EcoHealth

, Volume 6, Issue 4, pp 565–575

Impacts of Batrachochytrium dendrobatidis Infection on Tadpole Foraging Performance

  • Matthew D. Venesky
  • Matthew J. Parris
  • Andrew Storfer
Original Contribution

DOI: 10.1007/s10393-009-0272-7

Cite this article as:
Venesky, M.D., Parris, M.J. & Storfer, A. EcoHealth (2009) 6: 565. doi:10.1007/s10393-009-0272-7

Abstract

Pathogen-induced modifications in host behavior, including alterations in foraging behavior or foraging efficiency, can compromise host fitness by reducing growth and development. Chytridiomycosis is an infectious disease of amphibians caused by the fungus Batrachochytrium dendrobatidis (Bd), and it has played an important role in the worldwide decline of amphibians. In larval anurans, Bd infections commonly result in reduced developmental rates, however, the mechanism(s) responsible are untested. We conducted laboratory experiments to test whether Bd infections reduced foraging performance of Grey Treefrog (Hyla chrysoscelis) and Fowler’s Toad (Anaxyrus [= Bufo] fowleri) tadpoles. In the first experiment, we observed foraging behavior of Bd-infected and uninfected tadpoles to test for differences in foraging activity. In a second experiment, we tested for differences in the ingestion rates of tadpoles by examining the amount of food in their alimentary track after a 3-hour foraging period. We hypothesized that Bd-infected tadpoles would forage less often and less efficiently than uninfected tadpoles. As predicted, Bd-infected larvae forage less often and were less efficient at obtaining food than uninfected larvae. Our results show that Bd infections reduce foraging efficiency in Anaxyrus and Hyla tadpoles, and that Bd differentially affects foraging behavior in these species. Thus, our results provide a potential mechanism of decreased developmental rates of Bd-infected tadpoles.

Keywords

Anaxyrus fowleriBufochytridiomycosisforaging behaviorHyla chrysoscelispathogen

Introduction

There is increasing theoretical and empirical evidence of pathogen-induced modifications in host behavior (Altizer et al., 2003; Kluger et al., 1998; Moore, 2002). Pathogens can reduce survival in host species by directly causing mortality (Daszak et al., 1999; de Castro and Bolker, 2005) or indirectly by promoting behaviors that decrease host fitness while simultaneously increasing pathogen transmission (Dobson, 1988; Tierney et al., 1993). For example, snails (Potamopyrgus antipodarum) infected with the parasitic trematode (Microphallus) spend more time on rocks than uninfected snails, which leads to reduced foraging efficiency for snails and greater parasite transmission to the definitive host (Levri, 1999). In addition to causing mortality in host species, pathogen-induced changes in host behavior may negatively affect host life history traits, including growth and development (Forson and Storfer, 2006; Goater, 1994), and can alter population or community dynamics by disrupting intra- or interspecific interactions (Kohler and Wiley, 1997; Parris and Cornelius, 2004). Ultimately, pathogen impacts on host mortality or life history can reduce host fitness and even regulate host population size (Anderson and May, 1978).

One pathway by which pathogens alter host life history traits is through behavioral changes involving reductions in foraging performance (Levri and Lively, 1996). When pathogens reduce the foraging ability of individuals, host fitness may be compromised despite a sublethal infection (Gunn and Irvine, 2003). Although host species’ behavioral responses to pathogens may be similar, the mechanisms behind reduced foraging performance may differ. For example, pathogens can directly reduce foraging performance by decreasing the host’s foraging efficiency through impaired visual performance (Crowden and Broom, 1980), a reduced stomach capacity (Wright et al., 2006), or damage to foraging structures (Drake et al., 2007). Pathogens can also indirectly reduce foraging performance of hosts by reducing their activity levels so that infected individuals forage less often than uninfected individuals. For example, bumble bees (Bombus impatiens) infected with an intestinal protozoan (Crithidia bombi) forage less often than uninfected bees (Ottershatter et al., 2005).

