The Swiss Centre for Applied Ecotoxicology (Ecotox Centre) is the Swiss competence center for applied, practice-oriented ecotoxicology. It serves as the primary knowledge hub and discussion platform for research and development, consulting and education in that area (https://www.ecotoxcentre.ch). The Ecotox Centre has expertise in developing bioassay screening tools for environmental chemical mixtures and product leachates, and therefore is interested in new ways of analyzing complex samples. The current project, supported by the Swiss Food Safety and Veterinary Office and in collaboration with the Zurich University of Applied Sciences (Dr. Andreas Schönborn), is evaluating and applying HPTLC-based methods for the biological detection and subsequent identification of chemicals in food packaging and drinking water.
Introduction
Chemicals migrating into food from packaging materials can include unknown substances. To detect potentially toxic chemicals, we employ bioassays of effects including endocrine disruption. Bioassays such as the yeast estrogen screen (YES) are typically performed in 96-well microtiter plates (ISO standard [1]). Yeast cells modified for the YES produce ß-galactosidase upon induction of the embedded human estrogen receptor. This induction is ultimately quantified by monitoring ß-galactosidase-catalyzed production of colored or fluorescent substances.
In contrast to testing whole mixtures in microtiter plates, separation on HPTLC plates with subsequent bioassay detection has the potential to reveal multiple toxic chemicals, and help identify responsible entities. The planar-YES (P-YES) has been used to screen various samples on HPTLC plates [2–4], but there has been little to no comparison to the official method. Herein, the P-YES and the microtiter YES were compared in their sensitivity to screen for endocrine active compounds related to food packaging materials.
Standard solutions
Standards are prepared in ethanol at concentrations based on range-finding tests in both the P-YES and L-YES. Nine concentrations of each test chemical are prepared in 2-fold dilution series for application onto HPTLC plates.
Sample preparation
Chemical migration from coated metal cans is simulated according to European Commission guidelines for fatty foods [5]. Briefly, cans are loaded with 95% ethanol, sealed, and incubated at 60 °C for 10 days. Duplicate samples of three different can types are prepared simultaneously. Ninety-five percent ethanol in duplicate glass beakers mimick sample handling as negative controls. Migrates are concentrated under nitrogen to be tested in the bioassays at up to 2.4 mL migrate equivalents.
Chromatogram layer
HPTLC plate silica gel 60 (Merck), 20 x 10 cm, pre-washed with methanol
Sample application
Automatic TLC Sampler (ATS 4), application as bands, 15 tracks, band length 6.0 mm, distance from left edge 20.0 mm, at least 12.0 mm track distance, for tests without development, application in three rows at 15.0, 42.5, and 70.0 mm from the bottom of the plate, for tests with development, application at 8.0 mm from the lower edge, application volume of 5.0 μL for estradiol E2 standard solution, 20.0 μL for all other standard solutions and 40.0 μL of migrate samples, application of full concentration series (nine concentrations) of E2 and two test chemicals to each plate (n = 3)
Chromatography
AMD 2 development (isocratic) up to 80 mm with chloroform – acetone – petroleum ether 11:4:5 [4], followed by drying for 5 min with vacuum
Editor’s note: Development in the ADC 2 is also possible for this application.
P-YES
The bioassay is performed by spraying 2 mL yeast cells (McDonnell) [4] at 1000 ± 200 formazine attenuation units onto HPTLC plates (Derivatizer; red nozzle, spraying level 6) followed by incubation at 30°C for 3 hours. Then, plates are sprayed with 2 mL 0.5 mg/mL 4-methylumbelliferyl-ß-D-galactopyranoside (MUG) in buffer (Derivatizer; blue nozzle, spraying level 6), and incubated for 20 min at 37 °C.
Documentation
TLC Visualizer under white light and UV 366 nm
Microtiter bioassay (L-YES)
The L-YES (YES test assisted by enzymatic digestion with lyticase) is performed according to ISO standard [1]. Briefly, yeast cells are exposed to standard chemicals and samples in 96-well microtiter plates for 18 hrs. Detection occurred by monitoring cleavage of chlorophenol-red-ß-galactopyranoside.
Data analysis
Bioassay responses (peak height of fluorescence measurements for P-YES or absorbance at 540 nm for L-YES) are normalized to the maximum response of E2 and modelled in a four-parameter log logistic function. Median and 10 percent effective doses (ED50, ED10) are predicted with 95% confidence intervals.
Results and discussion
Our goal was to compare the P-YES with the standardized L-YES as a screening method for estrogenic effects of substances from food packaging. Towards that aim, we determined effect concentrations for 20 chemicals relevant to plastic packaging and screened example migrates of food packaging with both assays. Our interest was in the technical difference of the bioassays in microtiter plates on HPTLC plates. Therefore, we determined effective doses thoroughly without performing chromatography. This allowed us to eliminate chromatography as a factor in any differences we would see between the assays and increase the sample capacity on the plates. Separately, we also examined the effect of chromatography on model chemicals E2 and bisphenol A (BPA).
