Maria Sundström

A temporal trend study (1972-2008) of perfluorooctanesulfonate, perfluorohexanesulfonate, and perfluorooctanoate in pooled human milk samples from Stockholm, Sweden

The widespread presence of perfluorooctanesulfonate (PFOS), perfluorooctanoate (PFOA), and perfluorohexanesulfonate (PFHxS) in human general populations and their slow elimination profiles have led to renewed interest in understanding the potential human neonatal exposures of perfluoroalkyls (PFAs) from consumption of human milk. The objective of this study was to evaluate the concentrations of PFOS, PFHxS, and PFOA in pooled human milk samples obtained in Sweden between 1972 and 2008 (a period representing the most significant period of PFA production) and to see whether the time trend of these analytes parallels that indicated in human serum. Chemical analysis of PFOS, PFHxS, and PFOA was performed on pooled Swedish human milk samples from 1972 to 2008 after methodological refinements. The 20 samples which formed the 2007 pool were also analyzed individually to evaluate sample variations. Analyses were performed by HPLC-MS/MS. Due to the complexities of the human milk matrix and the requirement to accurately quantitate low pg/mL concentrations, meticulous attention must be paid to background contamination if accurate results are to be obtained. PFOS was the predominant analyte present in the pools and all three analytes showed statistically significant increasing trends from 1972 to 2000, with concentrations reaching a plateau in the 1990s. PFOA and PFOS showed statistically significant decreasing trends during 2001-2008. At the end of the study, in 2008, the measured concentrations of PFOS, PFHxS, and PFOA in pooled human milk were 75 pg/mL, 14 pg/mL, and 74 pg/mL, respectively. The temporal concentration trends of PFOS, PFHxS, and PFOA observed in human milk are parallel to those reported in the general population serum concentrations.

Comparative pharmacokinetics of perfluorohexanesulfonate (PFHxS) in rats, mice, and monkeys.

Perfluorohexanesulfonate (PFHxS) has been found in biological samples from wildlife and humans. The human geometric mean serum PFHxS elimination half-life has been estimated to be 2665days. A series of studies was undertaken to establish pharmacokinetic parameters for PFHxS in rats, mice, and monkeys after single administration with pharmacokinetic parameters determined by WinNonlin(®) software. Rats and mice appeared to be more effective at eliminating PFHxS than monkeys. With the exception of female rats, which had serum PFHxS elimination half-life of approximately 2 days, the serum elimination half-lives in the rodent species and monkeys approximated 1month and 4months, respectively, when followed over extended time periods (10-24weeks). Collectively, these studies provide valuable insight for human health risk assessment regarding the potential for accumulation of PFHxS in humans.

Tissue distribution of 35S-labelled perfluorooctane sulfonate in adult mice after oral exposure to a low environmentally relevant dose or a high experimental dose

The widespread environmental pollutant perfluorooctane sulfonate (PFOS), detected in most animal species including the general human population, exerts several effects on experimental animals, e.g., hepatotoxicity, immunotoxicity and developmental toxicity. However, detailed information on the tissue distribution of PFOS in mammals is scarce and, in particular, the lack of available information regarding environmentally relevant exposure levels limits our understanding of how mammals (including humans) may be affected. Accordingly, we characterized the tissue distribution of this compound in mice, an important experimental animal for studying PFOS toxicity. Following dietary exposure of adult male C57/BL6 mice for 1-5 days to an environmentally relevant (0.031 mg/kg/day) or a 750-fold higher experimentally relevant dose (23 mg/kg/day) of ³⁵S-PFOS, most of the radioactivity administered was recovered in liver, bone (bone marrow), blood, skin and muscle, with the highest levels detected in liver, lung, blood, kidney and bone (bone marrow). Following high daily dose exposure, PFOS exhibited a different distribution profile than with low daily dose exposure, which indicated a shift in distribution from the blood to the tissues with increasing dose. Both scintillation counting (with correction for the blood present in the tissues) and whole-body autoradiography revealed the presence of PFOS in all 19 tissues examined, with identification of thymus as a novel site for localization for PFOS and bone (bone marrow), skin and muscle as significant body compartments for PFOS. These findings demonstrate that PFOS leaves the bloodstream and enters most tissues in a dose-dependent manner.

