older than tadpoles at stage 25,
survived infections (Table I). Tadpoles grow substantially be-
tween these stages of development (average weight
6
SD, stage
25, 0.02
6
0.013 g; stage 27, 0.22
6
0.049 g), and presumably
have higher renal capacity associated with having larger pro-
nephroi (Viertel and Richter, 1999). Therefore, it appears that
tadpoles are at greatest risk of dying from echinostome infec-
tions during the period of development when renal capacity is
restricted to the pronephroi and constrained by small tadpole
size.
This observation that location of cercariae encystment de-

480 THE JOURNAL OF PARASITOLOGY, VOL. 89, NO. 3, JUNE 2003
pended largely on the developmental stage of the kidneys in
tadpoles, suggests that the difference in the locations of echi-
nostome encystment in Rana sp. tadpoles previously reported
by Thiemann and Wassersug (2000b) occurred because of the
different staged tadpoles used in their study. The average stages
of R. clamitans and R. sylvatica were Gosner (1960) 26 and 34,
respectively. In R. clamitans, the majority of metacercariae
were found in the head kidneys (
5
pronephroi), whereas in R.
sylvatica, most of the metacercariae were in the opisthonephric
kidneys (
5
mesonephroi). Therefore, their results were consis-
tent with our observations that cercariae predominately encyst
in the pronephroi of early tadpole stages and the mesonephroi
of later tadpole stages.
The shift in the location of cercariae encystment from the
pro- to the mesonephroi with development of the mesonephroi,
suggests that echinostome cercariae are responding to factors
associated with a physiologically active kidney, which is rea-
sonable given the other conditions under which Echinostoma
sp. and Echinoparyphium sp. cercariae encyst. In addition to
amphibian kidneys, these echinostome cercariae will encyst in
the kidneys of fish (Beaver, 1937) and have a high predilection
for the renal structures of snails (Najarian, 1954; Anderson and
Fried, 1987). Encystment, however, has also been reported in
other organs of snails (Najarian, 1954; Anderson and Fried,
1987), on snail mucus trails, and in vitro in Locke’s 1:1 solution
(Fried and Bennett, 1979), suggesting that factors that may be
encountered outside of renal systems are also involved. The
simple presence of trigger stimuli associated just with the kid-
neys, thus does not appear sufficient to induce echinostome cer-
cariae encystment, as was proposed for parasite site location
behaviors by Sukhdeo and Sukhdeo (1994). Cercariae contin-
ued to migrate to the pronephroi after the mesonephroi are func-
tioning in tadpoles of stages 31–32, even though the meso-
nephroi should be encountered before the pronephroi upon en-
tering the tadpole through the cloaca (Najarian, 1953). It may
be that encystment occurs only after a certain threshold of a
trigger stimulus has been encountered or that the response is
mediated by other factors, such as cues from other metacercar-
iae or density-dependent factors. For instance, more cercariae
encysted in regions outside of the pronephroi, specifically, in
regions where the mesonephroi would develop, as tadpoles of
stages 25–26 were exposed to higher doses of cercariae (Fig.
2). This pattern suggests that cercariae preferentially encyst in
the pronephroi of early tadpole stages but are able to encyst in
other locations if high densities of metacercariae are already
present in the pronephroi.
Thiemann and Wassersug (2000b) also reported that the dis-
tribution of echinostome metacercariae was biased in favor of
the right kidneys. In our study, cercariae encystment also ap-
peared to occur more frequently on 1 side of the tadpole than
the other, although the predominant side of encystment varied
across tadpole stages. In tadpoles exposed at stages 25–26, the
majority of metacercariae (35.7%) were found in the right pro-
nephros, whereas in tadpoles at stages 31–32, metacercariae
were most commonly found in the left pronephros (39.4%). In
tadpoles at stages 37–38, most metacercariae (55.4%) were in
the right mesonephros or nephric duct. An asymmetric distri-
bution of metacercariae may be an adaptive response that limits
the impacts of infections by sequestering metacercariae in only
1 portion of the nephric system or by allowing tadpoles to elim-
inate infections in the pronephroi after metamorphosis (Thie-
mann and Wassersug, 2000b). However, the degree of asym-
metry was poorly related to the probability of developing ede-
ma in our experiment (logistic regression,
x
2
5
2.22, P
5
0.136), and regardless of degree of asymmetry, the frequency
of edema increased with exposure dose (
x
2
5
5.54, P
5
0.019).
