Jonathan P. Benskin

Isomer profiles of perfluorochemicals in matched maternal, cord, and house dust samples: manufacturing sources and transplacental transfer.

Background: Perfluorochemicals (PFCs) are detectable in the general population and in the human environment, including house dust. Sources are not well characterized, but isomer patterns should enable differentiation of historical and contemporary manufacturing sources. Isomer-specific maternal–fetal transfer of PFCs has not been examined despite known developmental toxicity of perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA) in rodents. Objectives: We elucidated relative contributions of electrochemical (phased out in 2001) and telomer (contemporary) PFCs in dust and measured how transplacental transfer efficiency (TTE; based on a comparison of maternal and cord sera concentrations) is affected by perfluorinated chain length and isomer branching pattern. Methods: We analyzed matching samples of house dust (n = 18), maternal sera (n = 20), and umbilical cord sera (n = 20) by isomer-specific high-performance liquid chromatography tandem mass spectrometry. Results: PFOA isomer signatures revealed that telomer sources accounted for 0–95% of total PFOA in house dust (median, 31%). This may partly explain why serum PFOA concentrations are not declining in some countries despite the phase-out of electrochemical PFOA. TTE data indicate that total branched isomers crossed the placenta more efficiently than did linear isomers for both PFOS (p < 0.01) and PFOA (p = 0.02) and that placental transfer of branched isomers of PFOS increased as the branching point moved closer to the sulfonate (SO3–) end of the molecule. Conclusions: Results suggest that humans are exposed to telomer PFOA, but larger studies that also account for dietary sources should be conducted. The exposure profile of PFOS and PFOA isomers can differ between the mother and fetus—an important consideration for perinatal epidemiology studies of PFCs.

Disposition of perfluorinated acid isomers in sprague‐dawley rats; Part 1: Single dose

Perfluorinated acids (PFAs) and their precursors (PFA-precursors) exist in the environment as linear and multiple branched isomers. These isomers are hypothesized to have different biological properties, but no isomer-specific data are currently available. The present study is the first in a two-part project examining PFA isomer-specific uptake, tissue distribution, and elimination in a rodent model. Seven male Sprague-Dawley rats were administered a single gavage dose of approximately 500 microg/kg body weight perfluorooctane sulfonate (C(8)F(17)SO(3)(-), PFOS), perfluorooctanoic acid (C(7)F(15)CO(2)H, PFOA), and perfluorononanoic acid (C(8)F(17)CO(2)H, PFNA) and 30 microg/kg body weight perfluorohexane sulfonate (C(6)F(13)SO(3)(-), PFHxS). Over the subsequent 38 d, urine, feces, and tail-vein blood samples were collected intermittently, while larger blood volumes and tissues were collected on days 3 and 38 for isomer analysis by high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS). For all PFAs, branched isomers generally had lower blood depuration half-lives than the corresponding linear isomer. The most remarkable exception was for the PFOS isomer containing an alpha-perfluoromethyl branch (1m-PFOS), which was threefold more persistent than linear PFOS, possibly due to steric shielding of the hydrophilic sulfonate moiety. For perfluoromonomethyl-branched isomers of PFOS, a structure-property relationship was observed whereby branching toward the sulfonate end of the perfluoroalkyl chain resulted in increased half-lives. For PFHxS, PFOA, and PFOS, preferential elimination of branched isomers occurred primarily via urine, whereas for PFNA preferential elimination of the isopropyl isomer occurred via both urine and feces. Changes in the blood isomer profiles over time and their inverse correlation to isomer elimination patterns in urine, feces, or both provided unequivocal evidence of significant isomer-specific biological handling. Source assignment based on PFA isomer profiles in biota must therefore be conducted with caution, because isomer profiles are unlikely to be conserved in biological samples.

Isomer Profiling of Perfluorinated Substances as a Tool for Source Tracking: A Review of Early Findings and Future Applications

