1 NOVEL ANALYTICAL STRATEGIES FOR THE CHARACTERIZATION OF STEROIDS AND ENDOCRINE DISRUPTING COMPOUNDS (EDCS) IN BIOLOGICAL AND ENVIRONMENTAL SAMPLES By JOHN A. BOWDEN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009
2 2009 John A. Bowden
3 To my family
4 ACKNOWLEDGMENTS First, I would like to express my sincerest gr atitude to my research advisor, Richard A. Yost, for his thoughtful advice, never-ending patien ce, and constant support. Under his tutelage, I was given the necessary tool s and intellectual freedom duri ng my growth and development toward becoming an analytical chemist. I was extremely fortunate to have the opportunity to learn and mature und er his guidance. I would like to thank my wife, Laura, who is my continual inspiration and safety net. Her unwavering support and encouragement for my gradua te years has been an integral part of my success. She is equally important as my source of imagination, as she is the device to keeping my feet on the ground. I would like to thank my family, my parents John and Susan, my brother Michael, and my grandparents Melvin and Elaine Panfil. Through out my life, I have been blessed with the constant love and support from my family. I would not be in this position today if I did not have the foundation created by my family. I would also like to thank the unmentioned but invaluable family and friends who have suppor ted my efforts over the years. I would like to thank those who had direct im pacts in my research, specifically, Dominic Colosi, Diana Mora-Montero, Dr. Timothy Garrett, and those members of the Dr. Louis Guillette research group in the Zoology department. Both Dominic and Diana were key contributors to my research in both data collection and interpretatio n. Dominic, especially, played a vital role in stimulating and expanding my intellectual limits Beyond the research, mentoring both Dominic and Diana, provided an excellent forum for me to develop my mentoring skills. Dr. Garrett was a tremendous influence and sounding-board during my final years. Beyond the obvious and largely important gratitude I have for using his laboratory space, equipment, and time, I am forever grateful for the many discussions and advice he gave me over the past few years. Finally, I
5 would like to than the current and past member s of the Guillette research group, for the many samples they have collected and the many stimul ating discussions, beyond just the research, the collaboration was truly an integral part of my development. I would like to thank the curre nt and past members of the Yost group, a group of people that over the years have not been just resear ch colleagues, but also good friends. The Yosties have been an invaluable resource during group meetings, conferences, manuscript writing, instrumentation help, and with overall discussion s. I want to thank specifically Mike Napolitano, Frank Kero, Dodge Baluya, Dan Magparangalan, Jennifer Bryant, and Dave Pirman for their help on numerous occasions. I would also like to thank the members of my Ph. D. research committee, Dr. Dave Powell, Dr. Thomas Lyons, Dr. Ben Smith, and Dr. Louis Gu illette, for their time a nd encouragement, in an effort to help me progress as a fellow scie ntist. I would like to thank both Dr. Williams and Dr. Young, for their invaluable support and advi ce. Dr. Williams has been an excellent role model in her commitment to being a self-less mentor to others. And most importantly, I want to acknowledge my faith that has guided my life both personally and academically. I am very thankful for the blessings and opportunities that God has bestowed upon me. I will always aim to us e my talents to honor Him and his glory.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ........10 LIST OF FIGURES................................................................................................................ .......12 ABSTRACT....................................................................................................................... ............14 CHAPTER 1 INTRODUCTION................................................................................................................. .16 Overview of Endocrine Disruption.........................................................................................16 Background on Endocrine Disruption.............................................................................16 Mechanisms of Endocrine Disruption.............................................................................18 Types of Endocrine Disrupting Compounds (EDCs)......................................................19 Endocrine Disruption Effects.................................................................................................20 Consequences of Environmental Exposure.....................................................................20 The Alligator Model........................................................................................................21 The Human Impact..........................................................................................................22 Controversy of Endocrine Disruption.............................................................................24 Strategies for Endocrine Di srupting Characterization............................................................26 Issues Regarding Endocrine Activity..............................................................................26 Detection of Environmental EDCs..................................................................................27 Detection of Steroids an d EDC-Induced Responses.......................................................29 Background on EDC/Steroid Char acterization Strategies......................................................32 Analytical Methodology..................................................................................................32 Gas Chromatography.......................................................................................................34 Liquid Chromatography..................................................................................................35 Mass Spectrometry (MS) and Tandem Mass Spectrometry (MS/MS)...........................37 Sample Extraction...........................................................................................................39 Analytical Derivatization.................................................................................................41 Executive Summary.............................................................................................................. ..43 2 EVALUATION OF EXTRACTION ST RATEGIES FOR THE ANALYSIS OF STEROIDS AND ENDOCRINE DISRUP TING COMPOUNDS (EDCS) IN ENVIRONMENTAL AND BIOLOGICAL SAMPLES USING GAS CHROMATOGRAPHY/MASS SPECTROMETRY (GC/MS)............................................56 Introduction................................................................................................................... ..........56 GC/MS Analysis of EDCs and Steroids..........................................................................57 Liquid-Liquid Extraction (LLE)......................................................................................57 Solid-Phase Extraction (SPE)..........................................................................................58 Solid-Phase Microextraction (SPME).............................................................................58
