Immunomodulation in post-metamorphic northern leopard frogs

Immunomodulation in Post-metamorphic Northern Leopard Frogs,Lithobates pipiens, Following Larval Exposure to PolybrominatedDiphenyl EtherTawnya L. Cary,*,† Manuel E. Ortiz-Santaliestra,‡ and William H. Karasov†,‡

†Department of Zoology and ‡Department of Forest & Wildlife Ecology, University of Wisconsin, Madison, Wisconsin 53706, UnitedStates

ABSTRACT: Pollutants and disease are factors implicated inamphibian population declines, and it is hypothesized that thesefactors exert a synergistic adverse effect, which is mediated bypollutant-induced immunosuppression. Polybrominated diphenylethers (PBDEs) are ubiquitous pollutants that can exertimmunotoxicity, making them of interest to test effects onamphibian immune function. We orally exposed Lithobates (Rana)pipiens tadpoles to environmentally realistic levels (0−634 ng/gwet diet) of a pentabromodiphenyl ether mixture (DE-71) from assoon as they became free-swimming through metamorphic climax.To assess adaptive immune response in juvenile frogs, we used anenzyme-linked immunosorbent assay to measure specific IgYproduction following immunization with keyhole limpet hemocyanin (KLH). Specific KLH antibody response was significantlydecreased in juvenile frogs that had been exposed to PBDEs as tadpoles. When assessing innate immune responses, we foundsignificantly different neutrophil counts among treatments; however, phagocytic activity of neutrophils was not significantlydifferent. Secretion of antimicrobial skin peptides (AMPs) nonsignificantly decreased with increasing PBDE concentrations, andno significant effect of PBDE treatment was observed on efficacy of AMPs to inhibit chytrid fungus (Batrachochytriumdendrobatidis) growth. Our findings demonstrate that environmentally realistic concentrations of PBDEs are able to alter immunefunction in frogs; however, further research is needed to determine how these alterations impact disease susceptibility in L.pipiens.

■ INTRODUCTION

Global amphibian populations have been declining at analarming rate, and currently almost one-third of amphibianspecies are endangered.1,2 Pollution is a key factor contributingto the loss of amphibian biodiversity,2 and one class of flameretardants, polybrominated diphenyl ethers (PBDEs), hasgarnered increased attention over the past two decades.These compounds have been used extensively in plastics andtextiles and are ubiquitous in the environment in both bioticand abiotic matrices.3,4 Although production of the penta- andocta-BDE mixtures has been banned in the European Unionand voluntarily phased out in the United States, lower-brominated PBDE concentrations continue to persist at levelsreported to cause toxicity to humans and wildlife.5−7 This islikely due to environmental persistence of PBDEs and thedebromination of the deca-brominated congener BDE-209,8−10

which has remained in production in the United States(although production, sale, and import of decaBDE will bephased out by the end of 201311). Concentrations of PBDEs inGreat Lakes food webs range from 1.4 ng/g in zooplankton to>1000 ng/g tissue in top-predator fishes.3,12 Water concen-trations of PBDEs are relatively low due to the hydrophobicnature of PBDEs (54.9 pg BDE-47/L in San Francisco Baywatershed (no Great Lakes data available)13), while organic

matter contains greater concentrations and creates the basis forpotential biomagnification (tri- through hepta-BDEs in surficialsediments are 2.8 ng/g dry wt).14

Toxicity of PBDE exposure includes impaired neurologicalfunction (decreased spatial memory and learning15,16), thyroidhormone disruption,17−19 and developmental20−23 and repro-ductive toxicity.24,25 Additionally, alterations in immunefunction have been reported in humans and mice.26,27 Micesubchronically exposed to DE-71, a commercial pentaBDEmixture, had suppressed plaque-forming cell response againstsheep red blood cells and decreased thymus weight.28 Exposureto BDE-47, a major congener component of DE-71, reducedsplenocyte numbers in mice.29 Ranch mink, Mustela vison,exposed to DE-71 had altered secondary antibody productionto keyhole limpet hemocyanin as well as a higher percentage ofneutrophils compared to nonexposed individuals.30 In contrast,these same DE-71-exposed mink had lowered hematocrits andlowered percent lymphocytes compared to control mink.30

Amphibians have many of the same immune components as

Received: December 27, 2013Revised: March 30, 2014Accepted: April 15, 2014

Article

pubs.acs.org/est

© XXXX American Chemical Society A dx.doi.org/10.1021/es405776m | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

mammals,31,32 and therefore, PBDEs may have similar impactson anuran immune function. Also, thyroid hormone homeo-stasis is necessary for normal immunological development infrogs,33 and toxicants that disturb neuroendocrine control ofmetamorphosis may also alter immune function.It is hypothesized that direct or indirect suppression of the

immune system by contaminants could prevent adequateimmune responses in amphibians leading to increasedsusceptibility to pathogens.34 Because disease is anothermajor factor in amphibian population declines,2 it is of interestto assess immunotoxicity in amphibians, and recently, increaseddisease susceptibility following tadpole exposure to an herbicidewas shown in juvenile tree frogs.35 Although there has beenextensive research on the immune system of Xenopus laevis(reviewed in refs 32 and 36), only a few studies have tested theeffects of contaminants on amphibian immune function (e.g.,refs 37 and 38). Our goal was to assess whether PBDE exposureof larval Lithobates (Rana) pipiens resulted in long lasting, carry-over effects on the immune function of post-metamorphicfrogs. The present study tested for changes in measures of bothinnate and adaptive immunity in northern leopard frogs inresponse to dietary exposure of environmentally relevant levelsof PBDEs and, to our knowledge, is the first study thatinvestigated immunotoxic effects of PBDEs in amphibians.

■ METHODSStudy Organism. Five batches of Lithobates pipiens

embryos (∼200 embryos/batch) were purchased from Nasco(Ft. Atkinson, WI, USA), transferred to the laboratory, andimmediately placed in 0.5 L Nalgene containers filled with 300mL of dechlorinated, filtered tap water. Each container housed40 embryos, and water was changed twice daily until hatch.Nonviable embryos were removed daily to minimize bacterialgrowth. Following hatch, the number of tadpoles was culled to20 healthy tadpoles per container, and water was changed dailyuntil the tadpoles became free swimming, developedoperculum, and began foraging (developmental Gosner Stage25, GS 2539). All experimental procedures involving tadpolesand/or recent metamorphs were approved by the University ofWisconsin’s Institutional Animal Care and Use Committee.Dietary PBDE Exposure of Tadpoles. Polybrominated

diphenyl ethers are lipophilic compounds with high octanol−water partition coefficients (e.g., log KOW pentaBDE mixture =6.5840). Because PBDEs are readily detected in sediments andbiota and less so in dissolved and/or particulate aquaticphases,4 we used dietary exposure to better mimic the primaryecological exposure route. Beginning on February 22, 2009,free-swimming tadpoles (GS 25) were transferred to 18.9 Laquaria and fed diets without (control) and with a technicalpentaBDE mixture, DE-71, at four doses. The technical DE-71mixture consisted primarily of pentaBDE [56%], tetraBDE[35%], and hexaBDE [9%] and was purchased from WellingtonLaboratories (product TBDE-71; purity undetermined; percen-tages as determined by Wellington Laboratories). Control andPBDE-spiked tadpole food were made using powderized rabbitchow23 (Harlan Rabbit Chow catalog no. 2030, 3.3% lipids).Measured concentrations of DE-71 adsorbed to the diets were0 (nondetect), 1.1, 6.1, 71.4, and 634 ng DE-71/g diet wetweight.41 These exposure levels have been previously reportedto yield ecologically relevant tissue concentrations in tadpolesand froglets (≤660 ng ΣPBDEs/g wet mass42) and werechosen based on a previous PBDE exposure study with L.pipiens23 and reported PBDE body burdens in fish, crustaceans,

and zooplankton from the Laurentian Great Lakes’ waters.3,12

Specifically, tadpoles fed concurrently in our facility on two ofthe same diets described above (71.4 and 634 ng PBDE/g) hadbody burden residues of 46.3 and 632.4 ng PBDE/g tissue,respectively, after feeding for 50 days.43

