Abstract

In a recent egg injection study, we showed that in ovo exposure to perfluorohexane sulfonate (PFHxS) affects the pipping success of developing chicken (Gallus gallus domesticus) embryos. We also found evidence of thyroid hormone (TH) pathway interference at multiple levels of biological organization (i.e., somatic growth, messenger RNA expression, and circulating free thyroxine levels). Based on these findings, we hypothesize that PFHxS exposure interferes with TH-dependent neurodevelopmental pathways. This study investigates global transcriptional profiles in cerebral hemispheres of chicken embryos following exposure to a solvent control, 890 or 38,000ng PFHxS/g egg (n = 4–5 per group); doses that lead to the adverse effects indicated above. PFHxS significantly alters the expression (≥ 1.5-fold, p ≤ 0.001) of 11 transcripts at the low dose (890ng/g) and 101 transcripts at the high dose (38,000ng/g). Functional enrichment analysis shows that PFHxS affects genes involved in tissue development and morphology, cellular assembly and organization, and cell-to-cell signaling. Pathway and interactome analyses suggest that genes may be affected through several potential regulatory molecules, including integrin receptors, myelocytomatosis viral oncogene, and CCAAT/enhancer-binding protein. This study identifies key functional and regulatory modes of PFHxS action involving TH-dependent and -independent neurodevelopmental pathways. Some of these TH-dependent mechanisms that occur during embryonic development include tight junction formation, signal transduction, and integrin signaling, whereas TH-independent mechanisms include gap junction intercellular communication.

Perfluoroalkyl acids (PFAAs), specifically perfluorinated sulfonates (PFSAs) and carboxylates (PFCAs), are a family of synthetic substances used for their water- and stain-repellent properties. PFAAs have a unique fluorocarbon structure that renders them virtually nonbiodegradable and persistent in the environment. Globally, PFSAs and PFCAs have been detected in wildlife and humans and have a tendency to bioaccumulate and biomagnify in food webs (Conder et al., 2008). Due to their amphiphilic properties, PFAAs preferentially compartmentalize in protein-rich tissues and have been detected mainly in blood serum, liver, and egg samples of wild animal populations worldwide (Giesy and Kannan, 2001). Two commonly detected and studied PFAAs, perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA), have been largely phased-out of production in North America; however, certain international manufacturers continue to produce PFOS and its precursors (Wang et al., 2010b). Recent production has focused mainly on PFAAs with shorter carbon chain lengths (C < 8) to fill market demand.

Perfluorohexane sulfonate (PFHxS) is an example of a PFAA with a shorter carbon chain length (C = 6), for which biomonitoring data are limited in wild avian species. PFHxS has been detected in the plasma of herring gulls (Larus argentatus) from Lake Huron (Gebbink and Letcher, 2012) and in the livers of numerous avian species worldwide at average concentrations ranging from < 0.5 to 50ng/g wet weight (ww) (Houde et al., 2011; Kannan et al., 2002; Meyer et al., 2009). In herring gull colonies from the Great Lakes, PFHxS concentrations ranged from below detection limits to 3.8ng/g ww in whole eggs (Gebbink et al., 2011). Although PFHxS is detected in wild avian populations, very little is known about the toxicological effects of exposure. There is a growing body of evidence that suggests various PFAAs can impact neurodevelopment in birds and mammals (Johansson et al., 2008, 2009; Lau, 2009; Lau et al., 2003, 2004, 2007; Pinkas et al., 2010; Slotkin et al., 2008; Thibodeaux et al., 2003; Wang et al., 2010a; Yu et al., 2009). Chicken (Gallus gallus domesticus) egg injection studies report reduced hatching success in response to PFOS and PFOA and higher incidences of physical deformities (Molina et al., 2006; O’Brien et al., 2009a,b; Yanai et al., 2008). Posthatch cognitive behavior (i.e., imprinting behavior), immune alterations, and brain asymmetry changes have also been observed following in ovo PFOS and PFOA exposure (Peden-Adams et al., 2009; Pinkas et al., 2010). Furthermore, PFOS and PFOA have been reported to affect peroxisome proliferator-activated receptor alpha activation, lipid metabolism, gap junction intercellular communication (GJIC), and the thyroid hormone (TH) axis (DeWitt et al., 2009; Kudo et al., 1999; Lau et al., 2007; Upham et al., 2009).

The reported effects of PFAAs on neurodevelopment, in particular, led our laboratory to utilize an in vitro screening method to determine the effects of 11 short- and long-chained PFAAs on messenger RNA (mRNA) expression of TH-responsive genes in primary cultures of chicken and herring gull embryonic neuronal cells (Vongphachan et al., 2011). Effects on the TH pathway were assessed because previous studies reported reduced TH levels (i.e., triiodothyronine [T3] and thyroxine [T4]) in rodents exposed to PFAAs (Chang et al., 2008; Martin et al., 2007). Vongphachan et al. (2011) demonstrated that short-chained PFAAs (C < 8) altered the expression of TH-responsive genes, including type II and III 5′-deiodinases (d2 and d3), transthyretin (ttr), and neurogranin (rc3), in chicken embryonic neuronal cells to a greater extent than long-chained PFAAs (C ≥ 8). Furthermore, among 24 PFAAs examined, PFHxS had the strongest binding potency for ttr and was able to displace T4 from binding to ttr (Weiss et al., 2009). In Cassone et al. (2012), in ovo exposure to PFHxS reduced the pipping success of chicken embryos and affected the TH pathway at multiple levels of biological organization (i.e., reduced somatic growth, induced gene expression of d2, d3, rc3 and octamer motif-binding factor 1 [oct1], and decreased circulating free T4 levels). Because THs play an essential role in avian brain development (McNabb, 2007), disruption of this pathway via PFHxS exposure during neurodevelopment may result in harmful and irreversible effects.

Given that PFHxS affects TH homeostasis in chicken embryos (Cassone et al., 2012), we hypothesize that PFHxS exposure interferes with TH-dependent neurodevelopmental pathways. In order to test this hypothesis and to further characterize the effects of PFHxS in the brain, this study investigates global gene expression profiles in the cerebral hemisphere of chicken embryos from dose groups that demonstrated adverse effects, including a decrease in pipping success and disruptions of the TH system, in our previous egg injection study (Cassone et al., 2012). The main objectives of this study were to (1) assess the TH-dependent impacts of PFHxS exposure on gene expression in the brain and (2) identify novel neurodevelopmental modes of PFHxS action. This study identified potential key functional and regulatory events of PFHxS toxicity during avian neurodevelopment.

MATERIALS AND METHODS

Chemicals. Linear sodium PFHxS was purchased from Wellington Laboratories (Guelph, ON; > 98% pure). All stock solutions and serial dilutions were prepared in dimethyl sulfoxide (DMSO; Fisher Scientific, Ottawa, ON) to yield final in-egg concentrations of 890 and 38,000ng PFHxS/g egg. The concentrations of the solutions injected into the eggs were determined by high performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) as described previously (Cassone et al., 2012).

Egg injection and tissue collection. Tissues for gene expression analysis were collected from a PFHxS egg injection experiment described in Cassone et al. (2012). In brief, unincubated, fertilized White Leghorn chicken (G. g. domesticus) eggs were randomly distributed into the following control and treatment groups: DMSO control (n = 20), 890ng PFHxS/g egg (n = 20), and 38,000ng PFHxS/g egg (n = 20), henceforth referred to as the low dose (LD) and the high dose (HD), respectively. An uninjected control group was not included in this study; however, previous findings from our laboratory demonstrated a similar pipping success rate among DMSO and uninjected controls (O’Brien et al., 2009a,b). A small hole was drilled through the egg shell at the center of the air cell and DMSO or PFHxS (~1.0 μl/g egg) was injected into the air cell to attain the desired concentrations described above. After injection, the hole was sealed with filter tape and the eggs were placed horizontally into an incubator (Petersime, Model XI) set at 37.5°C and 58% humidity. During incubation, embryos were monitored frequently by candling and brought to pipping (day 21–22, stage 46; Hamilton and Hamburger, 1951), at which point embryos were euthanized by decapitation. The left and right cerebral hemispheres were collected; the left was used for chemical residue analysis as described in Cassone et al. (2012) and the right was immediately frozen in liquid nitrogen and stored at −80°C for subsequent RNA isolation.

RNA isolation and sample preparation. Total RNA was isolated from a 20–30mg section of the right cerebral hemisphere (n = 4–5 embryos per treatment group) using RNeasy Mini Kits according to the manufacturer’s instructions (Qiagen). Approximately 5 μg of total RNA was DNase treated using DNA-free kits according to the manufacturer (Ambion, Austin, TX). RNA was quantified with a NanoDrop 2000 Spectrophotometer (Thermo Scientific, Wilmington, DE) and RNA quality was assessed using a Bioanalyzer 2100 (Agilent Technologies, Mississauga, ON). Only samples with A260/A280 ratios > 1.8 and an RNA integrity number > 9 were used for downstream applications. Samples were then prepared by diluting 200ng RNA to a total volume of 8.3 μl with RNase-free water. A reference pool of RNA for microarray hybridizations was prepared from equal parts of all samples used for microarray analysis and all samples were stored at −80°C for subsequent experiments.

