Endocrine Disruption Screening by Protein and Gene Expression of

Jul 16, 2014 - conditions. Cryobiology 25, 323−330. (19) Chen, Z., Ding, Y., and Zhang, H. (2001) Cryopreservation of suckling pig hepatocytes. Ann...
1 downloads 0 Views 971KB Size
Download PDF
Article pubs.acs.org/crt

Endocrine Disruption Screening by Protein and Gene Expression of Vitellogenin in Freshly Isolated and Cryopreserved Rainbow Trout Hepatocytes Lauren K. Markell,† Robert T. Mingoia,† Heather M. Peterson, Jianhong Yao, Stephanie M. Waters, James P. Finn,‡ Diane L. Nabb, and Xing Han* DuPont Haskell Global Centers for Health and Environmental Sciences, Newark, Delaware 19714, United States S Supporting Information *

ABSTRACT: Xenobiotics may activate the estrogen receptor, resulting in alteration of normal endocrine functions in animals and humans. Consequently, this necessitates development of assay end points capable of identifying estrogenic xenobiotics. In the present study, we screened the potential estrogenicity of chemicals via their ability to induce vitellogenin (VTG) expression in cultured primary hepatocytes from male trout. A routine method for VTG detection measures the secretion of the protein by enzyme-linked immunosorbent assay (ELISA) in freshly isolated trout hepatocytes. However, this lengthy (6 days) culturing procedure requires that hepatocyte isolation is performed each time the assay is run. We optimized this methodology by investigating the utility of cryopreserved hepatocytes, shortening the incubation time, performing a quantitative real-time PCR (qPCR) method for VTG quantification, and verifying the model system with reference chemicals 17β-estradiol, estrone, diethylstilbestrol, hexestrol, genistein, and a negative control, corticosterone. To test the performance of both freshly isolated and cryopreserved hepatocytes, mRNA was collected from hepatocytes following 24 h treatment for VTG gene expression analysis, whereas cell culture media was collected for a VTG ELISA 96 h post-treatment. EC50 values were obtained for each reference chemical except for corticosterone, which exhibited no induction of VTG gene or protein level. Our results show linear concordance between ELISA and qPCR detection methods. Although there was approximately 50% reduction in VTG inducibility following cryopreservation, linear concordance of EC50 values was found between freshly isolated and cryopreserved hepatocytes, indicating that cryopreservation does not alter the functional assessment of estrogen receptor activation and therefore VTG expression. These studies demonstrate that qPCR is a sensitive and specific method for detecting VTG gene expression that can be used together with cryopreserved trout hepatocytes for screening estrogenic chemicals, resulting in a reduction of the time required to perform the assay and enabling greater access to the model system through the approach of cryopreservation.

■

understandably, it is not normally expressed in males.3,4 Therefore, VTG has been primarily used as a biomarker in male fish or in isolated hepatocytes from male fish5,6 for the detection of estrogenic xenobiotics. A well-established method for detection of VTG production is to measure the amount of VTG protein secreted into the cell culture media of freshly isolated rainbow trout hepatocytes.5,7−9 Because of its limited sensitivity, this method requires a lengthy incubation period (96 h) in order to obtain a reasonable dynamic range in VTG protein expression measured by enzyme-linked immunosorbent assay (ELISA). In addition, it is a disadvantage to isolate trout hepatocytes each time the assay is performed. In order to determine the correlation between VTG transcription and translation and to potentially improve upon this wellestablished method, we compared VTG gene (quantitative real-

INTRODUCTION Chemicals can cause endocrine disruption via interactions with various hormone systems (reproductive, thyroid, and the hypothalamus−pituitary−adrenal axis). Aberrations in these systems may alter development, growth, and reproduction, resulting in health effects in animals and humans. Therefore, methods for identifying the potential of endocrine disruption by chemicals have been continuously developed and improved. Vitellogenin (VTG) is a yolk glycolipoprotein that is normally secreted by the liver into the bloodstream during development of ovarian follicles. Found in oviparous animals including fish, reptiles, birds, and most insects, the production of VTG is initiated by endogenous 17β-estradiol binding to the estrogen receptor (ER). Activated by estrogens or estrogen-like agonists, the ER dimerizes and translocates into the nucleus, binding to estrogen response elements that are located upstream of the VTG gene.1 ER binding results in transcriptional activation2 and vitellogenin production. VTG is expressed in female fish, while © 2014 American Chemical Society