The two major manufacturing techniques for perfluorochemicals can be distinguished based on the isomeric profile of their products. ECF (major use from 1950s to 2002) results in a product containing both linear and branched isomers, while telomerization (major use from 2002 to present) typically yields an isomerically pure, linear product. Among the most important question today, which has implication for future regulation of these chemicals, is to what extent human and environmental exposure is from historical products (i.e., ECF) versus currently manufactured fluorochemicals (i.e., telomer). Perfluoroalkyl-chain branching can also affect the physical and chemical properties of these chemicals, which may influence their environmental transport and degradation, partitioning, bioaccumulation, pharmacokinetics, and toxicity. Unless perfluorinated substances are considered as individual isomers, much of this information will be overlooked or missed altogether, which could potentially lead to inaccuracies in human and environmental risk assessments. In this review, we have highlighted novel findings, current knowledge gaps, and areas for improvement based on early experiments on the disposition of PFA and PFA-precursor isomers in the environment. We have also emphasized the wealth of information that can potentially be gleaned from future work in this area, which renders routine adoption of isomer-specific methodologies an attractive and logical next step in the progression of fluorochemicals analysis. However, despite vast improvements in recent years, a fast and comprehensive method capable of separating all major PFA and PFA-precursor isomers, while removing interferences is still required before these methods becomes routine in most labs. Purified and characterized standards of PFOA and PFOS that have isomer profiles consistent with those of historically produced (i.e., 3M) PFOS and PFOA are also required. The limited data available on PFA isomer profiles that exist in the environment and the biological properties of each isomer suggest that examination of isomer profiles may yield clues on the source of PFA contamination to human and the environment. For example, contributions from historical versus current PFOA emissions can be quantified by examining the isomer profile in abiotic samples . Similarly, residual PFOS/PFOA in pre-2002 consumer products may be distinguished from directly emitted PFOS/PFOA by the existence of slight difference in isomer profile. PFOS signatures may also have the potential to distinguish between indirect exposure (via precursors) versus direct exposure (via the sulfonate), based on findings of isomer-specific and/or enantiospecific biotransformation in vitro. Isomer-specific monitoring extended to longer-chain PFAs may also be informative in determining current and historical exposure sources. Finally, given the recent increase of production of PFOSF-based chemicals, following their 2002 phase out, the ability of using isomer profiles to distinguish between historical and currently produced PFOS may also be possible.

Disposition of perfluorinated acid isomers in Sprague-Dawley rats; part 2: subchronic dose.

Two major industrial synthetic pathways have been used to produce perfluorinated acids (PFAs) or their precursors: Telomerization and electrochemical fluorination (ECF). Products of telomer and ECF origin can be distinguished by structural isomer profiles. A mixture of linear and branched perfluoroalkyl isomers is associated with ECF. Telomer products characteristically consist of a single perfluoroalkyl geometry, typically linear. In biota, it is unclear if the isomer profile is conserved relative to the exposure medium and hence whether PFA isomer profiles in organisms are useful for distinguishing environmental PFA sources. A companion study suggested isomer-specific disposition following a single oral gavage exposure to rats. To confirm these findings under a more realistic subchronic feeding scenario, male and female rats were administered PFA isomers by diet for 12 weeks, followed by a 12-week depuration period. The diet contained 500 ng/g each of ECF perfluorooctanoate (PFOA, approximately 80% n-PFOA), ECF perfluorooctane sulfonate (PFOS, approximately 70% n-PFOS), and linear and isopropyl perfluorononanoate (n- and iso-PFNA). Blood sampling during the exposure phase revealed preferential accumulation of n-PFOA and n-PFNA compared to most branched isomers. Female rats depurated all isomers faster than males. Both sexes eliminated most branched perfluorocarboxylate isomers more rapidly than the n-isomer. Elimination rates of the major branched PFOS isomers were not statistically different from n-PFOS. Two minor isomers of ECF PFOA and one branched PFOS isomer had longer elimination half-lives than the n-isomers. Although extrapolation of these pharmacokinetics trends in rats to humans and wildlife requires careful consideration of dosage level and species-specific physiology, cumulative evidence suggests that perfluorocarboxylate isomer profiles in biota may not be suitable for quantifying the relative contributions of telomer and ECF sources.

Perfluorinated acid isomer profiling in water and quantitative assessment of manufacturing source.

A method for isomer profiling of perfluorinated compounds (PFCs) in water was developed and applied to quantitatively assess the contributions from electrochemical (ECF) and telomer manufacturing processes around source regions of North America, Asia, and Europe. With the exception of 3 sites in Japan, over 80% of total perfluorooctanoate (PFOA, C(7)F(15)COO(-)) was from ECF, with the balance attributable to strictly linear (presumably telomer) manufacturing source(s). Comparing PFOA isomer profiles in samples from China, with PFOA obtained from a local Chinese manufacturer, indicated <3% difference in overall branched isomer content; thus, exclusive contribution from local ECF production cannot be ruled out. In Tokyo Bay, ECF, linear-telomer, and isopropyl-telomer sources contributed to 33%, 53%, and 14% of total PFOA, respectively. Perfluorooctane sulfonate (PFOS, C(8)F(17)SO(3)(-)) isomer profiles were enriched in branched content (i.e., >50% branched) in the Mississippi River but in all other locations were similar or only slightly enriched in branched content relative to historical ECF PFOS. Isomer profiles of other PFCs are also reported. Overall, these data suggest that, with the exception of Tokyo Bay, ECF manufacturing has contributed to the bulk of contamination around these source regions, but other sources are significant, and remote sites should be monitored.

