Chemicals. TBPH and TBMEHP were synthesized by AsisChem (Watertown, MA). We purchased MEHP from TCI America (Portland, OR); rosiglitazone from Cayman Chemical (Ann Arbor, MI); and 2-(4-(2-(1-(cyclohexanebutyl)-3-cyclohexylureido)ethyl)phenylthio)-2-methylpropionic acid (GW7647), a prototypic PPARα agonist, was purchased from Sigma-Aldrich (St. Louis, MO).
. We sampled the homes, offices, and cars of 31 participants in the Boston, Massachusetts, area during the winter of 2009. The institutional review board of the Boston University Medical Center approved the study protocol, and all participants gave their informed consent. Dust samples were collected by investigators using a cellulose extraction thimble (Whatman International, Piscataway, NJ) inserted between the crevice tool and vacuum tube extender of a Eureka Mighty-Mite vacuum cleaner (Allen et al. 2008
). Each main living area of the home and each office was vacuumed for approximately 10 min, capturing dust from the surface area of the room. Cars (n
= 20) were vacuumed for approximately 10 min, collecting dust from the entire surface of the front and back seats. The dashboard, floor, and other surfaces of the vehicles were not vacuumed. Dust samples were sieved to collect particles < 500 μm in size. The sieved samples were placed in clean amber glass jars and stored at –20°C until analysis. Sodium sulfate powder was used as a surrogate for dust in the collection of field blanks. Samples were analyzed for TBPH using gas chromatography–mass spectrometry (GC-MS) as previously described (Stapleton et al. 2008
). An indoor dust standard reference material (SRM 2585; National Institute of Standards and Technology, Gaithersburg, MD), field blanks (n
= 12) and laboratory blanks (sodium sulfate, n
= 3) were all run alongside the dust samples for quality control purposes. TBPH was not detected in any field or laboratory blank. The limit of detection (LOD) was based on the instrumental detection limit using a signal to noise ratio of 3 (2.5 ng). Concentrations below the LOD were substituted with a value of one-half the LOD. The concentration of TBPH in SRM 2585 was 779 ± 108 ng/g, which is within the range reported by two other research laboratories (Sahlström et al. 2011
; Van den Eede et al. 2012
. In vitro
metabolism experiments were performed using commercially available purified hepatic porcine esterase (Sigma-Aldrich) to assess the potential metabolism of TBPH and quantify the formation rate of TBMEHP. Purified human and rat esterases were not available. Enzymatic incubations were performed in 1 mL of 0.1 M potassium phosphate buffer (pH 7.4) for 2 hr at 37°C in the presence of 5.6 μM TBPH delivered in 1 μL of DMSO and 0.1 mg of porcine hepatic esterase. We selected the concentration of TBPH used to maximize detection of potential metabolites without greatly exceeding the aqueous solubility. Lower concentrations of TBPH were evaluated for metabolism under similar conditions but resulted in no detectable levels of TBMEHP. The concentration of esterases used was also maximized based on the stock solution available. Reactions were stopped by the addition of 1 mL of 1 M hydrochloric acid. Before extraction, 50 ng of tetrachloro-monohexyl phthalate (TCMHP; Sigma-Aldrich) was added to each sample as an internal quantitation standard. Solid-phase extraction was performed with Sampli-Q OPT cartridges (3 mL, 30 mg; Agilent Inc., Santa Clara, CA), which were first conditioned with 3 mL methanol (MeOH) and rinsed with 3 mL water. Samples were added to the cartridges, and the cartridges were rinsed with 3 mL water; the analytes were then eluted with 3 mL MeOH. Low rates of nonenzymatic formation of TBMEHP (< 2% of active samples) detected in control incubations without esterase were subtracted from values observed in active incubations. Analysis of TBMEHP was performed using liquid chromatography with tandem mass spectrometry (LC/MS/MS) with negative electrospray ionization. All samples were prepared in duplicate, and experiments were repeated on 2 separate days. In vitro
TBPH metabolism has been described in detail by Roberts et al. (2012)
In vivo rat experiments. Timed-pregnant Fischer rats were purchased from Charles River Laboratories (Wilmington, MA). Rats were housed at 21 ± 1°C on a 12-hr light/dark cycle; food and water were given ad libitum. All animal procedures were performed with approval from Brown University’s Institutional Animal Care and Use Committee, and animals were treated humanely and with regard for alleviation of suffering. Rats were assigned by randomized weight to three treatment groups: corn oil/ethanol control (n = 10), 200 mg/kg TBMEHP (low dose; n = 9), and 500 mg/kg TBMEHP (high dose; n = 10). Rats were housed in random pairs within groups. TBMEHP, which is a water-insoluble solid at room temperature, was dissolved in a solution of corn oil and 5% ethanol. Dosing was by gavage (2 mL/kg).
