Fluorinated alkyl substances and technical mixtures used in food paper-packaging exhibit endocrine-related activity in vitro.

Migration of chemicals from packaging materials to foods may lead to human exposure. Polyfluoroalkyl substances (PFAS) can be used in technical mixtures (TMs) for use in food packaging of paper and board, and PFAS have been detected in human serum and umbilical cord blood. The specific structures of the PFAS in TMs are often unknown, but polyfluorinated alkyl phosphate esters (PAPs) have been characterized in TMs, food packaging, and in food. PAPs can be metabolized into fluorotelomer alcohols (FTOHs) and perfluoroalkyl carboxylic acids (PFCAs). Some PFAS have endocrine activities, highlighting the need to investigate these effects. Herein, we studied the endocrine activity of less characterized PFAS, including short-chain PFCAs and FTOHs, PAPs, and TMs of unknown chemical composition. Long-chain PFCAs were also included. We applied seven assays covering effects on estrogen, glucocorticoid, androgen, and peroxisome proliferator-activated receptor (PPAR) activity, as well as steroidogenesis in vitro and ex vivo. In general, PAPs, FTOHs, TMs, and long-chain PFCAs showed estrogenic activity through receptor activation and/or increasing 17ß-estradiol levels. Furthermore, short- and long-chain PFCAs activated PPARα and PPARγ. Collectively, this means that (i) PAPs, FTOHs, and PFCAs exhibit endocrine activity through distinct and sometimes different mechanisms, (ii) two out of three tested TMs exhibited estrogenic activity, and (iii) short-chain FTOHs showed estrogenic activity and short-chain PFCAs generally activate both PPARα and PPARγ with similar potency and efficacy as long-chain PFCAs. In conclusion, several new and divergent toxicological targets were identified for different groups of PFAS.

Fluorochemicals used in food packaging inhibit male sex hormone synthesis.

Polyfluoroalkyl phosphate surfactants (PAPS) are widely used in food contact materials (FCMs) of paper and board and have recently been detected in 57% of investigated materials. Human exposure occurs as PAPS have been measured in blood; however knowledge is lacking on the toxicology of PAPS. The aim of this study was to elucidate the effects of six fluorochemicals on sex hormone synthesis and androgen receptor (AR) activation in vitro. Four PAPS and two metabolites, perfluorooctanoic acid (PFOA) and 8:2 fluorotelomer alcohol (8:2 FTOH) were tested. Hormone profiles, including eight steroid hormones, generally showed that 8:2 diPAPS, 8:2 monoPAPS and 8:2 FTOH led to decreases in androgens (testosterone, dehydroepiandrosterone, and androstenedione) in the H295R steroidogenesis assay. Decreases were observed for progesterone and 17-OH-progesterone as well. These observations indicated that a step prior to progestagen and androgen synthesis had been affected. Gene expression analysis of StAR, Bzrp, CYP11A, CYP17, CYP21 and CYP19 mRNA showed a decrease in Bzrp mRNA levels for 8:2 monoPAPS and 8:2 FTOH indicating interference with cholesterol transport to the inner mitochondria. Cortisol, estrone and 17ß-estradiol levels were in several cases increased with exposure. In accordance with these data CYP19 gene expression increased with 8:2 diPAPS, 8:2 monoPAPS and 8:2 FTOH exposures indicating that this is a contributing factor to the decreased androgen and the increased estrogen levels. Overall, these results demonstrate that fluorochemicals present in food packaging materials and their metabolites can affect steroidogenesis through decreased Bzrp and increased CYP19 gene expression leading to lower androgen and higher estrogen levels.

Primary aromatic amines (PAAs) in black nylon and other food-contact materials, 2004-2009.

Primary aromatic amines (PAAs) were analysed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) in migrates from 234 samples of food-contact materials, including black nylon (polyamide) kitchen utensils (n = 136), coloured plastics (28), and clear/printed multilayer film/laminates (41), from retailers, importers, and food producers. A further 29 utensils in use were obtained from colleagues. Very high PAA migration was found from black nylon kitchen utensils to the food simulant 3% acetic acid: the 'non-detectable' limit (20 microg aniline equivalents kg(-1) food) was exceeded by up to 2100 times. All the other materials were compliant. The majority of the non-compliant utensils came from China. The predominant PAAs were aniline and 4,4'-methylenedianiline (4,4'-MDA). The frequency of violations decreased from the year 2004 (55%) to the autumn of 2005 (13%), possibly due to increased demands for in-house documentation, but they remained almost constant from 2005 to 2009. The validity of the results was shown by recovery studies, participation in proficiency testing, and comparative testing of utensils by two laboratories. Migration modelling was used to compare how various compliance migration test conditions influenced the final test results. Long-term release of PAAs was fitted by diffusion modelling experiments and long-term release was also seen as expected from used utensils. Toxicologists consider these migration levels of the suspected carcinogenic PAAs as a problem of major concern.

Mathematical modelling of migration: a suitable tool for the enforcement authorities?

A few years ago, it became accepted that the plastics industry could use migration modelling for compliance testing. When a calculation confirms that the migration of a compound from a plastic material or article is below the specific migration limit, this is considered sufficient documentation for compliance with legislation. In the case of non-compliance, the result needs to be verified experimentally. The European Commission recommends that the enforcement authorities use migration modelling as well to avoid long and expensive analysis. The aim of the present work was to investigate the practical possibilities of implementing migration-modelling software as a tool in official food control and possibly in improving the own-check programmes of Danish plastic-converting plants. Food inspectors from nine regional food control centres initially attended a training course in the use of a commercial modelling software package and were supported further during the project period of about 1 year. They visited 40 producers of final plastic materials and articles, but mainly due to a lack of elementary knowledge of the detailed composition of the materials only 16 full migration calculations were performed. A major reason was a lack of information from those in the raw material supply chain who considered their products protected by commercial confidentiality. In general, the food inspectors were in favour of using migration modelling for future control visits.