Perfluoroalkyl substances (PFASs) are contaminants of emerging concern, and have been detected in drinking water, wildlife, humans, and the environment. Cyclodextrins (CDs), cyclic sugars composed of glucose monomers, are proposed as a potential remediation strategy. CDs can form host-guest complexes with hydrophobic molecules; this complexation could be capitalized on for PFAS removal and sequestration. These dissertation projects aim to study the fundamental host-guest interactions between a variety of PFASs and CDs for eventual applications in environmental and biological remediation. 1D and 2D Nuclear magnetic resonance (NMR) spectroscopic methods were employed to determine the strength, dynamics, and structure of the CD:PFAS host-guest complexes. Legacy and emerging PFASs were studied with the three native CDs (α-, β-, and γ-CDs) as well as β-CD derivatives. β-CD and its derivatives exhibit the strongest complexation with all studied PFASs, with association constants of 102-105 M-1, depending on PFAS chain length, functional groups, and branching. The host-guest complex was not significantly disturbed under different environmental conditions, such as changing pH, ionic strength, and in the presence of humic acid. A competition study between perfluorooctanoic acid (PFOA), β-CD, and human serum albumin (HSA), the most abundant protein in blood serum, was then conducted using NMR, circular dichroism, and fluorescence spectroscopies. Excess β-CD was able to totally reverse all PFOA binding to HSA. Finally, the host-guest complex was studied within a biological organism to test its viability as a remediation strategy. The attenuation of the toxicity of PFOA in zebrafish embryos, a model organism for toxicology studies, was tested with β-CD. Excess β-CD increased the LC50 (lethal concentration for 50 % of the population) of PFOA compared to PFOA in the absence of β-CD (p < 0.0001). These dissertation projects suggest that the encapsulation of PFASs by CDs has potential in PFAS remediation strategies.

Contexte : L’acide perfluorooctanoïque (PFOA) est présent dans le sang de presque tous les participants dans les études de biosurveillance. L’appréciation des risques pour la santé qui sont associés aux concentrations sanguines de PFOA est difficile, car les valeurs guides d'exposition (VGE) sont généralement exprimées en termes de dose externe. Des équivalents de biosurveillance (ÉB) compatibles avec les VGE pourraient faciliter l'interprétation des concentrations de PFOA dans le contexte du risque pour la santé. Objectifs : i) Dériver des ÉB pour le PFOA sérique ou plasmatique correspondant aux VGE de la Environmental Protection Agency des États-Unis (U.S. EPA), de la Agency for Toxic Substances and Disease Registry (ATSDR) et de Santé Canada, et ii) comparer les ÉB avec les concentrations de PFOA mesurées dans des enquêtes nationales de biosurveillance. Méthodes : En partant des points de départ à partir desquels les VGE ont été estimées, nous avons dérivé des ÉB à l'aide de données ou modèles pharmacocinétiques et de facteurs d'incertitude. À l'aide d'un modèle pharmacocinétique animal de gestation et de lactation, nous avons converti les points de départ chez les souris gestantes (U.S. EPA 2016, ATSDR) en concentrations sériques chez l’organisme en développement pour quatre mesures de dose (concentrations moyenne prénatale, moyenne postnatale, moyenne totale, et maximale). Les concentrations équivalentes chez le fœtus humain et l’enfant ont ensuite été converties en ÉB correspondants dans le sérum maternel humain au moment de la conception à l'aide d’une modèle pharmacocinétique humain de grossesse et d’allaitement. Le point de départ chez les rongeurs adultes (Santé Canada) a été converti en un ÉB dans le sérum humain adulte à l'aide de données expérimentales. Pour le point de départ basé sur l'épidémiologie (valeur provisoire de 2022 de l’U.S.EPA), un BE a été dérivé à l'aide d'un modèle pharmacocinétique humain de grossesse et d’allaitement. Les BE dérivés ont été comparés à des données de biosurveillance canadiennes et américaines. Résultats : Les ÉB étaient de 684 ng/mL pour Santé Canada, de 0.012 ng/mL