Rat dams were dosed once per day on gestational days (GDs) 18 and 19. Six hours after the final dose, dams were euthanized by isoflurane overdose and cervical dislocation. Blood was collected by cardiac puncture, and serum was isolated by centrifugation and then frozen at –80°C until further use. Organs were harvested and weighed, then fixed in 10% neutral buffered formalin. Fetuses were euthanized by decapitation, and sex was identified by internal dissection. Within each litter, both testes of male pups were numbered upon isolation. The first testis was placed into formalin, the second and third testes were snap frozen in liquid N2, and the fourth testis was incubated in media for a 3-hr testosterone production assay. With the fifth testis, the sequence was repeated, ensuring an equal and randomized assignment of testes within a litter for various analyses.
Dam ovaries, adrenal glands, kidneys, and livers were isolated and weighed. The thyroid was harvested by excising the upper part of the trachea. The kidneys, livers, and thyroids were embedded in paraffin, sectioned (5 µm), and stained with hematoxylin and eosin (H&E). We scored livers for the presence of mitotic cells by counting the number of characteristic events (cells with chromosomal condensation and aggregation, and nuclear envelope breakdown) in five randomly selected high-powered fields per liver section. The average number of mitotic cells per high-powered field per dam was then converted to a square millimeter unit area and used to compare across groups. This morphological assessment was complemented by immunohistochemical staining using a rabbit monoclonal antibody to the Ki67 (catalog no. ab16667; Abcam, PLC, Cambridge, MA) proliferation marker on paraffin-embedded liver sections from the 29 dams. Ki67 is expressed during all phases of the cell cycle but does not quantify how many cells will undergo mitosis. Slides were scanned using an Aperio CS Scanscope (Aperio Technologies Inc., Vista, CA); 20 randomly selected fixed-size fields from each section were selected for quantification by overlaying the tissue with a numbered grid. Cells were counted as proliferative events regardless of stain intensity if the stained area was > 3 µm in length, if at least half of the stained area was within the fixed-size area, and if there was a clear central area of the stain that was more intensely stained than the rest. We used the area of the fixed-size fields to calculate Ki67-positive staining per square millimeter, and then averaged the measurements for each slide. The slide averages were pooled within treatment groups and used to compare across groups.
Liver apoptosis was quantified by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining. Livers were embedded in paraffin; 5-µm tissue sections were stained using an ApopTag Peroxidase In Situ Apoptosis Detection Kit (catalog no. S7100; Chemicon International, Temecula, CA) according to the manufacturer’s protocol. The number of TUNEL-positive cells per high-powered field (400×) was determined in 20 randomly selected fields from the liver sections from each dam. The area of the high-powered field was used to calculate the number of TUNEL-positive cells per square millimeter. We then used the average number of positive cells per dam per square millimeter to compare across treatment groups.
Maternal serum was analyzed for liver enzymes (alkaline phosphatase, alanine aminotransferase, and aspartate aminotransferase), albumin, uric acid, total protein, glucose, triglycerides, carbon dioxide, creatinine, blood urea nitrogen, calcium, chloride, sodium, potassium, phosphorus, and cholesterol with a Beckman Coulter DxC analyzer (Beckman Coulter, Brea, CA). Thyroid hormones (T3 and T4) were measured using a Siemens ADVIA Centaur XP assay system (Siemens AG, Munich, Germany).