The fertilized embryos obtained from Nasco are offspring offield-collected adult males and females from southeasternMinnesota and Wisconsin, such that we expect the tadpoles inour study to be to representative of natural populations.However, information regarding genetic relatedness withinand/or across batches was not provided by the supplier, so as aprecaution batches of tadpoles were not mixed and each batchwas replicated twice at each treatment level. Therefore, eachdietary treatment was replicated in 10 different glass aquaria (5batches × 2 aquaria/batch) containing 20 tadpoles each in 12 Lof dechlorinated, filtered tap water. Tadpoles were housedcommunally at this density (1.7 tadpoles per 1 L water) inorder to balance healthy living conditions and resourcesnecessary to accommodate this number of tadpoles. Environ-mental conditions were 14:10 h light:dark light cycle, 23 ± 1°C water temperature, and humidity ≥30%. Static renewal ofthe aquaria water (≥80% water change) was performed fourtimes per week to provide satisfactory water quality: pH = 7.8−8.3, nitrite < 1.0 mg/L, total NH3 < 1 mg/L). Additionally, airstones in each aquarium provided sufficient aeration through-out the experiment (dissolved oxygen ≥75% saturation).Tadpoles were fed ad libitum each day by placing three blocksof the appropriate diet into the aquaria (the size of blocksincreased with tadpole growth but averaged 0.5 g wet mass;range of 0.25−0.95 g). Prior to water changing and the additionof fresh food, any remaining food from the previous day as wellas feces were removed.The present immune toxicity study was part of a broader

experiment designed to test for both immune and reproductivetoxicity of PBDE exposure. The findings of reproductivetoxicity have been reported previously.41 Tadpoles wereassigned to each component of the broader study by using alottery system within each replicated aquarium to ensurerandom placement. Therefore, only a portion of the frogsdescribed above was assigned to the present study to analyzeimmunotoxic end points [at metamorphic climax (GS 42), 218frogs were assigned to immunotoxic end points; however only102 frogs survived to 10 weeks post-metamorphosis for samplecollection].Dietary exposure continued until metamorphic climax. When

larvae reached metamorphic climax (emergence of forelimbs atGS 42), they were removed from the aquaria and placedindividually into a 500 mL polypropylene jar. Becausemetamorphosis and early post-metamorphosis are sensitiveperiods for frogs, we housed frogs individually to minimizemortality risks (e.g., competition, pathogen transmission,stress). Twenty-five milliliters of dechlorinated, filtered tapwater was added to the jars, and the jars were tilted to provideboth a wet and dry surface for the metamorphosing frogs. Frogswere not fed during metamorphosis as tail resorption and fatreserves provide adequate energy during this time period.Newly metamorphosed juveniles (GS 46) were fed uncon-taminated crickets and mealworms six times per week untilcollection at 10 weeks post-metamorphosis. In order to provideproper nutrition, crickets were dusted with calcium andvitamins prior to being fed to the froglets. The depuration ofcontaminant levels during this period (e.g., ref 43) is not amajor concern because the primary research question focuses

Environmental Science & Technology Article

dx.doi.org/10.1021/es405776m | Environ. Sci. Technol. XXXX, XXX, XXX−XXXB

on how PBDE exposure during larval development affectsimmune function post metamorphosis.Adaptive Immune Assay: Enzyme-Linked Immuno-

sorbent Assay (ELISA). Vaccination and Blood Collection.Seven weeks post-metamorphosis, frogs were intraperitoneal(i.p.) injected with 25 μL of the antigen [keyhole limpethemocyanin (2 mg KLH/25 mL PBS)] and Titer Max Gold(Sigma), an adjuvant previously used in L. pipiens and known tostimulate antibody production.44 Keyhole limpet hemocyanin isa large protein with many epitopes45 and is able to elicitantibody responses in frogs.44 Two weeks later, frogs wereboosted with a second injection of KLH (2 mg KLH/25 mLPBS; without adjuvant) to initiate a secondary adaptiveimmune response to the antigen. One week following thebooster injection, frogs were anesthetized in 0.2% MS-222, andblood was collected directly from the heart using a 28G needle.The blood was immediately transferred to heparin-coatedmicrohematocrit capillary tubes and centrifuged to separate theplasma from the red blood cells. The plasma was stored at −20°C until ELISA analysis.Specific Antibody Response to Keyhole Limpet Hemocya-

nin. Antibody response in 10-week-old L. pipiens wasdetermined using a modification of enzyme-linked immuno-sorbent assays (ELISA) previously reported using frogplasma.42 High binding, flat-bottomed microtiter plates(NUNC, cat. no. 442404) were coated with 100 μL of KLH(1 mg/mL) diluted in coating buffer (0.015 M Na2CO3, 0.035M NaHCO3, pH 9.6) and incubated at 4 °C overnight. Eachplate was washed with wash buffer (PBS with 0.05% Tween)using a Biotek ELx405 microplate washer and blocked withblocking buffer (5% nonfat dry milk with 0.5% Tween). Plateswere incubated for 1 h at 37 °C and washed with wash buffer.Diluted plasma samples (1:100 in blocking buffer) were thenadded to the appropriate wells and incubated at 37 °C for 1 h.Specific anti-KLH antibodies in the frog plasma were detectedby adding 100 μL of monoclonal mouse anti-Xenopus IgY(11D5, Xenopus laevis Research Resource for Immunology,University of Rochester Medical Center) to the wells. Plateswere incubated at 37 °C for 1 h. After washing, 100 μL ofhorseradish peroxidase-conjugated rabbit anti-mouse IgG(Sigma Chemicals, cat. no. A-9044) diluted 1:5000 (in blockingbuffer) was added to each well and again incubated for 1 h at 37°C. The plates were washed, and 150 μL of horseradishperoxidase substrate (1-Step ABTS, Pierce) was added to thewells. The reaction was allowed to develop for 45 min at roomtemperature and then stopped by adding 100 μL of 1% sodiumdodecyl sulfate to each well. Absorbance values were read at405 nm on a Wallac Victor2 microplate reader, and adjustedabsorbance values (mean sample absorbance minus the meannegative control absorbance value) were recorded. Plasma fromnon-immunized frogs was used as a negative control. Positivecontrols consisted of plasma from frogs that were previouslydetermined to produce anti-KLH IgY ≥ 2-fold that of thenegative control plasma. Both negative and positive controlsamples were included on each plate to validate and normalizeplate-to-plate differences.Total Antibody Response. Total IgY and IgM were

evaluated for the same juvenile frogs that were analyzed forspecific anti-KLH IgY. The ELISA method used was the sameas above for specific anti-KLH, except microtiter plates werecoated with 100 μL of diluted frog plasma in coating buffer andincubated at 4 °C overnight. Each plate was washed andblocked as above and the appropriate primary antibody [mouse

anti-Xenopus IgY (11D5) or mouse anti-Xenopus IgM (10A9)]was added to each well. Following incubation, the plates werewashed and the secondary antibody was added. Antibodydetection was determined as above. The negative control fortotal antibody response was plasma from a nonreactive species(Zebra finch collected by Tess Killpack, University of WI-Madison), and the positive control was Xenopus tropicalisplasma.