Microarray hybridization. Experimental RNA samples were labeled with Cyanine 5-CTP (Cy5) and chicken reference RNA was labeled with Cyanine 3-CTP (Cy3) using the Quick Amp Labeling Kit (Agilent Technologies), according to manufacturer’s instructions. Briefly, double-stranded complementary DNA (cDNA) was synthesized from 200ng total RNA using MMLV-RT with T7 promoter primer. Cyanine-labeled cRNA targets were then transcribed in vitro using T7 RNA polymerase. The synthesized cRNA was purified using the RNeasy Mini Kit (Qiagen), and labeled cRNA (825ng) was fragmented at 60°C for 30min with fragmentation solution. Cy3-reference cRNA and Cy5-sample cRNA were hybridized to Agilent 4X44K chicken gene expression microarrays (containing 44,000 60-mer oligonucleotide probes; Array ID 026441) at 65°C for 17h with Agilent hybridization solution and washed according to the manufacturer’s instructions. Arrays were scanned on an Agilent G2505B scanner at 5 μm resolution. Data were acquired with Agilent Feature Extraction Software, version 10.7.3.1.

Data analysis for microarrays. A reference design (Kerr, 2003; Kerr and Churchill, 2001) was used to analyze the gene expression data. Data were preprocessed using R software (http://www.R-project.org). The median signal intensities were normalized using the global locally weighted scatterplot smoothing method (Yang et al., 2002) using the transform.madata function in the microarray analysis of variance (MAANOVA) library (Wu et al., 2003). Probes with fluorescent intensity signals significantly greater than the local mean background plus 3 SDs of background intensity were identified as being expressed (denoted as present). Ratio intensity plots were constructed for the raw and normalized data for each array to identify outliers or microarrays with poor data quality. Differentially expressed genes were identified using the MAANOVA library. An ANOVA model including the main effect of treatment and the block effect of the slide was applied. The Fs statistic (Cui et al., 2005), a shrinkage estimator, was used for the gene-specific variance components and the associated p values for all the statistical tests were estimated using the permutation method (30,000 permutations with residual shuffling). The least-squares means (Goodnight and Harvey, 1978; Searle et al., 1980) were used to estimate the fold changes for each pairwise comparison. Probes were considered differentially expressed if they had absolute fold changes ≥ 1.5 relative to controls and p ≤ 0.001. The promoter regions (−8kb to +2kb from transcriptional start site) of all differentially expressed genes were scanned for potential TH response elements (TREs), as described in Paquette et al. (2011).

Hierarchical clustering was performed using GeneSpring GX version 11.0.2 (Agilent Technologies). Clustering was performed on both entities and conditions, using the Euclidian distance metric and the centroid linkage rule.

Functional and canonical pathway analysis was performed in Ingenuity Pathway Analysis (IPA), which identified biological functions/diseases or pathways that were most significant to the data set. Molecules from the data set that were recognized as human, mouse, or rat orthologs, were differentially expressed, and were associated with biological functions and/or diseases in Ingenuity’s Knowledgebase were considered for the analysis. A right-tailed Fisher’s exact test was used to calculate a p value determining the probability that each biological function/disease or pathway assigned to that data set was due to chance alone.

For interaction network generation in IPA, differentially expressed genes that mapped into IPA were overlaid onto a global molecular network developed from information contained in Ingenuity’s proprietary Knowledgebase. Networks of differentially expressed genes were then algorithmically generated based on their connectivity. Using the Agilent chicken reference set, only direct interactions between experimentally observed molecules (up to 70 human, rat, and mouse orthologs per network) that are expressed in nervous system tissues/primary cells and/or central nervous system (CNS) cell lines were considered. MicroRNAs were not included as a data source for network interactions.

In the network diagrams, genes are represented as nodes and the biological relationship between two nodes is represented as an edge (line). All edges are supported by at least one reference from the literature, a textbook, or from canonical information stored in the Ingenuity Knowledgebase. Nodes are displayed using various shapes that represent the functional class of the gene. The intensity of the node color indicates the degree of up- (red) or down- (green) regulation. Duplicate genes were resolved by using the maximum fold change value. White and yellow nodes are not differentially expressed and are added to the network as bridging/connector molecules. Any connector molecule that had direct interactions with four or more dysregulated genes was considered to have potential regulatory roles in the PFHxS response and was denoted as a yellow node. IPA was then used to identify all possible direct or indirect interactions between potential regulatory molecules and all dysregulated genes in each dose group.

Real-time RT-PCR. Total RNA (375ng) was reverse transcribed to cDNA using SuperScript II reverse transcriptase and random hexamer primers (Invitrogen Canada) as per the manufacturer’s instructions. Reactions containing an RNA template but lacking reverse transcriptase were run in parallel to verify the absence of contaminating genomic DNA (no-reverse transcriptase control). A 1:20 dilution of cDNA with diethylpyrocarbonate (DEPC)-treated water was prepared and stored at −80°C for subsequent real-time RT-PCR. Changes in mRNA expression were assessed by real-time RT-PCR using the Brilliant Q-PCR Core Reagent Kit (TaqMan assay) or Brilliant SYBR Green QPCR Master Mix (SYBR assay) (Agilent Technologies) and MX3000P or MX3005P PCR systems (Stratagene, La Jolla, CA). Primer pairs (Invitrogen) and TaqMan fluorogenic probes (Biosearch, Novato, CA) for the transcripts listed in Supplementary table 1 were designed and optimized for real-time RT-PCR.

TABLE 1

Enriched Functional Categories for Genes Differentially Expressed in Chicken Embryonic Cerebral Hemisphere Following Exposure to 890ng/g (LD) or 38,000ng/g (HD) PFHxS

Rank Category LD HD 
p value range No. of genes p value range No. of genes 
Tissue development and morphology 7.59E-04–4.49E-02 5.58E-04–4.61E-02 18 
Cellular assembly and organization 7.59E-04–4.62E-02 5.58E-04–4.61E-02 13 
Cell-to-cell signaling and interaction 7.57E-03–4.18E-02 5.58E-04–4.61E-02 
Cellular movement 2.28E-03–4.77E-02 6.06E-04–4.16E-02 11 
Nervous system development and function 7.59E-04–4.62E-02 6.06E-04–4.61E-02 
Cell development and morphology 7.59E-04–4.03E-02 7.56E-04–4.36E-02 19 
Molecular transport 1.52E-03–1.36E-02 2.17E-03–3.25E-02 
Lipid metabolism 7.59E-04–3.79E-03 2.17E-03–3.70E-02 
Cellular growth and proliferation 7.59E-04–4.03E-02 4.26E-03–4.67E-02 12 
10 Endocrine system development and disorders 7.59E-04–7.59E-04 4.67E-03–4.69E-02 10 
11 Protein synthesis and trafficking 1.36E-02–1.36E-02 4.71E-03–1.28E-02 
12 Gene expression 3.79E-03–3.79E-03 4.71E-03–3.25E-02 
13 Amino acid metabolism 7.59E-04–1.52E-03 4.71E-03–3.25E-02 
14 Embryonic development 7.59E-04–4.40E-02 4.71E-03–3.40E-02 14 
15 Nucleic acid metabolism 1.52E-03–1.36E-02 4.71E-03–3.70E-02 
16 Carbohydrate metabolism 1.52E-03–1.52E-03 9.39E-03–3.70E-02 
17 Cancer 4.40E-02–4.77E-02 9.39E-03–4.61E-02 19 
18 Cellular function and maintenance   1.41E-02–1.41E-02 
19 DNA replication, recombination, and repair   1.87E-02–1.87E-02 
20 Organismal survival   3.01E-02–3.01E-02 
Rank Category LD HD 
p value range No. of genes p value range No. of genes 
Tissue development and morphology 7.59E-04–4.49E-02 5.58E-04–4.61E-02 18 
Cellular assembly and organization 7.59E-04–4.62E-02 5.58E-04–4.61E-02 13 
Cell-to-cell signaling and interaction 7.57E-03–4.18E-02 5.58E-04–4.61E-02 
Cellular movement 2.28E-03–4.77E-02 6.06E-04–4.16E-02 11 
Nervous system development and function 7.59E-04–4.62E-02 6.06E-04–4.61E-02 
Cell development and morphology 7.59E-04–4.03E-02 7.56E-04–4.36E-02 19 
Molecular transport 1.52E-03–1.36E-02 2.17E-03–3.25E-02 
Lipid metabolism 7.59E-04–3.79E-03 2.17E-03–3.70E-02 
Cellular growth and proliferation 7.59E-04–4.03E-02 4.26E-03–4.67E-02 12 
10 Endocrine system development and disorders 7.59E-04–7.59E-04 4.67E-03–4.69E-02 10 
11 Protein synthesis and trafficking 1.36E-02–1.36E-02 4.71E-03–1.28E-02 
12 Gene expression 3.79E-03–3.79E-03 4.71E-03–3.25E-02 
13 Amino acid metabolism 7.59E-04–1.52E-03 4.71E-03–3.25E-02 
14 Embryonic development 7.59E-04–4.40E-02 4.71E-03–3.40E-02 14 
15 Nucleic acid metabolism 1.52E-03–1.36E-02 4.71E-03–3.70E-02 
16 Carbohydrate metabolism 1.52E-03–1.52E-03 9.39E-03–3.70E-02 
17 Cancer 4.40E-02–4.77E-02 9.39E-03–4.61E-02 19 
18 Cellular function and maintenance   1.41E-02–1.41E-02 
19 DNA replication, recombination, and repair   1.87E-02–1.87E-02 
20 Organismal survival   3.01E-02–3.01E-02 