The role that pathogens play in amphibian populations and communities is especially important considering worldwide amphibian declines (Daszak et al., 2003; Lips et al., 2006). Although the loss of amphibian species are associated with multiple factors (Collins and Storfer, 2003; Dodd and Smith, 2003; Pounds et al., 2006), evidence is accumulating for the role that infectious diseases play in population regulation (Daszak et al., 2003; de Castro and Bolker, 2005). Chytridiomycosis is an emerging infectious disease of amphibians caused by the pathogenic fungus Batrachochytrium dendrobatidis (Bd), which has played a major role in the decline of amphibian populations over the past two decades (Berger et al., 1998; Daszak et al., 2003). Bd infects keratinizing tissue, or those tissues fated to keratinize (Fellers et al., 2001), and colonization of Bd in tadpoles is restricted to the keratinized mouthparts. Although Bd does not generally cause mortality in anuran tadpoles, Bd often reduces growth and developmental rates (Garner et al., 2009; Parris, 2004; Parris and Baud, 2004; Parris and Cornelius 2004; but see Smith et al., 2007). However, the mechanism(s) underlying this result are untested. Rapid growth and development is critical for survival of the majority of amphibian species which are dependent on temporary aquatic habitats (Werner and Anholt, 1993; Wilbur and Collins, 1973). That is, reaching a minimum size for metamorphosis before pond drying increases the probability of successfully metamorphosing (Kiesecker and Skelly, 2001; Wilbur and Collins, 1973). Reductions in growth and development can also compromise larval fitness by decreasing the ability to escape gape-limited predators (Kurzava, 1998), and increasing intraspecific competition (Semlitsch et al., 1988).

Given the potential impacts that host behavior has on the disease transmission process, understanding the effects of Bd on amphibians with different behaviors and life histories is a key component in understanding the etiology of Bd. Between amphibian life stages (i.e., tadpoles and metamorphs), there are apparent differences in the pathogenicity of Bd (Carey et al., 2006; Parris and Cornelius, 2004). In the larval stage of anurans, there is interspecific variation in susceptibility to Bd (Blaustein et al., 2005; Woodhams and Alford, 2005), and Bd induced changes in growth and developmental rates (Parris and Cornelius, 2004), however, it is unknown whether Bd differentially affects species’ behaviors. If differing host foraging strategies or morphologies are differentially impacted by Bd infections, the overall impact of Bd may be species specific. Intraspecific effects of Bd also remain unclear; specifically, little is known as to how Bd affects tadpoles through ontogeny (but see Smith et al., 2007). For example, natural ponds often contain many species of amphibian larvae in various developmental stages due to variation in timing of the breeding season. Given the evidence that tadpoles respond to environmental stressors differently during ontogeny (e.g., Audo et al., 1995), disease may influence larval performance at some developmental stages more than others. To elucidate the mechanism for reduced growth rates observed in Bd-infected tadpoles, we conducted a series of laboratory experiments investigating the foraging behavior and foraging efficiency of anuran tadpoles through ontogeny. We used two species of tadpoles—Grey Treefrogs (Hyla versicolor) and Fowler’s Toads (Anaxyrus [= Bufo] fowleri). These two species are ideal focal taxa because they differ in behaviors associated with how and where they forage within the water column of ponds. Generally, Anaxyrus tadpoles aggregate in small groups and scrape food off the substrate of ponds (Beiswenger, 1977), whereas Hyla tadpoles forage actively from the middle portion of the water column (Wilbur and Alford, 1985). These differences in behavior are a critical component of our research design because differences in foraging behavior may differentially impact host fitness.

In the first experiment, we made repeated observations of foraging behavior at three stages during larval development in Bd-infected and uninfected A. fowleri and H. chrysoscelis tadpoles to test if Bd reduced foraging activity. In the second experiment, we tested the effects of Bd on foraging efficiency through development. We hypothesized that Bd-infected tadpoles would forage less often and less efficiently than uninfected tadpoles. If Bd differentially affects foraging performance at different stages in larval ontogeny, we also hypothesized that the effects of Bd on foraging performance would vary according to developmental stage.

