Use of a Novel Double-Crested Cormorant ... - ACS Publications

Feb 19, 2016 - University, Ottawa, ON Canada K1A 0H3. ‡. Royal Military College of Canada, Kingston, Ontario K7K 7B4, Canada. •S Supporting Inform...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/est

Use of a Novel Double-Crested Cormorant ToxChip PCR Array and the EROD Assay to Determine Effects of Environmental Contaminants in Primary Hepatocytes Doug Crump,*,† Amani Farhat,† Suzanne Chiu,† Kim L. Williams,† Stephanie P. Jones,† and Valerie S. Langlois‡ †

Ecotoxicology and Wildlife Health Division, Environment and Climate Change Canada, National Wildlife Research Centre, Carleton University, Ottawa, ON Canada K1A 0H3 ‡ Royal Military College of Canada, Kingston, Ontario K7K 7B4, Canada S Supporting Information *

ABSTRACT: In vitro screening tools and ‘omics methods are increasingly being incorporated into toxicity studies to determine mechanistic effects of chemicals and mixtures. To date, the majority of these studies have been conducted with wellcharacterized laboratory animal models. In the present study, well-established methods developed for chicken embryonic hepatocyte (CEH) studies were extended to a wild avian species, the double-crested cormorant (DCCO; Phalacrocorax auritus), in order to compare the effects of several environmental contaminants on cytotoxicity, ethoxyresorufin O-deethylase (EROD) activity, and mRNA expression. Five organic flame retardants and one plasticizer decreased cormorant hepatocyte viability in a similar manner to that observed in previous studies with CEH. EROD activity was induced in a concentration-dependent manner following exposure to two dioxin-like chemicals and the calculated EC50 values were concordant with domestic avian species from similar species sensitivity categories. Transcriptomic effects were determined using a novel DCCO PCR array, which was designed, constructed and validated in our laboratory based on a commercially available chicken PCR array. The DCCO array has 27 target genes covering a wide range of toxicity pathways. Gene profiles were variable among the 10 chemicals screened; however, good directional concordance was observed with regard to results previously obtained in CEH. Overall, the application of well-established methods (i.e., CEH and chicken PCR array) to the double-crested cormorant demonstrated the portability of the techniques to an indicator species of ecological relevance.



INTRODUCTION High-throughput in vitro screening techniques are a critical component of toxicity testing in the 21st century.1 They provide a means to generate toxicologically relevant data efficiently to help prioritize the allocation of resources toward compounds and/or mixtures presenting the greatest potential risk. Combined with in vitro assays, emergent ‘omic approaches are being incorporated into toxicity testing to elucidate modes of action at cellular/subcellular levels of biological organization that predict apical outcomes. Typically, these approaches are designed for model laboratory species in order to validate and normalize findings to species for which there is an abundance of information (e.g., toxicological effects, annotated genome, wellcharacterized development). Our laboratory has been developing and conducting in vitro screening assays that utilize embryonic hepatocytes derived from various avian species (predominantly domestic species) to screen for biochemical and transcriptomic effects of complex mixtures and individual environmental chemicals of concern. Published XXXX by the American Chemical Society

For example, the avian embryonic hepatocyte assay was utilized to compare species differences in sensitivity to dioxinlike chemicals (DLCs) based on ethoxyresorufin O-deethylase (EROD) activity and cytochrome P450 1A4/1A5 (Cyp1a4/ 1a5) mRNA induction.2,3 Hepatocytes were prepared from chicken (Gallus gallus), ring-necked pheasant (Phasianus colchicus), and Japanese quail (Coturnix japonica), species representing the three main sensitivity types (high, moderate, low, respectively) determined for birds.4 Biochemical/molecular results were predictive of in ovo LD50 values without requiring whole animal experimental studies.2,3 Porter et al.5 utilized a chicken embryonic hepatocyte (CEH) assay in conjunction with a chicken ToxChip polymerase chain reaction (PCR) array to determine the effects of 16 organic flame Received: December 17, 2015 Revised: February 17, 2016 Accepted: February 19, 2016

