Relationships between host body condition and ...

Parasitol Res DOI 10.1007/s00436-016-4957-x

ORIGINAL PAPER

Relationships between host body condition and immunocompetence, not host sex, best predict parasite burden in a bat-helminth system Elizabeth M. Warburton 1 & Christopher A. Pearl 1,3 & Maarten J. Vonhof 1,2

Received: 24 September 2015 / Accepted: 12 February 2016 # Springer-Verlag Berlin Heidelberg 2016

Abstract Sex-biased parasitism highlights potentially divergent approaches to parasite resistance resulting in differing energetic trade-offs for males and females; however, tradeoffs between immunity and self-maintenance could also depend on host body condition. We investigated these relationships in the big brown bat, Eptesicus fuscus, to determine if host sex or body condition better predicted parasite resistance, if testosterone levels predicted male parasite burdens, and if immune parameters could predict male testosterone levels. We found that male and female hosts had similar parasite burdens and female bats scored higher than males in only one immunological measure. Top models of helminth burden revealed interactions between body condition index and agglutination score as well as between agglutination score and host sex. Additionally, the strength of the relationships between sex, agglutination, and helminth burden is affected by body condition. Models of male parasite burden provided no support for testosterone predicting helminthiasis. Models that best predicted testosterone levels did not include parasite burden but instead consistently included month of capture and Electronic supplementary material The online version of this article (doi:10.1007/s00436-016-4957-x) contains supplementary material, which is available to authorized users. * Elizabeth M. Warburton [email protected]

1

Department of Biological Sciences, Western Michigan University, 1903 West Michigan Avenue, Kalamazoo, MI 49008-5410, USA

2

Environmental and Sustainability Studies Program, Western Michigan University, 1903 West Michigan Avenue, Kalamazoo, MI 49008-5410, USA

3

Present address: Department of Biomedical Sciences, Grand Valley State University, 1 Campus Drive, Allendale, MI 49401, USA

agglutination score. Thus, in our system, body condition was a more important predictor of immunity and worm burden than host sex. Keywords Sex-biased parasitism . Helminthiasis . Eptesicus fuscus . Immunocompetence . Body condition

Introduction Sex-biased parasitism highlights the potential for differing energetic trade-offs between males and females. Hosts can attempt to eliminate their parasites by mounting an energetically costly immune response (Lochmiller and Deerenberg 2000) and resources that might otherwise be used for selfmaintenance, or reproduction must be transferred into immune investment (Sheldon and Verhulst 1996; Schmid-Hempel and Ebert 2003; Hanssen et al. 2003; Hasu et al. 2006; Honkavaara et al. 2009; Allen and Little 2011; Gooderham and Schulte-Hostedde 2011; Moreno-Rueda 2011; Friesen et al. 2015). However, although they may be faced with similar challenges, patterns of infection, the strength and mode of response to parasites, and relative levels of investment in self-maintenance versus reproduction may differ between the sexes. Males are often expected to harbor more parasites than females, due to higher exposure via greater levels of activity, larger home ranges, or different dietary habits and/or greater susceptibility to parasites due to intrinsic characteristics that impact host susceptibility. Sex differences in immune function could result from differences in life histories and sexual behavior, with males investing more resources to mate attraction and mating success rather than immunity, while females may invest more energy into immunity to increase survival and ultimately their own reproductive success (Rolff 2002;

Parasitol Res

Stoehr and Kokko 2006; Forbes 2007; Zuk 2009; Nunn et al. 2009). The production of androgen steroid hormones, in particular, has been linked to suppressed immune responses in males, making them more susceptible to parasitic infection (Folstad and Karter 1992; Poulin 1996; Zuk and McKean 1996; Talleklint-Eisen and Eisen 1999; Ferrari et al. 2004; Lajeunesse et al. 2004; Hillegass et al. 2008; Perkins et al. 2008, Robinson et al. 2008; Krasnov et al. 2012; Friesen et al. 2015). However, while male-biased parasitism occurs in multiple vertebrate taxa (Moore and Wilson 2002) and there is some support for androgen suppression of immunity (Roberts et al. 2004, Hepworth et al. 2010), male-biased parasitism is not universal (Nakazawa et al. 1997; Schalk and Forbes 1997; McCurdy et al. 1998; Porteous and Pankhurst 1998, Dare and Forbes 2009; Khan et al. 2010; Rossin et al. 2010, MacIntosh et al. 2010; Robinson et al. 2010; Kiffner et al. 2013; Waterman et al. 2014). In addition, the relationship between testosterone and immune function may in fact be bidirectional, as induced immune response has been shown to suppress testosterone production (Boonekamp et al. 2008; Cai et al. 2009; Mills et al. 2009), reflecting a trade-off between male investment in self-maintenance and reproductive effort (Degen 2006; Marzal et al. 2007; Mills et al. 2010). If limited energy results in trade-offs between immune function and reproduction or growth, then the optimal strategy for a parasitized individual may be to bear the costs of parasitism rather than mount an immune response. Immunity involves energetic investment into a costly response that can increase host survival by eliminating parasites (Bordes et al. 2012) but may result in less energy being available for other important activities. Alternatively, bearing infection is less costly in terms of energy because it does not require parasite elimination, but rather, the host may limit pathology caused by parasites and prioritize self-maintenance (Bordes et al. 2012). Limited host resources may create energetic trade-offs that may favor costly immune responses when body condition is high or limit the immune response if body condition is low (Norris and Evans 2000; Martin et al. 2006, 2008; Ujvari and Madsen 2006; Forsman et al. 2008). Individual hosts therefore may address parasitic infection differently, with some individuals mounting an immune response rather than tolerating infection or vice versa, to optimize their survival or reproductive output. Our investigation focused on the relationships between parasite burden, sex, immune function, and body condition within wild populations of big brown bats (Eptesicus fuscus, Chiroptera: Vespertilionidae) and their helminth parasites. Helminths are easily counted discrete units that spend their adult lives in one single host. Big brown bats are relatively common, widely distributed throughout most of North America, and form aggregations in buildings or trees during the summer (Kurta and Baker 1990). During pregnancy and lactation, female bats will forage heavily throughout the night