We examined the effects of in utero exposure on the fetal testes. Formalin-fixed fetal testes were embedded in glycol methacrylate (Heraeus Kulzer, LLC, Wehrheim, Germany), sectioned (5 µm), and stained with H&E. The sections were scanned using an Aperio CS ScanScope, and the digitized images were used to measure seminiferous cord area. The testis slides were then scored under a microscope for the presence of MNGs; the number of MNGs per testis was normalized by seminiferous cord area and then pooled by litter.
We examined testosterone production in the fetal testis by incubating single fetal testes in M-199 media at 37°C for 3 hr. The media supernatant was collected in aliquots and analyzed for testosterone by immunoassay (Ligand Assay and Analysis Core, Center for Research in Reproduction, University of Virginia, Charlottesville, VA).
. We investigated the effects of TBMEHP on deiodinase activity using a competitive substrate assay developed by Butt et al. (2011)
. Briefly, rat liver microsomes (Invitrogen, Carlsbad, CA) were diluted to 1 mg protein/mL in 0.1 M potassium phosphate buffer (pH 7.4) containing dithiothreitol and NADPH. A specific mass (100 nM) of T4 was added to the samples and incubated at 37°C for 60 min in a shaking water bath with increasing concentrations (0–200 mg/mL) of TBMEHP. We selected substrate and protein concentrations such that the concentration of T4 was << Km
(Visser et al. 1979
). The conversion of T4 to its deiodinated metabolites [T3, 3,3´,5´-reverse triiodothyronine (rT3), 3,3´-diiodothyronine (3,3´-T2), and 3´-monoiodothyronine] was measured using LC/MS/MS as described by Butt et al. (2011)
. All samples were run in triplicate. Half maximal inhibitory concentration (IC50
) values were obtained using the one-site competition model in SigmaPlot (version 9.01; Systat Software Inc., Chicago, IL).
. We purchased NIH 3T3 L1 preadipocyte cells from ATCC (Manassas, VA). FAO cells stably transfected with mouse PPARα (FAO-PPARα cells; Shipley et al. 2004
) were kindly provided by D. Waxman (Boston University, Boston, MA). Stocks of NIH 3T3 L1 cells were maintained in DMEM (Dulbecco’s modified Eagle medium) with 10% calf serum (Sigma-Aldrich). Stocks of FAO-PPARα cells were maintained in DMEM with 5% fetal bovine serum (FBS) and 2 μg/mL puromycin (both from Sigma-Aldrich). Maintenance media were also supplemented with 5 μg/mL plasmocin (Invitrogen, San Diego, CA) and 20 mM l
-glutamine (Mediatech, Manassas, VA). Cultures were maintained at 37°C in a humidified 5% CO2
atmosphere. For experiments, NIH 3T3 L1 cells and FAO-PPARα cells were plated at 40,000 cells/well (24-well plates) or 200,000 cells/well (6-well plates) in DMEM with 5% FBS, plasmocin and l
-glutamine. Sodium pyruvate (1 mM; MP Biomedicals LLC, Solon, OH) was also included in NIH 3T3 L1 experimental medium. Cultures were allowed to become confluent over 3–4 days.
Prior to dosing, the medium was replaced and supplemented with insulin (0.5 μg/mL) and dexamethasone (10–9 M). Cultures received no treatment (naive) or were treated with vehicle (DMSO, 0.1%), the PPARγ ligand rosiglitazone (100 nM), the PPARγ ligand GW7647 (100 nM), MEHP (10–100 μM), TBPH (10–100 μM) or TBMEHP (10–100 μM). After treatment, cells were cultured for 24 hr (for mRNA expression) or 7 days (for lipid accumulation and perilipin expression). For perilipin expression experiments, medium was changed and the cultures were re-dosed on day 4. All other experiments received a single treatment.