Isomer-Specific Biotransformation Rates of a Perfluorooctane Sulfonate (PFOS)-Precursor by Cytochrome P450 Isozymes and Human Liver Microsomes

The exposure sources of perfluorooctane sulfonate (PFOS) in humans and wildlife are not well characterized. Human biomonitoring data show that PFOS profiles may consist of up to approximately 50% branched isomers, despite the fact that historical direct manufacturing of PFOS generally resulted in products containing no more than approximately 30% branched isomers. These observations cannot be explained based on what is known about the pharmacokinetics of branched PFOS isomers; thus, here we examined the relative isomer-specific biotransformation rates of a model PFOS-precursor (N-ethylperfluorooctane sulfonamide, NEtFOSA) with human microsomes and recombinant human cytochrome P450s (CYPs) 2C9 and 2C19. Using solid phase microextraction-gas chromatography-electron capture detection to monitor NEtFOSA disappearance, and liquid chromatography-tandem mass spectrometry to monitor product formation, we showed that, in general, human microsomes and CYP isozymes transformed the branched isomers more rapidly than linear NEtFOSA. Among branched isomers, perfluoroalkyl branching geometry significantly influenced the rate of biotransformation. As a result, PFOS isomer patterns in biota exposed predominantly to precursors could be much different than expected from the isomer pattern of the precursor. While these data are suggestive that the relatively high abundance of branched PFOS isomers present in some humans, or wildlife, may be explained by substantial exposure to PFOS-precursors, in vivo studies with other relevant PFOS-precursors are warranted to validate this as a biomarker of exposure source.

Perfluoroalkyl Acids in the Atlantic and Canadian Arctic Oceans

We report here on the spatial distribution of C(4), C(6), and C(8) perfluoroalkyl sulfonates, C(6)-C(14) perfluoroalkyl carboxylates, and perfluorooctanesulfonamide in the Atlantic and Arctic Oceans, including previously unstudied coastal waters of North and South America, and the Canadian Arctic Archipelago. Perfluorooctanoate (PFOA) and perfluorooctanesulfonate (PFOS) were typically the dominant perfluoroalkyl acids (PFAAs) in Atlantic water. In the midnorthwest Atlantic/Gulf Stream, sum PFAA concentrations (∑PFAAs) were low (77-190 pg/L) but increased rapidly upon crossing into U.S. coastal water (up to 5800 pg/L near Rhode Island). ∑PFAAs in the northeast Atlantic were highest north of the Canary Islands (280-980 pg/L) and decreased with latitude. In the South Atlantic, concentrations increased near Rio de la Plata (Argentina/Uruguay; 350-540 pg/L ∑PFAAs), possibly attributable to insecticides containing N-ethyl perfluorooctanesulfonamide, or proximity to Montevideo and Buenos Aires. In all other southern hemisphere locations, ∑PFAAs were <210 pg/L. PFOA/PFOS ratios were typically ≥1 in the northern hemisphere, ∼1 near the equator, and ≤1 in the southern hemisphere. In the Canadian Arctic, ∑PFAAs ranged from 40 to 250 pg/L, with perfluoroheptanoate, PFOA, and PFOS among the PFAAs detected at the highest concentrations. PFOA/PFOS ratios (typically ≫1) decreased from Baffin Bay to the Amundsen Gulf, possibly attributable to increased atmospheric inputs. These data help validate global emissions models and contribute to understanding of long-range transport pathways and sources of PFAAs to remote regions.

Perfluorooctane sulfonate toxicity, isomer-specific accumulation, and maternal transfer in zebrafish (Danio rerio) and rainbow trout (Oncorhynchus mykiss).