Received: May 23, 2014 Published: July 16, 2014 1450

dx.doi.org/10.1021/tx5002089 | Chem. Res. Toxicol. 2014, 27, 1450−1457

Chemical Research in Toxicology

Article

Figure 1. Experimental design for qPCR and ELISA analysis of cultured male trout hepatocytes treated with reference chemicals. For freshly isolated and cryopreserved hepatocytes, cells were seeded on day 1 and dosed on day 2 for qPCR and days 2 and 4 for ELISA. For the qPCR method, cells were collected and lysed for analysis on day 3, while the media for ELISA was collected on day 6. analysis. For ELISA, 48 h after initial dosing, 100 μL of media was replaced with another 100 μL of media containing the reference chemical at 1× target concentrations to allow for continued incubation for another 48 h. After a total of 96 h of treatment, 100 μL of supernatant was aliquoted onto an ELISA plate (catalog no. 442404, Thermo Scientific, Waltham, MA, USA) and stored at −80 °C until further processing. ELISA measurements were also collected 48 and 72 h after chemical incubation, although these results did not yield data sufficient for analysis. For each individual experiment, cells were treated with 12 doses in triplicate (qPCR) or quadruplicate (ELISA). Cytotoxicity Assay. Cytotoxicity of cells as a result of treatment by reference chemicals was assessed by measuring cell membrane integrity with Alamar blue staining using an adapted protocol.12 Cryopreserved hepatocytes were treated with the reference chemicals for either 24 or 96 h postdosing at selected concentrations equivalent to those for qPCR and ELISA experiments. For staining, 100 μL of cell culture media was replaced with an equal volume of Alamar blue cell viability stain (serumfree L-15 media, 1% penicillin−streptomycin, 20% Alamar blue reagent) to achieve a final concentration of 10% Alamar blue. Plates were incubated for 30 min at 19 °C. Fluorescence was measured using wavelength pairs 530/595 (excitation/emission). Viability was determined by generating a ratio of the fluorescence of solvent control cells exposed to DMSO and fluorescence at each individual reference chemical dose. VTG Protein Production Analysis. VTG protein production following treatment with reference chemicals was determined by ELISA. The following protocol was followed7,8 with alterations: plates containing cell culture VTG samples were removed from −80 °C and placed in a refrigerator to thaw prior to adding 100 μL of standards in triplicate consisting of purified VTG (BioSense Laboratories AS, Bergen, Norway). Plates were incubated in the dark at 4 °C for at least 18 h prior to performing ELISA and were then washed 3 times with trisbuffered saline (TBS) (25 mM Tris, 0.15 M NaCl, 0.05% Tween-20) and blocked with 1% w/v BSA in TBS for 1 h at 4 °C with gentle rocking. Plates were then washed 3 times with 1× TBS before adding 100 μL of primary antibody (monoclonal AB BN-5 mouse anti-salmon Vtg, Biosense Laboratories; 1:2000), which was followed by incubation for 2 h at 4 °C with rocking. After incubation, plates were washed 3 times in TBS, 200 μL of secondary antibody (EIA grade affinity purified goat anti-mouse IgG horseradish peroxidase conjugate, Bio-Rad, Hercules, CA, USA; 1:6000) was added, and plates were incubated at 4 °C for 1 h. After incubation with the secondary antibody, plates were washed 5 times in TBS before 100 μL of substrate (TMB, Thermo Scientific) was added. The plates were then capped and placed on a rotator for approximately 40 min. The enzyme reaction was stopped by adding 50 μL of 1 M H2SO4 to the wells. Absorbance values were read at 450 nm on a spectrophotometer (SpectraMax 384Plus, Molecular Devices, Sunnyvale, CA, USA). The absolute expression of VTG was determined on the basis of the standard curve. VTG mRNA analysis. RNA was isolated using the RNeasy 96 kit (Qiagen) under conditions optimized from the product protocol. Modified for use with a Sorvall Legend RT centrifuge and 96-well microplate carriers (catalog no. 50-869-250, Thermo Scientific), the spin speed was set at 4150g for 7 and 15 min, in place of 5600g for 4 and 10 min as described in the product protocol. In addition, RNase-free water (45−60 μL) was added with one elution step rather than duplicate elutions. Reverse transcription of 9 μL of the isolated RNA was carried out using the high-capacity RNA-to-cDNA Kit (Life Technologies,

time polymerase chain reaction; qPCR) and protein expression (ELISA) in freshly prepared and cryopreserved rainbow trout hepatocytes that were treated with a group of reference chemicals (17β-estradiol, estrone, diethylstilbestrol (DES), hexestrol, genistein, and corticosterone). These chemicals represent a range of potencies in estrogenicity.7,10 Our results demonstrate the ability to screen potential estrogens in cryopreserved hepatocytes, relying on qPCR for the sensitive detection of VTG gene expression.