For TaqMan assays, each 25 μl reaction contained 1X Core PCR buffer, 5mM MgCl2, 0.8mM dNTP mix, 8% glycerol, 75nM ROX reference dye, forward and reverse primers at optimized concentrations (Supplementary table 1), 200nM fluorogenic probe, 5 μl diluted cDNA (1:20), and 1.25U SureStart Taq Polymerase. The thermocycler program included an enzyme activation step at 95°C (10min) and 40 cycles of 95°C (30 s) and 60°C (1min). SYBR assays were used as an alternative to TaqMan assays when fluorogenic probes failed and amplification was nonexponential. Each 25 μl SYBR Green reaction contained 2X Brilliant SYBR Green QPCR Master Mix, 75nM ROX reference dye, forward and reverse primers at optimized concentrations (Supplementary table 1), and 5 μl diluted cDNA (1:20). The thermocycler program included an enzyme activation step at 95°C (10min) and 40 cycles of 95°C (30 s), 60°C (1min), and 72°C (30 s). All reactions were performed using cDNA from the same 4–5 embryos per treatment group used for microarray analysis. All PCR products were sequenced to verify that each primer pair was amplifying the intended gene target. All gene targets were normalized to β-actin (internal control) as its expression was invariable across treatment groups. An identical reaction was conducted, in parallel, for each assay by replacing cDNA template with DEPC-treated water (no template control) to monitor for contamination. Cycle threshold (Ct) data were normalized to β-actin using the 2−∆Ct equation (Schmittgen and Livak, 2008). The fold change in target gene mRNA abundance in PFAA treatment groups was expressed relative to the solvent control group. Statistically significant differences in mRNA expression were identified by performing a one-way ANOVA to 2−ΔCt-transformed data followed by a Bonferroni’s t-test for multiple comparisons versus the solvent control (SigmaStat v2.03; SPSS). Changes were considered statistically significant if p ≤ 0.05.

RESULTS

Differentially Expressed Genes

Changes in gene transcription in the cerebral hemisphere of chicken embryos exposed to PFHxS were identified using Agilent 4x44K chicken gene expression microarrays. The data discussed in this publication have been deposited in the National Centre for Biotechnology Information Gene Expression Omnibus (series accession GSE37339) (Edgar et al., 2002). A total of 106 probes (representing 78 unique genes) were differentially expressed (fold change ≥ 1.5, p ≤ 0.001) in the cerebral hemisphere following exposure to PFHxS. Of these differentially expressed probes, the majority (94/106 = 89%) were downregulated by PFHxS in the developing chicken brain, whereas only 12 probes (11%) were upregulated. A Venn diagram showing the number of probes that were up- or downregulated (i.e., dysregulated) by the 890ng/g (LD) and 38,000ng/g (HD) of PFHxS is depicted in Figure 1. A detailed list of all probes that were differentially expressed following PFHxS exposure is included in Supplementary table 2. Hierarchical clustering was conducted using the list of differentially expressed genes and revealed two main branches dividing the HD group from the solvent control group (Fig. 2).

TABLE 2

Enriched Canonical Pathways for Genes That Were Differentially Expressed in the Cerebral Hemisphere of Chicken Embryos Exposed to 38,000ng/g PFHxS. Only Pathways With Two or More Affected Genes Were Reported

Ingenuity canonical pathways p value No. of genes Genes No. of molecules in pathway 
Dendritic cell maturation 2.57E-03 pik3r1, hla-c, col3a1 110 
ILK signaling 1.20E-02 dsp, pik3r1, fn1, vim, pgf 159 
Virus entry via endocytic pathways 2.82E-02 pik3r1, hla-c 83 
Role of macrophages, fibroblasts, and endothelial
    cells in rheumatoid arthritis 3.63E-02 pik3r1, fn1, sfrp2, pgf 247 
IGF-1 signaling 3.72E-02 pik3r1, nov 93 
Hereditary breast cancer signaling 4.27E-02 pik3r1, slc19a1 109 
Relaxin signaling 4.57E-02 pik3r1, pde5a 127 
Hepatic fibrosis/hepatic stellate cell activation 4.90E-02 fn1, col3a1, pgf 122 
Ingenuity canonical pathways p value No. of genes Genes No. of molecules in pathway 
Dendritic cell maturation 2.57E-03 pik3r1, hla-c, col3a1 110 
ILK signaling 1.20E-02 dsp, pik3r1, fn1, vim, pgf 159 
Virus entry via endocytic pathways 2.82E-02 pik3r1, hla-c 83 
Role of macrophages, fibroblasts, and endothelial
    cells in rheumatoid arthritis 3.63E-02 pik3r1, fn1, sfrp2, pgf 247 
IGF-1 signaling 3.72E-02 pik3r1, nov 93 
Hereditary breast cancer signaling 4.27E-02 pik3r1, slc19a1 109 
Relaxin signaling 4.57E-02 pik3r1, pde5a 127 
Hepatic fibrosis/hepatic stellate cell activation 4.90E-02 fn1, col3a1, pgf 122 
FIG. 1.

Venn diagram illustrating the number of genes up-(↑) or down-(↓) regulated (fold change ≥ 1.5, p ≤ 0.001) by either 890ng/g (LD) or 38,000ng/g (HD) PFHxS in the cerebral hemisphere of embryonic chickens.

FIG. 1.

Venn diagram illustrating the number of genes up-(↑) or down-(↓) regulated (fold change ≥ 1.5, p ≤ 0.001) by either 890ng/g (LD) or 38,000ng/g (HD) PFHxS in the cerebral hemisphere of embryonic chickens.

FIG. 2.

Hierarchical clustering of expression profiles from the cerebral hemisphere of chicken embryos exposed to the DMSO solvent control, 890ng/g (LD), or 38,000ng/g (HD) PFHxS. Clustering was based on 78 unique and differentially expressed genes (fold change ≥ 1.5, p ≤ 0.001).

FIG. 2.

Hierarchical clustering of expression profiles from the cerebral hemisphere of chicken embryos exposed to the DMSO solvent control, 890ng/g (LD), or 38,000ng/g (HD) PFHxS. Clustering was based on 78 unique and differentially expressed genes (fold change ≥ 1.5, p ≤ 0.001).

The two largest significant decreases in expression in response to PFHxS exposure were for two splice variants of nephroblastoma overexpressed gene (nov), which were downregulated by 4.3- and 5.2-fold at the HD. Both nov splice variants were also decreased in the LD by 1.6- and 2-fold, although the expression changes were not statistically significant. The third largest significant decrease in expression in response to PFHxS exposure was for claudin 11 (cldn11), which was reduced by 3.2-fold at the HD. A duplicate probe for cldn11 was also decreased significantly by 2.4-fold at the HD. At the LD, both cldn11 probes were also decreased by 1.4-fold, but not significantly. Few of the differentially expressed genes determined by the microarray analysis were upregulated by PFHxS exposure. The largest significant increase in expression in response to PFHxS exposure was for heparan sulfate 6-O-sulfotransferase 2 (hs6st2), which was upregulated by 4.9- and 2.4-fold at the LD and HD, respectively.

The expression of several differentially expressed genes identified via microarray analysis was assessed using real-time RT-PCR. The expression of cldn11, desmoplakin (dsp), and nov was consistent between approaches (Fig. 3). An additional nine transcripts, selected for their role in the interaction networks discussed below, were also analyzed by real-time RT-PCR to assess concordance with microarray results (Supplementary fig. 1). A complete list of transcripts assessed is included in Supplementary table 1. Transcripts assessed by real-time RT-PCR were directionally consistent with microarray data for 17 out of 24 conditions (i.e., 12 genes at two dose groups). This level of concordance between real-time RT-PCR and microarray data is comparable to a study by Wang et al. (2010a), which showed that gene expression profiles in the rat brain were directionally consistent for 11 out of 14 conditions.

FIG. 3.

The relative mRNA expression levels of neuronal transcripts following in ovo exposure of chicken embryos to 890ng/g (LD) or 38,000ng/g (HD) PFHxS. mRNA levels of (A) claudin 11 (cldn11), (B) desmoplakin (dsp), and (C) nephroblastoma overexpressed gene (nov) were determined by microarray (MA, black bars) and real-time RT-PCR (qPCR, white bars) (n = 4–5; error bars represent SEM; *p ≤ 0.05; **p ≤ 0.001).

FIG. 3.

The relative mRNA expression levels of neuronal transcripts following in ovo exposure of chicken embryos to 890ng/g (LD) or 38,000ng/g (HD) PFHxS. mRNA levels of (A) claudin 11 (cldn11), (B) desmoplakin (dsp), and (C) nephroblastoma overexpressed gene (nov) were determined by microarray (MA, black bars) and real-time RT-PCR (qPCR, white bars) (n = 4–5; error bars represent SEM; *p ≤ 0.05; **p ≤ 0.001).

Functional Analysis and Canonical Pathway Mapping

For all probes on the chicken array, 29,985 out of 43,603 (69%) were recognized by IPA as gene orthologs in human, rat, or mouse. Of all dysregulated genes, 86 of 106 probes (81%) were mapped as orthologs in IPA (shown in Supplementary table 2). The majority of unmapped IDs correspond to probes for hypothetical proteins and ESTs with unknown function and therefore could not be included in downstream functional, pathway, and interactome analysis.

Functional enrichment analysis of the differentially expressed genes was performed using IPA. The relevant functional categories are summarized in Table 1. Detailed lists of the specific significantly enriched functions and genes within each category are included in Supplementary table 3. The top functional categories, based on significance (p ≤ 0.05), were tissue development and morphology, cellular assembly and organization, cell-to-cell signaling and interaction, cellular movement, and nervous system development and function. Many of the enriched tissue development and morphology functions were involved in adherence, adhesion, assembly, and cell-cell contact of astrocytes, glioma cells, and neuroepithelial cells.