A

DOI: 10.1021/acs.est.5b06181 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Wellington Laboratories (Guelph, ON)1,2-dibromo-4-(1,2dibromoethyl)-cyclohexane (DBE-DBCH; formerly abbreviated as TBECH; technical formulation is a 1:1 molar ratio of the α- and β-DBE-DBCH diastereomers), PCB126, decabrominated diphenyl ether (BDE-209), and tetradecabromo1,4-diphenoxybenzene (TeDB-DiPhOBz; Technical SAYTEX120, Lot# 0GN01-$I0); (3) Pfaltz and Bauer (Waterbury, CT)tris(1-chloro-2-propyl) phosphate (TCIPP); and (4) TCI America (Portland, OR)tris(1,3-dichloro-2-propyl) phosphate (TDCIPP). The preparation and analysis of TCDD stock solutions is provided elsewhere.2 Stock solutions of all test chemicals, with the exception of TeDB-DiPhOBz and BDE-209, were prepared by dissolution in dimethyl sulfoxide (DMSO; Sigma-Aldrich) to a final concentration of 60 mM. Prior to dissolution in DMSO, TeDB-DiPhOBz and BDE-209 powder was dissolved in 30% THF/n-hexane solution to achieve a final, nominal concentration of 300 μM. Ten mL solutions of these were either exposed to sunlight irradiation for 21 days (SI) or blown down to dryness under a gentle flow of nitrogen and redissolved in DMSO immediately (i.e., nonirradiated [NI]) following procedures detailed previously.11 DMSO stock solutions of 5 mM and 10 mM were prepared for SI-TeDB-DiPhOBz and SIBDE-209, respectively, and 0.38 and 1.8 μM for NI-TeDBDiPhOBz and NI-BDE-209, respectively. For all chemicals, stock solutions were diluted such that the final DMSO concentration in the aqueous cell culture medium was 0.5%. Preparation and Dosing of DCEH Cultures. Fertilized, unincubated DCCO eggs (n = 25) were collected from nests containing one egg on 20 May 2015, at Doucet Rock, Lake Huron (46° 8′22.09″N, 82°51′7.35″W) and transported in foam-lined coolers to the National Wildlife Research Centre in Ottawa. Immediately upon arrival (within 24 h of collection), eggs were incubated according to methods described elsewhere.12 Briefly, eggs were placed horizontally in a Petersime (Model XI) Incubator at 37.5 °C and 60% relative humidity. The egg trays were automatically rotated (approximately 90°) every 2 h. In addition, all eggs were turned 180° along their long axis once per day. On incubation day 26 (∼2 day prehatch), embryos were euthanized by decapitation and livers were removed and pooled. Cultured hepatocytes were prepared by collagenase digestion and filtration, as described previously.13,14 Twenty-five μL of the cell suspension was distributed into 48-well plates, which contained 500 μL of Medium 199 supplemented with sodium bicarbonate (2.24 g/L), penicillin (60 mg/L), streptomycin (100 mg/L), insulin (1 mg/L), and L-thyroxine (1 mg/L) (all reagents supplied by Sigma), and incubated at 37.5 °C and 5% CO2 for 24 h prior to dosing. The final nominal concentration ranges for the various chemicals listed above are indicated in Table 1. These concentrations were shown to elicit effects in previous studies with CEH2,3,5,11,15,16 and were included to permit cross-species comparisons. For EROD and cell viability plates, a total of n = 3 wells/treatment were included whereas n = 6 wells/treatment were included for plates to be used for PCR array. DCEH were incubated for 24 h following chemical administration and the medium was then aspirated and DCEH were frozen at −80 °C for subsequent RNA isolation or assayed immediately for cell viability. Plates used for EROD assays were rinsed with 200 μL/ well of phosphate-buffered saline-ethylenediaminetetraacetic acid prior to being flash frozen in powdered dry ice and stored at −80 °C.