Materials and Methods

Animal Collection and Husbandry

Anaxyrus fowleri and H. chrysoscelis eggs were collected from natural ponds within the Meeman Shelby State Park (in Shelby County, Tennessee, USA). All tadpoles used in our experiments were derived from field-collected egg masses of each species. On April 25, 2008, we collected five A. fowleri egg masses from three ponds; on May 2, 2008, we collected six H. chrysoscelis egg masses from two ponds. Immediately after collection, eggs were transported to the laboratory at The University of Memphis. Upon hatching, tadpoles were maintained in 4.2-l plastic containers at a density of five tadpoles/l for 2 days. After this period, we removed all tadpoles that had reached the free-swimming stage (stage 25; Gosner, 1960) and combined the tadpoles from the different clutches to evenly distribute potential genetic effects on the larval traits we measured. We then randomly selected our test subjects out of the remaining stock of tadpoles. Test subjects were placed individually in 1.5-l plastic containers filled with 1 l of aged tap water. Throughout the experiment, tadpoles were maintained on a 12 h light: 12 h dark photoperiod at 19°C (±1°C) and were fed a mixture of ground rabbit chow and Sera Micron® (Sera, Heinsburg, Germany) daily.

Full water changes were conducted once per week. We took the following precautions to avoid accidental Bd transmission between treatments. For each water change, we used a piece of mesh screen to remove the focal tadpole from its container and placed it in a temporary transfer container with the same dimensions and volume of water as the experimental container. Water from the experimental container was poured into a bucket and 1.0 l of aged tap water was then added to that container. The focal tadpole was then placed back into the experimental container. During each water change, an experimenter wore latex gloves to prevent the transmission of Bd. To minimize the possibility of Bd transmission between Bd-exposed and nonexposed (control) treatments, we always performed water changes on nonexposed tadpoles before exposed tadpoles. Additionally, we used different laboratory equipment (gloves, transfer container, mesh screen) between each developmental stage in both of the pathogen treatments. At the end of the experiment, we thoroughly disinfected all containers by adding bleach (6% sodium hypochlorite) to yield a 10% solution, which kills Bd (Johnson and Speare, 2003). Throughout the experiment, all equipment and water was disinfected in a similar fashion.

Batrachochytrium dendrobatidis Inoculation

Bd was grown in the laboratory on tryptone-gelatin hydrolysate-lactose (TGhL) agar in 9-cm Petri dishes according to standard protocol (Longcore et al., 1999). We harvested Bd zoospores by adding 10.0 ml of sterile water to the cultures and collected the zoospores that emerged from the zoosporangia after 30 min. At Gosner 25 (Day 0), all tadpoles of each species were randomly split into two experimental groups—a Bd-exposed group and a nonexposed (control) group. For the Bd-exposed group, we administered Bd by exposing A. fowleri (N = 40) and H. chrysoscelis (N = 37) to water baths containing infectious concentrations of fungal zoospores. Tadpoles were placed individually in 50-ml waterbaths and exposed to an infectious concentration of zoospores (2500 zoospores/ml) for 24 hours. For the nonexposed group, we followed the same protocol but used plates with only TGhL, and exposed an additional group of tadpoles (A. fowleri, N = 40; H. chrysoscelis, N = 37) to water baths with no Bd zoospores. Our design simulated transmission by water, one of the possible modes of Bd transmission in natural environments (Pessier et al., 1999). Six days after exposure to Bd, we conducted the two foraging experiments which ran concurrently.

Experiment 1: Tadpole Foraging Behavior

In Experiment 1, we tested whether Bd infections negatively affect foraging behavior of A. fowleri and H. chrysoscelis tadpoles through ontogeny. We made repeated observations of the foraging behavior of Bd-exposed (N = 10, per species) and nonexposed (N = 10, per species) tadpoles at three developmental stages—early (Gosner 28–29), middle (Gosner 30–33), and late (Gosner 36–38). At each developmental stage, we made a series of three behavioral observations of foraging activity on each individual on 3 consecutive days. In total, nine behavioral observations were made on each larva. On each observation date, the order of the observations was randomized to minimize the potential for any diel effects on tadpole foraging behavior. In addition, the infection status of each larva was unknown during the behavioral observations.

