An Analytical Toolbox for Investigating the Bioaccumulation Potential of Per- and Polyfluoroalkyl Substances in Aquatic Ecosystems

Abstract

Per- and polyfluoroalkyl substances (PFAS) are a diverse class of anthropogenic organofluorine chemicals with tens of thousands of potential structures. Thousands of sites across the United States (U.S.) are contaminated by PFAS from sources including manufacturing, waste disposal, and use of aqueous film-forming foams (AFFF). Exposure to PFAS has been associated with immune dysfunction, liver toxicity, reproductive and developmental effects, and various cancers. Fish and shellfish are a major dietary source of PFAS exposure for many populations. However, there are no federal guidelines for recommended limits on PFAS exposure from fish consumption. Several U.S. states have developed fish consumption advisories, but they only focus on one PFAS, perfluorooctane sulfonate (PFOS). A major challenge in extending guidelines to include total PFAS exposure is that standard targeted mass spectrometry methods only measure a small fraction of the total PFAS in environmental samples. Precursors to terminal PFAS which have been associated with adverse health effects often make up the majority of total PFAS in environmental samples and consumer products. However, precursor bioaccumulation in aquatic ecosystems is not well understood, in part due to limitations of prior measurements. A toolbox of analytical and statistical methods is needed to better constrain total PFAS exposures and assess the contribution of precursors to total PFAS in commonly consumed fish. This thesis focuses on the development and optimization of an analytical toolbox for measuring PFAS in aquatic ecosystems to better understand precursor bioaccumulation and PFAS exposures of fish consumers. Chapter 2 of this thesis applies analytical and statistical methods from the toolbox to identify and quantify precursors in freshwater fish species impacted by diffuse contamination sources. Specifically, I compare measured concentrations of precursors with ≤6 perfluorinated carbons (short-chain PFAS) using targeted analysis to concentrations estimated based on statistical inference of oxidizable precursors and semi-quantified from suspect screening analysis. This example emphasizes the benefits of using a toolbox of methods to better understand the robustness of any given measurement, particularly for compounds that lack analytical standards. Chapters 3 and 4 of this thesis use the analytical methods developed in Chapter 2 in addition to organofluorine analysis, to characterize the signature and bioaccumulation potential of PFAS in freshwater and estuarine fish/shellfish species collected along a hydrological gradient away from AFFF-contaminated source zones. In Chapter 3, I use extractable organofluorine measurements, suspect screening, and targeted analysis to construct a mass budget for total PFAS present in contaminated fish and shellfish. PFOS and perfluoroalkyl sulfonamides (FASA) account for the majority of organofluorine in fish. FASA are stable intermediate precursor degradation products of other sulfonamido precursors present in AFFF. The composition and concentrations of PFAS in surface water and fish change with hydrological distance from the source zones, largely due to degradation of precursors along the flow path. The flux of PFAS from the AFFF source zones is identifiable based on the composition of PFAS in fish > 8 km downgradient in the freshwater river. The AFFF signature becomes indistinguishable in fish when the river intersects the coastal estuary where diffuse PFAS sources are more apparent. This is important for elucidating the areal extent of PFAS exposure for fish adjacent to PFAS contamination sources. These results emphasize that fish consumers may be exposed to PFAS that exceed fish consumption limits across broad hydrological catchments due to transport of the source signal within a watershed. In Chapter 4, I use paired measurements of FASA in surface water and fish to compare their bioaccumulation potential to PFOS, the only PFAS included in fish consumption advisories. Bioaccumulation factors (BAF) for FASA with 4-8 perfluorinated carbons are 1 to 3 orders of magnitude higher than their terminal PFAA, including PFOS. These data reinforce the urgent need for fish consumption advisory programs to consider potential risks associated with FASA exposures. My dissertation provides insights into the propensity of precursors to bioaccumulate in fish species impacted by both diffuse and AFFF contamination sources. Chapters 2 and 3 emphasize the benefits of using a toolbox of methods to quantify total PFAS and assess the accuracy across measurements. Chapter 3 highlights the importance of hydrological proximity in understanding the composition and concentrations of PFAS in fish downgradient of contamination sources. Chapters 2, 3, and 4 show that FASA are an important class of bioaccumulative precursors that should be incorporated into routine monitoring and risk assessments. These findings are critical to inform regulatory strategies to avoid underestimating the magnitude of PFAS exposure for fish consumers.