Tissue distribution of (35)S-labelled perfluorooctane sulfonate (PFOS) in C57Bl/6 mice following late gestational exposure

Exposure of rodents in utero to perfluorooctane sulfonate (PFOS) impairs perinatal development and survival. Following intravenous or gavage exposure of C57Bl/6 mouse dams on gestational day (GD) 16 to (35)S-PFOS (12.5mg/kg), we determined the distribution in dams, fetuses (GD18 and GD20) and pups (postnatal day 1, PND1) employing whole-body autoradiography and liquid scintillation counting. In dams, levels were highest in liver and lungs. After placental transfer, (35)S-PFOS was present on GD18 at 2-3 times higher levels in lungs, liver and kidneys than in maternal blood. In PND1 pups, levels in lungs were significantly higher than in GD18 fetuses. A heterogeneous distribution of (35)S-PFOS was observed in brains of fetuses and pups, with levels higher than in maternal brain. This first demonstration of substantial localization of PFOS to both perinatal and adult lungs is consistent with evidence describing the lung as a target for the toxicity of PFOS at these ages.

Influence of contaminated drinking water on perfluoroalkyl acid levels in human serum – A case study from Uppsala, Sweden

In 2012 a contamination of drinking water with perfluoroalkyl acids (PFAAs) was uncovered in the City of Uppsala, Sweden. The aim of the present study was to determine how these substances have been distributed from the contamination source through the groundwater to the drinking water and how the drinking water exposure has influenced the levels of PFAAs in humans over time. The results show that PFAA levels in groundwater measured 2012-2014 decreased downstream from the point source, although high ΣPFAA levels (>100ng/L) were still found several kilometers from the point source in the Uppsala aquifer. The usage of aqueous film forming fire-fighting foams (AFFF) at a military airport in the north of the city is probably an important contamination source. Computer simulation of the distribution of PFAA-contaminated drinking water throughout the City using a hydraulic model of the pipeline network suggested that consumers in the western and southern parts of Uppsala have received most of the contaminated drinking water. PFAA levels in blood serum from 297 young women from Uppsala County, Sweden, sampled during 1996-1999 and 2008-2011 were analyzed. Significantly higher concentrations of perfluorobutane sulfonic acid (PFBS) and perfluorohexane sulfonic acid (PFHxS) were found among women who lived in districts modeled to have received contaminated drinking water compared to unaffected districts both in 1996-1999 and 2008-2011, indicating that the contamination was already present in the late 1990s. Isomer-specific analysis of PFHxS in serum showed that women in districts with contaminated drinking water also had an increased percentage of branched isomers. Our results further indicate that exposure via contaminated drinking water was the driving factor behind the earlier reported increasing temporal trends of PFBS and PFHxS in blood serum from young women in Uppsala.

Radiosynthesis of perfluorooctanesulfonate (PFOS) and perfluorobutanesulfonate (PFBS), including solubility, partition and adhesion studies

Here, we describe for the first time the synthesis of [(35)S] PFOS and [(35)S] PFBS with sulfur-35 enriched sulfur dioxide as the radiolabelled reagent, resulting in 2.5 and 2.3 mCi of product, respectively. Basic information concerning the physicochemical properties of perfluorooctanesulfonate (PFOS), perfluorobutanesulfonate (PFBS) and perfluorooctanoic acid (PFOA) are still limited. Hence, we utilized these radiolabelled perfluoroalkanesulfonates (PFSAs), as well as carbon-14 labelled perfluorooctanoic acid ([(14)C] PFOA) to determine some basic characteristics of physiological and experimental significance. The solubility of PFOS in buffered aqueous solutions at pH 7.4 was found to be severely reduced in the presence of potassium and sodium ions, which, however, did not reduce the solubility of PFOA or PFBS. PFOS was found to adhere to a small extent to polypropylene and polystyrene, whereas no such adhesion of PFOA or PFBS was detected. The extents of adhesion of PFOS and PFOA to glass were found to be 20% and 10%, respectively. For the first time, the partition coefficients for PFOS, PFBS and PFOA between n-octanol and water were determined experimentally, to be -0.7, -0.3, and 1.4, respectively, reflecting the difference in the amphiphilic natures of these molecules.

Tissue distribution of 35S-labelled perfluorobutanesulfonic acid in adult mice following dietary exposure for 1-5 days.

Perfluorobutanesulfonyl fluoride (PBSF) has been introduced as a replacement for its eight-carbon homolog perfluorooctanesulfonyl fluoride (POSF) in the manufacturing of fluorochemicals. Fluorochemicals derived from PBSF may give rise to perfluorobutanesulfonic acid (PFBS) as a terminal degradation product. Although basic mammalian toxicokinetic data exist for PFBS, information on its tissue distribution has only been reported in one study focused on rat liver. Therefore, here we characterized the tissue distribution of PFBS in mice in the same manner as we earlier examined its eight-carbon homolog perfluorooctanesulfonate (PFOS) to allow direct comparisons. Following dietary exposure of adult male C57/BL6 mice for 1, 3 or 5d to 16 mg (35)S-PFBS kg(-1) d(-1), both scintillation counting and whole-body autoradiography (WBA) revealed the presence of PFBS in all of the 20 different tissues examined, demonstrating its ability to leave the bloodstream and enter tissues. After 5d of treatment the highest levels were detected in liver, gastrointestinal tract, blood, kidney, cartilage, whole bone, lungs and thyroid gland. WBA revealed relatively high levels of PFBS in male genital organs as well, with the exception of the testis. The tissue levels increased from 1 to 3 d of exposure but appeared thereafter to level-off in most cases. The estimated major body compartments were whole bone, liver, blood, skin and muscle. This exposure to PFBS resulted in 5-40-fold lower tissue levels than did similar exposure to PFOS, as well as in a different pattern of tissue distribution, including lower levels in liver and lungs relative to blood.