retardants (OFRs) on mRNA expression of 27 target genes. Cluster analysis was used to identify the OFRs that elicited the greatest number of gene alterations, which were then identified as priorities for follow-up whole animal studies. Hickey et al.6 used the same CEH assay to screen 18 perfluoroalkyl acids for effects on cytotoxicity and mRNA expression. In addition to single chemical screening studies, the CEH assay has been used to compare biochemical (i.e., EROD activity and porphyrin concentration) and transcriptomic responses for complex mixtures of organohalogen contaminants derived from herring gull egg extracts collected from variably contaminated nesting colonies.7 While the substantial amount of data for domestic/model avian species is critical to establish a baseline response, our group has been challenged by government regulators with regard to the applicability of our findings in assessing risk to wild avian species that are actually exposed to these various chemical stressors. Double-crested cormorants (DCCO; Phalacrocorax auritus) are piscivorous birds that breed extensively across North America permitting their use as wildlife sentinels of aquatic ecosystem health from areas including the Great Lakes and marine/coastal environments. Ludwig et al.8 assessed embryonic death rates and deformities in DCCO embryos and hatchlings collected from the Upper Great Lakes and correlated the results with planar polychlorinated biphenyl (PCB) and dioxin congeners as well as EROD activity. Decreased embryonic viability was observed in chicken embryos following injection of organohalogen contaminant extracts prepared from eggs of DCCOs collected from contaminated breeding colonies near Green Bay, Lake Michigan.9 Additionally, hepatocytes were prepared from 1day old DCCO hatchlings to determine concentration− response characteristics of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on EROD induction and the median EC50 value was 13 nM; greater than the value for chicken hatchlings and less than that calculated for ring-billed gull, herring gull and tern hatchlings.10 The main overall objective of the present study was to compare end points measured in a domestic laboratory model species with a fish-eating bird that would be naturally exposed to environmental pollutants. We prepared primary hepatocytes from late-stage (i.e., day 26; 2 days prehatch) embryonic DCCOs and used an in vitro screening approach to determine effects of select priority environmental chemicals that have been shown to elicit significant cytotoxic, biochemical and/or transcriptomic effects in CEH. Transcriptomic effects were determined using a novel DCCO PCR array, which was designed, constructed and validated in our laboratory. The array has 27 target genes covering a wide range of toxicity pathways. The transcriptomic profiles were directionally concordant with previous chicken PCR array results highlighting the versatility of this technology and its application to an ecological indicator species of relevance.



EXPERIMENTAL SECTION Chemicals. Chemicals administered to double-crested cormorant embryonic hepatocyte (DCEH) cultures were purchased from the following suppliers: (1) Sigma-Aldrich (St. Louis, MO)bisphenol A (BPA), bisphenol S (BPS), tris(methylphenyl) phosphate (TMPP; formerly abbreviated as TCP), tris(2-butoxyethyl) phosphate (TBOEP), triethyl phosphate (TEP), tris(2,3-dibromopropyl) isocyanurate (TBC), and allyl 2,4,6-tribromophenyl ether (ATE); (2) B

DOI: 10.1021/acs.est.5b06181 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

wavelength: 460 nm) using a fluorescence plate-reader (Cytofluor 2350, Millipore, Bedford, MA). DCCO ToxChip PCR Array Development. (i). RNA Extraction and cDNA Synthesis. Total RNA was isolated from ∼30 mg of DCCO liver, which was collected from a prefledged juvenile (∼28-days old), using Qiagen RNeasy mini kits according to the manufacturer’s instructions. The concentration and purity of extracted RNA was quantified by determining the A260/A280 absorbance ratio with a NanoDrop 2000 spectrophotometer (Thermo Scientific, Wilmington, DE); samples with an A260/A280 > 1.8 were used for cDNA Synthesis. Total RNA (1 μg) was DNase treated using Turbo DNA-free kits (Ambion, Austin, TX) and reverse transcribed to cDNA using QuantiTect reverse transcription kits (Qiagen) for use in (a) coding sequence discovery (i.e., sequencing DCCO genes orthologous to those already identified in other avian species) and (b) validating custom-designed primers for specificity to DCCO cDNA. cDNA samples were diluted 1:10 with diethylpyrocarbonate (DEPC)-treated water prior to PCR. (ii). Determination of DCCO Coding Sequences. The coding sequence for each gene of interest (Table 2) was obtained from the National Centre for Biotechnology information (NCBI) nucleotide database for as many avian species as possible including, chicken, zebra finch (Taeniopygia guttata), turkey (Meleagris gallopavo), budgie (Melopsittacus undulates), great cormorant (P. carbo), mallard duck (Anas platyrhynchos), saker falcon (Falco cherrug), pigeon (Columba livia), and herring gull (Larus argentatus). Coding sequences were aligned using Clustal Omega and primers were designed (Primer 3 Plus) to achieve >80% match to all species and at least a 5 base pair match to all species at the 3′ end. Reverse transcription polymerase chain reactions (RT-PCR) were performed using an Eppendorf Mastercycler Pro S instrument. Each 25 μL reaction contained 1X PCR buffer, 3−4 mM MgCl2, 0.4 mM dNTPs, 2 U Platinum taq, 400 nM custom-designed forward and reverse primers, and 5 μL diluted cDNA (1:10) in DEPC water. The thermocycle program consisted of an enzyme activation step at 95 °C for 5 min followed by 40 cycles of 95 °C for 30 s, 55 °C for 1 min, and 72 °C for 30 s, and a final cycle held at 72 °C for 1 min. PCR products were separated using Invitrogen’s E-Gel iBase system on CloneWell 0.8%SYBR Safe gels, collected in sterile deionized water, and sequenced by StemCore Laboratories (Ottawa, ON). Sequences were compared to the NCBI nucleotide database to ensure the appropriate coding sequence was amplified. (iii). Primer Design and Validation for Array. Primers for the PCR array gene targets were designed using NCBI’s Primer BLAST function to yield amplicons between 50 and 210 base pairs in length and were checked for specificity against available avian taxa (see Supporting Information (SI) Table S1 for array primer sequences). Brilliant II SYBR Green Q-PCR Master Mix kits (Agilent Technologies) were used for real-time RT-PCR. Each 25 μL reaction contained 2X Brilliant SYBR Green QPCR Master Mix, 75 nM ROX reference dye, 200 nM forward and reverse primers, and 5 μL diluted cDNA (1:10). Reactions were performed on a Stratagene Mx3005P (La Jolla, CA) or an Eppendorf Mastercycler ep realplex4 instrument with the following thermal profile: 95 °C for 10 min followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min and ending with a dissociation curve segment specific to each instrument (Stratagene: 95 °C for 1 min, 55 °C for 30 s and 95 °C for