Innate Immune Assay: Leukocyte Recruitment andPhagocytic Activity. A lavage assay designed to collectperitoneal leukocytes was adapted from Vatnick et al.46 Tenweeks post-metamorphosis, frogs were given an i.p. injection ofthioglycollate medium (0.067 mL/g body mass) to stimulateleukocyte extravasation. This same inoculum containedfluorescently labeled 1 μm microbeads (Fluoresbrite CarboxyBB microspheres, 2.5 × 107 beads/mL) at a concentration of0.55 μL/mL thioglycollate. After 24 h, individuals wereanesthetized in 0.1% MS-222 (tricaine methanesulfonate),and phosphate buffered saline (PBS) was i.p. injected into thelower abdomen at a volume of 0.33 mL/g frog using a 27Gneedle. The abdomen was then gently massaged in order toflush the intraperitoneal cavity and suspend peritonealleukocytes. A 23G needle was used to aspirate the PBSsolution containing suspended cells. The cell suspension wasdiluted and stained with trypan blue for determination of cellviability, and total leukocyte counts (cells/mL) were obtainedusing a hemacytometer. Neutrophils comprised >95% of theleukocytes collected from the peritoneal cavity, and thus,leukocyte counts were characterized as the number ofneutrophils.Phagocytic activity of the neutrophils was determined by

counting 100 neutrophils and determining the percentage ofcells that contained fluorescently labeled microbeads. Fluo-rescently labeled microbeads within neutrophils were visualizedusing a Nikon fluorescent microscope with a UV filter under600× magnification. The assay was optimized for the propermicrobead to neutrophil ratio such that excess microbeadswould allow cells amble opportunity to engulf microbeads, butnot so high to create aggregates of microbeads.

Skin Peptide Collection and Quantification. A total of45 7-week-old frogs from all treatments were injected in thedorsal lymph sac with 5 μL per g body mass of 1 mMnorepinephrine (i.e., 5 nmol/g) as norepinephrine-bitartratesalt in amphibian phosphate buffered saline (APBS), whichstimulates antimicrobial peptide (AMP) secretion.47 Frogs werethen placed in 50 mL of collection buffer (2.92 g NaCl and 2.05g sodium acetate in 1 L of HPLC-grade water) for 15 min aspreviously described.47−49 Following collection, the frogs wereremoved from the buffer, and the buffer was acidified by adding1 mL of 50% HCl. Peptides were partially enriched over C18Sep-Paks (Waters Corp. WAT020515, Milford, MA, USA) asdescribed by Rollins-Smith et al.47 and quantified using theMicroBCA assay (Pierce, Rockford, IL, USA) according tomanufacturer’s instructions except that the peptide bradykinin(RPPGFSPFR) (Sigma Chemical, St. Louis, MO, USA) wasused as a standard.47−49 Serially diluted bradykinin solutionswere measured at 562 nm to establish a standard curve, andunknown samples were referenced against the standard curve.The mass of each frog was determined at the time of the

injection, and total secreted peptides were quantified. Toestimate the total amount of peptides recovered, the surfacearea was calculated according to the method of McClanahanand Baldwin50 [surface area (cm2) = 9.9 (mass in grams)0.56].

Environmental Science & Technology Article

dx.doi.org/10.1021/es405776m | Environ. Sci. Technol. XXXX, XXX, XXX−XXXC

The thickness of the mucus was assumed to be 50 μm,51 andtherefore the volume of mucus covering 1 cm2 of skin would be5 μL. As a result, the concentration of peptides in mucus (μg/mL) = total peptides (μg) per cm2 × 200 (because 1 mL = 5 μL× 20048,49).Determination of Antifungal Activity of Skin Pep-

tides. Collected skin secretions were freeze-dried and stored at−20 °C until analysis. Skin secretions are known to be a majorimmune mechanism of amphibians against the deadly fungusBatrachochytrium dendrobatidis (Bd).52 The antifungal efficacyof collected skin peptides and the potential effect of PBDEexposure on potency were estimated through the calculation ofthe minimum inhibitory concentration (MIC), i.e., the lowestpeptide concentration capable to significantly reduce fungalgrowth. Because of the low peptide amount in some secretionsamples, we pooled together the samples corresponding toseveral individuals (n = 3 or 4) from the same DE-71 treatment.Dried secretion samples were resuspended in PBS to anestimated concentration of 4000 μg equivalents of bradykinin/mL. The concentration in the resulting dilution was checkedagain using the MicroBCA assay, and serial dilutions of skinpeptides (2000−15.625 μg eq/mL) were prepared for eachsample. We added 25 μL of each dilution in triplicate to wells ofa 96-well microplate; in addition, we prepared three extra wellsof the highest peptide concentration to be used for the negativecontrol.Batrachochytrium dendrobatidis culture (JEL197) was re-

ceived from stocks maintained by Joyce Longcore at theUniversity of Maine. This culture corresponds to the typeisolate of Bd and was isolated from a Blue Poison Dart Frog(Dendrobates azureus) from the National Zoological Park(Washington, DC, USA) in 1997.53 Pieces of thalli takendirectly from the culture received were immersed in vialscontaining 1% tryptone broth and incubated at 16 °C.Zoospore harvesting was performed as described in previousstudies (e.g., ref 54). Briefly, we added ∼500 μL of this brothculture to 1% tryptone agar in 9 cm culture dishes andincubated them at 16 °C for 7 days. Dishes with growing thalliwere then flooded with 5 mL of water that was previouslydisinfected by passing through a 0.45 μm filter. After 30 min,we decanted the plates to collect the zoospore solution.Zoospore concentration was measured by counting a measuredaliquot on a hemocytometer, and sterile distilled water was thenadded to adjust for a final working dilution of 5 × 104

zoospores/mL.We added 25 μL of the working solution of Bd into each well

with the exception of those corresponding to negative controls,where we added 25 μL of the working solution that waspreviously autoclaved for 20 min at 120 °C to kill thezoospores. Microplates were then incubated at 16 °C for 7days. Fungal growth was estimated directly by measuring theoptical density at 492 nm in a Wallac Victor2 microplate reader.A linear regression between the skin peptide concentration

and the optical density (both log transformed to attainlinearity) was calculated for each pooled sample. MIC wascalculated as the concentration corresponding to the valuewhere this regression line crossed the upper limit of the 95%confidence interval of the optical density readings obtainedfrom the negative control (adapted from ref 55).Statistical Analyses. Statistical analyses were performed in

SAS (Version 8.01) or R (Version 2.10.21, R Foundation forStatistical Computing). Proportional data for survival andphagocytosis were adjusted with arcsine square-root trans-

formation. Gosner Stage data were rank transformed.Leukocyte counts and absorbance data were adjusted withlog10 transformation to ensure normality. A mixed model one-way analysis of variance (ANOVA) was performed todetermine differences in tadpole parameters among treatments(fixed factor was treatment, random factor was replicate tanknested in embryo batch, and response variables were percentsurvival, total length, and Gosner Stage). A one-way ANOVAwas performed to determine differences in innate immuneparameters among treatments (independent variable wastreatment, and response variables were leukocyte number,percent phagocytosis, extracted skin peptide mass per cm2 andMIC). An analysis of covariance was performed to determinedifferences in adaptive immune parameters among treatments(independent variable was treatment, covariate was snout−ventlength or mass of the juvenile frog, and response variable wasthe adjusted absorbance values for specific anti-KLH IgY, totalIgY, or total IgM). Akaike information criterion (AIC) was usedfor model selection. Post hoc multiple comparisons were madesimultaneously using Tukey’s HSD. Statistical significance wasaccepted for p < 0.05.