as they require much more energy during this period (Kurta et al. 1990) while males and non-reproductive females will often forage for much shorter periods 1 to 2 h after sunset (Grinevitch et al. 1995). Given that bats often contract helminths via ingestion of infected prey (Kumar 1999; Schell 1985), typical foraging behaviors would put female bats at higher risk of helminthiasis than males. However, if testosterone is immunosuppressive, then male bats in our host-parasite system should have higher helminth burdens than females, especially considering that male big brown bats produce some of the highest recorded testosterone levels in mammals (Mendonca et al. 1996). The sex-specific feeding behavior that increases female exposure to infective prey and the high male testosterone levels in E. fuscus thus decouple increased exposure to parasites from possible testosterone-mediated immunosuppression. This host-parasite system therefore provides the ideal opportunity to examine if (1) sex-biased parasitism occurred in our host-parasite system, either with females bearing an increased parasite burden due to elevated exposure or males hosting more parasites due to testosterone-related immunosuppression, (2) the sexes were investing equally into immunity or if body condition, regardless of host sex, better predicted parasite burdens due to the absence of sex-specific tradeoffs between immunity and condition, (3) if testosterone level was an important predictor of male parasite burden relative to other variables, and (4) if male testosterone levels were negatively correlated with measures of immune function. Ultimately, uncovering these relationships will allow us to determine if one sex disproportionately maintains helminth infection in a host population and if the sexes cope with parasitic infection differently, or if trade-offs between body condition and immunity within individual hosts better predict parasite burden.

Materials and methods Host and parasite collection One hundred forty adult E. fuscus (75 females, 65 males) were captured from seven colonies in Michigan, USA (Fig. 1, Online Resource 1), with mist nets as they emerged from their roosts or were hand-caught in the roost prior to nightly emergence. Bats captured in the same roost were considered to belong to the same site. Bats were captured at most colonies at least twice over a period of 2 years (2008 and 2009), with the exception of MI-1 which was sampled a single time because bats were excluded from the roost after initial sampling (see Online Resource 2 for a list of capture dates for each study site). All bats were confirmed to be adults (1 year and older) based on the degree of ossification of phalanges (Kunz and Anthony 1982).

Parasitol Res

Vaucher 1994; Lotz and Font 2008; Tkach 2008; Anderson et al. 2009) and then identified to species (Williams 1960; Rausch 1975; Lotz and Font 1983; Tkach et al. 2000; Guerrero et al. 2002). Preserved and stained frozen specimens were compared to preserved and stained fresh specimens from another dataset (Warburton and Vonhof unpublished data) in order to aid identification. Assessing parasite resistance

Fig. 1 Locations of study sites (N = 7) in the lower peninsula of Michigan (USA)

We recorded host sex, host age, capture date, and capture location via handheld GPS for all captured bats. We also recorded mass and right forearm length for each individual host. Bats were euthanized by cervical dislocation followed by exsanguination via cardiac puncture with a heparinized needle and then were frozen at −20 °C. Whole blood was centrifuged to separate plasma from red blood cells (RBCs), and resulting plasma was frozen at −80 °C until further analysis. All euthanasia procedures were authorized by Western Michigan University’s Institutional Animal Care and Use Committee and the Michigan Department of Natural Resources. Helminth burden was then assessed by necropsy after thawing overnight in a 4 °C refrigerator. The entire carcass and viscera were examined separately using a stereoscope. Trematodes and cestodes were counted, collected via forceps or dissecting needle as appropriate for their size, stained with Semichon’s acetocarmine, mounted, and examined with a compound light microscope. Nematodes were stored in glycerine-alcohol, cleared, and examined as temporary mounts with a phase-contrast microscope. Then, parasites were identified using dichotomous keys to genera (Czaplinski and

Given that simultaneous quantification of multiple immune indices can provide a deeper understanding of host immunocompetence (Biard et al. 2015), we used two immune assays: lytic assay and agglutination assay. Lysis score represents pathogen recognition by native antibodies (NAbs) and subsequent complement activation, part of the innate immune response (Matson et al. 2005). However, agglutination score depends on the ability of NAbs to recognize foreign antigens, marking them for phagocytosis; thus, NAbs are strongly associated with B cell functions necessary for acquired immunity (Palacios et al. 2012). After obtaining blood from cardiac puncture, plasma was used to assess lysis and agglutination by performing hemolysis/hemagglutination assays following a technique similar to Matson et al. (2005) with a few modifications. In short, 25 μL of plasma from each individual was pipetted into wells in the first and second columns of a 96-well round-bottom polystyrene tissue culture plate (Costar 3790, Corning). Plasma in the first column remained undiluted and served as a positive control. Then, 25 μL of phosphatebuffered saline (PBS) were pipetted into the well in the second through twelfth columns of the plate. This created a 1:1 dilution of plasma to saline in the second well that was serially diluted through to the 11th well. The 12th well contained only PBS and served as a negative control. After dilution, 25 μL of washed, laboratory-grade rabbit RBCs (Hemostat Laboratories) was added to each well to serve as foreign antigen. Plates were incubated at 37 °C for 90 min and then scanned as a full-size image at 600 pixels per inch using an Epson Perfection flatbed scanner to record lytic activity. After leaving the plate at room temperature for an additional 20 min, we recorded agglutination using the same procedure. All digital images were then scored manually by the same individual to determine the relative roles of complement and NAbs in functional immunocompetence. Individuals were scored as negative log2 of the last plasma dilution exhibiting each behavior with lysis reflecting the interaction of complement and NAbs while agglutination reflected the activity of NAbs only. Assessing host body condition We employed two methods to determine body condition: body mass to forearm length index and neutrophil to lymphocyte (N/L) ratio. Body condition index was calculated as the