Table 1. Chemicals Screened in Double-Crested Cormorant Embryonic Hepatocyte Cultures and the Corresponding Nominal Exposure Concentrations and/or Ranges Administered for the Various Assays Identified CASRN

common name

80−09−1 80−05−7 1330−78−5

bisphenol S [BPS] bisphenol A [BPA] tris(methylphenyl) phosphate [TMPP]

78−51−3

tris(2-butoxyethyl) phosphate (TBOEP) triethyl phosphate [TEP] tris(2,3-dibromopropyl) isocyanurate [TBC] allyl 2,4,6-tribromophenyl ether [ATE] α/β-1,2-dibromo-4-(1,2-dibromoethyl)-cyclohexane [DBEDBCH] polychlorinated biphenyl 126 [PCB 126] deca-brominated diphenyl ether [BDE-209] tetradecabromo-1,4-diphenoxybenzene [TeDB-DiPhOBz; SAYTEX-120] tris(1-chloro-2-propyl) phosphate [TCIPP] tris(1,3-dichloro-2-propyl) phosphate [TDCIPP] 2,3,7,8-Tetrachlorodibenzo-p-dioxin [TCDD]

78−40−0 52434−90−9 3278−89−5 3322−93−8

57465−28−8 1163−19−5 58965−66−5

13674−84−5 13674−87−8 1746−01−6

assay

[nominal]

PCR array cell viability cell viability, PCR array cell viability

300 μM 1, 300 μM 10, 300 μM

PCR array PCR array

1, 10, 300 μM 1, 10, 100 μM

PCR array

10 μM

cell viability, PCR array EROD

1, 10, 300 μM

PCR array PCR array

1, 300 μM

0.03−1000 nM 1 μM [NI] 1, 10 μM [SI] 1 μM [NI] 1, 10 μM [SI]