■ RESULTSGrowth and Development of Tadpoles. To determine

whether PBDE exposure impacted survival, growth, anddevelopment of the tadpoles, we assessed percent survival,total length, and Gosner Stage (GS) on day 53 of theexperiment. This time point corresponded to the longestexposure window before tadpoles began undergoing meta-morphosis, which would confound measurements and theability to make statistical comparisons among treatments. Seventadpoles per replicated aquaria were measured for total lengthand visually analyzed for GS development (N = 70 for 0, 1.1,and 634 ng/g; N = 56 for 6.1 and 71.4 ng/g because twoaquaria per treatment had to be removed from the sample dueto experimenter error). On day 53, percent tadpole survival wasnot significantly different across treatments (F4,26 = 0.52, p =0.721) and was ≥94.5% for all treatment groups. At this sametime point, total length of tadpoles differed significantly amongtreatments (F4,308 = 2.81, p = 0.026). The 71.4 ng PBDE/gtreated tadpoles were on average 10% shorter than the controltadpoles (p = 0.013; Figure 1a). Tadpole GS also differedsignificantly among treatments (F4,308 = 4.11, p = 0.003). The71.4 ng PBDE/g treatment had significantly slower develop-ment compared to the control tadpoles (p = 0.003; Figure 1b).

Adaptive Immune Function. Specific IgY antibodyresponse to KLH was significantly different among treatmentgroups (F4,44 = 2.73, p = 0.041; n = 6−13). Specifically, dietaryexposure of 71.4 ng PBDE/g significantly decreased anti-KLHIgY response compared to the control-treated frogs (p = 0.018;Figure 2a). We further tested the amount of total IgY and IgMlevels in these frogs to measure overall antibody response to allantigens (Figure 2b). Due to insufficient sample volume, wewere unable to test some individuals for total immunoglobulinlevels. This is reflected in the sample size difference betweenthe total IgY and IgM compared to the specific-KLH response.The ability of juvenile frogs to generate total immunoglobulinsof both isotypes did not differ significantly with PBDEtreatment [IgY: F4,39 = 0.66, p = 0.624 (n = 6−12); IgM:F4,39 = 0.70, p = 0.597 (n = 6−12)].

Innate Immune Function. The recruitment of neutrophils,which ranged from 7.0 × 104 to 5.6 × 105 cells/mL, differedsignificantly among treatments (F4,30 = 3.36, p = 0.022; n = 5−

Environmental Science & Technology Article

dx.doi.org/10.1021/es405776m | Environ. Sci. Technol. XXXX, XXX, XXX−XXXD

10 per treatment). Recruitment means were higher in the threehigher concentration PBDE treatments, and post-hoc Tukey’scomparisons determined that the 1.1 and 6.1 ng PBDE/gtreatments differed significantly from each other in number ofneutrophils (p = 0.046), but the mean number of neutrophils inthe control treatment was not statistically different from themean number of neutrophils in any of the PBDE treatments (p≥ 0.15; Figure 3a). Phagocytic activity was broken down intofour levels to discriminate between cells that engulfed no beads(none), 1−2 beads (low), 3−9 beads (medium), and 10 ormore beads (high). The percentage of neutrophils thatphagocytosed microbeads did not differ significantly amongtreatment groups for all four levels: none (F4,25 = 2.13, p =0.107), low (F4,25 = 1.77, p = 0.167), medium (F4,25 = 1.32, p =0.289), and high (F4,25 = 1.31, p = 0.294) (Figure 3b). In allsamples, excess microbeads were present in the interstitial fluid,and no aggregates of microbeads were observed outside of thecells.Neither the amount of antimicrobial skin peptides extracted

per cm2 (F4,10 = 0.832, p = 0.535) nor their potency to inhibitBd growth, estimated from the MIC (F4,8 = 0.507, p = 0.732)were significantly affected by larval exposure to PBDE (Figure4).

■ DISCUSSION

Growth and Development. Tadpoles were fed environ-mentally realistic concentrations of DE-71 in order to assesspotential immunotoxic effects of PBDEs; however, we alsomeasured survival, growth, and developmental end points tocompare with other studies. Survivorship of tadpoles untilexposure day 53 was greater than 94.5% for all treatmentgroups and was not significantly affected by exposure toPBDEs, although our earlier study did find a significant declinedue to PBDE exposure.23 As in our earlier study,23 both growthand development rate were significantly depressed, comparedwith controls, in some of the dose groups (Figure 1). Thenonlinear dose response in tadpole growth and development issimilar to previous findings with PBDE exposure23 and isconsistent with exposure to endocrine disrupting compounds inwhich low doses have large impacts at the hormone level.56

Tadpole growth and development is driven by circulatingthyroid hormone levels, and PBDEs have been shown to lowercirculating thyroid hormones,17,57 which may explain thegrowth and developmental delays demonstrated in this study.

Adaptive Immune Response. Amphibians possess manycomponents of adaptive immunity that have been characterizedin mammals including B and T lymphocytes, immunoglobulins(Ig), T-cell receptors, major-histocompatibility complex, andrecombination-activating genes.32 We tested the ability ofjuvenile frogs to mount a specific antibody response followingantigenic challenge to keyhole limpet hemocyanin (KLH).Larval PBDE exposure significantly lowered the ability of post-metamorphic frogs to mount an adaptive humoral response.Juvenile frogs exposed to 71.4 ng PBDE/g as tadpoles had92.4% lower levels of antibodies produced that were specific toKLH compared to control frogs. The other PBDE-exposedfrogs (1.1, 6.1, and 634 ng/g) had, on average, 89.2%, 48.8%,and 66.8% lower secondary antibody responses, respectively,compared to the controls. Exposure to PBDEs did not affecttotal production of either IgY or IgM isotypes in 10-week-oldfrogs. As total IgY and KLH-specific IgY both measure theability of frogs to mount a secondary antibody response, wemight expect both measures to follow a similar pattern. Becausewe found a significant decrease in specific-KLH IgY antibodyproduction with PBDE exposure, which was not apparent withtotal IgY measures, immunological memory against KLH was amore sensitive immunotoxic measure than total IgY. Measuringantibody responses to a specific antigen, such as KLH, controlsfor the timing of antibody class-switching and provides a morerefined measure of the organism’s adaptive antibody response.Total IgY includes all antibodies produced against all possibleantigens regardless of the timing of when the organism wasexposed, such that the response may be muddled, therefore notproviding as clear a response as specific-KLH IgY. Total IgMlevels were similar among treatments; this provides evidencethat PBDE exposure did not suppress the ability of juvenilefrogs to produce antibodies in general, but that perhaps theprocess of generating memory B cells with antibodies specific toKLH was impaired.In mammals, alteration of lymphocyte numbers, lymphoid

organs, and antibody responses following PBDE exposure havebeen reported (e.g., refs 27−30 and 58). For example, miceexposed subchronically to 1000 μg DE-71/g body mass hadsuppressed primary antibody response to sheep red blood cells(sRBC), which was associated with decreased thymus weight atthe same concentration,28 and in harbor porpoises, both thymic

Figure 1. (A) Tadpole length and (B) Gosner Stage (i.e.,developmental stage) on day 53 of the DE-71 exposure experiment.Data were analyzed using a mixed model analysis of variance.Treatment groups that share the same letter are not statisticallysignificant from each other (p > 0.05; Tukey post-hoc comparisons).Gosner Stage data were rank transformed for statistical analysis but arepresented here with observed Gosner Stage values for biologicalclarity. (B) N = 70 for 0, 1.1, and 634 ng/g; N = 56 for 6.1 and 71.4ng/g. Data is presented as the mean ± standard error of the mean.

