Perfluoroalkylated acids in the eggs of great tits (Parus major) near a fluorochemical plant in Flanders, Belgium.

Perfluoroalkyl acids (PFAAs) are highly persistent substances which have been detected in wildlife around the world, including birds. Although bird eggs have often been used to determine and monitor PFAAs levels in the marine environment, this has rarely been done in the terrestrial environment. In the present study we examined the concentrations and composition profile of 12 PFAAs (4 perfluoroalkyl sulfonic acids (PFSAs) and 8 perfluoroalkyl carboxylic acids (PFCAs) in the eggs of great tits (Parus major) collected at a fluorochemical plant and in three other areas, representing a gradient in distance from the pollution source (from 1 to 70 km), in Antwerp, Belgium. The PFSA concentrations measured at the site of the fluorochemical plant were among the highest ever reported in eggs with median concentrations of 10380 ng/g (extrapolated), 99.3 ng/g and 47.7 ng/g for PFOS, PFHxS and PFDS respectively. Furthermore, the median concentration of 19.8 ng/g for PFOA was also among the highest ever reported in bird eggs. Although these concentrations decreased sharply with distance from the fluorochemical plant, levels found in the adjacent sites were still high compared to what has been reported in literature. Moreover, based on what is known in literature, it is likely that these concentrations may cause toxicological effects. PFOS was the dominant contributor to the PFSA and PFAAs (63.4-97.6%) profile at each site, whereas for PFCAs this was PFOA at the plant site and the nearest locations (41.0-52.8%) but PFDoA (37.7%) at the farthest location. Although there is some evidence that PFAAs concentrations close to the plant site are decreasing in comparison with earlier measurements, which may be due to the phase out of PFOS, more research is necessary to understand the extent of the toxicological effects in the vicinity of this PFAAs hotspot.

Persistent organic pollutants in the Olifants River Basin, South Africa: Bioaccumulation and trophic transfer through a subtropical aquatic food web.

This study investigates the trophic transfer of persistent organic pollutants (POPs: PCBs, PBDEs, OCPs and PFASs) in the subtropical aquatic ecosystem of the Olifants River Basin (South Africa) by means of trophic magnification factors (TMFs). Relative trophic levels were determined by stable isotope analysis. POP levels in surface water, sediment and biota were low. Only ∑DDTs levels in fish muscle (1, indicating biomagnification of all detected POPs. Calculated TMFs for PCBs were comparable to TMF values reported from the tropical Congo River basin and lower than TMFs from temperate and arctic regions. For p,p'-DDT, a higher TMF value was observed for the subtropical Olifants River during the winter low flow season than for the tropical Congo river. TMFs of DDTs from the present study were unexpectedly higher than TMFs from temperate and arctic aquatic food webs. The fish species in the aquatic ecosystem of the Olifants River can be consumed with a low risk for POP contamination.

Occurrence of perfluorinated alkylated substances in cereals, salt, sweets and fruit items collected in four European countries.

In the context of a European project, 12 perfluoroalkyl acids (PFAAs) were determined in 14 food items collected in four European countries representing northern, southern, eastern and western Europe. This study presents the results of PFAAs measured in fruit, cereals, sweets and salt. Out of the 12 PFAAs, 10 PFAAs were detected in 67% of the samples. Overall, PFOA was the most abundant compound and the highest concentrations were found for PFOS but all were less than 1ngg(-1). When comparing the four countries, highest levels and detection frequencies were observed in Belgium (Western Europe), followed by the Czech Republic (Eastern Europe), Italy (Southern Europe) and finally Norway (Northern Europe). Comparison of profiles and levels is difficult due to variations in constitution of the food categories in the investigated countries and countries of origin of the food items. Dietary intake assessments for PFOS and PFOA show that the daily intake of PFAAs is far below the existing tolerable levels. However, they contribute to the total dietary intake and should therefore be included in future dietary exposure assessments.

Pathways and factors for food safety and food security at PFOS contaminated sites within a problem based learning approach.