Parasitol Res

residual of a significant ordinary least squares regression analysis (p < 0.0001) of right forearm length on mass (Schulte-Hostedde et al. 2005). Therefore, individuals with a negative index value possessed lower body condition for their size whereas individuals with a positive index value enjoyed higher body condition for their size. This method is a good approximation of body condition when compared with total body electrical conductivity for E. fuscus (Pearce et al. 2008). Although sexual size dimorphisms have the potential to confound mass-length allometries (Barnett et al. 2015) and influence sex-biased parasitism (Moore and Wilson 2002), an ANCOVA performed in the R 3.2.2 statistical environment did not indicate that the slopes of mass-forearm length regression lines differed between the sexes (F = 0.001, p = 0.97). The ratio of neutrophils to lymphocytes provides useful information on host condition with higher ratios indicating increased physiological stress, and those hosts experiencing higher stress could exhibit decreased immunocompetence (Davis et al. 2008). To determine N/L ratio, whole blood obtained from cardiac puncture was used to make at least two blood smears from each individual bat. These blood smears were examined via light microscopy to assess number and type of circulating leukocytes, and the ratio of neutrophils to lymphocytes was calculated (Superina and Sierra 2008). Assessing male testosterone levels Plasma testosterone concentrations were determined using a competitive binding ELISA following the manufacturer’s instructions (ENZO Life Sciences, Farmingdale, NY; ADI-901065). The range of the standard curve was 7.81 to 2000 pg/ml. Prior to loading into the ELISA, plasma was incubated with the steroid displacement reagent for 30 min and then brought to the final dilution in assay buffer. Samples were diluted to fall within the range of the standard curve. Assay sensitivity was approximately 12 pg/ml; intra- and inter-assay coefficients of variation were 2.6 and 3.6 %, respectively. Model building procedure We sought to model three different relationships: (1) overall host parasite burdens as functions of host sex, parasite resistance, and body condition; (2) male-only host parasite burdens as functions of parasite resistance, body condition, and testosterone level, and (3) male testosterone levels as functions of the immune response, body condition, and month of host capture. To determine which variables best predicted these relationships, we utilized the generalized linear models function in SPSS 17.0 to create candidate models via maximum likelihood estimation. As our data for parasite burdens were overdispersed, models predicting those data used a negative binomial distribution, log link, and a dispersion parameter, a

spread of values around the central tendency, that SPSS automatically estimated as 2.3. Our data for host testosterone violated normality due to a positive skew (Kolmogorov-Smirnov Z = 2.01, p = 0.001); therefore, we used a log10 transformation on testosterone levels which resulted in an approximately normal distribution (Kolmogorov-Smirnov Z = 1.02, p = 0.17). Generalized linear models for log10 of testosterone levels (LogT) used the normal distribution with an identity link. Candidate models included combinations of all predictor variables and all interactions between them. We used sex (for male and female data), lysis, agglutination, body condition index, N/L ratio, and testosterone level (for male data only) as predictors of parasite burden; we also included site of capture as well as month and year of capture to control for possible spatiotemporal variation in parasite burden. We used parasite burden, lysis, agglutination, body condition index, and N/L ratio as predictors of testosterone level for male hosts; we also included month of capture because seasonal peaks in testosterone levels are documented in vespertilionid bats (Mendonca et al. 1996). We also used an intercept-only model as a null model to provide a reference for our candidate models. The best-fit models were identified using Akaike’s information criterion (AIC; Mazerolle 2006) and then independently confirmed for fit using the omnibus loglikelihood test. If any interactions between variables consistently appeared in our best-fit models of parasite burden and testosterone level, we used the SPSS macro PROCESS (Hayes 2013) to perform tests for moderation in order to uncover the relationships between interacting variables. Moderation is the combined effect of two variables, a predictor and a moderator, on an outcome variable (see Online Resource 5 for a conceptual model). Tests for moderation essentially use linear regression; however, the data are centered or transformed into deviations around the mean (Field 2013). This allowed us to determine if the significance of the relationship between predictor and outcome variables changed as the moderator increased or decreased. We applied a √(x + 1) transformation to normalize overdispersed helminth burdens prior to performing tests of moderation.

Results Parasite and host data We collected 11 different species of parasites, many of which are common in vespertilionid bats of the Midwestern USA (Nickel and Hansen 1967; Blankespoor and Ulmer 1970; Coggins et al. 1982; Lotz and Font 1983; Lotz and Font 1985; Pistole 1988; McAllister and Bursey 2009), from our sample of 140 hosts. Ten of these species were intestinal helminths while one species (Litomosoides guiterasi, Nematoda:

Parasitol Res

Dipetalonematidae) was found in the body cavity. Digenetic trematodes from the family Lecithodendriidae were by far the most common helminths in our study (see Table 1 for component community data, Online Resource 2 for infracommunity data). Some lecithodendriids had relatively high prevalence (e.g., Paralecithodendrium naviculum, Paralecithodendrium transversum), while others had very low prevalence (e.g., Acanthatrium microacanthum, Glyptoporus noctophilus). We indentified only one nonlecithodendriid trematode species (Plagiorchis vespertilionis) as well as only one cestode species (Hymenolepis roudabushi) and one nematode species (L. guiterasi). These three species had relatively low prevalence, mean abundance, and mean intensity (sensu Bush et al. 1997) compared to the most common lecithodendriid species. Mean abundance and mean intensity were highest among species belonging to Paralecithodendrium with Pa. transversum exhibiting the highest mean abundance and Paralecithodendrium macnabi having the highest mean intensity. We found no evidence for sex-biased parasitism in our system. Male and female hosts had comparable parasite burdens (Online Resource 6). The sexes also had similar lysis scores (Online Resource 7) and N/L ratios (Online Resource 8). However, females possessed slightly higher body condition indices (Online Resource 9) and significantly higher agglutination scores (Online Resource 10), suggesting that females invested more in that measure of parasite resistance than males. Best-fit models Five models of helminth burden had better fit than the intercept-only model (Table 2). All five models included agglutination, body condition index, lysis, and year as well as Table 1 Taxonomic designation, prevalence, mean intensity, and mean abundance of helminth fauna infecting E. fuscus in our study