cell viability

1, 300 μM

cell viability

1, 300 μM

EROD

0.003−100 nM

Cell Viability Determination. Viability of DCEH was estimated with the ViaLight Plus kit (Lonza), according to the manufacturer’s instructions, for the following chemicals at 1 and 300 μM: TDCIPP, TCIPP, TBOEP, TMPP, DBE-DBCH, and BPA. These particular chemicals were selected because they decreased cell viability in chicken and/or herring gull embryonic hepatocytes in previous studies.5,15,16 Untreated cells were used as a negative control and cells dosed with a final, in-well concentration of 300 μM TDCIPP were used as a positive control (previous findings showed complete cell death at this concentration of TDCIPP in CEH15). There was one modification to the protocol: all medium was aspirated and 100 μL of fresh medium was added to each well prior to the addition of 50 μL cell lysis buffer to account for the difference in medium volume between the 48-well plates used for the hepatocyte culture and the 96-well plates used for the cell viability assay. Luminescence data from plates were read with a 1 s integrated reading time using the Luminoskan Ascent luminometer (Thermo Fisher Scientific) and resulting luminescence values were compared among wells to determine any treatment-related effects. EROD Activity. EROD assays were conducted as described previously;13,14 all reagents were obtained from Sigma-Aldrich. Briefly, DCEH were incubated at 37.5 °C in the presence of nicotinamide adenine dinucleotide phosphate (NADPH, reduced) and 7-ethyoxyresorufin for 7 min and the reaction was stopped by cold acetonitrile containing fluorescamine (0.15 mg/mL). Resorufin and protein (bovine serum albumin [BSA]) standard curves were prepared on each plate.14 Plates were analyzed for both EROD activity (excitation wavelength: 530 and emission wavelength: 590 nm) and total protein concentration (excitation wavelength: 400 nm and emission C

DOI: 10.1021/acs.est.5b06181 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

A subset (n = 5; Rpl4, Pah, Cyp1a4, Igf1, Eef1a1) of amplicons were sequenced as described above to verify that the amplicon matched the desired DCCO coding sequence. iv). Assay Optimization. Amplification efficiency was determined using an 8 point standard curve in a 2X dilution series (ranging from 0.4 to 50 ng of cDNA; assuming 100% RT efficiency) with the RT2 SYBR Green ROX qPCR Mastermix (Qiagen). Each 25 μL reaction contained 200−400 nM of the target specific forward and reverse primers. Assays with reaction efficiencies between 90 and 110% and R2 > 0.95 were deemed acceptable. All assays were linear within this range except for Fgf19, which failed to amplify at cDNA inputs of less than 3 ng. v). Quality Control. A no-reverse transcriptase control is included with every cDNA synthesis preparation (described below) and added to the no-reverse transcriptase well on the array (containing primers for the elongation factor 1-alpha [Eef1a1] reference gene) to test for genomic DNA contamination. A no template control (NTC) well, which contains 1 μL DEPC-water instead of template and Eef1a1 primers, is included on the array to control for reagent contamination. A positive PCR control (PPC; Bio Rad PrimePCR Positive Control SYBR Green Assay; 10025591) containing a synthetic DNA template that is not present in the DCCO genome is included to ensure consistency of PCR efficiency between different arrays. This assay was validated by Bio-Rad for human and mouse genomes. To ensure the positive PCR assay did not amplify a DCCO-specific template, the assay was performed in the presence (n = 4) and absence (n = 4) of DCCO cDNA. The cycle threshold (Ct) was the same regardless of the presence of DCCO template (average Ct = 20.66 ± 0.11). DCCO ToxChip PCR Array. Total RNA was extracted from DCEH using the Qiagen RNeasy 96-kit according to the manufacturer’s instructions (Qiagen, Mississauga, ON), including an on-column DNase treatment. To ensure sufficient RNA yield for PCR array analysis, two replicate wells of hepatocytes were combined prior to RNA extraction. A total of n = 3 technical replicates were included for the DMSO control and the treatment groups listed in Table 1. RNA quality and concentration were determined using a Nanodrop 2000 (Thermo Scientific). Total RNA (∼200 ng) was reverse transcribed using the QuantiTect Reverse Transcription kit (Qiagen) with modifications detailed in Porter et al.5 and resulting cDNA samples were diluted with DEPC-water and added directly to the RT2 SYBR Green Mastermix (Qiagen). Twenty-five μL of the Mastermix was aliquoted to each well of the DCCO PCR array containing primers at preoptimized concentrations. Each 96-well plate contained three identical sets of 32 genes and controls (Table 2) and therefore a single treatment group (n = 3/treatment) was screened using a single array. PCR array plates were run using the Stratagene MX3005P PCR system (Agilent Technologies) with the following thermal profile: 95 °C for 10 min followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min and ending with a dissociation curve segment of 95 °C for 1 min, 55 °C for 30 s and 95 °C for 30 s. Data Analysis. A one-way analysis of variance (ANOVA) with Tukey’s honestly significant differences (HSD) post hoc test was used to determine significant differences in cell viability between the two treatment concentrations (1 and 300 μM) of the six test chemicals described above and the vehicle control in GraphPad Prism v5.02 (San Diego, CA). EROD activity data were fit to a modified Gaussian curve as described previously14 for DCEH treated with TCDD and

Table 2. Pathways and Description of the 32 Gene Targets on the Double-Crested Cormorant ToxChip PCR Array pathways phase I and II metabolism

immune function

gene symbol

accession no.