Environmental Science & Technology Article

dx.doi.org/10.1021/es405776m | Environ. Sci. Technol. XXXX, XXX, XXX−XXXE

atrophy and splenic depletion were significantly correlated toincreased PBDE body burdens.58 Similar suppression ofhumoral immunity has been shown in frogs exposed topesticide mixtures, which has also been linked to decreasedsplenocyte cellularity.44,59 Loss of lymphocytes is likely thecause for the decreased primary antibody responses, and thismay be driven by depression of lymphocyte populations in boththe spleen and thymus. Because exposed frogs in the presentstudy retained the ability to generate total IgM at the same levelas that of unexposed frogs, we do not believe this is the mainmode of action for PBDE-induced immunomodulation in L.pipiens at environmentally relevant concentrations. It is possiblethat with increased PBDE concentration, systematic toxicityleads to apoptosis of lymphocytes and that our present studyobserved immunomodulation in frogs at a toxin level that didnot result in systemic toxicity. However, we are limited in this

discussion as we did not measure lymphoid organ masses (e.g.,spleen or thymus) or cellularity in our frogs.The suppression of humoral immunity in frogs exposed to

PBDEs is concerning because appropriate antibody responsesare necessary for protection against pathogens. Specifically,clearance of ranavirus infections in adult frogs has been linkedto humoral immunity,42,60 and antibodies specific to Bd areproduced upon infection;48 however, antibody responses to Bdare likely not as protective as innate immune responses due tosuppression of the adaptive immune response by thefungus.61,62

Innate Immune Response. In order to provide anaccurate assessment of an organism’s immune health, a rangeof assays to examine the functionality of an immune response isnecessary.63 Because of this we also assessed features of theinnate immune system in juvenile frogs previously exposed to

Figure 2. (A) Box-whisker plot of the level of specific anti-KLH IgY in plasma of frogs following a second (booster) injection of KLH. Thehorizontal line in each box is the median, the top and bottom of the box represent the 75th and the 25th percentile, respectively, and the whiskersdefine the 5th and 95th percentile observations. Specific anti-KLH IgY levels differed significantly across treatments (analysis of covariance, p =0.041). Treatments sharing the same letter (above box-whisker plot) are not significantly different (Tukey post-hoc comparisons, p > 0.05). Frogspreviously exposed to 71.4 ng PBDE/g as tadpoles had significantly lower specific anti-KLH IgY compared to control frogs (p = 0.018). (B) Totalimmunoglobulin (IgY and IgM) levels in juvenile frogs were not statistically different among treatments (p = 0.624 and 0.597, respectively). Data ispresented as the mean ± standard error of the mean.

Figure 3. (A) Leukocyte counts measuring the recruitment of neutrophils differed significantly among treatments (analysis of variance, p = 0.022),but PBDE-exposed groups were not significantly different from controls. Treatment groups sharing the same letter are not statistically different fromeach other (p > 0.05; Tukey post-hoc comparisons). *The highest dose showed a statistical trend for increased neutrophil recruitment compared tothe 1.1 ng/g treatment (p = 0.057). (B) Percentage of neutrophils that engulfed beads at different levels [none (0 beads), low (1−2 beads), medium(3−9 beads), or high (10+ beads)] for each treatment group. For all four levels of phagocytosis, the percent phagocytic neutrophils in each level didnot differ significantly among treatment groups (analysis of variance, p ≥ 0.107). Data is presented as the mean ± standard error of the mean.

Environmental Science & Technology Article

dx.doi.org/10.1021/es405776m | Environ. Sci. Technol. XXXX, XXX, XXX−XXXF

PBDEs as tadpoles. To assess innate immunity, we collectedperitoneal leukocytes following stimulation with fluid thio-glycollate46 and assessed phagocytic activity of the leukocytes.Because neutrophils act as the first responders in the innateimmune response, neutrophils were expected to be therepresentative leukocytes in the peritoneal lavage fluid, andthey were the dominating leukocyte type in our samples.Compared to control animals, frogs exposed to DE-71 did nothave significantly altered neutrophil recruitment; however,there was a statistically significant increase in neutrophil countsin the 6.1 ng PBDE/g treatment group (p = 0.046) and a trendfor an increase in the 634 ng PBDE/g (p = 0.057) whencompared to the lowest PBDE-exposed frogs. A dose-dependent increase in neutrophils with increasing PBDE levelshas also been reported in mammals exposed to PBDEs.30,64

Phagocytic activity of neutrophils did not differ significantlyacross treatment groups, and this is in contrast to loweredphagocytic activity when northern leopard frogs were exposedto low pH or the pesticide atrazine.46,65 However, it isnoteworthy to mention that on average the highest PBDEtreatment group had between 29% and 97% more phagocyticcells compared to all other treatment groups. It is possible thatour data did not have enough statistical power to determinesignificant differences in mean percentage phagocytic neu-trophils. Because all lavage samples had excess microbeads inthe interstitial fluid and were easily distinguishable fromengulfed microbeads, we do not think that the delivery ofmicrobeads affected our results. Larval DE-71 exposure did notsignificantly influence juvenile innate immune functioncompared to control treated frogs, but because increasedneutrophil recruitment occurred with increased PBDEconcentration, it is possible that immune stimulation occurredin L. pipiens with increasing PBDE exposure. Becausephagocytic cells are precursors to the inflammatory response,increased phagocyte counts are not necessarily indicative of ahealthier immune system and, in contrast, may indicatehypersensitive immune reactions, autoimmune disease, and/ora stress response.66

Skin peptides play a major role in amphibian immunedefenses, as they contribute to an integument free from

infecting microbes and protect against external parasites. In ourstudy, frogs fed ≥6.1 ng PBDE/g larval diets, secreted peptidesat a level that was 74 to 88% lower than control frogs; however,this decrease was not significantly different likely attributable toa small sample size and limited statistical power. The results ofother studies analyzing the impact of pollution on antimicrobialskin peptides offer differing conclusions. For example, Gibbleand Baer67 did not detect any effect of the herbicide atrazine ontotal peptide collection of skin secretions in post-metamorphicX. laevis, while Davidson et al.37 reported that exposure to theinsecticide carbaryl significantly reduced the total peptideconcentrations recovered from the skin secretions of Ranaboylii. Because AMPs act as one of the main barriers againstinfection by Bd, to the point that a correlation seems to existbetween the antifungal efficacy of skin secretion of amphibianspecies, and their tolerance against the lethal effects of Bd,52,68

it is important to determine how environmental pollutantsmight contribute to increased Bd infection. Our data suggeststhat PBDEs may be influencing AMP levels, but additionalresearch is necessary. Additionally, non-native Bd strains (likethe strain used in this study) may cause increased adverseeffects in native North American frog species.69 For greaterecological realism, future Bd-challenge studies with L. pipiensshould use native Bd strains.Lithobates pipiens is among the most tolerant amphibian

species against Bd infection,70 in part because of the highefficacy of peptides purified from its skin inhibiting growth ofthe fungus.71 We should thereby expect a low MIC of peptidesextracted from L. pipiens skin compared to other species.However, the average MIC found among control froglets (424μg/mL) is within the range of, or even higher than that foundfor other amphibian species (e.g., R. boylii: 25 μg/mL,37 Litoriasp.: 100−200 μg/mL;52 X. laevis: 500 μg/mL72). We did notdetect any significant effect of PBDE exposure on the anti-Bdeffectiveness of skin peptides, which generally matches withfindings of other studies analyzing the impact of environmentalpollutants on this innate immunity feature (e.g., refs 37 and67).Proper immune function is vital for disease resistance, and

although the present study found no evidence for increased

Figure 4. Box-whisker plots of (A) the amount of skin peptides extracter per cm2 of skin and (B) the minimum inhibitory concentration (MIC) ofskin peptides capable of limiting growth of Batrachochytrium dendrobatidis in vitro. The horizontal line in each box is the median, the top and bottomof the box represent the 75th and the 25th percentile, respectively, and the whiskers define the 5th and 95th percentile observations. The amount ofskin peptides extracted per cm2 of skin did not differ among treatments (analysis of variance, p = 0.535), nor did the MIC of skin peptides differamong treatment groups (analysis of variance, p = 0.732). All samples are pooled samples, each representing skin secretions from 3 or 4 frogs.