Furthermore, live metacercariae were found in the pronephroi
regions of frogs that had completed metamorphosis, suggesting
that degeneration of the pronephroi does not kill metacercariae.
More detailed investigations into the responses of cercariae to
the stage of kidney development and mechanisms involved in
stimulating encystment behavior would likely help explain the
asymmetric patterns of encystment.
Interestingly, our experiments also revealed that as tadpoles
age, they become less susceptible to echinostome infections.
Significantly, fewer cercariae successfully encysted in later tad-
pole stages than in early tadpole stages (Fig. 3), even though
later tadpole stages were in containers with smaller volumes
relative to their body size than were tadpoles in earlier stages
of development. Therefore, the extent of contact between cer-
cariae and tadpoles, as potentially influenced by the size of our
experimental containers, cannot explain this relationship be-
tween infection success and tadpole stage. On the contrary, our
observation suggests that the ability of cercariae to enter, en-
cyst, or both, declines as tadpoles age. Tadpoles often violently
shake their tails and accelerate their swimming speed upon con-
tact with cercariae (Thiemann and Wassersug, 2000a; A.
Schotthoefer, unpubl. data). As a result, older tadpoles, which
are larger, may have a greater ability to dislodge or escape from
cercariae than young tadpoles (A. Taylor and A. Wassersug,
pers. comm.). Alternatively, because the immune system of old-
er tadpoles is more developed (Du Pasquier et al., 1996), they
may have a greater ability to effectively kill encysting cercariae
than young tadpoles.
Despite this decline in infection success with tadpole stage,
frogs remained susceptible to infection after metamorphosis,
suggesting that both tadpoles and juvenile frogs are suitable
second intermediate hosts of echinostomes, even though the in-
fection success for metamorphic frogs was low (0.7%). It re-
mains possible that adult frogs develop immunologic resistance
to infections (Martin and Conn, 1990) or have the ability to
detect and respond behaviorally to the presence of cercariae
(e.g., Kiesecker and Skelly, 2000), which might reduce their
likelihood of becoming infected. It is clear, however, that phys-
iological changes that accompany metamorphosis do not totally
preclude echinostome cercariae from successfully encysting in
kidneys of R. pipiens.
With regard to the potential ecological consequences of echi-
nostome infections in tadpole populations, the observed stage-
specific pattern of tadpole mortality leads to the prediction that
tadpole populations composed largely of early-staged tadpoles
are at greatest risk of being negatively affected by echino-
stomes. As indicated by the survival experiment, infections ac-
quired early during development could result in 80–100% mor-
tality (Table I). Such mortality events in nature could severely
reduce the size of tadpole populations or result in the local
extinction of distinct cohorts. When amphibian life histories are
considered in relation to the seasonal dynamics of echinostome
infections in snail populations, however, it is not clear if tad-
poles at such critical stages of development are likely to be at

SCHOTTHOEFER ET AL.—ECHINOSTOME INFECTIONS IN TADPOLES 481
high risks of becoming infected in nature. Field investigations
have revealed that prevalence of echinostome infections in wild
snail populations tends to be low during the spring and increas-
es throughout the summer (Rosen et al., 1994; Sapp and Esch,
1994; Schmidt and Fried, 1997). Consequently, amphibian spe-
cies that breed early in the season and have shorter larval pe-
riods would likely experience lower risks of echinostome in-
fections than species that breed later and have longer larval
periods. For R. pipiens, breeding occurs from March to May
(Conant and Collins, 1998), and juveniles typically emerge
from ponds 3 to 4 mo later (Merrell, 1977). Thus, the stages of
tadpoles that are at the greatest risk of dying from infections,
i.e., stage 25, may not be the same stages of tadpoles that are
at the greatest risk of exposure to echinostome cercariae.
Differences in amphibian life histories with respect to timing
of breeding and the seasonal prevalence of infections in snails
also might help explain the observation by Thiemann and Was-
sersug (2000a) that R. clamitans tadpoles were less susceptible
to laboratory infections than R. sylvatica tadpoles. Rana clam-
itans breeds later in the season (April–August), and larvae can
take over a year to develop, in contrast to R. sylvatica, which
is an early (March–April) and explosive breeder, and have tad-
poles that develop to frogs within 3 mo (Conant and Collins,
1998). Being at greater risk of exposure to echinostome cer-
cariae historically could have promoted the development of re-
sistance (behavioral or immunologic) against infection in R.
clamitans, as proposed by Thiemann and Wassersug (2000a).