the interaction between agglutination and body condition index. In addition, four of the five models included host sex and three of the five included the interaction between agglutination and sex. To gain insight into the relationships between the dependent and independent variables, we examined the parameter estimates for the top candidate model that were not involved in any interactions. The coefficient for year (b = −0.773) indicated that bats caught in 2008 had heavier parasite burdens than bats caught in 2009. The coefficient for lysis score (b = −0.360) indicated that individuals with lower lysis scores also had higher parasite burdens. To better understand interactions between sex, BCI, and agglutination as well as their influence on parasite burden, we performed two tests for moderation. Given the evidence in the best-fit models, we proposed that BCI could be moderating the effect of host sex on agglutination. We determined that significance of the conditional effect of sex on agglutination score changed depending on the values for BCI (Online Resource 3). Body condition indices 2.22 standard deviations below the mean had no significant effect (p = 0.0605); however, body condition indices at the mean (p = 0.0027) or 2.22 standard deviations above the mean had a significant conditional effect of sex on agglutination score (p = 0.0251). This means that the strength of the relationship between sex and agglutination is affected by BCI values. We then proposed that, while holding the effect of sex constant, BCI could moderate the effect of agglutination score on parasite burden. We determined that significance of the conditional effect of agglutination score on parasite burden changed depending on the values for BCI (Online Resource 4). Body condition indices at the mean (p = 0.2269) or 2.24 standard deviations below the mean (p = 0.8423) had no significant effect; however, body condition indices from 2.24 standard deviations above the mean had a significant conditional effect (p = 0.0132) of

Species

Higher taxonomy

Prevalence (%)

Mean abundance (±SE)

Mean intensity (+SE)

A. eptesici A. microacanthum A. pipestrelli Pa. macnabi Pa. naviculum Pa. swansoni Pa. transversum G. noctophilus Pl. vespertilionis H. roudabushi

Trematoda: Lecithodendriidae Trematoda: Lecithodendriidae Trematoda: Lecithodendriidae Trematoda: Lecithodendriidae Trematoda: Lecithodendriidae Trematoda: Lecithodendriidae Trematoda: Lecithodendriidae Trematoda: Lecithodendriidae Trematoda: Plagiorchiidae Cestoda: Hymenolepididae

5.71 2.13 4.29 10.00 32.14 3.57 45.00 1.43 16.43 3.57

2.31 (±1.01) 0.17 (±0.10) 0.34 (±0.15) 2.62 (±1.23) 8.10 (±2.20) 1.54 (±1.41) 10.99 (±4.85) 0.11 (±0.10) 0.38 (±0.09) 0.42 (±0.31)

40.38 (±4.24) 8.00 (±0.68) 7.83 (±0.73) 26.21 (±3.89) 25.20 (±3.90) 43.20 (±7.51) 24.41 (±7.25) 8.00 (±0.85) 2.30 (±0.24) 11.80 (±1.62)

L. guiterasi

Nematoda: Dipetalonematidae

7.14

0.08 (±0.02)

1.10 (±0.09)

Sensu Bush et al. (1997). Note that standard error (SE) terms accompany both mean intensity and mean abundance

Parasitol Res Table 2

Terms and AIC values for the top five candidate models of parasite burden compared to the intercept-only model

Model terms

AIC

Agglutination

BCI

Lysis

Sex

Year

Agglutination × BCI

Agglutination

BCI

Lysis

Sex

Year

Agglutination × BCI

Agglutination Agglutination

BCI BCI

Lysis Lysis

Sex Sex

Year Year

Agglutination × BCI Agglutination × BCI

Agglutination Intercept-only

BCI

Lysis

Year

Agglutination × BCI

Agglutination × Sex

1089.588 1089.872

Agglutination × Sex Agglutination × Sex

Lysis × Sex Agglutination × Sex × BCI

1089.978 1090.073 1090.637 1101.455

Note that agglutination, body condition (BCI), lysis, and year, as well as the interaction between agglutination and BCI, appeared in all five top models. All models had an omnibus log-likelihood p value of 0.001 or less

Discussion

agglutination score on parasite burden. This means that the strength of the relationship between agglutination and parasite burden is affected by BCI values. The top five models of male-only helminth burden all had better fit than the intercept-only model (Table 3). All five models included agglutination and year while four of the five models also included lysis and/or BCI. The interaction between agglutination and BCI occurred in only one of the models. None of the top five models included host testosterone level as a predictor of parasite burden, and the model with the lowest AIC that included testosterone had worse fit than models that did not include this term. As there were no consistent interactions between predictor variables in the top models, we did not follow up with tests of moderation. The top five models of testosterone level (LogT) also had better fit than the intercept-only model (Table 4). All five models included month while four of the five models included agglutination and three of them included BCI. Models that included interactions between terms or that included parasite burden had worse fit than those that did not include this term. Given that none of the top five models included any interaction terms, we did not follow up with tests of moderation.

We found no evidence for sex-biased parasitism in our hostparasite system. Further, we found that both sexes were equally likely to mount an immune response and female bats scored higher than males in only one immunological measure (agglutination). However, our tests of moderation suggested that BCI moderated the effect agglutination score on parasite load, and since females were in slightly better condition than males, BCI was driving higher female agglutination scores. These moderating effects also accounted for the BCI-agglutination score interaction in our best-fit models of helminth burden. Lysis score, which was also present in all five top models and without interactions in four of those five models, did not differ significantly between the sexes or interact with measures of body condition in any of the top models. This suggests that individuals could be investing in different facets of the immune response based not on sex but on body condition. All individuals may have invested in lysis while only those with higher body condition resources could employ agglutination. Immune system trade-offs are well documented (e.g., Mallon et al. 2003; Schmid-Hempel 2003; Downs et al. 2013); therefore, individuals could invest in different facets of the immune system, depending on their body condition and

Table 3 Terms and AIC values for the top five candidate models of male host helminth burden compared to the intercept-only model

Table 4 Terms and AIC values for the top five candidate models of male host testosterone level (LogT) compared to the intercept-only model

Model Terms

AIC

Model terms

338.864 339.335 340.978 341.326 341.570 349.101

Agglutination Agglutination

Agglutination Agglutination Agglutination Agglutination Agglutination Intercept-only

Year Year Year Year Year

Lysis Lysis Lysis

BCI BCI BCI

Agglutination × BCI

Note that agglutination and year appeared in all five top models while host testosterone level did not appear in any of the top models. All models had an omnibus log-likelihood p value of 0.001 or less