Environmental Science & Technology Article

dx.doi.org/10.1021/es405776m | Environ. Sci. Technol. XXXX, XXX, XXX−XXXG

susceptibility to Bd, immunosuppression by environmentalpollutants may contribute to declining amphibian populationsby increasing susceptible to other pathogens. Immunomodu-lation in juvenile L. pipiens previously exposed to PBDEs astadpoles suggests that larval exposure can have long-lastingeffects on immune function of frogs. However, at 10 weekspost-metamorphosis, these animals had measurable bodyburdens of PBDEs,43 so it is not possible to determine fromthis study whether exposure as a tadpole or whether bodyburdens of juvenile frogs alone affected immune responses.Further studies are necessary to elucidate whether toxicantexposure at the larval and/or post-metamorphic life-stageresults in alterations of the anuran adult immunity.

■ AUTHOR INFORMATIONCorresponding Author*Phone: (608) 263-4344. Fax: (608) 262-9922. E-mail: [email protected] authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe thank Nathan Van Schmidt for assistance with bioassays.The present study was sponsored by the University ofWisconsin Sea Grant Institute under grants from the NationalSea Grant College Program, National Oceanic and Atmos-pheric Administration, U.S. Department of Commerce, andfrom the State of Wisconsin. Federal grant NA16RG2257,project R/EH-2.

■ REFERENCES(1) Stuart, S. N.; Chanson, J. S.; Cox, N. A.; Young, B. E.; Rodrigues,A. S. L.; Fischman, D. L.; Waller, R. W. Status and trends of amphibiandeclines and extinctions worldwide. Science 2004, 306, 1783−1786.(2) IUCN. The IUCN Red List of Threatened Species: www.iucnredlist.org (accessed Jul 8, 2013).(3) Hites, R. A. Polybrominated diphenyl ethers in the environmentand in people: A meta-analysis of concentrations. Environ. Sci. Technol.2004, 38, 945−956.(4) Law, R. J.; Herzke, D.; Harrad, S.; Morris, S.; Bersuder, P.;Allchin, C. R. Levels and trends of HBCD and BDEs in the Europeanand Asian environments, with some information for other BFRs.Chemosphere 2008, 73, 223−241.(5) Voorspoels, S.; Covaci, A.; Jaspers; Neels, H.; Schepens, P.Biomagnification of PBDEs in three small terrestrial food chains.Environ. Sci. Technol. 2007, 41, 411−416.(6) Shaw, S. D.; Berger, M. L.; Brenner, D.; Kannan, K.; Lohmann,N.; Papke, O. Bioaccumulation of polybrominated diphenyl ethers andhexabromocyclododecane in the northwest Atlantic marine food web.Sci. Total Environ. 2009, 407, 3323−3329.(7) Johnson, P. I.; Stapleton, H. M.; Sjodin, A.; Meeker, J. D.Relationships between polybrominated diphenyl ether concentrationsin house dust and serum. Environ. Sci. Technol. 2010, 44, 5627−5632.(8) Stapleton, H. M.; Brazil, B.; Holbrook, R. D.; Mitchelmore, C. L.;Benedict, R.; Konstantinov, A.; Potter, D. In vivo and in vitrodebromination of decabromodiphenyl ether (BDE 209) by juvenilerainbow trout and common carp. Environ. Sci. Technol. 2006, 40,4653−4658.(9) Van den Steen, E.; Covaci, A.; Jaspers, V. L. B.; Dauwe, T.;Voorspoels, S.; Eens, M.; Pinxten, R. Accumulation, tissue-specificdistribution and debromination of decabromodiphenyl ether (BDE209) in European starlings (Sturnus vulgaris). Environ. Pollut. 2007,No. 148, 648−653.(10) Huwe, J. K.; Smith, D. J. Accumulation, whole-body depletion,and debromination of decabromodiphenyl ether in male Sprague-

Dawley rats following dietary exposure. Environ. Sci. Technol. 2007, 41,2371−2377.(11) U.S. Environmental Protection Agency. Deca-BDE Phase-outInitiative http://www.epa.gov/oppt/existingchemicals/pubs/actionplans/deccabde.html (accessed Jul 8, 2013).(12) Stapleton, H. M.; Baker, J. E. Comparing polybrominateddiphenyl ether and polychlorinated biphenyl bioaccumulation in afood web in Grand Traverse Bay, Lake Michigan. Arch. Environ.Contam. Toxicol. 2003, 45, 227−234.(13) Oram, J. J.; McKee, L. J.; Werme, C. E.; Connor, M. S.; Oros, D.R.; Grace, R.; Rodigari, F. A mass budget of polybrominated diphenylethers in San Francisco Bay, CA. Environ. Int. 2008, 34, 1137−1147.(14) Qiu, X.; Marvin, C. H.; Hites, R. A. Dechlorane plus and otherflame retardants in a sediment core from Lake Ontario. Environ. Sci.Technol. 2007, 41, 6014−6019.(15) Moser, V. C.; Coburn, C. G.; Farmer, J. D.; Jarema, K. A.;MacPhail, R. C.; McDaniel, K. L.; Phillips, P. M.; Kodavanti, P. R. S.Neurobehavioral assessment using a functional observational batteryand motor activity in rats perinatally exposed to DE-71. Neuro-toxicology 2006, 27, 1166−1166.(16) Yan, T.; Xiang, L.; Xuejun, J.; Chengzhi, C.; Youbin, Q.; Xuelan,Y.; Yang, L.; Changyan, P.; Hui, C. Spatial learning and memory deficitof low level polybrominated diphenyl ethers-47 in male adult rat ismodulated by intracellular glutamate receptors. J. Toxicol. Sci. 2012,37, 223−233.(17) Meerts, I.; van Zanden, J. J.; Luijks, E. a. C.; van Leeuwen-Bol,I.; Marsh, G.; Jakobsson, E.; Bergman, A.; Brouwer, A. Potentcompetitive interactions of some brominated flame retardants andrelated compounds with human transthyretin in vitro. Toxicol. Sci.2000, 56, 95−104.(18) Schriks, M.; Roessig, J. M.; Murk, A. J.; Furlow, J. D. Thyroidhormone receptor isoform selectivity of thyroid hormone disruptingcompounds quantified with an in vitro reporter gene assay. Environ.Toxicol. Pharmacol. 2007, 23, 302−307.(19) Hallgren, S.; Sinjari, T.; Hakansson, H.; Darnerud, P. O. Effectsof polybrominated diphenyl ethers (PBDEs) and polychlorinatedbiphenyls (PCBs) on thyroid hormone and vitamin A levels in rats andmice. Arch. Toxicol. 2001, 75, 200−208.(20) Balch, G. C.; Velez-Espino, L. A.; Sweet, C.; Alaee, M.; Metcalfe,C. D. Inhibition of metamorphosis in tadpoles of Xenopus laevisexposed to polybrominated diphenyl ethers (PBDEs). Chemosphere2006, 64, 328−338.(21) Schriks, M.; Zvinavashe, E.; David Furlow, J.; Murk, A. J.Disruption of thyroid hormone-mediated Xenopus laevis tadpole tail tipregress ion by hexabromocyclododecane (HBCD) and2,2′,3,3′,4,4′,5,5′,6-nona brominated diphenyl ether (BDE206).Chemosphere 2006, 65, 1904−1908.(22) Carlsson, G.; Kulkarni, P.; Larsson, P.; Norrgren, L. Distributionof BDE-99 and effects on metamorphosis of BDE-99 and-47 after oralexposure in Xenopus tropicalis. Aquat. Toxicol. 2007, 84, 71−79.(23) Cary Coyle, T. L.; Karasov, W. H. Chronic, dietarypolybrominated diphenyl ether exposure affects survival, growth, anddevelopment of Rana pipiens tadpoles. Environ. Toxicol. Chem. 2010,29, 133−141.(24) Stoker, T. E.; Cooper, R. L.; Lambright, C. S.; Wilson, V. S.;Furr, J.; Gray, L. E. In vivo and in vitro anti-androgenic effects of DE-71, a commercial polybrominated diphenyl ether (PBDE) mixture.Toxicol. Appl. Pharmacol. 2005, 207, 78−88.(25) Mercado-Feliciano, M.; Bigsby, R. M. Hydroxylated metabolitesof the polybrominated diphenyl ether mixture DE-71 are weakestrogen receptor-alpha ligands. Environ. Health Perspect. 2008, 116,1315−1321.(26) Reistad, T.; Mariussen, E. A commercial mixture of thebrominated flame retardant pentabrominated diphenyl ether (DE-71)induces respiratory burst in human neutrophil granulocytes in vitro.Toxicol. Sci. 2005, 87, 57−65.(27) Fair, P. A.; Stavros, H.-C.; Mollenhauer, M. A. M.; DeWitt, J. C.;Henry, N.; Kannan, K.; Yun, S. H.; Bossart, G. D.; Keil, D. E.; Peden-Adams, M. M. Immune function in female B6C3F1 mice is modulated