■

MATERIALS AND METHODS

Caution: This study involves chemicals that are known endocrine disruptors and should be handled with care. Chemicals. 17β-estradiol, estrone, DES, genistein, hexestrol, and corticosterone (≥98% purity) were all purchased from Sigma-Aldrich (St. Louis, MO, USA). Cell culture media and serum (Dulbecco’s Modified Eagle Medium (DMEM), L-15 media, fetal bovine serum) were purchased from Gibco/Life Technologies (Carlsbad, CA, USA). CELLect bovine albumin, low fatty acid (catalog no. 0218057510) was purchased from MP Biomedicals (Santa Ana, CA, USA). Hepatocyte Isolation. Primary trout hepatocytes were isolated and cryopreserved according to a standard protocol from adult male rainbow trout (Oncorhynchus mykiss).11 Trout approximately one and a half years old and approximately 10−12 in. in length were purchased from Limestone Springs Fishing Preserve (Richland, PA). Trout hepatocytes were pooled from three fish after each isolation (batch). There were a total of three separate isolations. From each batch, a portion of the freshly isolated hepatocytes was assayed on the isolation day, and the remaining cells were cryopreserved, with recovery post-thaw at approximately 30%. Hepatocytes were cryopreserved for 1−8 weeks prior to use. Freshly isolated and cryopreserved hepatocytes from the same batch of pooled cells were compared. Cell Culture. For cryopreserved hepatocytes, cells were thawed in a room temperature water bath until only a small ice crystal remained, and they were then quickly added to a 15-fold volume of DMEM (containing 10% fetal bovine serum and 0.25% bovine serum albumin, BSA, pH7.8) at room temperature. The cells were washed twice with either ice-cold cell culture media (serum free L-15 with 0.29 mg/mL L-glutamine, 4.5 mM NaHCO3, 100 units/mL penicillin, 100 μg/L streptomycin, and 0.25 μg/mL amphotericin) for cryopreserved hepatocytes or with icecold L-15 alone for freshly isolated hepatocytes. The pH for both wash media was 7.8. For both fresh and cryopreserved preparations, trout hepatocytes were suspended in cell culture media and were counted in the presence of 0.04% trypan blue (ViCell XR, Beckman Coulter, Indianapolis, IN, USA) to evaluate cell viability, which was ≥85%. Freshly isolated and cryopreserved trout hepatocytes were plated in 96well Primaria plates (BD Biosciences, San Jose, CA, USA) at 0.5 × 106 cells/mL, 0.2 mL/well.7 Overview of Experimental Design. Figure 1 describes the study design. Following cell plating in 96-well tissue culture treated plates, hepatocytes were incubated at 15 °C for 1 day for cell attachment. Separate plates were prepared for ELISA and qPCR analysis. Starting on the second day, half the volume of media (100 μL) was replaced with dosing media containing the reference chemicals at 2 times the target concentrations. Twenty four hours after dosing, media was removed, and 150 μL of RLT buffer (Qiagen, Hilden, Germany) was added to each well before plates were stored in a −80 °C freezer for qPCR 1451

dx.doi.org/10.1021/tx5002089 | Chem. Res. Toxicol. 2014, 27, 1450−1457

Chemical Research in Toxicology

Article

Figure 2. VTG protein concentrations in both freshly isolated and cryopreserved trout hepatocytes as a function of chemical concentration. 17βEstradiol, estrone, and DES caused a dose-dependent increase in VTG expression, whereas nonmonotonic dose−responses were observed with hexestrol and genistein. Data is represented as VTG protein concentration based on a standard curve using commercially available VTG. VTG protein concentration was below the limit of detection when hepatocytes were treated with vehicle alone. Data points absent for corticosterone and other chemical doses indicate that VTG protein concentration was not detectable. Twelve doses per chemical were tested. Error bars were obtained from four replicate samples per treatment condition in batch 1. Carlsbad, CA USA) in a 20 μL reaction volume. qPCR was performed using SYBR Green PCR master mix (Life Technologies) with the following primers (300 nM final concentration): VTG forward 5′-TGC TGA AGC TTG TGA ACC CTG AGA-3′; VTG reverse 5′-TAC CCA CAA CAC CCT TGG CAT ACT-3′; RPL8 forward 5′-CAA AGC CCA CGT GAA GCA CAG AAA-3′; and RPL8 reverse 5′-ATG CCC TCT GCA GCA ATA AAC AGC-3′, which were designed using GeneBank Accession IDs AJ011691.1 and AY957563.1, respectively. RPL8 has previously been shown to be a suitable housekeeping gene when testing estrogenic compounds.13,14 For data analysis, relative VTG expression was normalized to RPL8, and fold-change was generated by comparing each treatment to control. Individual qPCR plates were normalized with internal standards of VTG/RPL8 expression that were included on each plate.

Data Analysis. EC50 values were generated from the dose−response curves using Origin Pro (version 9.0, OriginLabs Corporation, Northampton, MA, USA). For genistein and hexestrol, where a nonmonotonic dose−response was observed, EC50 values were estimated within a dose range up to the maximal response. Fold-change in EC50 values for each method was determined from ratios of cryopreserved/freshly isolated hepatocytes and ELISA/qPCR. Individual fold-change for each reference chemical was used to determine the overall average and standard deviation.