Prior to the start of the behavioral trial, the focal container of each test subject was removed from a laboratory shelf and placed on an observation table; 0.1 grams of the rabbit chow/SeraMicron® mixture was dispensed into the water and then the test subject was allowed to acclimate for 3 minutes to any disturbances associated with the movement of the container. After the acclimation period, an experimenter observed the foraging behavior of the test subject during a 12-minute behavioral trial. Each experimental trial was divided into 20-second intervals, and we recorded whether the test subject foraged at any point during the 20-second interval. The proportion of intervals during which the test subject foraged during the 12-minute experimental trial was calculated as an estimate of foraging activity. Similar measures have been used in other larval anuran behavior experiments, and are accurate indices of overall activity level (Han et al., 2008; Parris et al., 2006).

Because of the differences in foraging behavior and mode among the two study species, we conducted a series of preliminary observations on uninfected tadpoles to qualify foraging behavior in our experiment. In general, Anaxyrus tadpoles foraged exclusively on food that sank to the floor of the experimental container and foraged with their bodies parallel to the floor of the container. In contrast, Hyla generally foraged on food that remained floating on the surface of the water with their bodies’ perpendicular to the floor of the container. However, some H. chrysoscelis tadpoles foraged on large pieces of food that had sunk to the floor of the container. For both species, we qualified the test subject as foraging if it performed either of these behaviors. In addition, we qualified these behaviors as foraging only when the test subject was in close proximity to food in order to avoid erroneous classification of breathing as foraging behavior. A preliminary analysis of variance (ANOVA) revealed no significant differences in the percentage of time spent foraging between each of the three trial dates at each developmental stage (P > 0.100 in all cases). Accordingly, for each developmental stage, we took the grand mean of the three replicate trials and analyzed those data with repeated measures ANOVA. For each species, we used two-way repeated measures ANOVA to test for an effect of pathogen treatment (Bd-exposed or control) and developmental stage (early: Gosner 28–29; middle: Gosner 30–33; and late: Gosner 36–38) on tadpole foraging behavior. When appropriate, we used Holm–Sidak post hoc analyses. All statistical analyses were performed in SPSS (version 17; SPSS, Inc., Chicago, IL).

Experiment 2: Tadpole Foraging Efficiency

In Experiment 2, we tested whether Bd infections negatively affect foraging efficiency of A. fowleri and H. chrysoscelis tadpoles throughout ontogeny. We assessed the short-term foraging efficiency of Anaxyrus (N = 30; 10 at each developmental stage) and Hyla (N = 27; 9 at each developmental stage) tadpoles by examining the quantity of food consumed during one 3-hour trial. Test subjects in this experiment were different than the subjects used in Experiment 1, although both experiments ran concurrently. Trials were conducted at the same developmental stages as those used in Experiment 1—early (Gosner 28–29), middle (Gosner 30–33), and late (Gosner 36–38). Each subject was sacrificed immediately following the trial, therefore, different individuals were used at each of the three developmental stages. As in Experiment 1, the infection status of each larva was unknown to the experimenter during data collection.

Prior to the start of the foraging efficiency trial, tadpoles were fasted for 2 days to empty the intestine. In addition, we checked the focal containers periodically during the 2 days prior to the experiment to remove any fecal matter, which could be a food source for tadpoles. A preliminary experiment with H. versicolor tadpoles confirmed this method of emptying the intestine [M. Venesky, unpublished data]. At the start of the trial, we placed 0.30 g of cultured Anabaena in the container of each test subject. At the end of the trial, each test subject was removed from the container, sacrificed, and stored in 70% EtOH. All tadpoles were dissected on the same date of the experiment to accurately quantify Anabaena consumption during the trial.