TABLE 3

Molecules From Networks Generated in IPA That Had Interactions With Four or More Genes That Were Differentially Expressed (DE) in the Cerebral Hemisphere of Chicken Embryos Exposed to 38,000ng/g PFHxS

Molecule Entrez name FC p value No. of interacting DE genes 
Direct Indirect Total 
MYC Myelocytomatosis viral oncogene homolog (avian) −1.13 0.021 
SRC Sarcoma (Schmidt-Ruppin A-2) viral oncogene homolog (avian) 1.11 0.007 
HIF1α Hypoxia inducible factor 1 alpha 1.17 0.169 
HNF4α Hepatocyte nuclear factor 4 alpha −1.15 0.061 
C/EBPβ CCAAT/enhancer-binding protein beta −1.05 0.538 
SP1 Specificity protein 1 transcription factor −1.18 0.214 
ITGβ1 Integrin beta 1 −1.26 0.189 
THBS1 Thrombospondin 1 −1.48 0.002 
Molecule Entrez name FC p value No. of interacting DE genes 
Direct Indirect Total 
MYC Myelocytomatosis viral oncogene homolog (avian) −1.13 0.021 
SRC Sarcoma (Schmidt-Ruppin A-2) viral oncogene homolog (avian) 1.11 0.007 
HIF1α Hypoxia inducible factor 1 alpha 1.17 0.169 
HNF4α Hepatocyte nuclear factor 4 alpha −1.15 0.061 
C/EBPβ CCAAT/enhancer-binding protein beta −1.05 0.538 
SP1 Specificity protein 1 transcription factor −1.18 0.214 
ITGβ1 Integrin beta 1 −1.26 0.189 
THBS1 Thrombospondin 1 −1.48 0.002 

Note. Fold change (FC) and p-values from microarray are also reported.

Differentially expressed genes were mapped to canonical pathways using Ingenuity’s Knowledgebase in IPA, which revealed that exposure to PFHxS dysregulated genes belonging to several pathways. Canonical pathways that were significantly affected are summarized in Table 2. Pathways that were most significantly disrupted by PFHxS exposure included dendritic cell maturation, integrin-linked kinase (ILK) signaling (Supplementary fig. 2), and virus entry via endocytic pathways.

Interaction Networks and Potential Regulatory Molecules

Interaction networks were generated in IPA using mapped orthologs of differentially expressed genes from the HD treatment group. Interaction networks for the LD group were not identified as too few genes were dysregulated and connections could not be established. The network generated is shown in Figure 4, where only direct interactions with gene products that occur in the brain were considered. Indirect interactions were not considered because they do not require evidence that a physical interaction exists between genes or gene products and, therefore, can lead to uncertainties regarding the nature of the interaction because of the potential involvement of one or multiple intermediary factor(s). A summary of potential regulatory molecules (i.e., interactions with four or more genes that were differentially expressed by PFHxS exposure) is listed in Table 3 and a detailed list of the interacting genes is included in Supplementary table 4. The potential regulatory molecules identified through interactome analysis include integrin beta 1 (ITGβ1), myelocytomatosis viral oncogene (MYC), and CCAAT/enhancer-binding protein beta (C/EBPβ).

FIG. 4.

IPA-generated interaction network for genes dysregulated by exposure to 38,000ng/g PFHxS. Molecule shapes (nodes) represent different molecule types (see legend). Red nodes represent genes that were significantly upregulated and green nodes represent genes that were significantly downregulated. Connecting lines (edges) represent direct interactions between genes that are documented in the Ingenuity Knowledgebase. White and yellow shapes are genes that were added to the network by IPA based on their connectivity to dysregulated genes, where yellow shapes are considered potential regulatory molecules (see Table 3).

FIG. 4.

IPA-generated interaction network for genes dysregulated by exposure to 38,000ng/g PFHxS. Molecule shapes (nodes) represent different molecule types (see legend). Red nodes represent genes that were significantly upregulated and green nodes represent genes that were significantly downregulated. Connecting lines (edges) represent direct interactions between genes that are documented in the Ingenuity Knowledgebase. White and yellow shapes are genes that were added to the network by IPA based on their connectivity to dysregulated genes, where yellow shapes are considered potential regulatory molecules (see Table 3).

Discussion

This study determined the effects of exposure to two PFHxS doses on global transcription profiles in the cerebral hemisphere of developing chicken embryos. Avian toxicogenomics studies have also been conducted using Northern bobwhite (Colinus virginianus) and include the development, annotation, and utilization of microarrays to determine molecular perturbations following 2,6-dinitrotoluene exposure (Rawat et al., 2010a,b). Our analysis revealed 78 genes that were differentially expressed, the majority of which were responsive only at the HD (86%) and were primarily downregulated (89%). Cluster analysis on the differentially expressed genes separated DMSO-exposed cerebral hemisphere from all except one of the PFHxS-treated embryos. Hierarchical clustering could not separate two LD samples from the HD treatment groups. The clustering results demonstrate a clear dose-response relationship, with the HD clustering furthest from the DMSO group and the LD having an intermediate expression profile.

The nov transcript had the largest significant decrease in expression levels following PFHxS exposure, a result that was confirmed by real-time RT-PCR. Nephroblastoma overexpressed gene is involved in organogenesis and regulation of various cellular processes such as adhesion, migration, proliferation, differentiation, and survival (Le Dreau et al., 2010b). A microarray study on liver tissue of 14-week-old chickens revealed induction of nov expression after 4 weeks of PFOS and PFOA exposure at 0.1 and 0.5mg/ml, respectively, followed by 4 weeks of depuration (Yeung et al., 2007). The results from this study and Yeung et al. (2007) indicate that nov expression is altered following exposure to various PFAAs. The variability in direction of disregulation could be based on differences in life-history stage (pipping embryo vs. 14-week-old chickens), tissue type (cerebral hemisphere vs. liver), and PFAA (PFHxS vs. PFOS and PFOA) between the two studies. Furthermore, nov may play a role in GJIC, where the antiproliferative activity of nov affects reorganization of cellular contacts (Gupta et al., 2001). It is well documented that PFAAs can influence GJIC (Hu et al., 2002; Upham et al., 2009; Yoo et al., 2008) and chronic GJIC disruption could lead to serious neurological and endocrinological problems (Hu et al., 2002). In rat liver and dolphin kidney epithelial cell lines, PFHxS inhibits GJIC in a rapid and reversible dose-dependent manner, although the mechanism is not fully understood (Hu et al., 2002; Yoo et al., 2008). It is plausible that the effects of PFHxS on GJIC in the brain are related to nov downregulation. In addition, altered GJIC activity may partly explain the reduced growth observed in chicken embryos from the PFHxS egg injection study (Cassone et al., 2012) as GJIC is known to be essential for normal growth and development (Fu et al., 2004). Overall, the utility of nov expression as an indicator of PFAA exposure warrants further investigation (e.g., mechanistic in vitro studies) given the variable expression results in two avian studies following exposure to various PFAAs (this study vs. Yeung et al., 2007).

Claudin 11 is a transmembrane protein that functions within the myelin sheaths of the brain and its transcript levels were significantly diminished in response to the HD of PFHxS, a result that was confirmed by real-time RT-PCR. Claudin 11 plays a critical role in tight junction (TJ) formation. TJs are composed partly of claudins and when formed between the blood-brain barrier (BBB) endothelial cells, lead to high endothelial electrical resistance and low paracellular permeability (Stamatovic et al., 2008). Claudin 11-null mice exhibit a 60% decrease in nerve conduction (Devaux and Gow, 2008) and absent CNS myelin and TJ formation, leading to neurological deficiencies (Tiwari-Woodruff et al., 2001). PFOS treatment of human brain microvascular endothelial cells—major components of the BBB—results in disassembly (i.e., “opening”) of endothelial TJs and increases in permeability; claudin 5 distribution is also disrupted (Stamatovic et al., 2008). Desmoplakin, another important player in endothelial junctions, was also significantly decreased in the HD PFHxS group and validated by real-time RT-PCR. Desmoplakin is involved in cellular structure and is an indispensable component of functional desmosomes, intercellular junctions that tightly link adjacent cells. Desmoplakin knockout mice demonstrate increased embryonic death (Gallicano et al., 1998; Vasioukhin et al., 2001). Taken together, our results indicate that PFHxS may act on cldn11 and/or dsp, which affects TJ formation and may facilitate transport across the BBB. This hypothesis is in agreement with the dose-dependent PFHxS accumulation observed in the cerebral hemisphere of chicken embryos reported in Cassone et al. (2012).

Functional enrichment analysis identified the top functional categories to be tissue development and morphology, cellular assembly and organization, cell-to-cell signaling and interaction, cellular movement, and nervous system development and function. To our knowledge, only one other study has investigated gene expression profiles in the brain in response to PFAA treatment (Wang et al., 2010a). Wang et al. (2010a) demonstrated similar functional enrichment in the cortex of rat pups exposed prenatally to PFOS, where the top functional categories were cell cycle, TJ, and cell communication. As such, the functional profiles determined by the present avian study are consistent with the gene expression study using mammalian models.

