Parasite-induced modifications in host behavior are often categorized as responses that increase transmission rates and ultimately parasite fitness (Poulin, 2010) or as adaptive host responses to clear and/or eliminate their infection burden (often referred to as “sickness behaviors”; reviewed in Adelman and Martin, 2009). The behavior of parasitized animals, however, is not always adaptive in respect to the host or parasite (Poulin, 1995). Host behavior can change in response to physiological, immunological, and/or anatomic changes associated with pathology from the infection. For example, amphibian larvae infected with Ranavirus have behavioral deficits that include erratic swimming and lethargy (reviewed in Gray et al., 2009); this can impair their ability to perform as well as noninfected amphibians, which doesn't appear to increase host or parasite fitness. Thus, behavioral side effects of infection, such as lethargy, can reduce host fitness by decreasing a host's ability to acquire food and escape predators.

All hosts are not equal, and there is individual variation in how hosts respond to parasitism. Relatively few individuals in a population typically carry the heaviest infection burdens, and the majority of hosts are not infected (Woolhouse et al., 1997). Moreover, of those infected, individuals vary in their tolerance to infection, and the healthiest individuals (e.g., those in the best body condition) do not always carry the lowest infection intensity (Raberg et al., 2007). This variation in host species resistance to parasites, coupled with variation in host responses when they are infected, has generated considerable interest in the relationship between host performance and fitness change with increasing infection intensity. For example, Rohr et al. (2010) demonstrated that the relationship between change in host fitness and parasite infection intensity was not always negative. In other words, the most heavily infected tadpoles of Anaxyrus [= Bufo] americanus and Lithobates [= Rana] clamitans did not necessarily die earlier than tadpoles with lower infection burdens. This result emphasizes the need to move beyond categorizing animals as “infected” or “noninfected” to better understand the relationship between infection burden and host performance.

Larval anurans are an excellent model system to study parasite-mediated changes in host behavior because they are infected with a diversity of micro- and macroparasites (Sutherland, 2005) and exhibit many different parasite-induced behavioral changes (e.g., Taylor et al., 2004; Venesky et al., 2009; Koprivnikar et al., 2014). The amphibian chytrid fungus, Batrachochytrium dendrobatidis (hereafter “Bd”), is of particular interest to ecologists because it is associated with amphibian declines and extinctions on each continent on which it occurs (Fisher et al., 2009). Many biotic and abiotic factors interact to affect the outcome and severity of Bd infections, such as temperature (Raffel et al., 2013), amphibian biodiversity (Searle et al., 2011; Venesky et al., 2014), and immune responsiveness (Fites et al., 2013; McMahon et al., 2014). Once infected with Bd, the most common behavioral changes in tadpoles include reduced activity (Parris et al., 2006) and reduced feeding activity and food consumption (Venesky et al., 2009).

Previous work on Bd-induced behavioral changes did not quantify infection burden and considered infection status as a categorical variable (Parris et al., 2006; Venesky et al., 2009). In this study, we examined the relationship between parasite infection and feeding performance in Green Frog (L. clamitans) tadpoles. Our aim was to build upon the findings of Venesky et al. (2009) and test for a relationship between Bd-infection intensity and food consumption. If foraging performance correlates with Bd-infection intensity, we predicted there would be a negative relationship between Bd-infection intensity and food consumption.

Materials and Methods

Results

Upon examination of the data, we noticed the two tadpoles that consumed all food from the microscope slides also cleared the majority of the food from their intestines and, thus, were not representative of their low food consumption score. Because we arbitrarily selected only four microscope slides to photograph, we could not use the amount of food removed from each slide as a response variable. Thus, we removed these two individuals from the statistical analyses.

As predicted, tadpoles with a greater Bd infection consumed significantly less food than did tadpoles with less Bd but only after controlling for individual size differences in the volume of each tadpole's intestinal tract (F_1,9 = 9.31; P = 0.014; R² = 0.509; Fig. 1A). The absolute volume of food consumed was not predictable based on Bd-infection intensity (F_1,9 = 3.18; P = 0.108; R² = 0.260; Fig. 1B). None of the tadpoles had any noticeable mouthpart deformities (e.g., missing teeth or gaps in the keratinized jaw sheaths).

View Figure

• Download as Powerpoint
• Download Figure

Discussion

Our result corroborates and builds upon the findings of previous research by demonstrating that Bd infection reduces tadpole feeding efficiency and that feeding performance is negatively correlated with Bd-infection intensity (Venesky et al., 2009). In addition, we now have evidence that Bd infection reduces tadpole feeding efficiency in species of three different genera (Anaxyrus, Hyla, and Lithobates), suggesting this is a generalized pathology associated with Bd infection. Whether these generalities indicate a similar mechanism of reduced food consumption is unknown, and future studies are needed to elucidate this.

Our finding provides evidence that Bd-infected tadpoles suffer a reduced potential to obtain food that is independent of Bd-induced changes to the mouthparts of tadpoles. None of the tadpoles used in our experiment had any visible signs of mouthpart damage, consistent with previous work that found Bd-induced mouthpart deformations occurring 3–4 weeks post infection in another ranid tadpole (Rachowicz and Vredenburg, 2004). Our result, however, is not consistent with previous work, which showed reduced food consumption in tadpoles with chytridiomycosis is associated with Bd-induced deformations to the keratinized mouthparts (Venesky et al., 2009, 2010). Generalized pond-type tadpoles all have 2 anterior and 3 posterior tooth rows that function in concert to lift food from a substrate (Venesky et al., 2011a). That tadpoles with the highest Bd infections consumed less food than did tadpoles with lower infections, despite having a full complement of teeth, suggests that another mechanism (e.g., changes in tadpole behavior), also might contribute to reduced feeding efficiency observed in tadpoles with chytridiomycosis.

Our result that food intake decreases as a function of increasing Bd-infection intensity has important implications for amphibian disease ecology. Hosts in poor nutritional condition often have dampened immune responses and are more susceptible to parasite infection (Peck et al., 1992; Feder et al., 1997; Houdijk et al., 2000). If food intake is positively correlated with immunocompetence (as it is with dietary protein intake; Venesky et al., 2011b), our result suggests that tadpoles of L. clamitans with higher Bd infections could suffer a reduced potential to resist other species of parasites. Concurrent infections by multiple species of parasites have been documented in amphibian host species (Sutherland, 2005); yet, it is unclear how co-infections affect host pathology (Romansic et al., 2011; Johnson and Hoverman, 2012). Future research needs to examine the net effects of reduced food consumption on tadpole health and physiology and how this relates to resistance to multiple species of parasites.