To quantify the amount of Anabaena ingested during the experimental trial, we dissected and straightened the entire intestine on a dissection pan. We measured the length of the intestine, to the nearest 0.05 mm, with calipers. We also measured the diameter (mm) of the intestine in three locations—the midpoint, and the anterior and posterior ends. We took the mean of the three diameter measurements and then estimated the total volume of the intestine. Anabaena is green and provides a sharp contrast to an empty intestine, allowing us to calculate the percentage of the intestine filled with food that was consumed during the 3-hour trial. At each developmental stage, we used independent samples t-tests to assess species-specific foraging responses to Bd. In each analysis, we used the percentage of intestine with food as the independent variable and the pathogen treatment (Bd-exposed or control) as the dependent variable. Our data met all assumptions of parametric statistics. All statistical analyses were performed in SPSS (version 17).

Confirmation of Batrachochytrium dendrobatidis Infections

We confirmed the infection status of all Bd-exposed and control tadpoles used in both experiments using real-time quantitative polymerase chain reaction (qPCR) following methods in Boyle et al. (2004). DNA was extracted from the tissue of the entire oral apparatus, which was dissected from all tadpoles immediately after their respective experimental trials. All tissues were stored in 100% EtOH until qPCR analyses. Each sample was run in triplicate against a Bd standard titration (from 105 to 101 zoospores) using relative qPCR on an ABI 7300 real-time PCR machine (Applied Biosystems, Inc., CA, USA), and the pathogen treatment (Bd-exposed or control) was unknown to the experimenter. However, because quantification of infection intensity was beyond the scope of these experiments, we considered an animal as “infected” if it tested positive in at least two of the three qPCR replicates.

Results

No tadpoles from our control treatment tested positive for Bd infection, however, qPCR analyses revealed considerable interspecific variation in the infection status of Bd-exposed tadpoles. For Anaxyrus, 27.5% of the Bd-exposed tadpoles tested positive: Of the 11 A. fowleri infected with Bd, 4 were from the foraging behavior experiment and 7 were from the foraging efficiency experiment (4 tadpoles in the early and 3 tadpoles in the late developmental stages). For Hyla, 72.5% of the Bd-exposed tadpoles tested positive: Of the 29 H. chrysoscelis infected with Bd, 7 were from the foraging behavior experiment and 22 were from the foraging efficiency experiment (5, 9, and 8 individuals from the early, middle, and late developmental stages, respectively).

We designed our experiments to test for the effects of Bd infection on tadpole life history traits. However, the relatively high numbers of Bd-exposed, but not infected, tadpoles allowed us to opportunistically test if Bd exposure, in the absence of infection, reduces tadpole foraging performance. When appropriate (when N > 3 within treatments), we used two-way repeated measures ANOVA to test for an effect of pathogen treatment (Bd exposed and infected, Bd exposed but uninfected, or control) and developmental stage (early: Gosner 28–29; middle: Gosner 30–33; and late: Gosner 36–38) on tadpole foraging behavior. For the foraging efficiency experiments, we were unable to statistically analyze the three treatment groups because of low sample sizes. Thus, we excluded the data from tadpoles in the Bd-exposure treatment that did not test positive for Bd infection and tested for differences between Bd-infected and control tadpoles (as described previously).

Experiment 1: Tadpole Foraging Behavior

For A. fowleri, repeated measures ANOVA indicated a significant main effect of pathogen treatment on foraging behavior (F2,59 = 10.71; P < 0.001), with Bd-infected animals foraging significantly less often than control and Bd-exposed but uninfected animals (Fig. 1). Additionally, there was no significant difference between foraging activity of control and Bd-exposed but uninfected animals. There was no effect of developmental stage (F2,59 = 2.890; P = 0.069) or a pathogen treatment by developmental stage interaction (F4,59 = 0.611; P = 0.657) on foraging behavior. Holm–Sidak post hoc analysis indicated that Bd-exposed and infected tadpoles were significantly less active than tadpoles from the control treatment in the early (P = 0.006), middle (P < 0.0001), and late (P = 0.009) developmental stages (Fig. 1). Additionally, Bd-exposed and infected tadpoles were significantly less active than tadpoles from the Bd-exposed and uninfected treatment in the early (P = 0.009) and middle (P = 0.008) developmental stages, but not during the late developmental stage (P = 0.091). There were no significant differences in activity levels of tadpoles from the control and Bd-exposed but not infected treatments.
https://static-content.springer.com/image/art%3A10.1007%2Fs10393-009-0272-7/MediaObjects/10393_2009_272_Fig1_HTML.gif
Figure 1