CYP3A37 CYP1A4 UGT1A1

KT886910

SULT1E1

KT886920

KT886918

BATF3 IL16

KT886907

glucose and fatty acid metabolism

PDK4

KT886905

oxidative stress

MT4 GPX3 GSTM3 TXN

lipid/cholesterol homeostasis

ACSL5

KT886909

HMGCR

KT886908

description cytochrome P450 A 37 cytochrome P450 1A4 UDP glucuronosyltransferase 1 family, polypeptide A1 sulfotransferase family 1E, estrogen-preferring member1 basic leucine zipper transcription factor, ATF-like 3 interleukin 16 (lymphocyte chemoattractant factor) pyruvate dehydrogenase kinase, isozyme 4

metallothionein 4 glutathione peroxidase 3 glutathione s-transferase thioredoxin

SLCO1A2

acyl-CoA synthetase long-chain family member 5 3-hydroxy-3-methylglutarylCoenzyme A reductase solute carrier organic anion transporter family, member 1A2 fibroblast growth factor 19 fatty acid binding protein 1, liver cytochrome P450, family 7, subfamily B, polypeptide 1

FGF19 LBFABP CYP7B1

KT886904

TTR THRSP

KT886921

IGF1

KT886911

P53 P53r2

KT886913 KT886915

MDM2

KT886914

steroid metabolism

ALAS1 NCOA3

KT886906 KT886912

aminolevulinate, delta, synthase1 nuclear receptor coactivator 3

calcium homeostasis

RGN

KT886902

regucalcin

phenylalanine catabolism

PAH

KT886903

phenylalanine hydroxylase

control

EEF1A1

KT886916

RPL4 GDC NTC PPC

KT886917

eukaryotic translation elongation factor 1 alpha 1 ribosomal protein L4 genomic DNA Contamination no template control positive PCR control

thyroid hormone pathway

tumor regulation

KT886919

transthyretin thyroid hormone responsive (SPOT14 homologue, rat) insulin-like growth factor 1 (somatomedin C) tumor suppressor protein 53 P53 inducible ribonucleotide reductase mouse double minute 2 ProtoOncogene; E3 Ubiquitin Protein Ligase

30 s; Eppendorf: 95 °C for 15 s, 60 °C for 15 s and 95 °C for 15 s). Primer specificity was verified by a single melting curve peak and a single band on an agarose gel of the appropriate size. D

DOI: 10.1021/acs.est.5b06181 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Figure 1. Effects of tris(1,3-dichloro-2-propyl) phosphate (TDCIPP), tris(1-chloro-2-propyl) phosphate (TCIPP), tris(2-butoxyethyl) phosphate (TBOEP), tris(methylphenyl) phosphate (TMPP), 1,2-dibromo-4-(1,2-dibromoethyl)-cyclohexane (DBE-DBCH), and bisphenol A (BPA) at two nominal exposure concentrations, 1 and 300 μM, on cell viability of double-crested cormorant embryonic hepatocytes following 24 h of exposure. Viability data are presented as the percent viability (n = 3 wells per treatment group) compared to the DMSO solvent control and significant differences, determined by one-way ANOVA, are identified by * (p < 0.05). Error bars for each data set represent the SD.

(Figure 1). Of most interest were TCIPP and TMPP because they decreased cell viability of the two wild species, herring gulls (Porter et al.;5 LC50 of 170 and 31 μM, respectively) and double-crested cormorants, but not chickens, which are typically considered highly sensitive to a wide array of chemicals. It is important to note that actual concentrations were not quantified in cormorant cells or culture medium. This would be an important addition to future studies in order to determine the actual uptake and resulting exposure level as it relates to wild avian species. All six of the chemicals screened for cytotoxicity have been detected in biota. For example, TCIPP was detected in herring gull eggs from the Great Lakes at concentrations up to 0.03 μg/g ww18 and TMPP concentrations of up to 0.1 μg/g lipid weight (lw) were measured in freshwater fish in Sweden.19 TDCIPP, TBOEP, and DBEDBCH were quantifiable in herring gull eggs with concentrations up to 0.0002 μg/g, 0.004 μg/g, and 0.0005 μg/g ww, respectively.20,21 BPA concentrations up to 0.001 μg/g ww were measured in liver samples of a high-trophic level avian species, the great cormorant.22 Assuming 100% uptake by the cormorant hepatocytes, even the lowest concentration tested, 1 μM, is equivalent to ∼150−250 μg/g based on the weight of hepatocytes, which is several orders of magnitude greater than levels of these six chemicals in the environment. However, Farhat et al.23 demonstrated that the cellular concentration of TDCIPP after 36 h in CEH was substantially lower than the initial administered concentration in the medium (i.e., < 2%) indicating that the comparison between concentrations in cultured cells and those measured in whole animals is more