Environmental Science & Technology Article

dx.doi.org/10.1021/es405776m | Environ. Sci. Technol. XXXX, XXX, XXX−XXXH

by DE-71, a commercial polybrominated diphenyl ether mixture. J.Immunotoxicol. 2012, 9, 96−107.(28) Fowles, J.; Fairbrother, A.; Baechersteppan, L.; Kerkvliet, N.Immunological and Endocrine Effects of the Flame-RetardantPentabromodiphenyl Ether (de-71) in C57bl/6j Mice. Toxicology1994, 86, 49−61.(29) Thuvander, A.; Darnerud, P. O. Effects of polybrominateddiphenyl ether (PBDE) and polychlorinated biphenyl (PCB) on someimmunological parameters after oral exposure in rats and mice. Toxicol.Environ. Chem. 1999, 70, 229−242.(30) Martin, P. A.; Mayne, G. J.; Bursian, S. J.; Tomy, G.; Palace, V.;Pekarik, C.; Smits, J. Immunotoxicity of the commercial polybromi-nated diphenyl ether mixture DE-71 in ranch mink (Mustela vison).Environ. Toxicol. Chem. 2007, 26, 988−997.(31) Du Pasquier, L.; Robert, J.; Courtet, M.; Mussmann, R. B-celldevelopment in the amphibian Xenopus. Immunol. Rev. 2000, 175,201−213.(32) Robert, J.; Ohta, Y. Comparative and developmental study ofthe immune system in Xenopus. Dev. Dyn. 2009, 238, 1249−1270.(33) Rollins-Smith, L. A. Metamorphosis and the amphibian immunesystem. Immunol. Rev. 1998, 166, 221−230.(34) Carey, C.; Cohen, N.; Rollins-Smith, L. Amphibian declines: animmunological perspective. Dev. Comp. Immunol. 1999, 23, 459−472.(35) Rohr, J. R.; Raffel, T. R.; Halstead, N. T.; McMahon, T. A.;Johnson, S. A.; Boughton, R. K.; Martin, L. B. Early-life exposure to aherbicide has enduring effects on pathogen-induced mortality. Proc. R.Soc. B: Biol. Sci. 2013, 280, 1502.(36) Du Pasquier, L.; Schwager, J.; Flajnik, M. F. The immune systemof Xenopus. Annu. Rev. Immunol. 1989, 7, 251−275.(37) Davidson, C.; Benard, M. F.; Shaffer, H. B.; Parker, J. M.;O’Leary, C.; Conlon, J. M.; Rollins-Smith, L. A. Effects of chytrid andcarbaryl exposure on survival, growth and skin peptide defenses infoothill yellow-legged frogs. Environ. Sci. Technol. 2007, 41, 1771−1776.(38) Christin, M. S.; Menard, L.; Giroux, I.; Marcogliese, D. J.; Ruby,S.; Cyr, D.; Fournier, M.; Brousseau, P. Effects of agriculturalpesticides on the health of Rana pipiens frogs sampled from the field.Environ. Sci. Pollut. Res. 2013, 20, 601−611.(39) Gosner, K. L. A simplified table for staging Anuran embryos andlarvae with notes on identification. Herpetologica 1960, 16, 183−190.(40) Hardy, M. L. The toxicology of the three commercialpolybrominated diphenyl oxide (ether) flame retardants. Chemosphere2002, 46, 757−777.(41) Van Schmidt, N. D.; Cary, T. L.; Ortiz-Santaliestra, M. E.;Karasov, W. H. Effects of chronic polybrominated diphenyl etherexposure on gonadal development in the northern leopard frog Ranapipiens. Environ. Toxicol. Chem. 2012, 31, 347−354.(42) Maniero, G. D.; Morales, H.; Gantress, J.; Robert, J. Generationof a long-lasting, protective, and neutralizing antibody response to theranavirus FV3 by the frog Xenopus. Dev. Comp. Immunol. 2006, 30,649−657.(43) Cary, T. L.; Karasov, W. H. Toxicokinetics of polybrominateddiphenyl ethers across life stages in the northern leopard frog(Lithobates pipiens). Environ. Toxicol. Chem. 2013, 32, 1631−1640.(44) Gilbertson, M. K.; Haffner, G. D.; Drouillard, K. G.; Albert, A.;Dixon, B. Immunosuppression in the northern leopard frog (Ranapipiens) induced by pesticide exposure. Environ. Toxicol. Chem. 2003,22, 101−110.(45) Harris, J. R. Keyhole limpet hemocyanin (KLH): a biomedicalreview. Micron 1999, 30, 597−623.(46) Vatnick, I.; Andrews, J.; Colombo, M.; Madhoun, H.;Rameswaran, M.; Brodkin, M. A. Acid exposure is an immunedisruptor in adult Rana pipiens. Environ. Toxicol. Chem. 2006, 25,199−202.(47) Rollins-Smith, L. A.; Woodhams, D. C.; Reinert, L. K.;Vredenburg, V. T.; Briggs, C. J.; Nielsen, P. F.; Michael Conlon, J.Antimicrobial peptide defenses of the mountain yellow-legged frog(Rana muscosa). Dev. Comp. Immunol. 2006, 30, 831−842.