Agglutination Agglutination Intercept-only

AIC

BCI BCI BCI

Month Month Month Month Month

Lysis Lysis

Year

70.261 70.349 70.409 71.188 72.131 91.407

Note that month appeared in all five top models and models that included terms other than month only slightly improved the AIC value. All models had an omnibus log-likelihood p value of 0.001 or less

Parasitol Res

the relative costs of each strategy, to provide an optimal outcome regarding survival and/or reproductive output. Also, interestingly, N/L ratios were not included in any of the top models, indicating either that this measure of physiological stress was not related to helminth burden or that physiological stress was not impacting the hosts’ immune response. The main components of lysis and agglutination, complement and NAbs, are uniquely intertwined as NAbs can activate the complement cascade and complement level is positively correlated with NAb diversity; therefore, assessing the relative roles of each mechanism can be difficult (Matson et al. 2005). Lysis score, one of our measures of parasite resistance, is a holistic index in that it represents the result of all individual processes involved in lysis into one measure of functional immunocompetence (Matson et al. 2005). One of these processes, the complement cascade, is a key component of lysis score; however, investment in complement-mediated responses, especially the alternative complement pathway, is less costly than antibody-mediated defenses (Lee et al. 2008). Complement proteins mediate recruitment of leukocytes to target helminths by generating chemotactic factors C3a and C5a; then, they promote leukocyte attachment by depositing factors C3b and iC3b on the surface of the helminth (Gasque 2004; Giacomin et al. 2004). This initiates immune effector cell functions like eosinophil-mediated killing, mast cell degranulation, phagocytosis, and antigen presentation (Meeusen and Balic 2000; Chen et al. 2009). Like lysis score, agglutination score is a summative index of all individual processes involved in agglutination presented as one measure of functional immunocompetence (Matson et al. 2005). As pathogen recognition by NAbs can play a crucial role in complement activation, lysis score represents both the level of pathogen recognition by NAbs and subsequent complement activation; however, agglutination scores are typically higher than lysis scores, indicating that the level of circulating NAbs is not a limiting factor for lytic activity (Matson et al. 2005). Unlike lysis, the process of agglutination depends on the ability of NAbs, essentially non-adaptive immunoglobulins, to recognize foreign antigens and mark them for phagocytosis by other immune cells (Boyden 1965). Despite being part of the innate immune response, NAbs are strongly associated with B cell functions necessary for acquired immunity, possibly because of a common genetic mechanism (Palacios et al. 2012). Therefore, NAbs, like other immunoglobulins (Lee et al. 2008), may be more energetically expensive than other components of innate immunity and individuals in better body condition could possess higher levels of circulating NAbs. Mounting an immune response requires energy, and tradeoffs between facets of the immune system become increasingly apparent as body condition deteriorates (Muehlenbein et al. 2010). Body condition itself can influence the type of immune response that a host deploys (Russo and Madec 2013), and infected individuals may have a suite of ideal defenses that

differ depending on their energy reserves. Lysis is an evolutionarily ancient method of parasite resistance that pre-dates adaptive immunity (Fujita et al. 2004); therefore, it could be a default resistance strategy, regardless of body condition. Our findings that body condition moderates the effect of agglutination on parasite burden suggest that all hosts may be able to invest in lysis while only a host with good body condition could invest in elevated levels of circulating NAbs that would result in high agglutination scores. Higher male exposure to parasites or energetic trade-offs between the production of testosterone and immune defenses may explain male-biased parasitism in other systems, such as birds (e.g., Moore and Wilson 2002; Schroderus et al. 2010); however, we found that male and female bats had similar parasite burdens. Differential exposure to parasites via different dietary habits can drive sex-biased parasitism in mammalian systems (Friesen et al. 2015). In our system, female bats typically forage more than males because they undergo pregnancy and lactation, thus requiring more food (Kurta and Baker 1990). All but one of the helminths in our study (L. guiterasi) are acquired by ingestion; therefore, if diet was contributing to increased parasite burden, then females should have been more heavily infected than males but were not. Alternatively, male bats in our host-parasite system produce large amounts of testosterone (Mendonca et al. 1996), and if testosterone were immunosuppressive, then we would have expected males to be more heavily infected than females. We found no evidence of either phenomenon and although we acknowledge that both increased female exposure and increased male susceptibility could occur simultaneously thus possibly negating any sex-biased parasitism, we found that body condition significantly impacted helminth burden by moderating host agglutination response regardless of host sex. In addition, we discovered that the top male-only models of parasite burden did not include testosterone level as a predictor. Instead, we found support for the immune system itself affecting the male testosterone levels because agglutination score was a consistent predictor in models of LogT. Thus, the immune response could be suppressing testosterone production (Boonekamp et al. 2008; Cai et al. 2009; Mills et al. 2009) due to trade-offs between male investments in reproduction or growth. (Degen 2006; Marzal et al. 2007; Mills et al. 2010). Overall, our investigation found no evidence for sex-biased parasitism in our host-parasite system. Instead, we discovered that individual body condition was more likely to influence parasite resistance and mediated the interaction between sex and agglutination. This suggests that host energy reserves, rather than sex, regulate trade-offs between self-maintenance and immunity. Thus, individuals with similar parasite burdens could be managing their infections via different facets of the immune response depending on available energy reserves. We also discovered that host testosterone levels were poor

Parasitol Res

predictors of male parasite burdens. Similarly, seasonal and immunological variables consistently explained male host testosterone level, supporting a correlation between immunity and testosterone production. However, the directional relationship between testosterone production and immune response in our host-parasite system remains unknown. Our results highlight the need for more investigation into interactions between immune response and body condition in order to better understand how individual hosts manage energetic trade-offs and respond to the pressures exerted by exposure to parasites or establishment of parasites. To better achieve this end, we recommend that future studies of sex-biased parasitism explicitly quantify, at minimum, host parasite burden, testosterone levels, and immunological variables rather than assessing only two of these three measures and assuming the third. Given that we found the potential for host condition to modify immune response, we also contend that future investigation of sex-biased parasitism should ideally include a relevant measure of host body condition in order to specifically examine the potential for energetic trade-offs between condition and immunity. Acknowledgments The authors would like to thank J.W. Warburton, C.J. Warburton, D.J. Clarke, B.A. Hines, S.M. Warner, D.M. Courtney, B.K. Hubbard, L.M. Vanbladeren, H.E. LaFore, and E.M. Freed for their assistance in the field and in the lab. We appreciate the assistance of F.A. Jiménez in nematode identification. We would also like to thank J. Glatz for creating a map of our sites. Finally, we thank S.L. Kohler, S.A. Gill, and J.M. Lotz as well as two anonymous reviewers for their constructive comments on earlier versions of the manuscript. Funding was provided by the American Society of Mammalogists, the American Society of Parasitologists, the Annual Midwestern Conference of Parasitologists, and Western Michigan University.