The functions described above are essential to embryonic development; perturbation of the associated genes may result in abnormal maturation and function of the CNS. Developmental neurotoxicity has been reported previously in response to PFAA exposure (Lau et al., 2004, 2007). At a cellular level, PFAAs have negative effects on cell growth and replication as well as cell numbers and viability in undifferentiated and differentiated PC12 cells (a neuronotypic cell line used to characterize neurotoxicity) (Slotkin et al., 2008). These findings agree with the cellular growth and proliferation functional category that was significantly enriched by PFHxS exposure in this study. In neonatal mice given a single, oral dose of PFOS or PFOA, levels of proteins important in normal brain development (i.e., neuronal growth and synaptogenesis) were altered in the hippocampus and cerebral cortex (Johansson et al., 2009), which supports the functional categories enriched in this study (i.e., tissue development and morphology and nervous system development and function). Furthermore, prolonged PFAA treatment moderately inhibited neurite growth and dramatically suppressed synaptogenesis in cultured neurons in a chain length- and functional group-dependent manner, where PFHxS suppressed the neurite sum lengths by ~10% (Liao et al., 2009). This reduction suggests that long-term exposure may result in serious neurodevelopmental damage.

Integrin Receptors and Signaling

Integrins are transmembrane receptors essential to embryonic development and necessary for neuronal cells to attach, spread, migrate, and extend processes on extracellular matrix molecules and mediate cell-cell and cell-matrix adhesion events (Wu and Reddy, 2012). Mice lacking extracellular matrix components, such as fibronectin, or lacking β1 integrins die during the early stages of embryonic development. Fibronectin 1 (fn1) was significantly decreased by the HD of PFHxS via microarray. This decrease was confirmed by real-time RT-PCR. Integrins were also found to interact with many of the top differentially expressed genes in our study, providing further support to their involvement in PFHxS response. Nephroblastoma overexpressed gene is a putative ligand for integrin receptors, including β1 and β5 (ITGβ1 and ITGβ5, respectively), and mediates several cellular actions such as cell adhesion and proliferation (Le Dreau et al., 2010a,b; Sin et al., 2009). Claudin 11 forms a complex with ITGβ1 to regulate proliferation and migration of oligodendrocytes (Tiwari-Woodruff et al., 2001, 2004). Therefore, decreased fn1 transcript levels and integrin-mediated response genes may contribute to the reduced pipping success observed in chicken embryos at the HD (Cassone et al., 2012).

Among enriched canonical pathways, ILK signaling had the largest number of perturbed genes (i.e., dsp, phosphoinositide-3-kinase, regulatory subunit 1 [pik3r1], fn1, vimentin [vim], and placental growth factor [pgf]) following PFHxS treatment. Most of the genes that were disrupted are downstream products of ILK activation, as opposed to genes that interact directly with ILK. The expression of these downstream genes—dsp, fn1, vim, and pgf—is directly regulated, at the transcriptional level, by snail homolog 2 (SLUG) or hypoxia inducible factor 1 alpha (HIF1α) (Jethwa et al., 2008; Savagner et al., 1997; Tan et al., 2004; Vuoriluoto et al., 2011). SLUG is a transcription factor required for the development of neural crest cells, which give rise to many tissues and cells during embryonic development (Cohen et al., 1998). HIF1α is a transcription factor that functions in reduced oxygen conditions to induce the transcription of various target genes involved in tumor angiogenesis, invasion, cell survival, and glucose metabolism (Kaur et al., 2005). SLUG and HIF1α are two of the most overexpressed transcription factors in human glioblastomas when compared with nontumor brains (Yang et al., 2010). Furthermore, PFOS and perfluorodecanoic acid both stimulate glioblastoma cell proliferation (Merritt and Foran, 2007).

Detailed network analysis of our data set revealed that an integrin receptor (i.e., ITGβ1) may be a potential regulatory molecule (i.e., ITGβ1 had interactions with four or more genes that were differentially expressed by PFHxS exposure) driving the observed integrin-associated responses in PFHxS-exposed embryos. ITGβ1 has been implicated in regulating the morphology of cortical development and dendritic spines, formation of synapses in neurons, and mediating neurite outgrowth after injury. Taken together, these data provide further evidence to suggest that integrin receptors and signaling play an integral role in PFHxS action.

Thyroid-Dependent Effects

THs are important for the control of growth and development in birds because they directly trigger cellular differentiation and maturation in a number of tissues, but are especially important in neurodevelopment (McNabb, 2007). Previously, we reported that PFHxS adversely affects the TH pathway of developing chicken embryos at multiple levels of biological organization (Cassone et al., 2012). At concentrations ≥ 890ng/g, PFHxS induced the expression of TH-responsive genes (d2, d3, rc3, and oct1) in liver and the cerebral hemisphere and diminished circulating free T4 levels. It was hypothesized that increases in d2 and d3 levels led to reduced free T4 levels as a result of augmented, localized metabolism of available T4. Furthermore, at 38,000ng/g, PFHxS decreased pipping success and reduced the mass and tarsus length of surviving embryos. In this study, we examined the cerebral hemispheres from dose groups in Cassone et al. (2012) that demonstrated adverse TH-related effects using microarrays, thereby phenotypically anchoring gene expression data to these toxicologically relevant endpoints.

The top differentially expressed genes were hypothesized to be associated with TH-dependent effects. As nov expression is not known to be directly or indirectly mediated by TH, it is unclear whether this effect is a TH-specific or independent process. However, the promoter region of nov was scanned for TRE consensus sequences and eight potential TREs were found. Future studies investigating whether these TREs are active sites of TH receptor binding and transcriptional control would be valuable to further elucidate PFHxS mode of action. Claudin 11 and dsp also contain two potential TREs in the promoter region of each gene. Furthermore, TJ opening is hypothesized to occur via the phosphatidylinositol 3-kinase (PI3K) signaling pathway (Wang et al., 2011), which plays a role in nongenomic TH action. In addition to the perturbation of cldn11 expression observed in this study, pik3r1 was significantly induced in response to the HD of PFHxS, a result that was directionally consistent with real-time RT-PCR data. THs activate the PI3K pathway at the cell surface via integrins, which ultimately increases transcription of various genes (Davis et al., 2009; Di Liegro, 2008; Lu and Cheng, 2010; Moeller and Broecker-Preuss, 2011; Sheng et al., 2012). Thus, TH-driven perturbations in pik3r1 may lead to changes in cldn11 and dsp expression, and reflect important neurotoxicological events in the TH-dependent mode of action of PFHxS. Moreover, nongenomic mechanisms of TH action have been described in many tissues, including the brain, some of which appear to be mediated by integrins (Di Liegro, 2008; Moeller and Broecker-Preuss, 2011; Sheng et al., 2012). This suggests that disruption of ILK signaling in response to PFHxS may also result from TH-dependent pathways. Protein-binding studies with integrins and THs, in response to PFHxS exposure, would offer further clarification on this mechanism.

Enriched functional categories can also be linked to the TH effects observed in our previous egg injection study (Cassone et al., 2012). Tissue-specific differentiation and maturation is a TH-dependent function in birds (McNabb, 2007) and appears to be affected by the LD and HD of PFHxS. At the LD of PFHxS, few genes with known functions were altered. Wingless-type MMTV integration site family, member 5A (wnt5a) was significantly decreased and hs6st2 and gamma-glutamyltransferase 1 (ggt1) were significantly increased by PFHxS at the LD. Together, wnt5a, hs6st2, and ggt1 are involved in tissue development and proliferation of cells. These altered genes and enriched functions were affected at the same dose group that demonstrated a decrease in free T4 levels in chicken embryos (Cassone et al., 2012). Diminished free T4 levels may result in reduced embryonic growth, which corresponds with the enriched functions observed at the LD in this study, as well as effects observed at the HD. At the HD, tissue development and morphology was also enriched via interference with intercellular interactions. Furthermore, the cellular assembly and organization category had several enriched functions relating to formation of intercellular junctions, further supporting the role of PFHxS in disrupting cell-cell interactions. In addition, this category comprised many enriched functions involved in the reorganization of F-actin, morphology of actin filaments, formation of actin cytoskeleton, and organization of actin stress fibers. Diminished circulating free T4 levels can influence actin and tubulin expression (Smith et al., 2002), and further supports a TH-dependent mode of action. Because THs regulate neurodevelopment, it can be hypothesized that the reduced free T4 levels observed in response to PFHxS exposure could be a factor explaining the enriched functional categories described in this study.

The potential regulatory molecules identified in the current study are also linked to TH-dependent effects. Thyroxine is able to induce short-term responses in both neurons and astrocytes by binding to integrins; T4 induces integrin binding to its ligand, laminin (Farwell et al., 2005). Farwell et al. (2005) demonstrate that T4 (but not T3) directly regulates F-actin content of elongating neurites and promotes extensive granule cell migration and neuronal process outgrowth, which are attenuated by anti-ITGβ1 antibodies. These data suggest that T4 influences neuronal process outgrowth via integrin receptors. Overall, the results of our study support a causal association between THs and integrin signaling at the molecular level (i.e., individual gene expression changes and perturbed pathways) and phenotypic level (i.e., decreased embryo growth and reduced free T4 levels) in response to PFHxS exposure in developing chicken embryos (Cassone et al., 2012).