■

RESULTS VTG Protein Concentration as Determined by ELISA. Cultured trout hepatocytes were treated with reference chemicals 17β-estradiol, estrone, DES, hexestrol, genistein, and 1452

dx.doi.org/10.1021/tx5002089 | Chem. Res. Toxicol. 2014, 27, 1450−1457

Chemical Research in Toxicology

Article

Figure 3. VTG gene expression in both freshly isolated and cryopreserved trout hepatocytes as a function of chemical concentration. VTG gene expression was normalized to the housekeeping gene RPL8. Data is represented as fold-change vs solvent control (average expression of cells treated with vehicle alone from all six experiments). 17β-Estradiol, estrone, and DES caused a dose-dependent increase in VTG expression. A nonmonotonic dose−response was observed with genistein treatment, and reduction of VTG expression was observed only at the top dose for hexestrol. Twelve doses per chemical were tested. Error bars were obtained from triplicate samples per treatment condition in batch 1.

reduced the dynamic range of the measurement, which hindered accurate determination of EC50 values from the dose−response curves (data not shown). Cryopreservation resulted in an average reduction in VTG induction of 50.5 ± 6.2% across batches 1−3 when comparing the maximal response of cryopreserved to that of freshly isolated hepatocytes using the ELISA method. The Alamar blue staining method was used to determine cytotoxicity levels of the reference chemicals within the respective dose ranges that caused VTG production changes, and cytotoxicity levels were expressed as a percentage relative to the solvent control (Figure S2A, Supporting Information). Following 96 h treatment, no apparent cytotoxicity was observed for 17β-estradiol, estrone, DES, hexestrol, and corticosterone

corticosterone for a total of 96 h. After the treatment, concentrations of VTG that were secreted in culture media were determined by ELISA. Figure 2 shows VTG production in both freshly isolated and cryopreserved trout hepatocytes as a function of chemical concentrations. Only data from batch 1 are included in Figure 2 for clarity purposes. Data from all three separate batches can be found in the Supporting Information (Figure S1). All of the reference chemicals caused changes in VTG protein secretion in the cultures of both freshly isolated and cryopreserved trout hepatocytes except for the negative control corticosterone, where the protein concentration was below the limit of detection at the doses tested. In addition, reducing the chemical exposure time from 96 to 48 or 72 h significantly 1453

dx.doi.org/10.1021/tx5002089 | Chem. Res. Toxicol. 2014, 27, 1450−1457

Chemical Research in Toxicology

Article

within the range of doses tested. Cell viability for genistein at doses greater than 25 μM was reduced to less than 80%, which may contribute to the observed nonmonotonic dose−response (Figures 2 and S1, Supporting Information). VTG Gene Expression as Determined by qPCR. VTG gene expression in cultured trout hepatocytes after 24 h exposure to the reference chemicals was analyzed via qPCR. Figure 3 shows the fold increase of VTG gene expression as a function of chemical concentrations, relative to the expression level in hepatocytes dosed with the solvent control. Only data from batch 1 are included in Figure 3 for clarity. Data from all three separate batches can be found in the Supporting Information (Figure S3). Dose−responses, as reflected by VTG gene expression, were similar to those observed with ELISA analysis (Figure 2). In addition, results for corticosterone show no biologically relevant induction compared to DMSO-treated control hepatocytes with no dose−response observed up to 100 μM and a fold-change of less than 2. There was no change in cell viability for all dose levels, as measured by Alamar blue staining following 24 h treatment (Figure S2B, Supporting Information). The reduction in inducibility of VTG expression with cryopreserved hepatocytes was slightly less than that observed for ELISA, with an average reduction of 43.5 ± 10.4% for batches 1−3 (Figure S3, Supporting Information). Method Comparison: ELISA vs qPCR and Freshly Isolated vs Cryopreserved Hepatocytes. EC50 values from both ELISA (Table 1) and qPCR (Table 2) methods were

Table 2. EC50 Values from Batches 1−3 Following qPCR Analysisa freshly isolated hepatocytes

cryopreserved hepatocytes

batch

1

2

3

1

2

3

Estradiol Estrone DES Hexestrol Genistein Corticosteroneb

0.36 11.5 1.45 108 590

0.55 5.23 5.29 41.6 1202

1.27 21.4 26.6 609 2647

1.24 24.8 4.99 363 1856

5.24 28.9 46.4 184 2998

5.52 99.3 21.9 1843 4212

a

Freshly isolated and cryopreserved hepatocytes were treated with estrogenic reference chemicals and a negative control (corticosterone). EC50 values (nM) were generated using a nonlinear regression curve fit in Origin Pro (version 9.0). bNegative control. No dose response of VTG mRNA was observed.

from the three batches of cryopreserved and freshly isolated hepatocytes. Figure 4 shows a high correlation between ELISA