Perfluorooctanesulfonic acid (PFOS) and related substances have been listed in Annex B of the Stockholm Convention. The implementation requires inventories of use, stockpiles, and environmental contamination including contaminated sites and measures for (risk) reduction and phase out. In most countries monitoring capacity is not available and therefore other approaches for assessment of contaminated sites are needed. Available informations about PFOS contamination in hot spot areas and its bio-accumulation in the food webs have been merged to build up a worst-case scenario We model PFOS transfer from 1 to 100ngL(-1) range in water to extensive and free-range food producing animals, also via the spread of contaminated sludges on agriculture soils. The modeling indicates that forages represented 78% of the exposure in ruminants, while soil accounted for >80% in outdoor poultry/eggs and pigs. From the carry-over rates derived from literature, in pork liver, egg, and feral fish computed concentration falls at 101, 28 and 2.7ngg(-1), respectively, under the 1ngL(-1) PFOS scenario. Assuming a major consumption of food produced from a contaminated area, advisories on egg and fish, supported by good agriculture/farming practices could abate 75% of the human food intake. Such advisories would allow people to become resilient in a PFOS contaminated area through an empowerment of the food choices, bringing the alimentary exposure toward the current Tolerable Daily Intake (TDI) of 150ngkg(-1)bodyweightd(-1) proposed by the European Food Safety Authority (EFSA).

Perfluoroalkyl acid contamination of follicular fluid and its consequence for in vitro oocyte developmental competence.

Perfluoroalkyl acids (PFAAs) have been shown to induce negative effects in laboratory animals and in vitro experiments. Also, PFAAs have been detected in human tissues and body fluids. The ovarian follicle constitutes a fragile micro-environment where interactions between hormones, growth factors, the oocyte and surrounding somatic cells are essential to generate a fully competent oocyte. In vitro experiments suggest that PFAAs can influence this balance, but very scarce in vivo data are available to confirm this assumption. In fact, the potential PFAA-presence in the follicular micro-environment is currently unknown. Therefore, we investigated if PFAAs are present in human follicular fluid and if their presence could be a risk factor for in vivo exposed developing oocytes. Furthermore, we compared the PFAA-distribution within serum and follicular fluid. PFAAs were analyzed by LC/MS in follicular fluid (n=38) and serum (n=20) samples from women undergoing assisted reproductive technologies (ARTs). Statistical models were used to investigate PFAA-distribution in both body fluids, to compare this behavior with the distribution of lipophilic organic pollutants and to explore the relationship between patient characteristics, ART-results and follicular fluid contamination. Perfluorooctane sulfonate (PFOS) was the PFAA found in the highest concentration in follicular fluid [7.5 (0.1-30.4) ng/mL] and serum [7.6 (2.8-12.5) ng/mL]. A new variable, Principal Component 1, representing the overall PFAA-contamination of the follicular fluid samples, was associated with a higher fertilization rate (p<0.05) and a higher proportion of top embryos relative to the amount of retrieved oocytes (p<0.05), after adjusting for age, estradiol-concentration, BMI, male subfertility and the presence of other organic pollutants as explanatory variables. To conclude, overall higher PFAA-contamination in the follicular micro-environment was associated with a higher chance of an oocyte to develop into a high quality embryo. Also, PFAAs have different distribution patterns between serum and follicular fluid compared to the lipophilic organic pollutants. Further research is of course crucial to confirm these new observations.

Characterisation of perfluorooctane sulfonate (PFOS) in a terrestrial ecosystem near a fluorochemical plant in Flanders, Belgium.

Bioaccumulation of perfluorooctane sulfonate (PFOS) in a restricted terrestrial food chain was investigated with the omnivorous wood mouse (Apodemus sylvaticus) on top of the studied food chain. The levels detected are very high compared with literature as a result of the presence of fluorochemical plant in the immediate vicinity of the study area. Soil, surface water, fruits of European elder and common blackberry, invertebrates, bank vole and wood mouse were collected at two sites, e.g. Blokkersdijk, adjacent to the fluorochemical plant, and Galgenweel, a reference site 2 km further away. In wood mouse, the highest PFOS concentrations were found in the liver followed by the pancreas, lungs and kidneys, with the spleen having the lowest levels. In the liver, the concentrations ranged from 787 to 22,355 ng/g ww at Blokkersdijk and these were significantly correlated with those detected in the kidneys (13.7-4,226 ng/g ww). If current results are compared to the findings of a previous study conducted in 2002 at the same sites, a significant decrease of PFOS in livers of wood mouse is observed. To the best of our knowledge, so far no studies reported levels of PFOS in terrestrial invertebrates under field conditions. At Blokkersdijk, PFOS was detected in all invertebrate species ranging from 28 to 9,000 ng/g. Soil and water were also contaminated with levels of respectively 68 ng/g and 22 ng/L. Biota-to-soil accumulation factors ranged from 0.11 to 68 for earthworms. Biomagnification factors (BMFs) of liver wood mouse/berries were as high as 302. BMFs for invertebrates were remarkably lower (up to 2).

Occurrence of perfluoroalkyl substances (PFASs) in various food items of animal origin collected in four European countries.