Other regulatory molecules identified include MYC and C/EBPβ, receptor pathways that were suppressed in response to PFHxS exposure. MYC is a transcription factor that plays an important role in growth control, differentiation and apoptosis and is believed to regulate the expression of 10–15% of all cellular genes (Hoffman and Liebermann, 2008). The MYC protein is one of the most frequently affected in a variety of cancers and its target genes participate in functions including cell cycle, survival, protein synthesis, and cell adhesion (Hoffman and Liebermann, 2008). The majority of genes that were dysregulated by PFHxS and that interact with MYC are involved in tissue development and apoptosis functions. Although MYC expression was not significantly altered in this study, it was implicated as a regulatory molecule in response to PFHxS treatment. PFOS exposure was found to upregulate MYC expression in the spleen of adult male mice (Dong et al., 2012). The variability in gene expression response between this study and the Dong et al. (2012) study may be based on differences in species (i.e., chicken vs. mouse) and life-history stage (i.e., embryo vs. adult). In addition, MYC expression can be directly regulated by THs: neuroblastoma cells treated with T3 rapidly downregulate MYC (Puzianowska-Kuznicka et al., 2006). C/EBPβ protein plays a pivotal role in the control of cellular growth, proliferation, and differentiation (Ramji and Foka, 2002). In the brain, C/EBPβ functions in neuronal differentiation, learning and memory processes, glial or neuronal cell functions, and synaptic plasticity (Calella et al., 2007; Sterneck and Johnson, 1998). All of the genes that were dysregulated by PFHxS and interact with C/EBPβ are involved in cell death (i.e., downregulation of these genes increases apoptosis). In mammalian cells, C/EBPβ binds the promoter region of the MYC gene and represses expression (Berberich-Siebelt et al., 2006; Gutsch et al., 2011; Sebastian et al., 2005). Furthermore, C/EBPβ plays a role in TH-dependent transactivation and its expression can also be mediated by THs (Wang et al., 2009). These data taken together suggest a TH-dependent mode of action in the regulation of cellular growth and survival mediated via MYC and C/EBPβ following PFHxS exposure in the developing chicken brain.

CONCLUSION

Due to the voluntary phase-out of PFOS and PFOA, short-chain PFAAs are being manufactured as replacement substances. PFHxS is an example of a short-chained PFSA that has been detected in the environment and biota, for which few toxicological studies exist. Previously, we reported TH-disrupting effects of PFHxS in developing chicken embryos (Cassone et al., 2012) with a lowest observable adverse effect level (890ng/g) that was 18 times greater than the highest reported mean concentration in avian wildlife (50ng/g ww in liver of grey herons [Meyer et al., 2009]). Effects on the TH pathway at 890ng/g PFHxS led to the genome-wide evaluation of transcription profiles of PFHxS action in this study; however, effects on gene expression and molecular mechanisms were minimal in the cerebral hemisphere at 890ng/g PFHxS. At 38,000ng/g, functional enrichment analysis revealed effects on genes associated with tissue development and morphology and cellular assembly and organization—two categories that are TH-dependent. Interactome analysis further suggested that genes may be affected through integrin receptors and integrin signaling pathways, which can be TH-dependent via nongenomic mechanisms. These transcriptional responses support our hypothesis that TH-dependent neurodevelopmental pathways are affected in developing chicken embryos exposed to PFHxS. One TH-independent mode of PFHxS action identified was GJIC via nov downregulation. Overall, the use of microarray technology enabled the identification of novel modes of PFHxS action, including developmental and cellular processes and expanded our understanding concerning the molecular effects of PFHxS.

SUPPLEMENTARY DATA

Supplementary data are available online at http://toxsci.oxfordjournals.org/.

FUNDING

Environment Canada: (1) Chemicals Management Plan (CMP), (2) Strategic Technology Applications of Genomics for the Environment (STAGE) and (3) the Ecotoxicology and Wildlife Health Division; the University of Ottawa; Center for Advanced Research in Environmental Genomics (CAREG).

References

Berberich-Siebelt
F.,
Berberich
I.,
Andrulis
M.,
Santner-Nanan
B.,
Jha
M. K.,
Klein-Hessling
S.,
Schimpl
A.,
Serfling
E.
2006
SUMOylation interferes with CCAAT/enhancer-binding protein beta-mediated c-myc repression, but not IL-4 activation in T cells.
J. Immunol.
 
176
4843
4851
Calella
A. M.,
Nerlov
C.,
Lopez
R. G.,
Sciarretta
C.,
von Bohlen und Halbach
O.,
Bereshchenko
O.,
Minichiello
L.
2007
Neurotrophin/Trk receptor signaling mediates C/EBPalpha, -beta and NeuroD recruitment to immediate-early gene promoters in neuronal cells and requires C/EBPs to induce immediate-early gene transcription.
Neural Dev.
 
2
4
Cassone
C. G.,
Vongphachan
V.,
Chiu
S.,
Williams
K. L.,
Letcher
R. J.,
Pelletier
E.,
Crump
D.,
Kennedy
S. W.
2012
In ovo effects of perfluorohexane sulfonate and perfluorohexanoate on pipping success, development, mRNA expression, and thyroid hormone levels in chicken embryos.
Toxicol. Sci. 127, 216–224.
 
Chang
S. C.,
Thibodeaux
J. R.,
Eastvold
M. L.,
Ehresman
D. J.,
Bjork
J. A.,
Froehlich
J. W.,
Lau
C.,
Singh
R. J.,
Wallace
K. B.,
Butenhoff
J. L.
2008
Thyroid hormone status and pituitary function in adult rats given oral doses of perfluorooctanesulfonate (PFOS).
Toxicology
 
243
330
339
Cohen
M. E.,
Yin
M.,
Paznekas
W. A.,
Schertzer
M.,
Wood
S.,
Jabs
E. W.
1998
Human SLUG gene organization, expression, and chromosome map location on 8q.
Genomics
 
51
468
471
Conder
J. M.,
Hoke
R. A.,
De Wolf
W.,
Russell
M. H.,
Buck
R. C.
2008
Are PFCAs bioaccumulative? A critical review and comparison with regulatory criteria and persistent lipophilic compounds.
Environ. Sci. Technol.
 
42
995
1003
Cui
X.,
Hwang
J. T.,
Qiu
J.,
Blades
N. J.,
Churchill
G. A.
2005
Improved statistical tests for differential gene expression by shrinking variance components estimates.
Biostatistics
 
6
59
75
Davis
P. J.,
Davis
F. B.,
Lin
H. Y.,
Mousa
S. A.,
Zhou
M.,
Luidens
M. K.
2009
Translational implications of nongenomic actions of thyroid hormone initiated at its integrin receptor.
Am. J. Physiol. Endocrinol. Metab.
 
297
E1238
E1246
Devaux
J.,
Gow
A.
2008
Tight junctions potentiate the insulative properties of small CNS myelinated axons.
J. Cell Biol.
 
183
909
921
DeWitt
J. C.,
Shnyra
A.,
Badr
M. Z.,
Loveless
S. E.,
Hoban
D.,
Frame
S. R.,
Cunard
R.,
Anderson
S. E.,
Meade
B. J.,
Peden-Adams
M. M.,
et al
2009
Immunotoxicity of perfluorooctanoic acid and perfluorooctane sulfonate and the role of peroxisome proliferator-activated receptor alpha.
Crit. Rev. Toxicol.
 
39
76
94
Di Liegro
I.
2008
Thyroid hormones and the central nervous system of mammals (review)
Mol. Med. Report
 
1
279
295
Dong
G. H.,
Zhang
Y. H.,
Zheng
L.,
Liang
Z. F.,
Jin
Y. H.,
He
Q. C.
2012
Subchronic effects of perfluorooctanesulfonate exposure on inflammation in adult male C57BL/6 mice.
Environ. Toxicol.
 
27
285
296
Edgar
R.,
Domrachev
M.,
Lash
A. E.
2002
Gene Expression Omnibus: NCBI gene expression and hybridization array data repository.
Nucleic Acids Res.
 
30
207
210
Farwell
A. P.,
Dubord-Tomasetti
S. A.,
Pietrzykowski
A. Z.,
Stachelek
S. J.,
Leonard
J. L.
2005
Regulation of cerebellar neuronal migration and neurite outgrowth by thyroxine and 3,3’,5’-triiodothyronine.
Brain Res. Dev. Brain Res.
 
154
121
135
Fu
C. T.,
Bechberger
J. F.,
Ozog
M. A.,
Perbal
B.,
Naus
C. C.
2004
CCN3 (NOV) interacts with connexin43 in C6 glioma cells: Possible mechanism of connexin-mediated growth suppression.
J. Biol. Chem.
 
279
36943
36950
Gallicano
G. I.,
Kouklis
P.,
Bauer
C.,
Yin
M.,
Vasioukhin
V.,
Degenstein
L.,
Fuchs
E.
1998
Desmoplakin is required early in development for assembly of desmosomes and cytoskeletal linkage.
J. Cell Biol.
 
143
2009
2022
Gebbink
W. A.,
Letcher
R. J.
2012
Comparative tissue and body compartment accumulation and maternal transfer to eggs of perfluoroalkyl sulfonates and carboxylates in Great Lakes herring gulls.
Environ. Pollut.
 
162
40
47
Gebbink
W. A.,
Letcher
R. J.,
Hebert
C. E.,
Chip Weseloh
D. V.
2011
Twenty years of temporal change in perfluoroalkyl sulfonate and carboxylate contaminants in herring gull eggs from the Laurentian Great Lakes.
J. Environ. Monit.
 
13
3365
3372
Giesy
J. P.,
Kannan
K.
2001
Global distribution of perfluorooctane sulfonate in wildlife.
Environ. Sci. Technol.
 
35
1339
1342
Goodnight
J. H.
Harvey
W. R.
1978
Least-squares means in the fixed-effects general linear models. R-103.SAS Technical Report
SAS Institute Inc.,
Cary, NC.
Gupta
N.,
Wang
H.,
McLeod
T. L.,
Naus
C. C.,
Kyurkchiev
S.,
Advani
S.,
Yu
J.,
Perbal
B.,
Weichselbaum
R. R.
2001
Inhibition of glioma cell growth and tumorigenic potential by CCN3 (NOV).
Mol. Pathol.
 