Table 1. EC50 Values from Batches 1−3 Following ELISA Analysisa freshly isolated hepatocytes

cryopreserved hepatocytes

batch

1

2

3

1

2

3

Estradiol Estrone DES Hexestrol Genistein Corticosteroneb

4.50 41.9 5.13 126 865

1.2 33.3 23.4 169 1963

10.2 84 95.1 1065 8924

18.5 124 33.6 802 8435

12.7 81.7 79.6

144 1184 244 1422 28 919

c

7367

a

Freshly isolated and cryopreserved hepatocytes were treated with estrogenic reference chemicals and a negative control (corticosterone). EC50 values (nM) were generated using a nonlinear regression curve fit in Origin Pro (version 9.0). bNegative control. VTG protein concentrations were below the limit of detection by the ELISA method. cHigh background with UV/vis measurements due to inadequate washing of ELISA plate inhibited EC50 value determination. Figure 4. Correlation plots for EC50 values of estrogenic chemicals comparing ELISA and qPCR methods for (A) freshly isolated and (B) cryopreserved trout hepatocytes. Data from all three batches were plotted. R2 values were generated from the average EC50 value of each reference chemical across the three batches of freshly isolated or cryopreserved hepatocytes using a linear fit in OriginPro (version 9.0).

generated for each of the three batches using freshly isolated and cryopreserved hepatocytes. For hexestrol and genistein, which exhibited nonmonotonic dose−responses, the EC50 value was generated by using the data points up to the maximal response. As expected, an EC50 value could not be generated for corticosterone because the VTG concentration was below the limit of detection (ELISA) or no dose−response was observed (qPCR). On the basis of the EC50 values, the compounds were ranked the same way (17β-estradiol > DES ≥ estrone > hexestrol > genistein > corticosterone) regardless of the analytical method (ELISA or qPCR) or type of cells (freshly isolated or cryopreserved). In order to demonstrate the correlation between methods, correlation curves were plotted using each individual EC50 value

and qPCR methods in both freshly isolated (Figure 4A) and cryopreserved (Figure 4B) hepatocytes, with R2 values of 0.991 and 0.935, respectively. Figure 5 shows a high correlation between the use of freshly isolated and cryopreserved hepatocytes in qPCR and ELISA methods, with R2 values of 0.982 and 0.904, respectively. 1454

dx.doi.org/10.1021/tx5002089 | Chem. Res. Toxicol. 2014, 27, 1450−1457

Chemical Research in Toxicology

Article

which the semiadherent hepatocytes must be dosed, with the potential for erroneous protein measurements due to inadvertent cell loss. In addition to exploring the utility of cryopreserved hepatocytes for detecting VTG induction, establishing an assay that provides ease of conduct through a shortened chemical incubation was a main driver for this research. Yet, efforts to optimize the ELISA assay with a shorter incubation time were unsuccessful, as the lack of dynamic range for VTG production made it difficult to discern dose-dependent changes (data not shown). Therefore, qPCR was investigated as an alternative method for detecting changes in VTG. When testing a xenobiotic in this in vitro assay, theoretically, there is a potential for a lack of concordance between VTG transcription and translation. This requires consideration of the differences between the two detection methods (qPCR vs ELISA). Quantitation of mRNA transcript by qPCR measures relative VTG gene expression by comparing the mRNA levels of treated hepatocytes to that of vehicle-treated controls. The qPCR method was streamlined with a 96-well plate based RNA extraction, allowing for multiple chemicals and dosing concentrations to be tested, which would otherwise be impossible with a phenol−chloroform extraction method. The lack of transcriptome maps for research with nontraditional species has certainly limited the scientific community from taking advantage of qPCR methods. The reference transcriptome for rainbow trout (Oncorhynchus mykiss) is incomplete,17 although the VTG gene sequence is mapped and searchable on open source databases (e.g., National Center for Biotechnology Information). In contrast, the ELISA measures absolute protein concentrations, quantifying VTG secreted into cell culture media using primary antibody binding, a secondary antibody conjugated to horseradish peroxidase (HRP), and a chromogenic reporter. Whereas species specificity is achieved with the qPCR method, an anti-salmon (Salmo salar) VTG antibody with crossreactivity to rainbow trout, Arctic char (Salvelinus alpinus), brown trout (Salmo trutta), and wrasse (Ctenolabrus rupestris) was suitable for these studies. In the present study, we demonstrated that the correlation plots and EC50 values show a high linear concordance between the different methodologies tested (Figures 4 and 5). The potency rank order for each chemical (17β-estradiol > DES ≥ estrone > hexestrol > genistein) is maintained when comparing both VTG gene expression and protein concentration and with both types of cells. The actual EC50 values were higher for those obtained by ELISA (relative to qPCR) or for those obtained in cryopreserved hepatocytes (relative to freshly isolated hepatocytes) (Tables 1, 2, S1, and S2, Supporting Information). The exact mechanism for these variances is unknown and was not explored in the present study. Trout hepatocytes maintain metabolic capacity in both fresh and cryopreserved preparations.11 Differences in collection time points (24 h, qPCR; 96 h, ELISA) may result in different levels of metabolism of these compounds and therefore metabolism may contribute to different EC50 values from the two methods (qPCR vs ELISA). An interesting finding is that EC50 values, when obtained in cryopreserved hepatocytes by the qPCR method, were actually quite comparable to those from freshly isolated hepatocytes by the ELISA method (Supporting Information, Table S3). More importantly, even though absolute EC50 values from our reference chemicals were dependent on the analytical method and cell type, the fact that the potencies of the chemicals in VTG induction were well-maintained with high linear concordance (Figures 4 and 5) demonstrated that the relative estrogenic