For transduction, lentiviral particles were prepared with the MISSION™ Non-target shRNA (short hairpin RNA) control vector (SHC002) or a MISSION™ PPARγ shRNA vector (TRCN0000001658 or TRCN0000001660; Sigma-Aldrich), as previously described (Yanik et al. 2011
). Medium on established NIH 3T3 L1 cultures was replaced with medium containing 8 μg/mL polybrene. Cultures were not transduced or they were transduced with either the nontarget lentivirus [multiplicity of infection (MOI) of 30:1] or with PPARγ-shRNA lentivirus (MOI of 15:1 for each virus) and incubated for 3 days. Before dosing, the medium was replaced with medium containing 10 nM dexamethasone and 0.5 μg/mL insulin. Cultures received no treatment (naive) or were treated with vehicle (DMSO, 0.1%) or TBMEHP (80 μM). On day 3, medium was replaced and the cultures were re-treated. Cells were harvested on days 4 and 7 of treatment.
To determine lipid accumulation, cells were washed once in cold phosphate-buffered saline (PBS) and stained with an aqueous solution of Nile Red (1 μg/mL in PBS) for 10 min (Greenspan et al. 1985
). Fluorescence [excitation 485 (20-nm bandwidth), emission 530 nm (25-nm bandwidth)] was measured using a Synergy2 multifunction plate reader (BioTek, Winooski, VT). Fluorescence in the experimental wells was normalized by subtracting the fluorescence measured in wells containing naive cells.
To determine perilipin expression, cells were washed once in cold PBS, collected, lysed in Cell Lysis Buffer (Cell Signaling Technology, Beverly, MA), and sonicated. The lysates were cleared by centrifugation, and the supernatants were used for protein expression analyses. Protein concentrations were determined by the Bradford method (Bradford 1976
). Total proteins (15–45 μg) were resolved on 10% gels, transferred to a 0.2-µm nitrocellulose membrane, and incubated with polyclonal rabbit anti-perilipin (catalog no. 3470; Cell Signaling Technology). Immunoreactive bands were detected using horseradish peroxidase–conjugated secondary antibodies (Biorad, Hercules, CA) followed by electrochemiluminescence. To control for equal protein loading, blots were re-probed with a β-actin–specific antibody (catalog no. A5441; Sigma-Aldrich) and analyzed as described above.
To determine changes in mRNA expression, cells were washed once in cold PBS and frozen at –80°C. Total RNA was extracted and genomic DNA was removed using the RNeasy Plus Mini Kit (Qiagen, Valencia, CA). cDNA was prepared using the GoScript™ Reverse Transcription System (Promega, Madison, WI), with a 1:1 mixture of random and oligo (dT)15
primers. Real-time quantitative polymerase chain reaction (qPCR) was performed using the GoTaq® qPCR Master Mix System (Promega). The following validated primers were purchased from Qiagen Inc.: mouse PPAR
[Genbank accession no. NM_011146 (http://www.ncbi.nlm.nih.gov/genbank/
)], mouse FABP4
(fatty acid binding protein 4; NM_024406), mouse 18S RNA
(NR_003278), rat AOX
(acyl co-A oxidase; NM_017340), and rat 18S/28S RNA
(M11188). qPCR reactions (in duplicate) were performed using an Applied Biosystems 7300 Real-Time PCR System (Applied Biosystems, Carlsbad, CA) as follows: hot-start activation at 95°C for 2 min, 40 cycles of denaturation (95°C for 15 sec), and annealing/extension (55°C for 60 sec). Relative gene expression was determined using the comparative CT
method, using the threshold value for 18S RNA or 18S/28S RNA for normalization. The CT
value for untreated samples was used as the reference point.
Statistics. All statistical analyses were performed using using GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA); differences between treatment groups were considered statistically significant at p < 0.05. Data from clinical chemistry, quantitative values from histopathology and immunohistochemistry, and in vitro lipid accumulation and gene expression data were analyzed by one-way analysis of variance (ANOVA) followed by Dunnett’s post hoc test. Data from transduction experiments were analyzed by two-way ANOVA followed by the Bonferroni post hoc test. Normality or log-normality of dust concentration data were assessed using histograms and Shapiro–Wilks tests. Because dust data were neither log-normal nor normal, concentrations in different microenvironments are reported using medians and were compared using nonparametric Wilcoxon rank sum tests.