This study summarises the results of the levels of 21 perfluoroalkyl substances (PFASs) in 50 selected pooled samples representing 15 food commodities with the special focus on those of animal origin, as meat, seafood, fish, milk, dairy products and hen eggs, which are commonly consumed in various European markets, e.g. Czech, Italian, Belgian and Norwegian. A new, rapid sample preparation approach based on the QuEChERS extraction procedure was applied. Ultra-performance liquid chromatography (UHPLC) coupled to tandem mass spectrometry (MS/MS) employing electrospray ionisation (ESI) in negative mode was used for the quantification of target analytes. Method quantification limits (MQLs) were in the range of 1-10 ng kg(-1) (ng l(-1)) for fish, meat, hen eggs, cheese and milk, and in the range of 2.5-125 ng kg(-1) for butter. Only 16 of the group of 21 PFASs were found in at least one analysed sample. From 16 PFASs, perfluorooctane sulfonate (PFOS) was the most frequently detected analyte present in approximately 50% of samples (in the range of 0.98-2600 ng kg(-1)). PFCAs with C8-C14 carbon chain were presented in approximately 20% of samples. The concentration ranges of individual compounds in the respective groups of PFASs were: 2.33-76.3 ng kg(-1) for PFSAs (without PFOS), 4.99-961 ng kg(-1) for PFCAs, 10.6-95.4 ng kg(-1) for PFPAs, and 1.61-519 ng kg(-1) for FOSA. The contamination level in the analysed food commodities decreased in the following order: seafood > pig/bovine liver >> freshwater/marine fish > hen egg > meat >> butter. When comparing the total contamination and profiles of PFASs in food commodities that originated from various sampling countries, differences were identified, and the contents decreased as follows: Belgium >> Norway, Italy > Czech Republic.

Perfluorinated alkylated substances in vegetables collected in four European countries; occurrence and human exposure estimations.

The human diet is recognised as one possible major exposure route to the overall perfluorinated alkylated substances (PFAS) burden of the human population, resulting directly from contamination of dietary food items, as well as migration of PFAS from food packaging or cookware. Most European countries carry out national monitoring programs (food basket studies) to monitor contamination with pollutants. Usually, for PFASs, non-coordinated approaches are used in Europe, since food basket studies are mainly carried out by national authorities following national requirements and questions, making comparisons between different countries difficult. A harmonised sampling campaign collecting similar food items in a uniform procedure enabling direct comparison between different regions in Europe was designed. We selected four countries (Belgium, Czech Republic, Italy and Norway), representing the four regions of Europe: West, East, South and North. In spring 2010 and 2011, 20 different types of vegetables were sampled in Belgium, Czech Republic, Italy and Norway. Perfluorinated carboxylic acids (PFCAs) were the main group of detected PFASs, with perfluorinated octanoic acid (PFOA) as the most abundant PFCA (with exception of samples from Czech Republic), followed by perfluorinated hexanoic acid and perfluorinated nonanoic acid. Dietary intake estimates for PFOA show only low human exposure due to vegetable consumption for adults and children, mostly governed by high intake of potatoes.

Ecotoxicoproteomics in gills of the sentinel fish species, Cottus gobio, exposed to perfluorooctane sulfonate (PFOS).

The environmental persistence, bioaccumulative tendency and potential toxicity of perfluorooctane sulfonate (PFOS) have generated great concern. This study aimed at evaluating the toxicity of short-term PFOS exposure in gills of the European bullhead Cottus gobio, a candidate sentinel species, by monitoring the response of some enzymes (citrate synthase CS, cytochrome c oxidase CCO, and lactate dehydrogenase LDH), and by undertaking a proteomic analysis using 2D-DIGE. First, a 96-h exposure to 1mg PFOS/L significantly altered the activity of mitochondrial CS and CCO. Second, 2D-DIGE gels were used to compare gills from the control fish group with tissues from fish exposed for 96h to either 0.1 or 1mg PFOS/L. From the 27 protein spots displaying significant changes in abundance following PFOS exposure, a total of 20 different proteins were identified using nano LC-MS/MS and the Peptide and Protein Prophet of Scaffold software. The differentially expressed proteins that were identified are involved in the general stress response, ubiquitin-proteasome system, energy metabolism, and actin cytoskeleton, which provide clues on the cellular pathways and components mainly affected by PFOS. Moreover, our results showed that most proteins were differentially expressed at the low but not at the high PFOS concentration. This work provides insights into the biochemical and molecular events in PFOS-induced toxicity in gill tissue, and suggests that further studies on the identified proteins could provide crucial information to better understand the mechanisms of PFOS toxicity in fish.