Tissue uptake, distribution and elimination of 14C-PFOA in zebrafish (Danio rerio)

Perfluorooctanoic acid (PFOA) is a long-chain perfluorinated chemical that has been shown to be non-degradable and persistent in the environment. Laboratory studies on bioconcentration and compound-specific tissue distribution in fish can be valuable for prediction of the persistence and environmental effects of the chemicals. In the present study male and female zebrafish (Danio rerio) were continuously exposed to 10μg/L of radiolabeled perfluorooctanoic acid ((14)C-PFOA) for 40 days, after which the exposed fish were transferred to fresh clean water for another 80 days wash-out period. At defined periodic intervals during the uptake and wash-out, fish were sampled for liquid scintillation counting and whole body autoradiography to profile the bioconcentration and tissue distribution of PFOA. The steady-state concentration of (14)C-PFOA in the zebrafish was reached within 20-30 days of exposure. The concentration-time course of (14)C-PFOA displayed a bi-exponential decline during washout, with a terminal half-life of approximately 13-14 days. At steady-state the bioconcentration of (14)C-PFOA into whole-body fish was approximately 20-30 times greater than that of the exposure concentration, with no differences between females and males. The bioconcentration factors for liver and intestine were approximately 100-fold of the exposure medium, while in brain, ovary and gall bladder the accumulation factors were in the range 15-20. Whole-body autoradiograms confirmed the highest labeling of PFOA in bile and intestines, which implies enterohepatic circulation of PFOA. The (14)C-PFOA was also observed in maturing vitellogenic oocytes, suggesting chemical accumulation via yolk proteins into oocytes with plausible risk for adverse effects on early embryonic development and offspring health. The bioconcentration at several (14)C-PFOA exposure concentrations were also investigated (0.3-30μg/L). This showed that bioconcentration increased linearly with tank exposure in the present in vivo model under steady-state conditions. From this model tissue concentrations of PFOA can be predicted when the external exposure level is known. The present study has generated experimental data on PFOA kinetics in zebrafish that can be valuable for aquatic environmental risk assessment.

Synthesis of 2-(4-chlorophenyl)-2-(4-chloro-3-thiophenol)-1,1-dichloroethene (3-SH-DDE) via Newman-Kwart rearrangement - A precursor for synthesis of radiolabeled and unlabeled alkylsulfonyl-DDEs

For the first time, a pathway for synthesis of 2-(4-chlorophenyl)-2-(4-chloro-3-thiophenol)-1,1-dichloroethene (3-SH-DDE), is presented. The compound is of particular interest as a precursor for synthesis of alkylsulfonyl-DDE containing different alkyl groups to discover structural activity relationships, and to promote synthesis of radiolabeled methylsulfonyl-DDE. 2-Chloro-5-methylphenol was first methylated and further oxidized to the corresponding benzoic acid. The acid was reduced to the corresponding aldehyde (4-chloro-3-methoxy benzaldehyde) via 4-chloro-3-methoxy-benzene methanol. A lead/aluminium bimetal system was used to carry out the reductive addition of tetrachloromethane to 4-chloro-3-methoxy benzaldehyde to obtain 2,2,2-trichloro-1-(4-chloro-3-methoxyphenyl)ethanol, the desired starting material to synthesize the DDT-analogue (2-(4-chlorophenyl)-2-(4-chloro-3-methoxy-phenyl)-1,1,1-trichloroethane). Elimination of hydrochloric acid and removal of the methyl group led to the 3-OH-DDE. The Newman-Kwart rearrangement was applied to convert 3-OH-DDE to 3-SH-DDE via the dimethylcarbamothioate derivative. 3-SH-DDE is then used as a precursor for the radiolabel synthesis. The overall yield to acquire 3-SH-DDE after 11 steps was 3%. The step with the lowest yield was the DDT-analog synthesis with a yield of 30%. All other step had a yield of >50%. 3-SH-DDE was methylated with (14)C-labeled iodomethane and oxidized by hydrogen peroxide to obtain 3-[(14)C]MeSO(2)-DDE in an overall yield of 30%.