PCB126. For each chemical, three EROD curves were generated from data derived from separate cell culture plates. EC50 values are presented as the mean value of the three replicates ± SEM. PCR array data were analyzed using MxPro v4.10 software (Agilent Technologies). Cycle threshold (Ct) data were normalized to two internal control genes (Eef1a1 and ribosomal protein L4 [Rpl4]) and the fold change of target gene mRNA abundance relative to the vehicle control was calculated using the 2−ΔΔCt method.17 A one-way ANOVA with Tukey’s HSD post hoc test (GraphPad Prismv5.02) was used to determine significant (p < 0.05) fold change differences. Hierarchical clustering of the fold change data was performed on R Statistics 3.0.2 (R Development Core Team) using the “gplots” package. Nonsignificant fold changes (p > 0.05) and those less than or equal to 1.5 were set to 0 to minimize noise.



RESULTS AND DISCUSSION Cell Viability. Double-crested cormorant hepatocytes were exposed to two concentrations (1 and 300 μM) of five organic flame retardants (TDCIPP, TCIPP, TBOEP, TMPP, and DBEDBCH) and BPA in order to generate some basic in vitro toxicity data. The inclusion of a full concentration response curve was not possible for the six compounds due to the limited pool of cormorant hepatocytes; however, the highest concentration was selected because it decreased cell viability of chicken and/or herring gull hepatocytes in previous studies and the lowest concentration had no effect on viability.5,15,16 All six of the chemicals screened significantly decreased viability of DCEH at 300 μM, whereas no effects were observed at 1 μM E

DOI: 10.1021/acs.est.5b06181 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Avian species sensitivity to dioxin-like compounds has been shown to be variable and is associated with the identity of amino acids at positions 324 and 380 within the AhR ligandbinding domain.4 This amino acid variability results in 3 main categories of species sensitivity−highly sensitive type 1 (Ile324_Ser380); moderately sensitive type 2 (Ile324_Ala380); and least sensitive type 3 (Val324_Ala380). Double-crested cormorants are a type 3 species4 and as predicted, the TCDD EC50 was similar to another type 3 species, Japanese quail, and greater than type 1 and 2 species (Table 3). Interestingly, for PCB126 exposure, the EROD EC50 value in DCEH was more similar to a type 2 species, ring-necked pheasant, and approximately 1 order of magnitude lower than Japanese quail, a type 3 species (Table 3). Ligand-dependent differences in binding with AhR types could help explain the variability in relative species sensitivity for PCB126 in this study and could be evaluated by conducting AhR binding assays. Regardless, the concentration-dependent induction of EROD activity in DCEH following exposure to TCDD and PCB126 revealed that the cultured cells were responding as expected. DCCO ToxChip PCR Array. The custom-designed DCCO ToxChip PCR array was used to screen for transcriptomic effects in DCEH at administered concentrations of environmental contaminants identified in Table 1, which were previously shown to alter mRNA expression levels in CEH.5,11,16 The array has two internal control genes, Eef1a1 and Rpl4, and their expression was invariable across all treatment groups indicating their suitability as controls. In addition, there was no amplification in the genomic DNA contamination control or no template control wells, and the positive PCR control met the appropriate quality control and assurance criteria (i.e., consistent Ct values across all plates included in the study). The complete list of DCCO gene targets and their respective fold changes for all chemicals and administered concentrations are included in SI Tables S2 and S3. Cluster analysis was performed with PCR array data for the following compounds at the highest noncytotoxic dose: DBEDBCH (10 μM), TBC (10 μM), TEP (300 μM), ATE (10 μM), TMPP (10 μM), and BPS (300 μM) (Figure 3). TMPP and BPS altered the most target genes, 9 and 10 of 27, respectively, and formed a related cluster. The second distinct cluster contained a main branch with DBE-DBCH (5/27 genes altered) separated from TBC (0/27 genes altered), TEP (0/27 genes altered), and ATE (2/27 genes altered). The sunlightirradiated flame retardants could not be included on the same heat map because the DMSO solvent control was prepared in a different manner (i.e., originally contained THF/hexane) as indicated in the Experimental Section. The NI-TeDB-DiPhOBz and NI-BDE-209 treatment groups clustered together whereas their sunlight-irradiated counterparts formed a separate cluster, further divided based on the nominal exposure concentration (Figure 4). One of the key motivations for developing the DCCO ToxChip PCR array was to enable the comparison of transcriptomic responses between a wild species and a laboratory model. Although the gene targets on the chicken and DCCO PCR arrays are not identical, there is enough overlap to assess whether the same chemicals dysregulate mRNA expression in a domestic avian species and a wild species that could be naturally exposed to the various chemicals screened. Overall, there was good concordance between the two species with regard to directional response of those gene