Figure 5. Correlation plots for EC50 values of estrogenic chemical comparing freshly isolated and cryopreserved trout hepatocyte preparations in (A) qPCR and (B) ELISA analysis methods. Data from all three batches were plotted. R2 values were generated from the average EC50 value of each reference chemical across the three batches of freshly isolated or cryopreserved hepatocytes using a linear fit in OriginPro (version 9.0).

Tables 1 and 2 also show apparent variances in EC50 values determined in the different methods and cell types. A general trend can be established in that EC50 values determined in cryopreserved hepatocytes were higher than those from freshly isolated hepatocytes regardless of the analytical method (ELISA and qPCR). Similarly, EC50 values obtained by ELISA were always higher than those obtained from the qPCR method, regardless of the cell type. In an attempt to quantify these differences, the ratios of the EC50 values for each comparison are included in the Supporting Information, and fold differences averaged between 4.1 and 7.1 (Tables S1 and S2). As a result of the above-described trends, it was observed that the apparent EC50 values obtained from two separate conditions, i.e., freshly isolated hepatocytes by ELISA method vs cryopreserved hepatocytes by qPCR method, were quite comparable (Tables 1 and 2). Their ratios were averaged at 1.4 (Supporting Information, Table S3).

■

DISCUSSION VTG detection by ELISA and qPCR using freshly isolated and cryopreserved trout hepatocytes was compared to determine the acceptability of each method for identifying estrogenic chemicals. Detecting both receptor agonist7,15 and antagonist ligands,5,6,16 the ELISA method using freshly isolated hepatocytes has been widely published,6,7,16 yielding acceptance within the literature. However, this is a lengthy method that includes multiple days in 1455

dx.doi.org/10.1021/tx5002089 | Chem. Res. Toxicol. 2014, 27, 1450−1457

Chemical Research in Toxicology

Article

selection of dosing concentrations was especially important for the weak estrogens hexestrol and genistein. Here, we found that these two chemicals both demonstrate dose−response curves with increasing VTG gene expression and protein secretion (up to 1 μM for hexestrol and 10 μM for genistein) but with a decreasing slope at the highest dose concentrations. The appearance of nonmonotonic dose responses has been reported previously.21 Naturally occurring hormones are typically found within the body at picomolar to nanomolar concentrations,21 and both hexestrol and genistein were tested at doses up to 1000 times higher than the endogenous VTG inducer estradiol. Cytotoxicity measurements following 96 h treatment do not rule out the possibility that a reduction in cell number could be contributing to a reduction in secreted VTG protein. It has been suggested that cytotoxicity has potentially resulted in false negative responses for weak estrogens (xenoestrogens) that require a high dosing concentration for a response to be noted.22,23 However, any variability in cell number is adjusted when normalizing VTG gene expression to the housekeeping gene and near 100% viability following 24 h treatment excludes cytotoxicity as a likely hypothesis. The implementation of cell viability analysis within the same experiment as VTG measurements12 will be pursued in the future in order to reduce the number of cells required for testing. In conclusion, in the present study we have successfully demonstrated that qPCR is a sensitive and specific method for detecting VTG gene expression in cryopreserved hepatocytes, with great utility toward screening for estrogenic xenobiotics.

