Phenylalanine as a probe for mineral-mediated hydroxyl radical formation

Abstract

Mineral-induced formation of reactive oxygen species (ROS) can lead to transformations of organic compounds in aqueous solutions and has been tied to human-health related illnesses resulting from inhalation of mineral or pyritic coal dust. This dissertation reports the use of phenylalanine (Phe) as a biologically-relevant probe to better characterize the conditions that control the formation and fate of hydroxyl radical (.OH) in suspensions of pyrite, coal, and manganese oxide minerals (MnOx). An HPLC-MS method involving minimal sample processing was developed that allows for sensitive analysis of both Phe and its specific phenyl hydroxylation product mixture of ortho-, meta-, and para-tyrosine (Tyr) - with o-Tyr and m-Tyr unique to reaction of Phe with .OH. In pyrite mixtures with pure water, .OH was produced at substantial rates for up to ten days in proportion to the surface area of the pyrite. It was shown for the first time that reaction kinetics were much less than first-order in initial Phe concentration, despite the fact that Phe disappearance as a function of time was generally exponential. These results could be rationalized and modeled by considering that the effective concentration of .OH in solution is limited by the flux of hydrogen peroxide (H2O2) from the pyrite surface, levels of .OH decrease with higher levels of reactant, and that an increasing fraction of .OH is consumed by Phe-degradation products as a function of time. Reaction rates were greatly reduced in pyrite suspension containing simulating lung fluid due to the effects of iron chelating agents (e.g., citrate), higher pH, adsorption of complex ions to active surface iron sites (e.g., phosphate), and competition of solutes for .OH (e.g., glycine and chloride). The effects of these components on Phe degradation rates in pyrite slurries were tested individually and compared to their expected competition factors, with the difference between Phe and Tyr kinetics predicted by the model attributed to decreased ROS formation. Similarly, suspensions of pyrite in simulated seawater slowed Phe degradation rates relative to salt concentration (8 - 32 parts-per-thousand) indicating competition for .OH. An initial lag in Phe loss in pyrite slurries with added sea salts was attributed to a higher initial pH of 6.5 to 8.2. The addition of 200 mg L-1 and 400 mg L-1 of dissolved humic acids to pyrite slurries slowed the hydroxylation of Phe in manner consistent with competition for .OH, with the modeled exponential decay curve fitting well based on second-order rate constants derived in previous studies. Phe was readily transformed to the three isomers of Tyr in slurries of pyritic coals. Degradation rates normalized to pyrite mass were faster in pyritic coal than in pure pyrite, despite the fact that the coal organic matrix can act to compete for .OH produced, which may be related to a greater surface area of framboidal pyrite present in coal. The effect of pH was characterized with a chemostat, where a constant pH was maintained by titration with sodium hydroxide. Constant pH experiments indicated that .OH in a slurry of pyritic coal can be formed to a greater extent at pH 4.5 than when pH was allowed to drift to around pH 3.3. On the other hand, a large initial burst of .OH was observed in a slurry at constant pH 7.4 (the pH at which physiological fluid is buffered), after which there was little evidence for .OH production. Phe loss was only observed in presence of synthetic MnOx with manganese oxidation states of +3 (Mn2O3) or +4 (MnO2), but the lack of Tyr formation could not support a mechanism involving .OH formation as indicated in experiments with a different probe. Rather, identification of an oxidative decarboxylation product was consistent with a direct electron transfer mechanism as proposed in prior work with MnOx. These findings show that specificity of Phe conversion to the three Tyr isomers can be useful in excluding false readings of .OH formation when other, less-selective probes implicate .OH.