54
293
299
Gutsch
R.,
Kandemir
J. D.,
Pietsch
D.,
Cappello
C.,
Meyer
J.,
Simanowski
K.,
Huber
R.,
Brand
K.
2011
CCAAT/enhancer-binding protein beta inhibits proliferation in monocytic cells by affecting the retinoblastoma protein/E2F/cyclin E pathway but is not directly required for macrophage morphology.
J. Biol. Chem.
 
286
22716
22729
Hamilton
V.
Hamburger
H. L.
1951
A series of normal stages in the development of the chick embryo.
J. Morphol.
 
88
49
92
Hoffman
B.,
Liebermann
D. A.
2008
Apoptotic signaling by c-MYC.
Oncogene
 
27
6462
6472
Houde
M.,
De Silva
A. O.,
Muir
D. C.,
Letcher
R. J.
2011
Monitoring of perfluorinated compounds in aquatic biota: An updated review.
Environ. Sci. Technol.
 
45
7962
7973
Hu
W.,
Jones
P. D.,
Upham
B. L.,
Trosko
J. E.,
Lau
C.,
Giesy
J. P.
2002
Inhibition of gap junctional intercellular communication by perfluorinated compounds in rat liver and dolphin kidney epithelial cell lines in vitro and Sprague Dawley rats in vivo.
Toxicol. Sci.
 
68
429
436
Jethwa
P.,
Naqvi
M.,
Hardy
R. G.,
Hotchin
N. A.,
Roberts
S.,
Spychal
R.,
Tselepis
C.
2008
Overexpression of Slug is associated with malignant progression of esophageal adenocarcinoma.
World J. Gastroenterol.
 
14
1044
1052
Johansson
N.,
Eriksson
P.,
Viberg
H.
2009
Neonatal exposure to PFOS and PFOA in mice results in changes in proteins which are important for neuronal growth and synaptogenesis in the developing brain.
Toxicol. Sci.
 
108
412
418
Johansson
N.,
Fredriksson
A.,
Eriksson
P.
2008
Neonatal exposure to perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) causes neurobehavioural defects in adult mice.
Neurotoxicology
 
29
160
169
Kannan
K.,
Choi
J. W.,
Iseki
N.,
Senthilkumar
K.,
Kim
D. H.,
Giesy
J. P.
2002
Concentrations of perfluorinated acids in livers of birds from Japan and Korea.
Chemosphere
 
49
225
231
Kaur
B.,
Khwaja
F. W.,
Severson
E. A.,
Matheny
S. L.,
Brat
D. J.,
Van Meir
E. G.
2005
Hypoxia and the hypoxia-inducible-factor pathway in glioma growth and angiogenesis.
Neuro Oncol.
 
7
134
153
Kerr
M. K.
2003
Design considerations for efficient and effective microarray studies.
Biometrics
 
59
822
828
Kerr
M. K.,
Churchill
G. A.
2001
Statistical design and the analysis of gene expression microarray data.
Genet. Res.
 
77
123
128
Kudo
N.,
Mizuguchi
H.,
Yamamoto
A.,
Kawashima
Y.
1999
Alterations by perfluorooctanoic acid of glycerolipid metabolism in rat liver.
Chem. Biol. Interact.
 
118
69
83
Lau
C.
2009
Perfluoroalkyl acids: Recent activities and research progress.
Reprod. Toxicol.
 
27
209
211
Lau
C.,
Anitole
K.,
Hodes
C.,
Lai
D.,
Pfahles-Hutchens
A.,
Seed
J.
2007
Perfluoroalkyl acids: A review of monitoring and toxicological findings.
Toxicol. Sci.
 
99
366
394
Lau
C.,
Butenhoff
J. L.,
Rogers
J. M.
2004
The developmental toxicity of perfluoroalkyl acids and their derivatives.
Toxicol. Appl. Pharmacol.
 
198
231
241
Lau
C.,
Thibodeaux
J. R.,
Hanson
R. G.,
Rogers
J. M.,
Grey
B. E.,
Stanton
M. E.,
Butenhoff
J. L.,
Stevenson
L. A.
2003
Exposure to perfluorooctane sulfonate during pregnancy in rat and mouse. II: Postnatal evaluation.
Toxicol. Sci.
 
74
382
392
Le Dreau
G.,
Kular
L.,
Nicot
A. B.,
Calmel
C.,
Melik-Parsadaniantz
S.,
Kitabgi
P.,
Laurent
M.,
Martinerie
C.
(
2010
).
NOV/CCN3 upregulates CCL2 and CXCL1 expression in astrocytes through beta1 and beta5 integrins.
Glia
 
58
1510
1521
Le Dreau
G.,
Nicot
A.,
Benard
M.,
Thibout
H.,
Vaudry
D.,
Martinerie
C.,
Laurent
M.
(
2010
).
NOV/CCN3 promotes maturation of cerebellar granule neuron precursors.
Mol. Cell Neurosci.
 
43
60
71
Liao
C.,
Wang
T.,
Cui
L.,
Zhou
Q.,
Duan
S.,
Jiang
G.
2009
Changes in synaptic transmission, calcium current, and neurite growth by perfluorinated compounds are dependent on the chain length and functional group.
Environ. Sci. Technol.
 
43
2099
2104
Lu
C.,
Cheng
S. Y.
2010
Thyroid hormone receptors regulate adipogenesis and carcinogenesis via crosstalk signaling with peroxisome proliferator-activated receptors.
J. Mol. Endocrinol.
 
44
143
154
Martin
M. T.,
Brennan
R. J.,
Hu
W.,
Ayanoglu
E.,
Lau
C.,
Ren
H.,
Wood
C. R.,
Corton
J. C.,
Kavlock
R. J.,
Dix
D. J.
2007
Toxicogenomic study of triazole fungicides and perfluoroalkyl acids in rat livers predicts toxicity and categorizes chemicals based on mechanisms of toxicity.
Toxicol. Sci.
 
97
595
613
McNabb
F. M.
2007
The hypothalamic-pituitary-thyroid (HPT) axis in birds and its role in bird development and reproduction.
Crit. Rev. Toxicol.
 
37
163
193
Merritt
R. L.,
Foran
C. M.
2007
Influence of persistent contaminants and steroid hormones on glioblastoma cell growth.
J. Toxicol. Environ. Health A
 
70
19
27
Meyer
J.,
Jaspers
V. L.,
Eens
M.,
de Coen
W.
2009
The relationship between perfluorinated chemical levels in the feathers and livers of birds from different trophic levels.
Sci. Total Environ.
 
407
5894
5900
Moeller
L. C.,
Broecker-Preuss
M.
2011
Transcriptional regulation by nonclassical action of thyroid hormone.
Thyroid Res.
 
4
(
Suppl. 1
)
S6
Molina
E. D.,
Balander
R.,
Fitzgerald
S. D.,
Giesy
J. P.,
Kannan
K.,
Mitchell
R.,
Bursian
S. J.
2006
Effects of air cell injection of perfluorooctane sulfonate before incubation on development of the white leghorn chicken (Gallus domesticus) embryo.
Environ. Toxicol. Chem.
 
25
227
232
O’Brien
J. M.,
Carew
A. C.,
Chu
S.,
Letcher
R. J.,
Kennedy
S. W.
(
2009
).
Perfluorooctane sulfonate (PFOS) toxicity in domestic chicken (Gallus gallus domesticus) embryos in the absence of effects on peroxisome proliferator activated receptor alpha (PPARalpha)-regulated genes.
Comp. Biochem. Physiol. C Toxicol. Pharmacol.
 
149
524
530
O’Brien
J. M.,
Crump
D.,
Mundy
L. J.,
Chu
S.,
McLaren
K. K.,
Vongphachan
V.,
Letcher
R. J.,
Kennedy
S. W.
(
2009
).
Pipping success and liver mRNA expression in chicken embryos exposed in ovo to C8 and C11 perfluorinated carboxylic acids and C10 perfluorinated sulfonate.
Toxicol. Lett.
 
190
134
139
Paquette
M. A.,
Dong
H.,
Gagne
R.,
Williams
A.,
Malowany
M.,
Wade
M. G.,
Yauk
C. L.
2011
Thyroid hormone-regulated gene expression in juvenile mouse liver: Identification of thyroid response elements using microarray profiling and in silico analyses.
BMC Genomics
 
12
634
Peden-Adams
M. M.,
Stuckey
J. E.,
Gaworecki
K. M.,
Berger-Ritchie
J.,
Bryant
K.,
Jodice
P. G.,
Scott
T. R.,
Ferrario
J. B.,
Guan
B.,
Vigo
C.,
et al
2009
Developmental toxicity in white leghorn chickens following in ovo exposure to perfluorooctane sulfonate (PFOS).
Reprod. Toxicol.
 
27
307
318
Pinkas
A.,
Slotkin
T. A.,
Brick-Turin
Y.,
Van der Zee
E. A.,
Yanai
J.
2010
Neurobehavioral teratogenicity of perfluorinated alkyls in an avian model.
Neurotoxicol. Teratol.
 
32
182
186
Puzianowska-Kuznicka
M.,
Pietrzak
M.,
Turowska
O.,
Nauman
A.
2006
Thyroid hormones and their receptors in the regulation of cell proliferation.
Acta Biochim. Pol.
 
53
641
650
Ramji
D. P.,
Foka
P.
2002
CCAAT/enhancer-binding proteins: Structure, function and regulation.
Biochem. J.
 