Perfluorinated substances in human food and other sources of human exposure.

The widespread distribution and degradation of PFCs in the environment results in a very complex exposure pattern, which makes it difficult to define the relative contribution to human exposure from different exposure pathways. The present review is designed to provide an overview of the existing data on levels of PFCs measured in the human diet and in drinking water. Data on levels of PFCs in the human diet are rather scarce, but the level in the fish appear to be well documented. Among PFCs, PFOS and PFOA are the best studied compounds in fish from both experimental and monitoring studies. Recently, the number of publications that address other PFCs has increased, but the total number available is still limited. In general, we discovered that care should be exercised when using the reviewed data, because, in the majority of publications, quality control and/or details on analysis are, at least partly, lacking. It has been well documented that PFOA and PFOS have the potential to accumulate in fish and concentrations up to 7 and 170 ng/g wwt, respectively in edible fish species have been found. PFOS is the most crucial and prominent compound identified, followed by the PFOA. Also, in aquatic invertebrate such as shrimps, mussels, clams, and oysters, high PFOS levels have been reported (up to 387 ng/g wwt). However in most publications PFC level reported in molluscs were less than 1 ng/g wwt. Positive correlations were found between PFC body burden and self reported fish consumption. In recognition of the potential for human exposure to PFCs via fish consumption, the Minnesota Department of Health has recently issued fish consumption advisories for contaminated sections of the Mississippi River. It is interesting to note that 79% of the reviewed publications on PFCs in the whole fish homogenates exceed the that threshold. Moreover, five of the PFC concentration reported in muscles tissue exceeded the advisory level of 38 ng/g wwt. Even though several authors concluded that consumption of contaminated food and drinking water constitutes the major exposure pathway for humans, only a few reports on PFCs in composite food exist. Food can be contaminated in an indirect way, because PFCs are widely used in food-packaging coatings and cooking materials. On the other hand, PFCs can also enter food organisms via environmental routes such as inhalation or adsorption from air. In a few studies, composite samples, duplicate diet samples, or other food items were analyzed for several PFCs, PFOS and PFOA, PFHpA, PFHxA, and PFHxS were meAsured and displayed concentrations ranging from-detected up to 15 ng/g wwt. In one study, a very high PFOA concentration of 118 ng/g were reported, but overall, PFC levels are below 10 ng/g wwt. It is important to note that, among all studies reviewed, PFCs were found in a maximum of 50% of the analyzed samples and generally only in 10% or less of samples analyzed. In contrast to what is observed in fish and other food items PFOA levels in drinking water (ND - 50 ng/L) and other PFCs (1-3 ng/L). In one study, extremely high values (519 ng/L) were measured in drinking water of a contaminated area in the Ruhr region. In Spain, bottled water was analyzed and four PFCs (PFOA, PFNA,PFDA and PFHpA) were found at low levels (<1 ng/L). Because of higher levels found in drinking water at several locations, some provisional drinking water guideline values for PFOS and PFOA have already been established, e.g., in the UK, Bavaria, and Minnesota. Since PFCs are present both in food and drinking water, Tolerable Daily Intake values for PFOS and PFOA have also been proposed by several institutes in Europe and in the USA. The ingestion of dust through hand-to-mouth transfer from indoor house dust can also be a potential source of PFC exposure, especially for toddlers and children. In publications on PFCs in indoor dust, the mean PFOS and PFOA levels varied between 39 and 1,200 ng/g and between 11 and 220 ng/g, respectively. Overall, it is clear that there is still lack of PFC exposure data for food and beverages, which renders the assessment of the contribution of the diet to total human PFC exposure uncertain. It is, therefore, appropriate that several scientific projects have recently been launched that addresses the assessment of human exposure to PFCs and related compounds from dietary sources.

Brominated flame retardants and perfluorinated compounds in indoor dust from homes and offices in Flanders, Belgium.