Percentage of time spent foraging for Batrachochytrium dendrobatidis exposed and infected (Bd+), B. dendrobatidis exposed and uninfected (Bd-), and uninfected (Control) Fowler’s Toad (Anaxyrus fowleri) tadpoles observed through ontogeny. Significant differences (P < 0.05) were detected between pathogen treatments within early (Gosner 28–29), middle (Gosner 30–33), and late (Gosner 36–38) developmental stages. Values at each developmental stage are the treatment means of three observations +1 SE.

In contrast to A. fowleri, Bd differentially affected tadpole foraging behavior in H. chrysoscelis. Repeated measures ANOVA revealed no main effect of pathogen treatment on foraging behavior (F2,59 = 1.97; P = 0.170), with Bd-exposed and infected, Bd-exposed but uninfected, and control animals spending similar amounts of time foraging (Fig. 2). There was a significant effect of developmental stage on time spent foraging (F2,59 = 4.038; P = 0.027), however, there was no pathogen treatment × developmental stage interaction (F4,59 = 0.406; P = 0.803).
https://static-content.springer.com/image/art%3A10.1007%2Fs10393-009-0272-7/MediaObjects/10393_2009_272_Fig2_HTML.gif
Figure 2

Percentage of time spent foraging for Batrachochytrium dendrobatidis exposed and infected (Bd+), B. dendrobatidis exposed and uninfected (Bd-), and uninfected (Control) Grey Treefrog (Hyla chrysoscelis) tadpoles observed within early (Gosner 28–29), middle (Gosner 30–33), and late (Gosner 36–38) developmental stages. Values at each developmental stage are the treatment means of three observations +1 SE.

Experiment 2: Tadpole Foraging Efficiency

In general, Bd infection had similar overall effects on tadpole foraging efficiency as it did on foraging behavior, however, there were clear intra- and interspecific differences in the foraging efficiency of Bd-infected tadpoles. Bd-infected and uninfected A. fowleri did not exhibit differences in foraging efficiency early in development (t12 = 1.07; P = 0.308; Fig. 3a) but Bd significantly reduced foraging efficiency late in development (t10 = 4.68; P = 0.001; Fig. 3b). Bd infection in H. chrysoscelis reduced foraging efficiency in the early and middle developmental stages (early: t12 = 3.65; P = 0.003; middle: t16 = 4.50; P = 0.003; Fig. 4a, b) but had no effect late in development (t15 = 1.47; P = 0.162; Fig. 4c).
https://static-content.springer.com/image/art%3A10.1007%2Fs10393-009-0272-7/MediaObjects/10393_2009_272_Fig3_HTML.gif
Figure 3

Percentage of food consumed for uninfected (Control) and Batrachochytrium dendrobatidis infected (Bd+) Fowler’s Toad (Anaxyrus fowleri) tadpoles during (a) early and (b) late developmental stages. Different letters above histograms indicate significant differences between pathogen treatment groups (P < 0.05). Values are treatment means +1 SE.

https://static-content.springer.com/image/art%3A10.1007%2Fs10393-009-0272-7/MediaObjects/10393_2009_272_Fig4_HTML.gif
Figure 4

Percentage of food consumed for uninfected (Control) and Batrachochytrium dendrobatidis infected (Bd+) Grey Treefrog (Hyla chrysoscelis) tadpoles during (a) early, (b) middle, and (c) late developmental stages. Different letters above histograms indicate significant differences between pathogen treatment groups (P < 0.05). Values are treatment means +1 SE.

Discussion

Ecologists increasingly recognize the potential for pathogens to modify host behavior which can disrupt predator–prey dynamics (Parris et al., 2006; Reed and Dobson, 1993), food webs (Lafferty et al., 2006), and compromise foraging abilities of infected hosts (Levri and Lively, 1996). Recent evidence suggests that tadpoles exhibit reduced growth and developmental rates when infected with Bd (Parris and Cornelius, 2004), yet the mechanism underlying this response was previously untested. Our results provide evidence that Bd infections reduce foraging efficiency in Anaxyrus and Hyla tadpoles and that Bd differentially affects foraging behavior in these species. Namely, Bd infections reduce the foraging activity of Anaxyrus tadpoles but not Hyla. In addition, uninfected tadpoles exposed to Bd were as active as control tadpoles.