7 Solid-Phase Microextraction with On -Fiber Derivatization (SPME-OFD)....................59 Experimental Section........................................................................................................... ...60 Steroids and Endocrine Di srupting Compounds (EDCs)................................................60 Liquid-Liquid Extraction (LLE) Method........................................................................61 Solid-Phase Extraction (SPE) Methods...........................................................................62 Offline Derivatization Strategy for LLE and SPE Extracts.............................................62 Relative Response Factor (RRF) and Relativ e Extraction Efficiency (REE) Values for the LLE and SPE Methods.....................................................................................63 Solid-Phase Microextra ction (SPME) Method................................................................63 Solid-Phase Microextraction On-Fiber Derivatization ( SPME-OFD) Method...............64 SPME Fiber Monitoring..................................................................................................64 Spiked Human Plasma (SPM E with and without OFD).................................................64 Gas Chromatography/Mass Spectrometry (GC/MS).......................................................65 Results and Discussion......................................................................................................... ..65 Method Development......................................................................................................65 The RRF Values and Extraction Efficiency for Spiked Water Using SPE.....................66 The RRF Values and Extraction Efficiency for Spiked Plasma Using SPE...................68 Analysis of Spiked Water Using the LLCE Cartridges...................................................68 SPME Analysis of Water.................................................................................................69 SPME Analysis of Plasma...............................................................................................69 Monitoring Fiber Condition............................................................................................70 Conclusion..................................................................................................................... .........70 3 ENHANCEMENT OF CHEMICAL DERI VATIZATION OF STEROIDS BY GAS CHROMATOGRAPHY/MASS SPECTROMETRY (GC/MS)............................................83 Introduction................................................................................................................... ..........83 GC/MS and Derivatization..............................................................................................83 Silylation Strategies.........................................................................................................8 4 Experimental Section........................................................................................................... ...85 Chemicals and Reagents..................................................................................................85 GC/MS Analysis..............................................................................................................86 Derivatization Experiments.............................................................................................86 Derivatization overview...........................................................................................86 Derivatization time and temperature........................................................................87 Derivatization enhancing experiments.....................................................................87 Results and Discussion......................................................................................................... ..88 Derivatization Optimization............................................................................................88 Derivatization time and temperature........................................................................88 Derivatization reagent..............................................................................................90 Derivatization Enhancers.................................................................................................90 Solvent enhancement................................................................................................90 Microwave-accelerated derivatization (MAD)........................................................91 Sonication-assisted derivatization (SAD)................................................................92 Conclusion..................................................................................................................... .........92
8 4 ENHANCED ANALYSIS OF STEROIDS BY GAS CHROMATOGRAPHY/ MASS SPECTROMETRY (GC/MS) USI NG MICROWAVE-ACCELERATED DERIVATIZATION (MAD)..................................................................................................99 Introduction................................................................................................................... ..........99 Microwave-Accelerated Derivatization (MAD)............................................................100 MAD with Polar Organic Solvents................................................................................102 MAD with Several Derivatization Strategies................................................................102 Experimental Section........................................................................................................... .104 Steroids, Solvents, and Reagents...................................................................................104 Instruments and Apparatus............................................................................................105 Traditional thermal derivatization..........................................................................105 Synthesis microwave system..................................................................................105 Domestic microwave oven.....................................................................................105 Gas chromatography/mass spectrometry (GC/MS)...............................................105 Derivatization Procedures.............................................................................................106 Maximum temperatures..........................................................................................106 Derivatization overview.........................................................................................106 Traditional methods................................................................................................107 Synthesis microwave system vs. traditional heating methods...............................107 Synthesis microwave system vs domestic microwave oven.................................108 Solvent enhancement with MAD...........................................................................108 Microwave heating of plasma extract....................................................................109 Results and Discussion......................................................................................................... 