Widespread occurrence of estrogenic UV-filters in aquatic ecosystems in Switzerland

Abstract

We performed a trace analytical study covering nine hormonally active UV-filters by LC–MS/MS and GC–MS in river water and biota. Water was analysed at 10 sites above and below wastewater treatment plants in the river Glatt using polar organic chemical integrative samplers (POCIS). Four UV-filters occurred in the following order of decreasing concentrations; benzophenone-4 (BP-4) > benzophenone-3 (BP-3) > 3-(4-methyl)benzylidene-camphor (4-MBC) > 2-ethyl-hexyl-4-trimethoxycinnamate (EHMC). BP-4 ranged from 0.27 to 24.0 μg/POCIS, BP-3, 4-MBC and EHMC up to 0.1 μg/POCIS. Wastewater was the most important source. Levels decreased with higher river water flow. No significant in-stream removal occurred. BP-3, 4-MBC and EHMC were between 6 and 68 ng/L in river water. EHMC was accumulated in biota. In all 48 macroinvertebrate and fish samples from six rivers lipid-weighted EHMC occurred up to 337 ng/g, and up to 701 ng/g in 5 cormorants, suggesting food-chain accumulation. UV-filters are found to be ubiquitous in aquatic systems.

Introduction

Personal care products enter aquatic systems either directly or via effluents of wastewater treatment plants (WWTPs). UV-absorbing chemicals (UV-filters) are added to sunscreens and a wide variety of cosmetics (skin and hair care products, lotions, creams, fragrances). Currently, 27 UV-filters are listed in the Cosmetics Directive, and a further 43 chemicals are listed as UV-filters in ingredients used in cosmetics in the EU (Wahie et al., 2007). In sunscreens the concentration of a specific UV-filter varies between 0.5 and 10%, but may reach 25% (Hauri et al., 2003). The added amount of UV-filters is increasing, because higher sunlight protection factors are used. Mainly two groups are applied, often in combination; UV-absorbing organic chemicals, and inorganic ZnO and TiO2 (nano)particles that scatter and reflect light. In addition, many organic chemicals are used as UV-absorbers in textiles, household products, fabrics, plastics, optical products, agricultural chemicals and many other materials by application of 0.05 up to 2% on or into the product to protect against UV-irradiation.

UV-filters enter the aquatic environment either directly via wash-off from skin and cloth during recreational activities, or indirectly via wastewater or swimming pool waters, where high levels of benzophenone-3 (BP-3), 4-methylbenzylidene camphor (4-MBC), and 2-ethyl-hexyl-4-trimethoxycinnamate (EHMC) of up to 40 μg/L have been detected (Zwiener et al., 2007). Widespread use of BP-3 in personal care products was documented in a survey in the U.S. population (Calafat et al., 2008). Several UV-filters including EHMC, BP-3, 4-MBC and BP-4 were found in untreated municipal wastewater in different countries at levels between 1.5 and 19 μg/L (Balmer et al., 2005, Li et al., 2007, Rodil et al., 2008). Lower levels of these UV-filters including octocrylene (OC) in the range of 0.01–2.7 μg/L occurred in treated wastewater (Balmer et al., 2005, Rodil et al., 2008), indicating some removal. UV-filters also occur in digested sewage sludge (Plagellat et al., 2006), sediments (Schlenk et al., 2005, Cuderman and Health, 2007), landfill leachates (Giokas et al., 2004), and they are released from coatings from building parts (Plagellat et al., 2006).

Several UV-filters have been detected in lakes at concentrations of up to 345 ng/L (Poiger et al., 2004, Cuderman and Health, 2007, Jeon et al., 2006), and up to 849 ng/L of benzophenone-4 (BP-4) in rivers (Rodil et al., 2008). High concentrations of up to 799 ng/L of 4-MBC have been detected in coastal areas at beaches (Langford and Thomas, 2008). Surprisingly, even at very remote environments such as the Pacific Ocean (Polynesia) EHMC, 4-MBC, BP-3 and 3-benzylidene-camphor (3-BC) occurred in the surface microlayer (Goksoyr et al., 2009).

Fish are important organisms to monitor the occurrence of persistent lipophilic contaminants. The presence of lipophilic UV-filters (4-MBC, EHMC, octocrylene (OC), BP-3, homosalate (HMS)) was reported in fish with lipid-weight concentrations of up to 3100 ng/g in Germany (Nagtegaal et al., 1997). Recently up to 2400 ng/g (OC) were detected in fish from rivers (Buser et al., 2006), and lower levels in lakes in Switzerland (Balmer et al., 2005).

Some UV-filters have hormonal activity in fish. Nine UV-filters displayed estrogenic and additional hormonal activity in vitro (Kunz and Fent, 2006a), and benzophenone-1 (BP-1), benzophenone-2 (BP-2), 3-benzylidene camphor (3-BC), and diethyl-4-amonobenzoate (Et-PABA) were estrogenic in fish (Kunz et al., 2006a). 3-Benzylidene camphor (Kunz et al., 2006b) and BP-2 (Weisbrod et al., 2007) showed adverse effects on fecundity and reproduction. We also found that mixtures of UV-filters showed synergistic interactions in vitro (Kunz and Fent, 2006b) and additive to antagonistic activity in vivo (Kunz and Fent, 2009).

In light of increasing concerns that some UV-filters may act as endocrine disruptors and because of the significant amounts of these chemicals being used today, there is a need for a better understanding of the occurrence of UV-filters. The current knowledge on the environmental occurrence and effects of UV-filters is reviewed by Fent et al. (2008). Currently, data on the occurrence of UV-filters in environmental and biological samples are still scarce. Previous studies focused primarily on lipophilic UV-filters, but there is a lack of data on polar UV-filters, and on compounds with proven hormonal activity. Recently, we developed a trace analytical method to determine in parallel nine UV-filters that display estrogenic activity, and spanning a large range of polarity using GC–MS/MS and LC–MS/MS techniques (Zenker et al., 2008). In the present study, we explored the origin of these contaminants by probing water and fish from seven representative Swiss rivers, all receiving inputs from wastewater treatment plants (WWTP). We analysed water and biota for seven UV-filters (BP-1, BP-2, BP-3, benzophenone-4 (BP-4), 3-BC, Et-PABA and 4,4′-dihydroxybenzophenone (4-DHB)) displaying estrogenic activity in vitro (Kunz and Fent, 2006a), as well as 4-MBC and EHMC, showing estrogenic activity in fish (Inui et al., 2003). The aim was to collect data on the current contamination situation in Switzerland, most notably on polar UV-filters (BP-2, BP-4, 4-DHB and Et-PABA). Environmental concentrations of polar UV-filters are largely unknown, although they are applied in materials protection and in cosmetics. Particular emphasis was placed on the analysis of biota at different trophic levels to evaluate the bioaccumulation in the food chain. The obtained data will serve as an important basis for the assessment of the environmental risks associated with these contaminants (Fent et al., 2008).