Perfluorooctane sulfonate (PFOS; C(8)F(17)SO(3) (-)) bioaccumulation and toxicity have been demonstrated in both aquatic and terrestrial organisms. The majority of investigations have examined total PFOS concentrations in wildlife and in toxicity testing, but isomer-specific monitoring studies are less common, and no laboratory-based study of PFOS isomer accumulation in fish has been reported. The present study examined accumulation and maternal transfer of PFOS isomers in zebrafish and tissue-specific accumulation of PFOS isomers in trout parr. A median lethal dose (LC50) of 22.2 and 2.5 mg/L was calculated for adult zebrafish and trout parr, respectively. A two-week PFOS exposure resulted in tissue-specific PFOS accumulation in trout, with maximum concentrations identified in the liver tissue (>50 microg/g). Prior exposure to PFOS as alevin did not affect the accumulation of PFOS in tissues later in life. In both species, accumulation of branched PFOS isomers generally occurred to a lesser extent than linear PFOS, which may explain the relative deficiency of branched PFOS isomers in some aquatic species in the field. Analysis of exposed trout tissues indicated that isomer discrimination may occur at the level of elimination or uptake and elimination processes in the kidney or gill, respectively. When zebrafish underwent a reproductive cycle in the presence of PFOS, approximately 10% (wt) of the adult PFOS body burden was transferred to the developing embryos, resulting in a higher total PFOS concentration in eggs (116 +/- 13.3 microg/g) than in the parent fish (72.1 +/- 7.6 microg/g). The isomer profile in eggs was not significantly different from that of adults, suggesting that the maternal transfer of branched and linear PFOS isomers in fish is largely nonisomer specific.

Biodegradation of N‑Ethyl Perfluorooctane Sulfonamido Ethanol (EtFOSE) and EtFOSE-Based Phosphate Diester (SAmPAP Diester) in Marine Sediments

Investigations into the biodegradation potential of perfluorooctane sulfonate (PFOS)-precursor candidates have focused on low molecular weight substances (e.g., N-ethyl perfluorooctane sulfonamido ethanol (EtFOSE)) in wastewater treatment plant sludge. Few data are available on PFOS-precursor biodegradation in other environmental compartments, and nothing is known about the stability of high-molecular-weight perfluorooctane sulfonamide-based substances such as the EtFOSE-based phosphate diester (SAmPAP diester) in any environmental compartment. In the present work, the biodegradation potential of SAmPAP diester and EtFOSE by bacteria in marine sediments was evaluated over 120 days at 4 and 25 °C. At both temperatures, EtFOSE was transformed to a suite of products, including N-ethyl perfluorooctane sulfonamidoacetate, perfluorooctane sulfonamidoacetate, N-ethyl perfluorooctane sulfonamide, perfluorooctane sulfonamide, and perfluorooctane sulfonate. Transformation was significantly more rapid at 25 °C (t(1/2) = 44 ± 3.4 days; error represents standard error of the mean (SEM)) compared to 4 °C (t(1/2) = 160 ± 17 days), but much longer than previous biodegradation studies involving EtFOSE in sludge (t(1/2) ∼0.7-4.2 days). In contrast, SAmPAP diester was highly recalcitrant to microbial degradation, with negligible loss and/or associated product formation observed after 120 days at both temperatures, and an estimated half-life of >380 days at 25 °C (estimated using the lower bounds 95% confidence interval of the slope). We hypothesize that the hydrophobicity of SAmPAP diester reduces its bioavailability, thus limiting biotransformation by bacteria in sediments. The lengthy biodegradation half-life of EtFOSE and recalcitrant nature of SAmPAP diester in part explains the elevated concentrations of PFOS-precursors observed in urban marine sediments from Canada, Japan, and the U.S, over a decade after phase-out of their production and commercial application in these countries.

Per- and Polyfluoroalkyl Substances in Landfill Leachate: Patterns, Time Trends, and Sources

Concentrations and isomer profiles for 24 per- and polyfluoroalkyl substances (PFASs) were monitored over 5 months (February-June, 2010) in municipal landfill leachate. These data were used to assess the role of perfluoroalkyl acid (PFAA) precursor degradation on changes in PFAA concentrations over time. The influence of total organic carbon, total suspended solids, pH, electrical conductivity (EC), leachate flow rates, and meteorological data (precipitation, air temperature) on leachate PFAS concentrations was also investigated. Perfluoropentanoate and perfluorohexanoate were typically the dominant PFASs in leachate, except for March-April, when concentrations of perfluorooctane sulfonate, perfluorooctanoate, and numerous PFAA-precursors (i.e., (N-alkyl) perfluorooctane sulfonamides and fluorotelomer carboxylic acids) increased by a factor of 2-10 (~4 μg/L to ~36 μg/L ΣPFASs). During this time, isomer profiles of PFOA became increasingly dominated by the linear isomer, likely from transformation of linear, telomer-manufactured precursors. While ΣPFAA-precursors accounted for up to 71% of ΣPFASs (molar basis) in leachate from this site, leachate from a second landfill displayed only low concentrations of precursors (<1% of ΣPFASs). Overall, degradation of PFAA-precursors and changes in leachate pH, EC, and 24-h precipitation were important factors controlling PFAS occurrence in leachate. Finally, 8.5-25 kg/yr (mean 16 kg/yr) of ΣPFASs was estimated to leave the landfill via leachate for subsequent treatment at a wastewater treatment plant.