Search

Chronic antibiotic exposure in humans can promote the evolution of antibiotic resistant microbes that can directly transfer to humans or host antibiotic resistant genes that can transmit to other infectious human pathogens. Sources of human antibiotic exposure vary, but farmed seafood is of great concern because the use of medicinal antibiotics in aquaculture for prophylactic purposes may be associated with residual levels of antibiotics in seafood products, and ultimately, exposure to humans. Recent studies have also documented the presence of antibiotics in wild seafood, suggesting environmental contamination, but a direct and comprehensive comparison of antibiotic profiles and concentrations in wild versus farmed seafood has not been systematically assessed with validated methods. Additionally, most seafood is cooked prior to human consumption, but detailed analysis of the thermal stability of antibiotics found in seafood remains unknown. The overall objective of this thesis was to validate common methods used for antibiotic extraction from seafood, and apply optimized procedures to test the hypothesis that farmed seafood will contain more antibiotics than wild seafood, and that thermal treatment will degrade antibiotics present in seafood. Method validation involved testing the stability of antibiotic standards stored as mixtures, and checking the extent of seafood matrix effects (i.e. ion suppression or enhancement) on extracted antibiotics measured with ultra-high pressure liquid chromatography coupled to tandem mass spectrometry (UPLC-MS/MS). Thus, in my first experiment, I investigated the stability of antibiotics stored as mixture in water: methanol for one week at different temperatures, pHs, water: methanol ratios and storage container types (i.e. glass vs. silanized glass), because prior studies had inconclusively suggested that these conditions might affect the stability of antibiotics (Experiment 1). I then explored whether the extraction of antibiotics from salmon, as a representative seafood matrix, is associated with matrix effects that can potentially be minimized with clean-up methods involving column or dispersive solid phase extraction (Experiment 2). For final experiment (Experiment 3), I used the information gained from my method validation efforts to measure antibiotic residues in both wild-caught and farm-raised fish and shrimp samples produced locally in U.S. and imported from other countries. In addition, I assessed the effect of thermal processing on the degradation of antibiotics in seafood matrices with varying fat levels, as lipids may protect antibiotics from thermal degradation. I found that antibiotics prepared as a mixture were not stable during one week storage in water: methanol irrespective of temperature and pH and that silanization of glass vials improved the storage stability of some quinolones and macrolides but deteriorated the stability of other antibiotic classes including some amphenicols, B-lactams, macrolides, sulfonamides and dihydrofolate reductase inhibitors (Experiment 1). This led me to conclude that antibiotics should be freshly mixed before use on UPLC-MS/MS. In Experiment 2, I found that salmon matrix components are associated with significant matrix effects, which were not improved with column or dispersive solid phase extraction clean-up. However, using appropriate internal standards that match the polarity of the antibiotics resulted in accurate quantitation of antibiotics despite losses in sensitivity. I therefore used appropriate internal standards for antibiotics quantitation in the seafood survey study in Experiment 3 (n=125), and found that both wild-caught and farm-raised seafood locally produced in U.S. or imported from other countries contained antibiotic residues. I found higher detection frequencies of antibiotics in farmed than wild-caught seafood and in imported than locally produced seafood. Surprisingly, antibiotic concentrations were higher in wild-caught than farm-raised seafood. Finally, I discovered that several antibiotics (quinolones, amphenicols, some macrolides, dihydrofolate reductase inhibitors, lincosamides and sulfonamides) were relatively stable in various fish matrices, irrespective of lipid content, under heat treatment. B-lactams, tetracyclines and a few macrolides were unstable under thermal treatment. Overall, using validated methods, this study provides new unexpected evidence of widespread contamination of antibiotics in both farmed and wild seafood and that thermal treatment does not degrade several antibiotic classes. The impact of chronic human exposures from seafood on the development of antibiotic resistance warrants immediate investigation.

Cocoa beans (Theobroma cacao) are a highly concentrated source of dietary flavanols- bioactive compounds associated with the health protective properties of cocoa. Cocoa beans undergo processing steps, such as fermentation, roasting, winnowing, grinding, pressing, etc., to produce a final product with specific desirable sensory attributes. It is well established that these processing steps, specifically fermentation and roasting, result in dramatic degradation of cocoa's native flavanols, but it is possible that these processing steps may generate compounds with novel activities, potentially preserving or enhancing bioactivity. Raw unfermented cocoa beans were processed by way of a partial factorial approach to produce cocoa powders from the same batch of raw beans using various combinations of fermentation [unfermented, cool fermented (maximum 46°C), hot fermented (maximum 60°C))] and roasting [unroasted, cool roasted (120°C), hot roasted (170°C)]. To simulate cocoa fermentation in a highly controlled environment, a pilot-scale fermentation model system was employed to eliminate many external unknowns and ensure that the differences between our cocoa powders were due to our various treatments, rather than unknown factors occurring during fermentation and roasting. Low and high molecular weight fractions (8-10 kDa cutoff) were produced from cocoa powder extracts (CPE) of each treatment to quantify Maillard reaction products (MRP). A HILIC-UPLC MS/MS method was developed to more efficiently and sensitively quantify cocoa flavanols with high degrees of polymerization (DP) produced during processing. Overall, cocoa processing significantly (p