Overall, our results, in addition to those reported by Thie-
mann and Wassersug (2000a, 2000b), raise questions regarding
the role of parasitism in the evolution of frog life histories, as
well as what effects current environmental changes might be
having on echinostome–tadpole interactions. As suggested by
our survival experiment, infections have the potential to nega-
tively impact tadpole populations. However, tadpoles appear to
develop tolerance to infections fairly early in development and,
therefore, the relevance of infections to the success of tadpole
populations in nature still must be investigated in the contexts
of amphibian life histories, timing, inoculum associated with
infections, and the stage-specific susceptibilities of tadpoles to
infection. Moreover, transmission success of echinostome cer-
cariae to tadpoles will likely be influenced by environmental
and anthropogenic factors, in addition to tadpole stage. For in-
stance, the presence of a predator increased the risk of infection
in R. clamitans tadpoles (Thiemann and Wassersug, 2000a), and
the removal of aquatic submergent vegetation by herbicides co-
incided with higher prevalences of severe infections in A. cre-
pitans populations (Beasley et al., 2003). In addition, factors
that increase the time to metamorphosis or impair tadpole
growth, such as exposure to environmental contaminants (Diana
et al., 2000; Boone et al., 2001; Boone and Semlitsch, 2002)
or interactions with competitors (Holomuzki and Hemphill,
1996; Lefcort et al., 1999), have the potential to change the
degree to which tadpole populations are affected by echino-
stome infections. The current study underscores the need to
identify conditions in nature under which periods of high-in-
fection risk overlap with highly susceptible developmental stag-
es of tadpoles or are such that infection intensities become high
enough to decrease the survival of less susceptible tadpoles.
ACKNOWLEDGMENTS
We thank Anson Koehler for help in collecting and caring for snails,
Jon Ingram for assistance with metamorphic frog care and exposures,
Melinda Brady for statistical advice, and Uriel Kitron, Scott Wright,
Christine Bunck, and 2 anonymous reviewers for their comments on
drafts of the manuscript. This research was supported by funding from
an EPA grant R825867 to V. R. Beasley, a Sigma Xi Grant-in-Aid of
Research Award to A. M. Schotthoefer, and the United States Geolog-
ical Survey. Although the research described was funded in part by the
EPA, it has not been subjected to any EPA review and therefore does
not reflect the views of the Agency, and no official endorsement should
be inferred.
LITERATURE CITED
A
NDERSON
,J.W.,
AND
B. F
RIED
. 1987. Experimental infection of Physa
heterostropha, Helisoma trivolvis, and Biomphalaria glabrata
(Gastropoda) with Echinostoma revolutum (Trematoda) cercariae.
Journal of Parasitology 73: 49–54.
B
ARINAGA
, M. 1990. Where have all the froggies gone? Science 247:
1033–1034.
B
EASLEY
, V. R., S. A. F
AEH
,B.W
IKOFF
,J.E
ISOLD
,D.N
ICHOLS
,R.C
OLE
,
A. M. S
CHOTTHOEFER
,C.S
TAEHLE
,M.G
REENWELL
,
AND
L. E.
B
ROWN
. 2003. Risk factors and the decline of the cricket frog, Acris
crepitans: Evidence for involvement of herbicides, parasitism, and
habitat modifications. In Status and conservation of U.S. amphib-
ians, M. J. Lannoo (ed.). University of Chicago Press, Chicago,
Illinois. [In press.]
B
EAVER
, P. C. 1937. Experimental studies with Echinostoma revolutum
(Froelich), a fluke from birds and mammals. Illinois Biological
Monographs 34: 1–96.
B
OONE
, M. D., C. M. B
RIDGES
,
AND
B. B. R
OTHERMEL
. 2001. Growth
and development of larval green frogs (Rana clamitans) exposed
to multiple doses of an insecticide. Oecologia 129: 518–524.
———,
AND
R. D. S
EMLITSCH
. 2002. Interactions of an insecticide with
competition and pond drying in amphibian communities. Ecologi-
cal Applications 12: 307–316.
C
ONANT
, R.,
AND
J. T. C
OLLINS
. 1998. A field guide to reptiles and
amphibians: Eastern and central North America, 2nd ed., expanded.
Houghton Mifflin Co., New York, New York, 616 p.
