Survey of plastic products
Fifty reusable PC-replacement products made from seven types of resins (acrylic, COC, COP, PES, PETG, PS, Tritan™) were obtained from 2010–2013 by purchase at retail outlets. These products included many items for which BPA-free resins are commonly used such as baby bottles, reusable water bottles, food storage, packaging, cups, medical supplies, and labware. These items were branded by well-known firms such as Camelbak, Nalgene, Dr. Weil, Born Free, AVENT, Costar, Crate and Barrel, Green-to-Grow, and LocknLock. There was no consistent price differential for products made with different resins. We tested each product with various (not necessarily the same) combinations of extraction solvents and stresses using two different assays, MCF-7 and BG1Luc (see below). We often stopped testing a product if a particular type of extraction solvent or stress showed that the product leached chemicals having significant levels of EA. That is, the aim of this study was not to perform an exhaustively-complete study of responses to all possible stresses and extraction conditions for a few products, but rather to survey a larger sample of PC-replacement products and assess whether some released chemicals with EA whereas others were potentially EA-free.
MCF-7 and BG1Luc cells
MCF-7:WS8 (MCF-7) cells were a gift from Dr. V. Craig Jordan, then at Northwestern Medical School, now at Georgetown Medical School. Every 2–3 months the MCF-7 cells were replaced with stocks from the same MCF-7 primary source to maintain more uniform MCF-7 cell characteristics throughout the study. BG1Luc cells were licensed from the University of California, Davis. Media and medium supplements (RPMI (Roswell Park Memorial Institute)-1640 Medium, DMEM, FBS, nonessential amino acids, l-glutamine, penicillin, streptomycin) used to initially grow and maintain the MCF-7 and BG1Luc cells were purchased from Invitrogen (Grand Island, NY, USA). Insulin was purchased from Sigma (St. Louis, MO, USA).
MCF-7 and BG1Luc cells were maintained in polystyrene T-75 flasks (BD Falcon, BD Biosciences, San Jose, CA, cat#353136) and polystyrene T-25 flasks (CytoOne, USA Scientific, Ocala, FL, cat#CC7682-4825). MCF-7 cells were seeded into 96-well flat bottom PS polystyrene plates (BD Falcon, cat#353075) and BG1Luc cells were seeded into 96-well white wall/clear bottom plates (Greiner Bio-One, Monroe, NC, cat#655098). Ethanol was obtained as 100% from various sources (OmniPur, EMD-Millipore, Billerica, MA; Acros Organics/Fisher Scientific, Pittsburgh, PA or Sigma-Aldrich, St. Louis, MO). Water was distilled on-site in an all-glass system and collected directly into glass before use in extractions. Extractions were performed in borosilicate glass tubes.
We used Labconco Class II Biosafety Hoods (Kansas City, MO, USA) equipped with a 254 nm fluorescent bulb to enclose EpMotion 5070 robotic workstations (Eppendorf, Hamburg, Germany) for serial dilutions of test chemicals, cell seeding, and media changes in 96-well plates . A Tristar Luminometer (Brethold Technology, Germany) was used to measure luminescence in BG1Luc assays. A Bio-Tek PowerWavex and a Bio-Rad 96-well plate reader spectrophotometer were used to measure DNA content in MCF-7 assays, as previously described [3, 18].
Protocols to stress plastic products
There are no regulatory protocols for stressing plastic products to test for leaching of chemicals with hormonal activity. Hence, we devised microwave, autoclave, and UV stresses described below to simulate various aspects of the short- or long-term effects of such common-use stresses of microwaving, boiling (moist high heat) or germicidal or UV exposure experienced by various types of PC plastics such as food containers, water bottles, or baby bottles. Some of these stresses have been previously described .
For most microwave stresses, 4x4 mm square pieces of plastic were placed into glass beakers in a 1200 W microwave oven set on “high” for two minutes, and then allowed to rest for 30minutes. The cycle was repeated 10 times. Some samples were placed in EA-free polypropylene (PP) tubes, and then microwaved on “high” setting for three minutes with a resting time 30minutes between stresses. The cycle was repeated 5 times. We did not detect consistent differences in leaching after the two protocols.
For autoclave stresses, plastic products were enclosed in individually crimped packets of EA-free aluminum foil and placed in a Tuttnauer autoclave at 134°C for 8 minutes.
For UV stresses, the UV radiation in sunlight is often classified  as UVC (100–280 nm), UVB (280–315 nm), and UVA (315–400 nm); visible wavelengths are from 400–750 nm. Most UVC wavelengths are filtered by the ozone layer before reaching the earth’s surface. However, UVC wavelengths are used in some germicidal UV devices, e.g., to sterilize baby bottles. For UV stresses, the protocols were as follows:
(1) Long wavelength (315–400 nm) UVA stresses that simulate many aspects of UV in sunlight: Samples were placed in a Q-Lab QUV unit containing UVA-340 nm bulbs to simulate exposure to moisture-free sunlight between 295 nm and 365 nm for 80 hours at 45–50°C.
(2) Short wavelength (100–280 nm) UVC stresses that simulate many aspects of UV in germicidal sterilizers: Samples were placed on aluminum foil in a Labconco Biosafety hood about 24” from a germicidal fluorescent light (maximum intensity wavelength of 254 nm) for 24 hours.