365
(
Pt 3
)
561
575
Rawat
A.,
Gust
K. A.,
Deng
Y.,
Garcia-Reyero
N.,
Quinn
M. J.,
Jr
Johnson
M. S.,
Indest
K. J.,
Elasri
M. O.,
Perkins
E. J.
(
2010
).
From raw materials to validated system: The construction of a genomic library and microarray to interpret systemic perturbations in Northern bobwhite.
Physiol. Genomics
 
42
219
235
Rawat
A.,
Gust
K. A.,
Elasri
M. O.,
Perkins
E. J.
(
2010
).
Quail Genomics: A knowledgebase for Northern bobwhite.
BMC Bioinformatics
 
11
(
Suppl. 6
)
S13
Savagner
P.,
Yamada
K. M.,
Thiery
J. P.
1997
The zinc-finger protein slug causes desmosome dissociation, an initial and necessary step for growth factor-induced epithelial-mesenchymal transition.
J. Cell Biol.
 
137
1403
1419
Schmittgen
T. D.,
Livak
K. J.
2008
Analyzing real-time PCR data by the comparative C(T) method.
Nat. Protoc.
 
3
1101
1108
Searle
S. R.
Speed
F. M.
Milliken
G. A.
1980
The population marginal means in the linear model: An alternative to least squares means.
Am. Statistician
 
34
216
221
Sebastian
T.,
Malik
R.,
Thomas
S.,
Sage
J.,
Johnson
P. F.
2005
C/EBPbeta cooperates with RB:E2F to implement Ras(V12)-induced cellular senescence.
EMBO J.
 
24
3301
3312
Sheng
Z. G.,
Tang
Y.,
Liu
Y. X.,
Yuan
Y.,
Zhao
B. Q.,
Chao
X. J.,
Zhu
B. Z.
2012
Low concentrations of bisphenol a suppress thyroid hormone receptor transcription through a nongenomic mechanism.
Toxicol. Appl. Pharmacol.
 
259
133
142
Sin
W. C.,
Tse
M.,
Planque
N.,
Perbal
B.,
Lampe
P. D.,
Naus
C. C.
2009
Matricellular protein CCN3 (NOV) regulates actin cytoskeleton reorganization.
J. Biol. Chem.
 
284
29935
29944
Slotkin
T. A.,
MacKillop
E. A.,
Melnick
R. L.,
Thayer
K. A.,
Seidler
F. J.
2008
Developmental neurotoxicity of perfluorinated chemicals modeled in vitro.
Environ. Health Perspect.
 
116
716
722
Smith
J. W.,
Evans
A. T.,
Costall
B.,
Smythe
J. W.
2002
Thyroid hormones, brain function and cognition: A brief review.
Neurosci. Biobehav. Rev.
 
26
45
60
Stamatovic
S. M.,
Keep
R. F.,
Andjelkovic
A. V.
2008
Brain endothelial cell-cell junctions: How to “open” the blood brain barrier.
Curr. Neuropharmacol.
 
6
179
192
Sterneck
E.,
Johnson
P. F.
1998
CCAAT/enhancer binding protein beta is a neuronal transcriptional regulator activated by nerve growth factor receptor signaling.
J. Neurochem.
 
70
2424
2433
Tan
C.,
Cruet-Hennequart
S.,
Troussard
A.,
Fazli
L.,
Costello
P.,
Sutton
K.,
Wheeler
J.,
Gleave
M.,
Sanghera
J.,
Dedhar
S.
2004
Regulation of tumor angiogenesis by integrin-linked kinase (ILK).
Cancer Cell
 
5
79
90
Thibodeaux
J. R.,
Hanson
R. G.,
Rogers
J. M.,
Grey
B. E.,
Barbee
B. D.,
Richards
J. H.,
Butenhoff
J. L.,
Stevenson
L. A.,
Lau
C.
2003
Exposure to perfluorooctane sulfonate during pregnancy in rat and mouse. I: Maternal and prenatal evaluations.
Toxicol. Sci.
 
74
369
381
Tiwari-Woodruff
S. K.,
Buznikov
A. G.,
Vu
T. Q.,
Micevych
P. E.,
Chen
K.,
Kornblum
H. I.,
Bronstein
J. M.
2001
OSP/claudin-11 forms a complex with a novel member of the tetraspanin super family and beta1 integrin and regulates proliferation and migration of oligodendrocytes.
J. Cell Biol.
 
153
295
305
Tiwari-Woodruff
S. K.,
Kaplan
R.,
Kornblum
H. I.,
Bronstein
J. M.
2004
Developmental expression of OAP-1/Tspan-3, a member of the tetraspanin superfamily.
J. Neurosci. Res.
 
77
166
173
Upham
B. L.,
Park
J. S.,
Babica
P.,
Sovadinova
I.,
Rummel
A. M.,
Trosko
J. E.,
Hirose
A.,
Hasegawa
R.,
Kanno
J.,
Sai
K.
2009
Structure-activity-dependent regulation of cell communication by perfluorinated fatty acids using in vivo and in vitro model systems.
Environ. Health Perspect.
 
117
545
551
Vasioukhin
V.,
Bowers
E.,
Bauer
C.,
Degenstein
L.,
Fuchs
E.
2001
Desmoplakin is essential in epidermal sheet formation.
Nat. Cell Biol.
 
3
1076
1085
Vongphachan
V.,
Cassone
C. G.,
Wu
D.,
Chiu
S.,
Crump
D.,
Kennedy
S. W.
2011
Effects of perfluoroalkyl compounds on mRNA expression levels of thyroid hormone-responsive genes in primary cultures of avian neuronal cells.
Toxicol. Sci.
 
120
392
402
Vuoriluoto
K.,
Haugen
H.,
Kiviluoto
S.,
Mpindi
J. P.,
Nevo
J.,
Gjerdrum
C.,
Tiron
C.,
Lorens
J. B.,
Ivaska
J.
2011
Vimentin regulates EMT induction by Slug and oncogenic H-Ras and migration by governing Axl expression in breast cancer.
Oncogene
 
30
1436
1448
Wang
F.,
Liu
W.,
Jin
Y.,
Dai
J.,
Yu
W.,
Liu
X.,
Liu
L.
2010
Transcriptional effects of prenatal and neonatal exposure to PFOS in developing rat brain.
Environ. Sci. Technol.
 
44
1847
1853
Wang
S.,
Zhang
S.,
Zhao
B.,
Lun
L.
2009
Up-regulation of C/EBP by thyroid hormones: A case demonstrating the vertebrate-like thyroid hormone signaling pathway in amphioxus.
Mol. Cell Endocrinol.
 
313
57
63
Wang
X.,
Li
B.,
Zhao
W. D.,
Liu
Y. J.,
Shang
D. S.,
Fang
W. G.,
Chen
Y. H.
2011
Perfluorooctane sulfonate triggers tight junction “opening” in brain endothelial cells via phosphatidylinositol 3-kinase.
Biochem. Biophys. Res. Commun.
 
410
258
263
Wang
Y.,
Fu
J.,
Wang
T.,
Liang
Y.,
Pan
Y.,
Cai
Y.,
Jiang
G.
2010
Distribution of perfluorooctane sulfonate and other perfluorochemicals in the ambient environment around a manufacturing facility in China.
Environ. Sci. Technol.
 
44
8062
8067
Weiss
J. M.,
Andersson
P. L.,
Lamoree
M. H.,
Leonards
P. E.,
van Leeuwen
S. P.,
Hamers
T.
2009
Competitive binding of poly- and perfluorinated compounds to the thyroid hormone transport protein transthyretin.
Toxicol. Sci.
 
109
206
216
Wu
H.
Kerr
M. K.
Cui
X.
Churchill
G. A.
2003
MAANOVA: A software package for the analysis of spotted cDNA microarray experiments.
In
The Analysis of Gene Expression Data: Methods and Software
  (
Parmigiani
G.
Garrett
E. S.
Irizarry
R. A.
Zeger
S.
, Eds.), pp.
313
431
Springer-Verlag, New York, NY.
Wu
X.,
Reddy
D. S.
2012
Integrins as receptor targets for neurological disorders.
Pharmacol. Ther.
 
134
68
81
Yanai
J.,
Dotan
S.,
Goz
R.,
Pinkas
A.,
Seidler
F. J.,
Slotkin
T. A.,
Zimmerman
F.
2008
Exposure of developing chicks to perfluorooctanoic acid induces defects in prehatch and early posthatch development.
J. Toxicol. Environ. Health A
 
71
131
133
Yang
H. W.,
Menon
L. G.,
Black
P. M.,
Carroll
R. S.,
Johnson
M. D.
2010
SNAI2/Slug promotes growth and invasion in human gliomas.
BMC Cancer
 
10
301
Yang
Y. H.,
Dudoit
S.,
Luu
P.,
Lin
D. M.,
Peng
V.,
Ngai
J.,
Speed
T. P.
2002
Normalization for cDNA microarray data: A robust composite method addressing single and multiple slide systematic variation.
Nucleic Acids Res.
 
30
e15
Yeung
L. W.,
Guruge
K. S.,
Yamanaka
N.,
Miyazaki
S.,
Lam
P. K.
2007
Differential expression of chicken hepatic genes responsive to PFOA and PFOS.
Toxicology
 
237
111
125
Yoo
H.,
Kannan
K.,
Kim
S. K.,
Lee
K. T.,
Newsted
J. L.,
Giesy
J. P.
2008
Perfluoroalkyl acids in the egg yolk of birds from Lake Shihwa, Korea.
Environ. Sci. Technol.
 
42
5821
5827
Yu
W. G.,
Liu
W.,
Jin
Y. H.
2009
Effects of perfluorooctane sulfonate on rat thyroid hormone biosynthesis and metabolism.
Environ. Toxicol. Chem.
 
28
990
996