D
IANA
, S. G., W. J. R
ESETARITS
J
R
., D. J. Schaeffer, K. B. Beckmen,
and V. R. Beasley. 2000. Effects of atrazine on amphibian growth
and survival in artificial aquatic communities. Environmental Tox-
icology and Chemistry 19: 2961–2967.
D
U
P
ASQUIER
, L., M. W
ILSON
,
AND
J. R
OBERT
. 1996. The immune system
of Xenopus: With special reference to B cell development and im-
munoglobulin genes. In The biology of Xenopus, R. C. Tinsley and
H. R. Kobel (eds.). Oxford University Press, Oxford, U.K., p. 301–
313.
F
OX
, H. 1963. The amphibian pronephros. The Quarterly Review of
Biology 38: 1–25.
F
RIED
, B. 1985. Maintenance of Echinostoma revolutum (Trematoda) in
the laboratory. Proceedings of the Pennsylvania Academy of Sci-
ence 59: 27–28.
———,
AND
M. C. B
ENNETT
. 1979. Studies on encystment of Echinos-
toma revolutum cercariae. Journal of Parasitology 65: 38–40.
———, B. A. F
RAZER
,
AND
I. K
ANEV
. 1998. Comparative observations
on cercariae and metacercariae of Echinostoma trivolvis and Echin-
oparyphium sp. Journal of Parasitology 84: 623–626.
———, P. L. P
ANE
,
AND
A. R
EDDY
. 1997. Experimental infection of
Rana pipiens tadpoles with Echinostoma trivolvis cercariae. Para-
sitology Research 83: 666–669.
G
OSNER
, K. L. 1960. A simplified table for staging anuran embryos and
larvae with notes on identification. Herpetologica 16: 183–190.
H
OLOMUZKI
,J.R.,
AND
N. H
EMPHILL
. 1996. Snail-tadpole interactions in
streamside pools. American Midland Naturalist 136: 315–327.
J
AFFEE
, O. C. 1954a. Morphogenesis of the pronephros of the Leopard
frog (Rana pipiens). Journal of Morphology 95: 109–123.
———. 1954b. Phenol red transport in the pronephros and mesoneph-
ros of the developing frog (Rana pipiens). Journal of Cellular and
Comparative Physiology 44: 347–363.
J
OHNSON
, P. T. J., K. B. L
UNDE
,E.G.R
ITCHIE
,
AND
A. E. L
AUNER
. 1999.

482 THE JOURNAL OF PARASITOLOGY, VOL. 89, NO. 3, JUNE 2003
The effect of trematode infection on amphibian limb development
and survivorship. Science 284: 802–804.
———, K. B. L
UNDE
,E.M.T
HURMAN
,E.G.R
ITCHIE
,S.N.W
RAY
,D.
R. S
UTHERLAND
,J.M.K
APFER
,T.J.F
REST
,J.B
OWERMAN
,
AND
A.
R. B
LAUSTEIN
. 2002. Parasite (Ribeiroia ondatrae) infection linked
to amphibian malformations in the western United States. Ecolog-
ical Monographs 72: 151–168.
K
IESECKER
, J. M.,
AND
D. K. S
KELLY
. 2000. Choice of oviposition site
by gray treefrogs: The role of potential parasitic infection. Ecology
81: 2939–2943.
K
OSTADINOVA
, A.,
AND
D. I. G
IBSON
. 2000. The systematics of the echi-
nostomes. In Echinostomes as experimental models for biological
research, B. Fried and T. K. Graczyk (eds.). Kluwer Academic Pub-
lishers, Dordrecht, The Netherlands, p. 31–57.
L
EFCORT
, H., S. M. T
HOMSON
,E.E.C
OWLES
,H.L.H
AROWICZ
,B.M.
L
IVAUDAIS
,W.E.R
OBERTS
,
AND
W. F. E
TTINGER
. 1999. Ramifica-
tions of predator avoidance: Predator and heavy-metal-mediated
competition between tadpoles and snails. Ecological Applications
9: 1477–1489.
M
ARTIN
,T.R.,
AND
D. B. C
ONN
. 1990. The pathogenicity, localization,
and cyst structure of echinostomatid metacercariae (Trematoda) in-
fecting the kidneys of the frogs Rana clamitans and Rana pipiens.
Journal of Parasitology 76: 414–419.