References Allen DE, Little TJ (2011) Identifying energy constraints to parasite resistance. J Evol Biol 24:224–229 Anderson RC, Chabaud AG, Willmott S (2009) Keys to the nematode parasites of vertebrates. Archival Volume. CAB International, Wallingford, UK Barnett CA, Suzuki TN, Sakaluk SK, Thompson CF (2015) Mass-based condition measures and their relationship with fitness: in what condition is condition? J Zool 296:1–5 Biard C, Monceau K, Motreuil S, Moreau J (2015) Interpreting immunological indices: the importance of taking parasite community into account. An example in blackbirds Turdus merula. Methods Ecol Evol 6:960–972 Blankespoor HD, Ulmer MJ (1970) Helminths from six species of Iowa bats. Proc Iowa Acad Sci 77:200–206 Boonekamp JJ, Ros AHF, Verhulst S (2008) Immune activation suppresses plasma testosterone level: a meta-analysis. Biol Lett 4: 741–744 Bordes F, Ponlet N, de Bellocq JG, Ribas A, Krasnov BR, Morand S (2012) Is there sex-biased resistance and tolerance in Mediterranean wood mouse (Apodemus sylvaticus) populations facing multiple helminth infections? Oecol 170:123–135

Boyden SV (1965) Natural antibodies and the immune response. Adv Immunol 5:1–28 Bush A, Lafferty K, Lotz J, Shostak A (1997) Parasitology meets ecology on its own terms: Margolis et al. revisited. J Parasitol 83:575–583 Cai XQ, Yang M, Zhong WQ, Wang DH (2009) Humoral immune response suppresses reproductive physiology in male Brandt’s voles (Lasiopodomys brandtii). Zool 112:69–75 Chen J, Crispin JC, Tedder TF, Lucca JD, Tsokos GC (2009) B cells contribute to ischemia/reperfusion-mediated tissue injury. J Autoimmun 32:195–200 Coggins JR, Tedesco JL, Rupprecht CE (1982) Seasonal changes and overwintering of parasites in the bat, Myotis lucifugus (Le Conte), in a Wisconsin hibernaculum. Am Mid Nat 107:305–315 Czaplinski B, Vaucher C (1994) Family Hymenolepididae Ariola, 1899. In: Khalil LF, Jones A, Bray RA (eds) Keys to the cestode parasites of vertebrates. CAB International, Wallingford, UK, pp 595–663 Dare OK, Forbes MR (2009) Patterns of infection by lungworms, Rhabdias ranae and Haematoloechus spp., in northern leopard frogs: a relationship between sex and parasitism. J Parasitol 95: 275–280 Davis AK, Maney DL, Maerz JC (2008) The use of leukocyte profiles to measure stress in vertebrates: a review for ecologists. Funct Ecol 22: 760–772 Degen AA (2006) Effect of macroparasites on the energy budget of small mammals. In: Morand S, Krasnov BR, Poulin R (eds) Micromammals and macroparasites: from evolutionary ecology to management. Springer, Tokyo, pp 372–399 Downs CJ, Brown JL, Wone B, Donovan ER, Hunter K, Hayes JP (2013) Selection for increased mass-independent maximal metabolic rate suppresses innate but not adaptive immune function. Proc Roy Soc B 280 Ferrari N, Cattadori IM, Nespereira J, Rizzoli A, Hudson PJ (2004) The role of host sex in parasite dynamics: field experiments on the yellow-necked mouse Apodemus flavicollis. Ecol Lett 7:88–94 Field A (2013) Discovering statistics using IBM SPSS statistics. Sage, New York Folstad I, Karter AJ (1992) Parasites, bright males, and the immunocompetence handicap. Amer Nat 139:603–622 Forbes MR (2007) On sex differences in optimal immunity. Trends Ecol Evol 22:111–113 Forsman AM, Vogel LA, Sakaluk SK, Grindstaff JL, Thompson CF (2008) Immune-challenged house wren broods differ in the relative strengths of their responses among different axes of the immune system. J Evol Biol 21:873–878 Friesen OC, Roth JD, Graham LC (2015) Sex-biased parasitism in monogamous arctic foxes is driven by diet. J Mammal 96:417–424 Fujita T, Matsushita M, Endo Y (2004) The lectin-complement pathway—its role in innate immunity and evolution. Immunol Rev 198:185–202 Gasque P (2004) Complement: a unique innate immune sensor for danger signals. Mol Immunol 41:1089–1098 Giacomin PR, Wang H, Gordon DL, Dent LA (2004) Quantitation of complement and leukocyte binding to a parasitic helminth species. J Immunol Meth 289:201–210 Gooderham K, Schulte-Hostedde A (2011) Macroparasitism influences reproductive success in red squirrels (Tamiasciurus hudsonicus). Behav Ecol 22:1195–1200 Grinevitch L, Holroyd SL, Barclay RMR (1995) Sex-differences in the use of daily torpor and foraging time by big brown bats (Eptesicus fuscus) during the reproductive season. J Zool 235:301–309 Guerrero R, Martin C, Gardner SL, Bain O (2002) New and known species of Litomosoides (Nematoda: Filarioidea): important adult and larval characters and taxonomic changes. Comp Parasit 69: 177–195