Extracts of plastic products
Two to five grams of unstressed or stressed samples of PC-replacement products were added to sterile glass test tubes. The tubes were placed under a germicidal UV light for 30 minutes to sterilize the samples before adding an extraction solvent to produce a final concentration of 1.0 g/mL. Such brief UVC exposures do not alter leaching . The extraction solvents consisted of saline-based solution (saline: RPMI-1640 Medium without phenol red), 100% EtOH, 10-50% aqueous EtOH, or distilled water. Most samples were extracted at 40°C for 240 hours in an incubator shaker. Saline extracts were diluted 2× with 2× estrogen-free medium (EFM) in a 1:1 ratio and then further diluted 1-4× with EFM so that the highest starting product concentration applied to wells was 0.125-0.500 g product/mL. [EFM was modified from cell maintenance media (see below) by replacing 10% FBS with 1% charcoal-stripped FBS and 4% charcoal-stripped calf serum and phenol red-free RPMI-1640.] EtOH extracts were concentrated 10× by evaporation and then diluted 100× with EFM to produce a highest starting concentration of 0.1 g product/mL to be applied to wells.
An MCF-7 cell line was used in a robotized version of the MCF-7 cell proliferation assay [3, 18] that has been employed for decades in manual format to reliably assess EA [16, 22]. The assay is currently undergoing validation for international use by ICCVAM/NICEATM . Chemicals with EA bind to ERs and activate the transcription of estrogen-responsive genes, which leads to proliferation of MCF-7 cells.
As previously described in detail [3, 18], cell maintenance media was used to grow and maintain the MCF-7 cells. This media consisted of RPMI-1640 media with non-essential amino acids, 10 μg/mL phenol red, 4 mM l-glutamine, 6 ng/mL insulin, 100 units/mL penicillin, 100 μg/mL streptomycin, and 10% fetal bovine serum (FBS). EA assays were performed in EFM. Each test extract at each concentration was added in triplicate or quadruplicate to 96-well plates containing MCF-7 cells in EFM. After six days exposure to test chemicals or extracts, the cell culture medium was aspirated and the amount of DNA/well, an indication of cell numbers, was assayed using a microplate modification of the diphenylamine assay [3, 18].
The BG1Luc4E2 cell line (aka BG1Luc) responds to estrogenic chemicals with the induction of firefly luciferase . The BG1Luc assay has been approved as a screening method for estrogenic chemicals by OECD, EPA, and ICCVAM/NICEATM . BG1Luc cells were maintained in cell culture medium that consisted of phenol red-free DMEM with 8% FBS, 100 units/mL penicillin, and 100 μg/mL streptomycin, L-glutamine and sodium pyruvate. Prior to assaying for EA, BG1Luc cells were placed for 3 days in EFM that was modified from cell culture medium by replacing 8% FBS with 4.5% charcoal-stripped FBS and substituting phenol red-free DMEM for phenol red-free containing DMEM. Acclimated cells were then seeded at 10,000 cells per well in 100 μL EFM in 96-well plates for 24 hours, followed by a 24 ± 6 hours incubation with test extracts in triplicate.
Cytotoxicity was assessed as described by ICCVAM/NICEATM  and Yang et al. :
Cells were visually observed under an inverted light microscope immediately before terminating incubation. Cellular cytotoxicity was assessed using scoring parameters 1 = normal cell morphology, 2 = low cytotoxicity (10 - 50% of cells with altered morphology), 3 = moderate cytotoxicity (50- 90% of cells had altered morphology), and 4 = high cytotoxicity (few or no cells visible). Test substance concentrations with a cytotoxicity score of 2 or higher were excluded from further analyses.
Cell culture medium was aspirated, cells were lysed, and luciferase was measured in an automated microplate luminometer (Tristar Berthold Technology, Germany) with the Promega Luciferase Assay System (Promega, Madison, WI, USA) following the manufacturer’s protocol.
Calculation of EA
The estrogenic effect of a test chemical or extract on cell proliferation or luciferase activity was calculated as %RME2, a percentage of the maximum DNA/well (or the Relative Luminescence Unit, RLU) produced by the maximum response relative to 17β-estradiol (E2, positive control). This value was corrected for the background (DNA or RLU in MCF-7 or BG1Luc, respectively) response to the vehicle (negative) control. We incorporated both a vehicle control (VC) and “sham” control (SC) in each experiment. The VC was the vehicle used for that particular assay. The SC was the vehicle taken through all steps that were used to assay the test sample/test extract. For both SCs and test extracts, %RME2 was calculated by subtraction of the VC value from the SC or test sample value, followed by normalization of each adjusted value to the maximum E2 value measured in the experiment (set at 100%) and the VC (set at 0%). Typical value of an SC was 0% ± 10% RME2. However, if the EA of an SC were greater than 15% RME2, then the entire experiment was rejected.
The EA of a test chemical or extract was classified as detectable if the EA effect was greater than 15%RME2, which was greater than three standard deviations (SD) of the SC response for that experiment. Therefore, 15%RME2 is a conservative measure of EA detectability. Stimulation of MCF-7 proliferation or BG1Luc Luciferase expression by test chemicals or extracts was confirmed as estrogenic (rather than non-specific effects) by suppression of the EA by co-incubation with the anti-estrogen (ICI 182,780 (ICI) at 10-7 - 10-8 M.
We saw no examples of an unsuppressed agonist response using either assay. That is, these in vitro assays rarely produce false positive responses [3, 18, 24]. Note also that we have limited our analyses to whether the plastic or chemical exhibited EA that was statistically significantly (p < 0.01) greater than any EA detected in VC or SC samples. We have not attempted to statistically compare absolute %RME2 values for different extracts or plastic products because extraction procedures, while similar, were not exactly the same for the two cell lines and the chemicals, much less their concentrations, in extracts were not known.