potency of a compound is independent of the analytical methods and cell types applied in this study. By using the qPCR method, turn-around time and ease of assay conduct is improved for screening potential estrogenic chemicals. This method is also more sensitive at detecting changes in VTG expression, with data points generated for each dose concentration used, in contrast to ELISA data where many doses at low concentrations were below detection limits. The increased sensitivity of the qPCR method may also aid in testing where chemical solubility or hepatocyte cytotoxicity limits the maximum dosing concentration used. Although chemicals with a range of efficacies were tested, the sensitivity of the assay was not challenged with very weak estrogenic chemicals or chemicals with mechanisms known to result in false positive responses, limiting the current analysis of assay utility. It was also observed that EC50 values obtained by ELISA (Table 1) and qPCR (Table 2) were up to 18-fold higher for batch 3 compared to that for batches 1 or 2. Batch-to-batch variability in VTG inducibility with male trout hepatocytes has been noted in similar studies.9 Although all hepatocyte isolations for these experiments originated from the same shipment of trout, the isolation for batch 3 was performed 5 months after the trout were first acclimated in our facility. We postulate that the age of the fish at time of sacrifice and duration in fish tanks as well as interindividual differences could account for the variability in estrogen receptor responsiveness found in batch 3. To address the known interassay variability, others have normalized VTG levels relative to the maximum protein production obtained for a positive control chemical such as 17β-estradiol.6,7 However, normalization is not necessary for the purpose of this study, as we sought to compare the performance and concordance of ELISA and qPCR methods and hepatocytes from freshly isolated and cryopreserved preparations. Similar to changes in hepatocyte enzymatic activity,11 hepatocyte-specific functions may also be impacted following cryopreservation. Therefore, the difference in VTG induction between freshly isolated and cryopreserved trout hepatocytes was also considered. Indeed, regardless of cell density, others have found that the total protein secretion rate was 50−60% that of freshly isolated hepatocytes, as measured by tritiated leucine incorporation.18,19 Our studies show that a reduction in VTG induction following cryopreservation was not limited to protein synthesis. Cryopreservation resulted in a reduction in VTG inducibility of approximately 50% at both the gene and protein levels among the three separate experiments, demonstrating the importance of validating cryopreserved testing models. It will be important to ensure that future studies verify that very weak estrogenic chemicals may still be detected. Yet, similar to ELISA vs qPCR analysis, the correlation plots and R2 values comparing the EC50 values of freshly isolated and cryopreserved hepatocyte showed high concordance (Figure 5). Isolating approximately 100 million viable hepatocytes from a single perfused trout liver11 is an obvious benefit to cryopreservation, reducing the need to order, transport, and house fish.20 We have demonstrated that, although a large portion of hepatocytes are lost during the cryopreservation process, the recovered cells (with a yield of 25− 45%) still maintained nearly 100% viability.11 Producing far too many cells to be used immediately, cryopreservation may also improve the standardization of assay implementation. Although cryopreserved hepatocytes show diminished inducibility that could potentially mask the response of weak estrogens, the qPCR method has enhanced sensitivity for detecting small but significant changes in VTG expression. The

■

ASSOCIATED CONTENT

* Supporting Information S

Protein concentration and gene expression dose−response curves from the reference chemicals for batches 1−3 of freshly isolated and cryopreserved trout hepatocytes, cytotoxicity analysis following 24 and 96 h treatment, EC50 value foldchange determinations from the ratios of ELISA/qPCR, cryopreserved/freshly isolated hepatocytes and freshly isolated hepatocytes analyzed by ELISA/cryopreserved hepatocytes analyzed by qPCR. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +1 302 451-5808; Fax: +1 302 451-3568; E-mail: xing. [email protected]. Present Address ‡

(J.P.F.) Teva Pharmaceuticals, 145 Brandywine Parkway, West Chester, Pennsylvania 19380, United States. Author Contributions †

These authors contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ABBREVIATIONS BSA, bovine serum albumin; DES, diethylstilbestrol; DMEM, Dulbecco’s modified Eagle’s media; ER, estrogen receptor; ELISA, enzyme-linked immunosorbent assay; HRP, horseradish peroxidase; qPCR, quantitative real-time polymerase chain reaction; RPL8, ribosomal protein L8; TBS, tris-buffered saline; VTG, vitellogenin 1456

dx.doi.org/10.1021/tx5002089 | Chem. Res. Toxicol. 2014, 27, 1450−1457

Chemical Research in Toxicology

■

Article

(18) Chesne, C., and Guillouzo, A. (1988) Cryopreservation of isolated rat hepatocytes: a critical evaluation of freezing and thawing conditions. Cryobiology 25, 323−330. (19) Chen, Z., Ding, Y., and Zhang, H. (2001) Cryopreservation of suckling pig hepatocytes. Ann. Clin. Lab. Sci. 31, 391−398. (20) Ferraris, M., Radice, S., and Dosanjh, M. K. (1998) Cryopreservation of isolated trout hepatocytes: viability and function in primary culture. CryoLetters 19, 55−64. (21) Vandenberg, L. N., Colborn, T., Hayes, T. B., Heindel, J. J., Jacobs, D. R., Jr., Lee, D. H., Shioda, T., Soto, A. M., vom Saal, F. S., Welshons, W. V., Zoeller, R. T., and Myers, J. P. (2012) Hormones and endocrinedisrupting chemicals: low-dose effects and nonmonotonic dose responses. Endocr. Rev. 33, 378−455. (22) Islinger, M., Pawlowski, S., Hollert, H., Volkl, A., and Braunbeck, T. (1999) Measurement of vitellogenin-mRNA expression in primary cultures of rainbow trout hepatocytes in a non-radioactive dot blot/ RNAse protection-assay. Sci. Total Environ. 233, 109−122. (23) Toomey, B. H., Monteverdi, G. H., and Di Giulio, R. T. (1999) Octylphenol induces vitellogenin production and cell death in fish hepatocytes. Environ. Toxicol. Chem. 18, 734−739.