Parasitol Res Hanssen SA, Folstad I, Erikstad KE, Oksanen A (2003) Costs of parasites in common eiders: effects of antiparasite treatment. Oikos 100:105– 111 Hasu T, Valtonen ET, Jokela J (2006) Costs of parasite resistance for female survival and parental care in a freshwater isopod. Oikos 114:322–328 Hayes AF (2013) Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. The Guilford Press, New York, NY Hepworth MR, Hardman MJ, Grencis RK (2010) The role of sex hormones in the development of Th2 immunity in a gender-biased model of Trichuris muris infection. Eur J Immunol 40:406–416 Hillegass MA, Waterman JM, Roth JD (2008) The influence of sex and sociality on parasite loads in an African ground squirrel. Behav Ecol 19:1006–1011 Honkavaara J, Rantala MJ, Suhonen J (2009) Mating status, immune defence, and multi-parasite burden in the damselfly Coenagrion armatum. Entomol Exp Appl 132:165–171 Khan MN, Sajid MS, Khan MK, Iqbal Z, Hussain A (2010) Gastrointestinal helminthiasis: prevalence and associated determinants in domestic ruminants of district Toba Tek Singh, Punjab, Pakistan. Parasitol Res 107:787–794 Kiffner C, Stanko M, Morand S, Khoklova IS, Shenbrot GI, Laudisoit A, Leirs H, Hawlena H, Krasnov BR (2013) Sex-biased parasitism is not universal: evidence from rodent-flea associations from three biomes. Oecol 173:1009–1022 Krasnov BR, Bordes F, Khokhlova IS, Morand S (2012) Gender-biased parasitism in small mammals: patterns, mechanisms, consequences. Mammalia 76:1–13 Kumar V (1999) Lecithodendriid infections Trematode infections and diseases of man and animals. Kluwer Academic Publishers, Boston, MA, p 363 Kunz TH, Anthony ELP (1982) Age estimation and postnatal-growth in the bat Myotis lucifugus. J Mammal 63:23–32 Kurta A, Baker R (1990) Eptesicus fuscus. Mamm Species 356:1–10 Kurta A, Kunz TH, Nagy KA (1990) Energetics and water flux of freeranging big brown bats (Eptesicus fuscus) during pregnancy and lactation. J Mammal 71:59–65 Lajeunesse MJ, Forbes MR, Smith BP (2004) Species and sex biases in ectoparasitism of dragonflies by mites. Oikos 106:501–508 Lee KA, Wikelski M, Robinson WD, Robinson TR, Klasing KC (2008) Constitutive immune defences correlate with life-history variables in tropical birds. J Anim Ecol 77:356–363 Lochmiller RL, Deerenberg C (2000) Trade-offs in evolutionary immunology: just what is the cost of immunity? Oikos 88:87–98 Lotz JM, Font WF (1983) Review of the Lecithodendriidae (Trematoda) from Eptesicus fuscus in Wisconsin and Minnesota. Proc Helminthol Soc Wash 50:83–102 Lotz JM, Font WF (1985) Structure of enteric helminth communities in 2 populations of Eptesicus fuscus (Chiroptera). Can J Zool 63:2969– 2978 Lotz JM, Font WF (2008) Family Lecithodendriidae. In: Bray RA, Gibson DI, Jones A (eds) Keys to the Trematoda, 3rd edn. CABI Publishing, Wallingford, UK, pp 527–536 MacIntosh AJJ, Hernandez AD, Huffman MA (2010) Host age, sex, and reproductive seasonality affect nematode parasitism in wild Japanese macaques. Primates 51:353–364 Mallon EB, Loosli R, Schmid-Hempel P (2003) Specific versus nonspecific immune defense in the bumblebee, Bombus terrestris L. Evolution 57:1444–1447 Martin LB, Han P, Lewittes J, Kuhlman JR, Klasing KC, Wikelski M (2006) Phytohemagglutinin-induced skin swelling in birds: histological support for a classic immunoecological technique. Funct Ecol 20:290–299

Martin LB, Weil ZM, Nelson RJ (2008) Seasonal changes in vertebrate immune activity: mediation by physiological trade-offs. Phil Trans Roy Soc B 363:321–339 Marzal A, Reviriego M, de Lope F, Moller AP (2007) Fitness costs of an immune response in the house martin (Delichon urbica). Behav Ecol Sociobiol 61:1573–1580 Matson KD, Ricklefs RE, Klasing KC (2005) A hemolysishemagglutination assay for characterizing constitutive innate humoral immunity in wild and domestic birds. Dev Comp Immunol 29:275–286 Mazerolle MJ (2006) Improving data analysis in herpetology: using Akaike’s Information Criterion (AIC) to assess the strength of biological hypotheses. Amphib-Reptil 27:169–180 McAllister C, Bursey C (2009) New host and geographic distribution records for helminths (Trematoda, Nematoda) in three species of vespertilionid bats (Chiroptera) from the Pine Ridge of Dawes County, Nebraska, USA. Comp Parasitol 76:117–121 McCurdy DG, Shutler D, Mullie A, Forbes MR (1998) Sex-biased parasitism of avian hosts: relations to blood parasite taxon and mating system. Oikos 82:303–312 Meeusen ENT, Balic A (2000) Do eosinophils have a role in the killing of helminth parasites? Parasitol Today 16:95–101 Mendonca MT, Chernetsky SD, Nester KE, Gardner GL (1996) Effects of gonadal sex steroids on sexual behavior in the big brown bat, Eptesicus fuscus, upon arousal from hibernation. Horm Behav 30: 153–161 Mills SC, Grapputo A, Jokinen I, Koskela E, Mappes T, Oksanen TA, Poikonen T (2009) Testosterone-mediated effects on fitness-related phenotypic traits and fitness. Amer Nat 173:475–487 Mills SC, Grapputo A, Jokinen I, Koskela E, Mappes T, Poikonen T (2010) Fitness trade-offs mediated by immunosuppression costs in a small mammal. Evolution 64:166–179 Moore SL, Wilson K (2002) Parasites as a viability cost of sexual selection in natural populations of mammals. Science 297:2015–2018 Moreno-Rueda G (2011) Trade-off between immune response and body mass in wintering house sparrows (Passer domesticus). Ecol Res 26: 943–947 Muehlenbein MP, Hirschtick JL, Bonner JZ, Swartz AM (2010) Toward quantifying the usage costs of human immunity: altered metabolic rates and hormone levels during acute immune activation in men. Am J Human Biol 22:546–556 Nakazawa M, Fantappie MR, Freeman GL, Eloi Santos S, Olsen NJ, Kovacs WJ, Secor WE, Colley DG (1997) Schistosoma mansoni: susceptibility differences between male and female mice can be mediated by testosterone during early infection. Exp Parasit 85: 233–240 Nickel PA, Hansen MF (1967) Helminths of bats collected in Kansas, Nebraska, and Oklahoma. Am Mid Nat 78:481–486 Norris K, Evans MR (2000) Ecological immunology: life history tradeoffs and immune defense in birds. Behav Ecol 11:19–26 Nunn CL, Lindenfors P, Pursall ER, Rolff J (2009) On sexual dimorphism in immune function. Phil Trans Roy Soc 364:61–69 Palacios MG, Cunnick JE, Winkler DW, Vleck CM (2012) Interrelations among immune defense indexes reflect major components of the immune system in a free-living vertebrate. Physiol Biochem Zool 85:1–10 Pearce RD, O’Shea TJ, Wunder BA (2008) Evaluation of morphological indices and total body electrical conductivity to assess body composition in big brown bats. Acta Chiropter 10:153–159 Perkins SE, Ferrari MF, Hudson PJ (2008) The effects of social structure and sex-biased transmission on macroparasite infection. Parasitol 135:1561–1569 Pistole D (1988) A survey of helminth parasites of chiropterans from Indiana. Proc Helminth Soc Wash 55:270–274