pour l’U.S. EPA (valeur provisoire de 2022), et variaient de 15 à 29 ng/mL pour l’U.S. EPA (valeur de 2016) et de 6 à 10 ng/mL pour l’ATSDR. Les 95ème centiles des concentrations sériques dans l'Enquête canadienne sur les mesures de la santé (ECMS) de 2018-2019 et dans la National Health and Nutrition Examination Survey (NHANES) de 2017-2018 étaient légèrement inférieures à l’ÉB de l’ATSDR et deux ordres de grandeur plus élevées que l'ÉB pour la valeur provisoire de 2022 de l’U.S. EPA. Conclusion : Les ÉB couvraient quatre ordres de grandeur. Les concentrations de PFOA mesurés dans des enquêtes nationales de biosurveillance canadiennes et américaines étaient supérieures ou similaires à certains de ces ÉB. / Background: Perfluorooctanoic acid (PFOA) is detected in the blood of virtually everyone participating in biomonitoring studies. Assessing the health risks associated with blood PFOA levels is challenging because exposure guidance values (EGVs) are typically expressed in terms of external dose. Biomonitoring equivalents (BEs) consistent with EGVs could facilitate interpreting biomarker levels in a health risk context. Objective: To i) derive BEs for serum/plasma PFOA corresponding to the EGVs of the U.S. Environmental Protection Agency (U.S. EPA), the Agency for Toxic Substances and Disease Registry (ATSDR) and Health Canada, and ii) compare with PFOA levels measured in national biomonitoring surveys. Methods: Starting from the points of departure from which EGVs were estimated, we derived BEs using pharmacokinetic data/models and uncertainty factors. Using an animal gestational/lactational pharmacokinetic model, we converted points of departure in pregnant dams (U.S. EPA 2016, ATSDR) into serum concentrations in the developing organism for four dose metrics: average prenatal, average postnatal, average overall, and maximum concentrations. Equivalent human fetus and child concentrations were then converted into corresponding BEs in maternal serum at time of conception using a pharmacokinetic model of human gestation/lactation. The point of departure in adult rodents (Health Canada) was converted into a BE in adult serum using experimental data. For the epidemiology-based point of departure (U.S.EPA 2022, draft), a BE was derived using a human pharmacokinetic model of gestation/lactation. BEs were compared with Canadian and U.S. biomonitoring data. Results: BEs (ng/mL) were 684 for Health Canada, 0.012 for the U.S. EPA (2022 draft), and ranged from 15-29 for the U.S. EPA (2016) and from 6-10 for ATSDR. Ninety-fifth percentiles of serum levels from the 2018-2019 Canadian Health Measures Survey (CHMS) and the 2017-2018 National Health and Nutrition Examination Survey (NHANES) were slightly below the BE for the ATSDR, and two orders of magnitude above the BE for the U.S. EPA (2022 draft). Conclusion: BEs varied over four orders of magnitude. Levels measured in Canadian and U.S. national surveys were higher than or close to some of these BEs.

A new class of treatment processes called Advanced Reduction Processes (ARP) has been proposed. The ARPs combine activation methods and reducing agents to form highly reactive reducing radicals that degrade oxidized contaminants. Batch screening experiments were conducted to identify effective ARP by applying several combinations of activation methods (ultraviolet light, ultrasound, electron beam, microwaves) and reducing agents (dithionite, sulfite, ferrous iron, sulfide) to degradation of five target contaminants (perchlorate, nitrate, perfluorooctanoic acid, 2,4 dichlorophenol, 1,2 dichloroethane) at 3 pH levels (2.4, 7.0, 11.2). These experiments identified the combination of sulfite activated by ultraviolet light produced by a low pressure mercury vapor lamp as an effective ARP. More detailed kinetic experiments were conducted with nitrate and perchlorate as target compounds and nitrate was found to degrade more rapidly than perchlorate. The effects of pH, sulfite concentration, and light intensity on perchlorate and nitrate degradation were investigated. The effectiveness of the sulfite/UV-L treatment process improved with increasing pH for both perchlorate and nitrate.