Prevention of pathogenesis is costly to the host (Woodhams et al., 2007) and may reduce the host’s ability to maintain normal somatic growth and development. Given the high energetic costs of metamorphosis (Steiner and van Buskirk, 2008), inadequate growth rates during the larval period can negatively affect adult fitness. Therefore, with or without infection, exposure to a pathogen may impose physiological costs on the host and thereby negatively affect performance or survival. For example, Garner et al. (2009) found that Bd exposure always reduced growth and developmental rates of Bufo bufo tadpoles, even if they did not exhibit infection at the time of death. However, we found no evidence that Bd exposure, without pathogenesis, negatively affects foraging behavior in these species. Thus, Bd infections negatively impact the foraging performance of infected tadpoles of some species, and may decrease growth and developmental rates, effects which may have cascading effects on host survival in larval (Kurzava, 1998; Semlitsch et al., 1988) and adult (Berven, 1990; Smith, 1987) phases.

In temporary aquatic habitats, the optimal strategy for tadpoles is to obtain the most food in the shortest period of time within the biotic and abiotic constraints of the given habitat (Wilbur and Collins, 1973). Ultimately, low activity levels and reduced foraging behavior in tadpoles may lead to reduced food intake (Anholt et al., 1996; Skelly, 1994), which can impact growth and developmental rates and cause larval mortality due to pond drying (Skelly and Werner, 1990), or decrease adult fitness in terrestrial environments because of smaller metamorphic sizes (Goater, 1994). In our foraging behavior experiments, Bd-infected Anaxyrus tadpoles reduced the amount of time they spent foraging compared to uninfected tadpoles, however, Bd had no effect on foraging time in Hyla tadpoles. The general reduction in overall tadpole activity levels we observed is consistent with recent experiments examining pathogen-induced changes in host behavior. For example, Parris et al. (2006) examined activity levels of Bd-infected Southern Leopard Frog (Lithobates sphenocephalus) tadpoles within an antipredatory context and found that Bd-infected tadpoles decreased their activity levels across all treatments, regardless of whether a predator was present or absent. Low activity levels may be a host’s general response to infection (Giles, 1983), rather than a specific response to a predator (Parris et al., 2006) or reduced foraging ability (this study). It is important to note that the link between foraging activity and growth rates is complex (Steiner, 2007), and factors such as metabolic rate (McPeek, 2004), gut morphology (Relyea and Auld, 2004), and immune response (Giles, 1983; Roy and Kirchner, 2000) may interact to mediate the tradeoffs between foraging activity and growth rates.

Many temperate zone tadpoles have keratinized jaw sheaths and labial teeth (Altig and McDiarmid, 1999) which allow them to facultatively forage on attached or suspended phytoplankton (Seale and Wassersug, 1979). These specializations in feeding morphology are critical for successful tadpole foraging. Oral deformities in tadpoles occur in response to Bd (Fellers et al., 2001), environmental contaminants (Blaustein and Johnson, 2003), and eutrophication (Johnson and Chase, 2004). Although the relationship between the occurrence of oral deformities and the effect(s) of these deformations on food acquisition is unclear, damaged foraging structures (i.e., keratinized jaw sheaths and labial teeth) likely compromise the foraging abilities of tadpoles (Rowe et al., 1996). In our experiments, Bd-infected Anaxyrus and Hyla tadpoles consumed less food during experimental trials than uninfected tadpoles, suggesting that Bd-infected tadpoles are less efficient at foraging. Recent experiments have documented a strong association between Bd infection and the loss of keratinized jaw sheaths and labial teeth (Drake et al., 2007; Fellers et al., 2001; Marantelli et al., 2004), and we documented these deformities in a subset of the tadpoles we observed. In general, Bd-infected tadpoles were missing portions of labial tooth rows. Given the functional role that labial teeth have in the foraging process (Wassersug and Yamashita, 2001), damage to labial tooth rows likely affects how tadpoles grasp and rake food while foraging.