complicated than assuming 100% uptake by the cells. Despite the difficulties associated with comparing in vitro exposure concentrations and those measured in biota, cytotoxicity data provide an effective means of comparing overt adverse effects among different chemicals and avian species. Effects of TCDD and PCB126 on EROD Activity. TCDD and the coplanar, dioxin-like PCB, PCB126, induced EROD activity in a concentration-dependent manner in DCEH (Figure 2). The decrease in EROD activity at higher

Figure 2. Concentration-dependent effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD; filled squares) and polychlorinated biphenyl 126 (PCB126; open circles) on 7-ethoxyresorufin O-deethylase (EROD) activity in cultured double-crested cormorant embryonic hepatocytes exposed for 24 h. Data points for EROD activity represent the mean value obtained from triplicate cell culture plates ± SEM.

concentrations was expected based on earlier studies2,3 and is due to the competitive inhibition of the reaction by the inducer(s) and not cytotoxicity. EROD EC50 values for TCDD and PCB126 in DCEH were 0.08 ± 0.03 and 0.26 ± 0.05 nM, respectively (Table 3). Table 3. Ethoxyresorufin O-Deethylase (EROD) EC50 Values (± SEM) in Chicken, Ring-Necked Pheasant, Japanese Quail, And Double-Crested Cormorant Hepatocyte Cultures Exposed to TCDD and PCB126 for 24 h

a

species

type

chickena ring-necked pheasanta Japanese quaila double-crested cormorant

1 2 3 3

TCDD

PCB126

EC50 ± SEM

EC50 ± SEM

0.027 0.047 0.13 0.08

± ± ± ±

0.004 0.007 0.04 0.03

0.043 0.43 3.8 0.26

± ± ± ±

0.001 0.06 0.73 0.05

Data from Manning et al.

The EROD assay has been utilized to determine dioxinassociated effects of individual compounds and complex environmental mixtures in cell culture studies with a variety of avian species to date.2,3,7 In fact, Sanderson et al.10 assessed the EROD-inducing potential of TCDD in double-crested cormorant hepatocytes and determined that the EC50 was 1.1 nM (following a correction factor to account for the use of isooctane instead of DMSO as a solvent). This corrected EC50 value is more than an order of magnitude greater than the EC50 calculated in the present study (0.08 nM). In addition to the use of a different solvent for TCDD administration (i.e., isooctane vs DMSO), Sanderson et al.10 prepared hepatocytes from 1-day old hatchling DCCOs as opposed to late-stage embryos, which could also have contributed to the variability in calculated EC50 values for TCDD. F

DOI: 10.1021/acs.est.5b06181 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Figure 3. Heat map depicting significant fold changes of 27 genes on the double-crested cormorant ToxChip PCR array clustered by the following chemicals at the highest administered noncytotoxic concentration (10 or 300 μM): 1,2-dibromo-4-(1,2-dibromoethyl)-cyclohexane (DBE-DBCH), tris(2,3-dibromopropyl) isocyanurate (TBC), triethyl phosphate (TEP), allyl 2,4,6-tribromophenyl ether (ATE), tris(methylphenyl) phosphate (TMPP), and bisphenol S (BPS). Red and green hues indicate significant up- and down-regulation, respectively, and fold-change values