(48) Ramsey, J. P.; Reinert, L. K.; Harper, L. K.; Woodhams, D. C.;Rollins-Smith, L. A. Immune defenses against Batrachochytriumdendrobatidis, a fungus linked to global amphibian declines, in theSouth African clawed frog, Xenopus laevis. Infect. Immun. 2010, 78,3981−3992.(49) Pask, J. D.; Woodhams, D. C.; Rollins-Smith, L. A. The ebb andflow of antimicrobial skin peptides defends northern leopard frogs(Rana pipiens) against chytridiomycosis. Global Change Biol. 2012, 18,1231−1238.(50) McClanahan, L., Jr.; Baldwin, R. Rate of water uptake throughthe integument of the desert toad, Bufo punctatus. Comp. Biochem.Physiol. 1969, 28, 381−389.(51) Brucker, R. M.; Harris, R. N.; Schwantes, C. R.; Gallaher, T. N.;Flaherty, D. C.; Lam, B. A.; Minbiole, K. P. C. Amphibian chemicaldefense: Antifungal metabolites of the microsymbiont Janthinobacte-rium lividum on the salamander Plethodon cinereus. J. Chem. Ecol. 2008,34, 1422−1429.(52) Woodhams, D. C.; Ardipradja, K.; Alford, R. A.; Marantelli, G.;Reinert, L. K.; Rollins-Smith, L. A. Resistance to chytridiomycosisvaries among amphibian species and is correlated with skin peptidedefenses. Anim. Conserv. 2007, 10, 409−417.(53) Longcore, J. E.; Pessier, A. P.; Nichols, D. K. Batrachochytriumdendrobatidis gen. et sp. nov., a chytrid pathogenic to amphibians.Mycologia 1999, 91, 219−227.(54) Ortiz-Santaliestra, M. E.; Rittenhouse, T. A. G.; Cary, T. L.;Karasov, W. H. Interspecific and post-metamorphic variation insusceptibility of three North American Anurans to Batrachochytriumdendrobatidis. J. Herpetol. 2013, 47, 286−292.(55) Sheafor, B.; Davidson, E. W.; Parr, L.; Rollins-Smith, L.Antimicrobial peptide defenses in the salamander, Ambystoma tigrinum,against emerging amphibian pathogens. J. Wildl. Dis. 2008, 44, 226−236.(56) Welshons, W.; Thayer, K.; Judy, B.; Taylor, J.; Curran, E.; vomSaal, F. Large effects from small exposures. I. Mechanisms forendocrine-disrupting chemicals with estrogenic activity. Environ.Health Perspect. 2003, 111, 994−1006.(57) Zhou, T.; Taylor, M. M.; DeVito, M. J.; Crofton, K. A.Developmental exposure to brominated diphenyl ethers results inthyroid hormone disruption. Toxicol. Sci. 2002, 66, 105−116.(58) Beineke, A.; Siebert, U.; McLachlan, M.; Bruhn, R.; Thron, K.;Failing, K.; Muller, G.; Baumgartner, W. Investigations of the potentialinfluence of environmental contaminants on the thymus and spleen ofharbor porpoises (Phocoena phocoena). Environ. Sci. Technol. 2005, 39,3933−3938.(59) Christin, M. S.; Menard, L.; Gendron, A. D.; Ruby, S.; Cyr, D.;Marcogliese, D. J.; Rollins-Smith, L.; Fournier, M. Effects ofagricultural pesticides on the immune system of Xenopus laevis andRana pipiens. Aquat. Toxicol. 2004, 67, 33−43.(60) Gantress, J.; Maniero, G. D.; Cohen, N.; Robert, J. Developmentand characterization of a model system to study amphibian immuneresponses to iridoviruses. Virology 2003, 311, 254−262.(61) Stice, M. J.; Briggs, C. J. Immunization is ineffective atpreventing infection and mortality due to the amphibian chytridfungus Batrachochytrium dendrobatidis. J. Wildl. Dis. 2010, 46, 70−77.(62) Fites, J. S.; Ramsey, J. P.; Holden, W. M.; Collier, S. P.;Sutherland, D. M.; Reinert, L. K.; Gayek, A. S.; Dermody, T. S.; Aune,T. M.; Oswald-Richter, K.; et al. The invasive chytrid fungus ofamphibians paralyzes lymphocyte responses. Science 2013, 342, 366−369.(63) Demas, G. E.; Zysling, D. A.; Beechler, B. R.; Muehlenbein, M.P.; French, S. S. Beyond phytohaemagglutinin: assessing vertebrateimmune function across ecological contexts. J. Anim. Ecol. 2011, 80,710−730.(64) Neale, J. C. C.; Gulland, F. M. D.; Schmelzer, K. R.; Harvey, J.T.; Berg, E. A.; Allen, S. G.; Greig, D. J.; Grigg, E. K.; Tjeerdema, R. S.Contaminant loads and hematological correlates in the harbor seal(Phoca vitulina) of San Francisco Bay, California. J. Toxicol.Environmental Health, Part A 2005, 68, 617−633.

Environmental Science & Technology Article

dx.doi.org/10.1021/es405776m | Environ. Sci. Technol. XXXX, XXX, XXX−XXXI

(65) Brodkin, M. A.; Madhoun, H.; Rameswaran, M.; Vatnick, I.Atrazine is an immune disruptor in adult northern leopard frogs (Ranapipiens). Environ. Toxicol. Chem. 2007, 26, 80−84.(66) Latimer, K.; Mahaffey, E.; Prasse, K. Duncan and Prasse’sVeterinary Laboratory Medicine; Clinical Pathology, 4th ed.; Iowa StateUniversity Press: Ames, IA, 2003.(67) Gibble, R. E.; Baer, K. N. Effects of atrazine, agricultural runoff,and selected effluents on antimicrobial activity of skin peptides inXenopus laevis. Ecotoxicol. Environ. Saf. 2011, 74, 593−599.(68) Pask, J. D.; Cary, T. L.; Rollins-Smith, L. A. Skin peptidesprotect juvenile leopard frogs (Rana pipiens) against chytridiomycosis.J. Exp. Biol. 2013, 16, 2908−2916.(69) Gahl, M. K.; Longcore, J. E.; Houlahan, J. E. Varying responsesof northeastern North American amphibians to the chytrid pathogenBatrachochytrium dendrobatidis: Varying susceptibility to Bd. Conserv.Biol. 2012, 26, 135−141.(70) Woodhams, D. C.; Hyatt, A. D.; Boyle, D. G.; Rollins-Smith, L.A. The northern leopard frog Rana pipiens is a widespread reservoirspecies harboring Batrachochytrium dendrobatidis in North America.Herpetol. Rev. 2008, 39, 66.(71) Rollins-Smith, L. A.; Doersam, J. K.; Longcore, J. E.; Taylor, S.K.; Shamblin, J. C.; Carey, C.; Zasloff, M. A. Antimicrobial peptidedefenses against pathogens associated with global amphibian declines.Dev. Comp. Immunol. 2002, 26, 63−72.(72) Gibble, R. E.; Rollins-Smith, L.; Baer, K. N. Development of anassay for testing the antimicrobial activity of skin peptides against theamphibian chytrid fungus (Batrachochytrium dendrobatidis) usingXenopus laevis. Ecotoxicol. Environ. Saf. 2008, 71, 506−513.

Environmental Science & Technology Article

dx.doi.org/10.1021/es405776m | Environ. Sci. Technol. XXXX, XXX, XXX−XXXJ