109 Maximum Temperatures of Reagent, Solvent, and Reagent:Solvent Combinations....109 Overview of MAD Analysis..........................................................................................110 MAD with MSTFA.......................................................................................................111 MAD with BSTFA/TMCS............................................................................................113 MAD with BSA.............................................................................................................114 MAD with Methoxylamine/BSTFA/TMCS Strategy...................................................114 MAD with MTBSTFA..................................................................................................115 MAD with Polar Organic Solvents................................................................................115 MAD of Spiked Plasma Extracts...................................................................................117 Conclusion..................................................................................................................... .......118 5 EVALUATION OF DERIVATIZATION STRATEGIES FOR THE COMPREHENSIVE ANALYSIS OF EN DOCRINE DISRUPTING COMPOUNDS USING GAS CHROMATOGRAPHY/ MASS SPECTROMETRY (GC/MS)...................135 Introduction................................................................................................................... ........135 Experimental Section........................................................................................................... .138 Chemicals and Solutions...............................................................................................138 Derivatization Setup......................................................................................................139 Block heating..........................................................................................................140 Microwave heating.................................................................................................140 Gas Chromatography/Mass Spectrometry.....................................................................140 Semi-Quantitative Calibration.......................................................................................141
9 Pilot-Study for Water Analysis.....................................................................................142 Results and Discussion......................................................................................................... 143 Characterization of Potential EDC Compounds............................................................143 Comprehensive EDC Profile.........................................................................................143 Non-Derivatized EDCs..................................................................................................144 Derivatized Steroid EDCs.............................................................................................144 Other Hydroxylated EDCs............................................................................................145 Microwave Derivatization.............................................................................................145 Semi-Quantitative Calibration Analysis........................................................................146 Spiked Water Analysis..................................................................................................147 Lake Apopka Water.......................................................................................................147 Conclusion..................................................................................................................... .......148 6 INVESTIGATION OF ESTROGENS IN ALLIGATOR PLASMA BY GAS CHROMATOGRAPHY/MASS SPECTROMETRY (GC/MS) COUPLED WITH SOLID-PHASE MICROEXTRACTION ON-FIBER DERIVATIZATION (SPMEOFD)........................................................................................................................... ..........159 Introduction................................................................................................................... ........159 Steroid Analysis by IA-Based Techniques....................................................................159 Steroid Analysis by Liquid Chroma tography/Tandem Mass Spectrometry (LC/MS/MS)..............................................................................................................160 Steroid Analysis by Gas Chromat ography/Mass Spectrometry (GC/MS)....................160 Analysis of Steroid by Solid-Phase Microe xtraction with On-Fiber Derivatization (SPME-OFD).............................................................................................................161 Experimental Section........................................................................................................... .162 Steroid Standards...........................................................................................................162 Liquid Chromatography/Tandem Mass Spectrometry (LC/MS/MS ) ............................162 Solid-Phase Microextraction with On -Fiber Derivatization (SPME-OFD)..................162 Gas Chromatography/Mass Spectrometry (GC/MS).....................................................163 Analysis of E2-Injected Alligator Plasma......................................................................164 Results and Discussion......................................................................................................... 164 Overall SPME-OFD GC/MS Procedure........................................................................164 SPME-OFD GC/MS Calibration...................................................................................164 Differences Between RIA and GC/MS (with SPME-OFD)..........................................165 Confirmation with LC/MS/MS Analysis.......................................................................166 17 -E2...........................................................................................................................166 Conclusion..................................................................................................................... .......167 7 CONCLUSION AND FUTURE WORK.............................................................................175 Conclusions.................................................................................................................... .......175 Future Work.................................................................................................................... ......178 LIST OF REFERENCES............................................................................................................. 179 BIOGRAPHICAL SKETCH.......................................................................................................207