Indeed, Bd-infected Anaxyrus and Hyla tadpoles with missing teeth have inferior feeding kinematics compared to uninfected tadpoles. The keratinized labial tooth rows of Bd-infected Anaxyrus and Hyla tadpoles spend less time raking over an algal covered substrate relative to tadpoles with undamaged dentition [Venesky et al., in review]. To compensate, Bd-infected tadpoles increase the rate in which their mouths open and close while foraging. Thus, Bd-infected tadpoles may suffer a reduced potential to obtain sufficient food or may expend more energy while foraging—both of which can reduce growth and developmental rates. It is important to note that we did not test for the degree of oral deformities in Bd-infected tadpoles and are unable to quantify how specific patterns of mouthpart damage may influence foraging efficiency.

There is strong evidence that food availability can influence the size at, and timing of, metamorphosis (Blouin, 1992; Semlitsch and Caldwell, 1982), and some experiments have tested the effect of food deprivation on growth and developmental rates through ontogeny (Audo et al., 1995; Hensley, 1993; Leips and Travis, 1994). These studies confirm that food availability affects size at metamorphosis, but developmental rates become fixed during late stages in ontogeny (Hensley, 1993; Leips and Travis, 1994). Food deprivation early in ontogeny reduces carbohydrate stores in tadpoles (Audo et al., 1995), which can reduce growth and developmental rates if food availability remains low. Therefore, it is possible that decreased foraging abilities and food acquisition in Bd-infected tadpoles has a direct effect on growth and developmental rates.

One key finding of our experiments was that Bd differentially affected the foraging performance of Anaxyrus and Hyla. In our behavior experiments, Bd reduced the amount of time Anaxyrus spent foraging but did not affect the amount of time Hyla spent foraging. In our efficiency experiments, both Anaxyrus and Hyla responded to Bd with reduced foraging efficiency. Without an understanding of other components of the tadpole response to Bd (i.e., physiological), it is unclear why Bd reduced the foraging behavior of Anaxyrus tadpoles but not Hyla. When considering the behavioral differences between tadpoles of these two species, Anaxyrus tadpoles differ from Hyla in that they are more social and are often found in aggregations (Beiswenger, 1977; Wilbur and Alford, 1985). Thus, the response observed in individual host behaviors may differ from behaviors exhibited when hosts are in groups (Han et al., 2008). Additional experiments testing the potential of pathogens to alter social cues would contribute to the biology of infectious diseases.

Acknowledgments

We thank M. Ferkin, M. Rensel, S. Schoech, and T. Wilcoxen for reviewing previous versions of this manuscript. R. Altig, K. Warkentin, and R. Wassersug provided suggestions on the foraging efficiency experiment. J. Longcore kindly provided the Bd isolate. We thank M. Beck for providing the materials and space for karyotyping Hyla tadpoles. We also thank N. Hobbs, M. Takahashi, L. Venesky, and T. Wilcoxen for assisting in collecting anuran eggs. Collection permits from TN were obtained prior to collecting the animals used in these experiments, and all experimental procedures were approved by the University of Memphis IACUC. All research conforms to the legal requirements of the United States of America. This publication was developed, in part, under a GRO Research Assistance Agreement No. MA-916980 awarded by the U.S. Environmental Protection Agency to M. Venesky. It has not been formally reviewed by the EPA. The views expressed in this document are solely those of the authors and the EPA does not endorse any products or commercial services mentioned in this publication. The University of Memphis Ecological Research Center Grant in Aid of Research also provided financial support to M. Venesky during portions of this study.

Copyright information

© International Association for Ecology and Health 2010

Authors and Affiliations

  • Matthew D. Venesky
    • 1
  • Matthew J. Parris
    • 1
  • Andrew Storfer
    • 2
  1. 1.Department of BiologyUniversity of MemphisMemphisUSA
  2. 2.School of Biological SciencesWashington State UniversityPullmanUSA