Dose responses of E2 and BPA are shown for the P-YES (without development) and L-YES. The doses producing 10% of the maximal E2 response (ED10) interpolated from the dose response curves were lower for the P-YES than the L-YES. This demonstrates greater sensitivity of the P-YES. We also examined assay parameters such as strain of yeast and indicator solution [6]. They were not able to explain the difference in sensitivity.
Thirteen of 20 chemicals were active at levels reaching at least 10% of the maximum E2 response. The ED10 of most of these chemicals were lower in the P-YES than the L-YES. This demonstrates that the P-YES is generally more sensitive than the L-YES, although that trend started to break down as potency of the chemicals decreases (higher EDs). The potency of each chemical relative to E2 is within a factor of two of its corresponding relative potency in the L-YES.
The P-YES is stable over time. We were able to obtain the same ED50 for E2 after one year.We also evaluated the effect of chromatographic development as shown with an asterisk. With tight 95% confidence intervals, the ED50 is different than without development on the same day. However, both with and without development were within historical variability of the P-YES without development. The historical range is consistently lower than for the L-YES (median L-YES ED50 > 1 x 10-14).
An example migrate of a metal can tested in the P-YES is shown below. This migrate had no estrogenic effect in the L-YES but did result in reduced cell growth. An estrogenic band of (an) unknown substance(s) in the migrate was revealed by testing with the P-YES. Spiked chemicals were also detectable in the migrate. BPA was detected at a concentration of 27 ng (1.2 x 10-10 mol), which is lower than would be required to detect BPA at its specific migration limit (2 μg, 8 x 10-9 mol/band) given the sample preparation used in this study.
This study showed that
- P-YES is more sensitive than L-YES
- P-YES results can be repeated up to one year later
- P-YES with chromatographic development can reveal substances hidden in L-YES
- P-YES and L-YES were sensitive enough to detect BPA within the European Commission specific migration limit [5]
Dose response curves of E2 and BPA: P-YES (without development) measured in fluorescence at 366 nm illumination and L-YES measured in absorbance at 540 nm; shaded regions exhibit the modelled 95% confidence intervals; horizontal dotted lines show the 10, 50, and 100% effect levels; vertical dotted lines indicate the corresponding ED10 and ED50; replicate is a band in P-YES, and a microtiter well in L-YES. Reproduced from [6] (https://creativecommons.org/licenses/by/4.0/legalcode)
The ED10 of chemicals that activated the estrogen receptor. Error bars (95% confidence interval of predicted value, dose response modeling based on triplicate plates) are often obscured by the data markers; replicate is a band in P-YES, and a microtiter well in L-YES. 2,2’-dihydroxy-4-methoxybenzophenone and benzylbutylphthalate produced a fluorescence response in P-YES but did not reach a 10% effect level. Reproduced with modifications from [6] (https://creativecommons.org/licenses/by/4.0/legalcode)
Repeatability of the median effective dose (ED50) of the reference chemical E2, over time in P-YES without development. Experiment days are shown that were performed within one month of each other, and up to a year later. Asterisk (*) indicates test with development (n=3). Reproduced from [6] (https://creativecommons.org/licenses/by/4.0/legalcode)
Detection of estrogen-active compounds in a migrate of a food contact material. Retention factor (RF) shown on left axis. Application position and the solvent front indicated by lower and upper dashed lines, respectively. Track 1: mixture of E2 (1 pg), 17∝-ethinylestradiol (1 pg), and estrone (10 pg) with increasing RF values, track 2: migrate of metal can, track 3: spiked migrate of metal can, track 4: spiked control migrate; samples on track 3 and 4 spiked with BPA (27 ng), benzophenone- 3 (140 ng), and estrone (0.2 ng); Zones marked with (a) native fluorescence of chemicals (i.e. not estrogenicity), (b) native estrogenicity, (c) co-eluting spiked chemicals estrone and benzophenone-3, (d) BPA.
[1] ISO. 19040-1. Water quality -- Determination of the estrogenic potential of water and waste water -- Part 1: Yeast estrogen screen (Saccharomyces cerevisiae), Geneva, Switzerland (2018)
[2] I. Klingelhofer, G.E. Morlock. Anal Chem (2015) 87(21):11098–104
[3] S. Buchinger et al. Anal Chem (2013) 85(15):7248-56
[4] A. Schoenborn et al. J Chromatogr A (2017) 1530:185-91
[5] European Commission. Commission Regulation (EU) No 10/2011 on plastic materials and articles intended to come into contact with food, Official Journal of the European Union (2011)
[6] A. J. Bergmann et al. Anal Bioanal Chem (2020) 412: 4527–4536
Further information is available online and on request from the authors.
Contact: Dr. Alan Bergmann, Swiss Centre for Applied Ecotoxicology, Eawag, Überlandstrasse 133, 8600 Dübendorf, Switzerland, alanjames.bergmann@oekotoxzentrum.ch