M
C
A
LPINE
,D.F.,
AND
D. B. B
URT
. 1998. Helminths of bullfrogs, Rana
catesbeiana, green frogs, R. clamitans, and leopard frogs, R. pi-
piens in New Brunswick. Canadian Field-Naturalist 112: 50–68.
M
C
C
LURE
, C. F. 1919. On the experimental production of edema in
larval and adult anura. Journal of General Physiology 1: 261–267.
———. 1928. On the structure of the pronephros in oedematous larvae
of anura. Anatomical Record 39: 349–358.
M
ERRELL
, D. J. 1977. Life history of the leopard frog, Rana pipiens,in
Minnesota. Bell Museum of Natural History. Occasional Papers 15:
1–22.
N
AJARIAN
, H. H. 1953. The entrance of the cercaria of Echinoparyphium
flexum (Linton) Dietz, 1910, into tadpole kidneys. Journal of Par-
asitology 39(Suppl.): 22.
———. 1954. Developmental stages in the life cycle of Echinopary-
phium flexum (Linton, 1892) Dietz, 1910 (Trematoda: Echinosto-
matidae). Journal of Morphology 94: 165–198.
———. 1955. Trematodes parasitic in the Salientia in the vicinity of
Ann Arbor, Michigan. American Midland Naturalist 53: 195–197.
N
ETER
, J., W. W
ASSERMAN
,
AND
M. H. K
UTNER
. 1990. Applied linear
statistical models: Regression, analysis of variance, and experimental
designs. Richard D. Irwin, Inc. Homewood, Illinois, 1181 p.
P
RUDHOE
, S.,
AND
R. A. B
RAY
. 1982. Platyhelminth parasites of the
Amphibia. Oxford University Press, Oxford, U.K., 217 p.
R
OSEN
, R. B., J. M. I
LAGAN
,J.S.L
AW
,M.A
SUNCION
,M.E.D
ENTON
,
AND
M. L. S
AN
. 1994. Seasonal prevalence of three species of di-
genetic trematodes in the snail Helisoma trivolvis at Owsley Fork
Reservoir, Kentucky. Transactions of Kentucky Academy of Sci-
ences 55: 32–35.
S
APP
, K. K.,
AND
G. W. E
SCH
. 1994. The effects of spatial and temporal
heterogeneity as structuring forces for parasite communities in Hel-
isoma anceps and Physa gyrina. American Midland Naturalist 132:
91–103.
S
CHMIDT
, K. A.,
AND
B. F
RIED
. 1997. Prevalence of larval trematodes
in Helisoma trivolvis (Gastropoda) from a farm pond in Northamp-
ton County, Pennsylvania, with special emphasis on Echinostoma
trivolvis (Trematoda) cercariae. Journal of the Helminthological So-
ciety of Washington 64: 157–159.
S
HUMWAY
, W. 1940. Stages in the normal development of Rana pipiens.
Anatomical Record 78: 139–147.
S
UKHDEO
,M.V.K.,
AND
S. C. S
UKHDEO
. 1994. Optimal habitat selection
by helminths within the host environment. Parasitology 109: S41–
S55.
T
HIEMANN
,G.W.,
AND
R. J. W
ASSERSUG
. 2000a. Patterns and conse-
quences of behavioural responses to predators and parasites in Rana
tadpoles. Biological Journal of the Linnean Society 71: 513–528.
———,
AND
———. 2000b. Biased distribution of trematode metacer-
cariae in the nephric system of Rana tadpoles. Journal of Zoology,
London 252: 534–538.
U
LMER
, M. J. 1970. Studies on the helminth fauna of Iowa I. Trematodes
of amphibians. American Midland Naturalist 83: 38–64.
V
IERTEL
, B.,
AND
S. R
ICHTER
. 1999. Anatomy: Viscera and endocrines.
In Tadpoles: The biology of anuran larvae, R. W. McDiarmid, and
R. Altig (eds.). University of Chicago Press, Chicago, Illinois, p.
92–148.
W
RIGHT
, K., L. B
ERGER
,D.K.N
ICHOLS
,R.S
PEARE
,M.J.S
REDL
,A.
P
ESSIER
,
AND
B. J
OHNSON
. 2001. Amphibian population decline.
Journal of Herpetological Medicine and Surgery 11: 14–27.
Z
AR
, J. H. 1996. Biostatistical analysis. Prentice Hall, Inc., Upper Sad-
dle River, New Jersey, 662 p.