REFERENCES

(1) Ing, N. H., and O’Malley, B. W. (1995) The steroid hormone receptor superfamily: molecular mechanisms of action. In Molecular Endocrinology: Basic Concepts and Clinical Correlations (Weintraub, B. D., Ed.) pp 195−215, Raven Press, New York. (2) Martinez, E., Givel, F., and Wahli, W. (1987) The estrogenresponsive element as an inducible enhancer: DNA sequence requirements and conversion to a glucocorticoid-responsive element. EMBO J. 6, 3719−3727. (3) Tyler, C. R., Eerden, B., Jobling, S., Panter, G., and Sumpter, J. P. (1996) Measurement of vitellogenin, a biomarker for exposure to oestrogenic chemicals, in a wide variety of cyprinid fish. J. Comp. Physiol., B 166, 418−426. (4) Tyler, C. R., Jobling, S., and Sumpter, J. P. (1998) Endocrine disruption in wildlife: a critical review of the evidence. Crit. Rev. Toxicol. 28, 319−361. (5) Petersen, K., and Tollefsen, K. E. (2012) Combined effects of oestrogen receptor antagonists on in vitro vitellogenesis. Aquat. Toxicol. 112−113, 46−53. (6) Tollefsen, K. E., Eikvar, S., Finne, E. F., Fogelberg, O., and Gregersen, I. K. (2008) Estrogenicity of alkylphenols and alkylated nonphenolics in a rainbow trout (Oncorhynchus mykiss) primary hepatocyte culture. Ecotoxicol. Environ. Saf. 71, 370−383. (7) Petersen, K., and Tollefsen, K. E. (2011) Assessing combined toxicity of estrogen receptor agonists in a primary culture of rainbow trout (Oncorhynchus mykiss) hepatocytes. Aquat. Toxicol. 101, 186−195. (8) Tollefsen, K. E., Mathisen, R., and Stenersen, J. (2003) Induction of vitellogenin synthesis in an Atlantic salmon (Salmo salar) hepatocyte culture: a sensitive in vitro bioassay for the oestrogenic and antioestrogenic activity of chemicals. Biomarkers 8, 394−407. (9) Navas, J. M., and Segner, H. (2006) Vitellogenin synthesis in primary cultures of fish liver cells as endpoint for in vitro screening of the (anti)estrogenic activity of chemical substances. Aquat. Toxicol. 80, 1− 22. (10) Scholz, S., Renner, P., Belanger, S. E., Busquet, F., Davi, R., Demeneix, B. A., Denny, J. S., Leonard, M., McMaster, M. E., Villeneuve, D. L., and Embry, M. R. (2013) Alternatives to in vivo tests to detect endocrine disrupting chemicals (EDCs) in fish and amphibians screening for estrogen, androgen and thyroid hormone disruption. Crit. Rev. Toxicol. 43, 45−72. (11) Mingoia, R. T., Glover, K. P., Nabb, D. L., Yang, C. H., Snajdr, S. I., and Han, X. (2010) Cryopreserved hepatocytes from rainbow trout (Oncorhynchus mykiss): a validation study to support their application in bioaccumulation assessment. Environ. Sci. Technol. 44, 3052−3058. (12) Schreer, A., Tinson, C., Sherry, J. P., and Schirmer, K. (2005) Application of Alamar blue/5-carboxyfluorescein diacetate acetoxymethyl ester as a noninvasive cell viability assay in primary hepatocytes from rainbow trout. Anal. Biochem. 344, 76−85. (13) Filby, A. L., and Tyler, C. R. (2007) Appropriate ‘housekeeping’ genes for use in expression profiling the effects of environmental estrogens in fish. BMC Mol. Biol. 8, 10. (14) Lange, A., Katsu, Y., Miyagawa, S., Ogino, Y., Urushitani, H., Kobayashi, T., Hirai, T., Shears, J. A., Nagae, M., Yamamoto, J., Ohnishi, Y., Oka, T., Tatarazako, N., Ohta, Y., Tyler, C. R., and Iguchi, T. (2012) Comparative responsiveness to natural and synthetic estrogens of fish species commonly used in the laboratory and field monitoring. Aquat. Toxicol. 109, 250−258. (15) Olsen, C. M., Meussen-Elholm, E. T., Hongslo, J. K., Stenersen, J., and Tollefsen, K. E. (2005) Estrogenic effects of environmental chemicals: an interspecies comparison. Comp. Biochem. Physiol., Part C: Toxicol. Pharmacol. 141, 267−274. (16) Navas, J. M., and Segner, H. (2000) Antiestrogenicity of betanaphthoflavone and PAHs in cultured rainbow trout hepatocytes: evidence for a role of the arylhydrocarbon receptor. Aquat. Toxicol. 51, 79−92. (17) Sanchez, C. C., Weber, G. M., Gao, G., Cleveland, B. M., Yao, J., and Rexroad, C. E., 3rd (2011) Generation of a reference transcriptome for evaluating rainbow trout responses to various stressors. BMC Genomics 12, 626. 1457

dx.doi.org/10.1021/tx5002089 | Chem. Res. Toxicol. 2014, 27, 1450−1457