Parasitol Res Porteous IS, Pankhurst SJ (1998) Social structure of the mara (Dolichotis patagonum) as a determinant of gastro-intestinal parasitism. Parasitol 116:269–275 Poulin R (1996) Helminth growth in vertebrate hosts: does host sex matter? Int J Parasitol 26:1311–1315 Rausch R (1975) Cestodes of the genus Hymenolepis Weinland, 1858 (sensu lato) from bats in North America and Hawaii. Can J Zool 53:1537–1551 Roberts ML, Buchanan KL, Evans MR (2004) Testing the immunocompetence handicap hypothesis: a review of the evidence. Anim Behav 68:227–239 Robinson SA, Forbes MR, Hebert CE, McLaughlin JD (2008) Malebiased parasitism by common helminths is not explained by sex differences in body size or spleen mass of breeding cormorants Phalacrocorax auritus. J Avian Biol 39:272–276 Robinson SA, Forbes MR, Hebert CE, McLauglin JD (2010) Male biased parasitism in cormorants and relationships with foraging ecology on Lake Erie, Canada. Waterbirds 33:307–313 Rolff J (2002) Bateman’s principle and immunity. Proc Roy Soc B 269: 867–872 Rossin MA, Malizia AI, Timi JT, Poulin R (2010) Parasitism underground: determinants of helminth infections in two species of subterranean rodents (Octodontidae). Parasitol 137:1569–1575 Russo J, Madec L (2013) Linking immune patterns and life history shows two distinct defense strategies in land snails (Gastropoda, Pulmonata). Physiol Biochem Zool 86:193–204 Schalk G, Forbes MR (1997) Male biases in parasitism of mammals: effects of study type, host age, and parasite taxon. Oikos 78:67–74 Schell SC (1985) Handbook of trematodes of North America north of Mexico. Idaho Research Foundation, Moscow, ID Schmid-Hempel P (2003) Variation in immune defence as a question of evolutionary ecology. Proc Roy Soc B 270:357–366 Schmid-Hempel P, Ebert D (2003) On the evolutionary ecology of specific immune defence. Trends Ecol Evol 18:27–32 Schroderus E, Jokinen I, Koivula M, Koskela E, Mapps T, Mills SC, Oksanen TA, Poikonen T (2010) Intra- and intersexual trade-offs

between testosterone and immune system: implications for sexual and sexually antagonistic selection. Amer Nat 176:E90–E97 Schulte-Hostedde AI, Zinner B, Millar JS, Hickling GJ (2005) Restitution of mass-size residuals: validating body condition indices. Ecology 86:155–163 Sheldon BC, Verhulst S (1996) Ecological immunology: costly parasite defences and trade-offs in evolutionary ecology. Trends Ecol Evol 11:317–321 Stoehr AM, Kokko H (2006) Sexual dimorphism in immunocompetence: what does life-history theory predict? Behav Ecol 17:751–756 Superina M, Sierra R (2008) Hematology and serum chemistry values in captive and wild pichis, Zaedyus pichiy (Mammalia, Dasypodidae). J Wildl Dis 44:902–910 Talleklint-Eisen L, Eisen RJ (1999) Abundance of ticks (Acari: Ixodidae) infesting the western fence lizard, Sceloporus occidentalis, in relation to environmental factors. Exp Appl Acarol 23:731–740 Tkach VV (2008) Family Plagiorchiidae. In: Bray RA, Gibson DI, Jones A (eds) Keys to the Trematoda, 3rd edn. CAB International, Wallingford, UK, pp 296–325 Tkach VV, Pawlowski J, Sharpilo VP (2000) Molecular and morphological differentiation between species of the Plagiorchis vespertilionis group (Digenea, Plagiorchiidae) occurring in European bats, with a re-description of P. vespertilionis (Muller, 1780). Syst Parasitol 47: 9–22 Ujvari B, Madsen T (2006) Age, parasites, and condition affect humoral immune response in tropical pythons. Behav Ecol 17:20–24 Waterman JM, Macklin GF, Enright C (2014) Sex-biased parasitism in Richardson’s ground squirrels (Urocitellus richardsonii) depends on the parasite examined. Can J Zool 92:73–79 Williams RR (1960) Acanthatrium lunatum n. sp., a parasite of the big brown bat and a key to the described species of Acanthatrium (Trematoda: Lecithodendriidae). Ohio J Sci 60:323–326 Zuk M (2009) The sicker sex. PLoS Pathog 5:3 Zuk M, McKean KA (1996) Sex differences in parasite infections: patterns and processes. Int J Parasitol 26:1009–1023