The increasing time spent indoors combined with the abundant usage of diverse indoor chemicals led to concerns involving the impact of these compounds on human health. The current study focused on two groups of important indoor contaminants i.e. Brominated flame retardants (BFRs) and Perfluorinated compounds (PFCs). Concentrations of both compound classes have been measured in Flemish indoor dust samples from homes and offices. ΣPolybrominated diphenyl ethers (PBDEs) (BDE 47, 99, 100, 154, 153, 197, 196 and 203) and BDE 209 in homes ranged between 4-1214 ng g(-1)dw (median 35) and <5-5295 ng g(-1)dw (median 313), respectively. Hexabromocyclododecane (ΣHBCD) levels ranged from 5 to 4,2692 ng g(-1)dw (median 130), with α-HBCD being the major isomer (mean 59%). In addition, tetrabromobisphenol A (TBBPA) ranged between <3 and 419 ng g(-1)dw (median 12). For all BFRs, median levels in office dust were up to an order of magnitude higher than in home dust. ΣPFCs (sum of perfluorobutane sulfonate (PFBS), perfluorohexane sulfonate (PFHxS), perfluorooctane sulfonic acid (PFOS), perfluorobutanoic acid (PFBA), perfluorohexanoic acid (PFHxA), perfluorooctanoic acid (PFOA), perfluorononanoic acid (PFNA) and perfluorodecanoic acid (PFDA)) concentrations in homes ranged from 0.2 to 336 ng g(-1) (median 3.0 ng g(-1)). Levels in office dust were higher (p<0.01) than in house dust with ΣPFCs ranging between 2.2 and 647 ng g(-1) (median 10 ng g(-1)) and median (PFOA) and perfluorooctane sulfonate values of 2.9 and 2.2 ng g(-1), respectively. The congener pattern was dominated by PFOA, followed by PFOS. Calculated human exposure was below the reference dose values set by the US-EPA for BDE 209, HBCD and below the provisional tolerable daily intakes proposed by European Food Safety Authority for PFOS and PFOA.

Brominated flame retardants and perfluorinated chemicals, two groups of persistent contaminants in Belgian human blood and milk.

We assessed the exposure of the Flemish population to brominated flame retardants (BFRs) and perfluorinated compounds (PFCs) by analysis of pooled cord blood, adolescent and adult serum, and human milk. Levels of polybrominated diphenyl ethers (PBDEs) in blood (range 1.6-6.5 ng/g lipid weight, lw) and milk (range 2.0-6.4 ng/g lw) agreed with European data. Hexabromocyclododecane ranged between <2.1-5.7 ng/g lw in milk. Perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) dominated in blood and ranged between 1 and 171 ng/mL and <0.9-9.5 ng/mL, respectively. Total PFC levels in milk ranged between <0.5-29 ng/mL. A significant increase in PBDE concentrations was detected from newborns (median 2.1) to the adolescents and adults (medians 3.8 and 4.6 ng/g lw, respectively). An identical trend was observed for PFOS, but not for PFOA. We estimated that newborn exposure to BFRs and PFCs occurs predominantly post-natally, whereas placental transfer has a minor impact on the body burden.

Exposure of the Flemish population to brominated flame retardants: model and risk assessment.

Human exposure to brominated flame retardants (BFRs) varies widely throughout the world as it depends on country-related usage, production and legislation of these chemicals. US and UK exposure assessments show very diverse levels and patterns which in turn, are likely to differ from those in background exposed countries such as Belgium, where levels tend to be about an order of magnitude lower. The current study assessed human exposure to BFRs through the indoor and outdoor environment (e.g. dust, soil, and air) and food for all age groups in Flanders, Belgium. Most relevant food groups were identified based on a national food consumption survey and food items with Flemish origin were collected. Dust samples were collected using a standardized protocol in 43 homes and 10 offices throughout Flanders. Food, human milk and dust samples were analysed for their polybrominated diphenylethers (PBDE) and hexabromocyclodecane (HBCD) content using GC/MS and LC/MS-MS. An exposure model was developed including all analysed data, complemented with literature data. The model covered human exposure of infants, children and adults through human milk, food, dust/soil ingestion and air inhalation. Total human exposure was compared to the existing toxicological criteria and previous exposure estimates. In general, the exposure levels through human milk are consistent with those of a background exposed European population, whereas dust and food intake are at the low end of what has been reported in previous European intake assessments. Total average intake of SigmaHBCD and SigmaBDE(5) at 50th percentile (P50) levels by newborns equals 3.1 and 12.0ng/kg body weight (bw) day, respectively. This intake increases to 15.2 and 20.9ng/kgbwday for SigmaHBCD and SigmaBDE(5), for higher exposed newborns (95th percentile=P95 levels). Due to the limited database on health-based limit values for PBDEs and HBCD, it is difficult to assess the immediate health concern for any of the age groups, although the higher intake of newborns indicates the need for ongoing monitoring. For median exposed individuals, the average SigmaHBCD intake peaked at the age 3 to 6years with an intake of 6.59ng/kgbwday and declines to approximately 1ng/kgbwday at later age. SigmaBDE(5) intake exhibited a different profile compared to SigmaHBCD with maximal levels for newborns and a decline to approximately 0.7ng/kgbwday at adulthood.