10 LIST OF TABLES Table page 2-1 Relative extraction effi ciencies (REE) for steroid/EDC mixture in water using the selected extraction strategies..............................................................................................72 2-2 Relative extraction effici encies (REE) for steroid/EDC mixture in plasma using the selected extraction strategies..............................................................................................73 2-3 The R2 values and linear concentration range (in ppb) obtained from SPME analysis of spiked water (without OFD)..............................................................................................74 2-4 Signal-to-noise (S/N) values of the polar steroids/EDCs extracted from spiked plasma (200 ng mL-1) using SPME (with and without OFD)........................................................75 3-1 Targeted derivatives and their ch aracteristic ions and retention times...................................94 3-2 The RRF values obtained using various time, temperature and derivatizing reagent combinations................................................................................................................... ...95 3-3 The RRF changes due to various deri vatization enhancements with BSTFA/TMCS............96 3-4 The RRF changes due to various de rivatization enhancements with MSTFA.......................97 3-5 The RRF changes due to various derivatization enhancements with BSA............................98 4-1 Maximum temperatures (in C) achieved with the synthesis microwave system for reagent and reagent:solvent combinations.......................................................................120 4-2 Targeted derivatives and ions summed to measure peaks areas for all the derivatization reagent strategies............................................................................................................. .121 4-3 Comparison of the RRF values for the synthesis microwave system (MW) normalized to thermal derivatization methods using MSTFA............................................................122 4-4 Percent relative st andard deviation (RSD, in %) of RRF values for each derivatized steroid using the domestic microwave ove n and the synthesis microwave system.........123 4-5 Comparison of the RRF values for the synthesis microwave system (MW) normalized to thermal derivatization methods using BSTFA/TMCS.................................................124 4-6 Comparison of the RRF values for the synthesis microwave system (MW) normalized to thermal derivatization methods using BSA.................................................................125 4-7 Comparison of the RRF values for the s ynthesis microwave system normalized to those for the thermal derivatization methods using MOX and MTBSTFA reagent.................126
11 4-8 The RRF change using the synthesis microwave system, derivatization reagent (BSTFA/TMCS), and organi c solvent (1:1, v/v).............................................................127 4-9 The RRF change in using the synthesis microwave system for steroids spiked into plasma at 200 ng mL-1......................................................................................................128 5-1 Characteristic ions fo r the underivatized endocrine disrupting compounds (EDCs)...........149 5-2 Characteristic ions for the deriva tized endocrine disrupting compounds (EDCs)...............150 5-3 Approximate detection limits using comprehensive EDC method......................................151 6-1 Estimated concentration levels (in ppb) for the E2-injected alligato rs using both RIA and GC/MS (with SPME-OFD) methods........................................................................168 6-2 Estimated percentage of -E2 in comparison to the -E2 in the E2-injected alligators using GC/MS coupled with SPME-OFD.........................................................................169
12 LIST OF FIGURES Figure page 1-1 Examples of endogenous sex steroids....................................................................................45 1-2 Examples of endocrine disrupting compounds (EDCs).........................................................46 1-3 Analytical methodology for the determina tion of steroids and EDCs from plasma and water samples.................................................................................................................. ...47 1-4 Structures of the steroids investigated in this study...............................................................48 1-5 Structures of the EDCs investigated in this study..................................................................49 1-6 Electron ionization (EI) process.......................................................................................... ...50 1-7 Atmospheric pressure chem ical ionization (APCI) process...................................................51 1-8 Liquid-liquid extraction steps............................................................................................ .....52 1-9 Solid-phase extraction steps.............................................................................................. .....53 1-10 Solid-phase microextraction step s (with on-fiber derivatization)........................................54 1-11 Acylation (PFA) and silylation (TMS) reaction mechanisms for 17 -estradiol..................55 2-1 Liquid-liquid cartridge ex traction (LLCE) procedure............................................................76 2-2 Peak area ratios of the non-polar chlorina ted EDCs in spiked water were obtained by dividing the peak area of the analyte by th e peak area of the surrogate, anthracene.........77 2-3 Peak area ratios of the polar EDCs in sp iked water were obtaine d by dividing the peak area of the analyte by the peak ar ea of the surrogate, anthracene......................................78 2-4 Peak area ratios of the remaining EDCs in spiked water were obtained by dividing the peak area of the analyte by the peak area of the surrogate, anthracene.............................79 2-5 Peak area ratios of the steroids in spiked water were obtained by dividing the peak area of the analyte by the peak area of the surrogate, anthracene.............................................80 2-6 Peak area ratios of the polar EDCs in spik ed plasma were obtaine d by dividing the peak area of the analyte by the peak ar ea of the surrogate, anthracene......................................81 2-7 Deteriorated SPME fiber. The extraction phase on the se ction indicated deteriorated over several extractions (with OFD)..................................................................................82 4-1 Strategy for two-step methoxime/t rimethylsilyl (MO/TMS) derivatization........................129
13 4-2 Strategy for derivatization using the MTBSTFA reagent....................................................130 4-3 Overhead view of the S-Class synthesis microwave system................................................131 4-4 Absolute relative response factors (RRF s) for MSTFA (A) and BSTFA/TMCS (B) for both the synthesis microwave system (CEM at 300 W) and the domestic microwave oven (DOM, at 300 and 1000 W)....................................................................................132 4-5 Absolute RRF values for BSTFA/TMCS at different microwave reaction times using the synthesis microwave system (at 300 W)....................................................................133 4-6 Absolute RRF values for BSTFA/TMCS with varying amounts of ACN using the the synthesis microwave system (at 300 W) for one minute.................................................134 5-1 Chromatograms of comprehensive ED C profile with (a) no derivatization, (b) derivatization using block h eating of 70 C for 30 minut es, and (c) derivatization using microwave heating 900 watts for 1 minute............................................................152 5-2 Relative response factors (RRFs) for the underivatized (GC -ready) EDCs in the comprehensive profile with and without the presence of derivatizing reagent (each at a concentration of 5 g mL-1)..........................................................................................154 5-3 Relative response factors (RRFs) for the derivatized steroi ds with and without derivatization (each at a concentration of 5 g mL-1)......................................................155 5-4 Relative response factors (RRFs) for the other derivatized polar EDCs with and without derivatization (each at a concentration of 5 g mL-1)......................................................156 5-5 Chromatogram of 100 ppb (0.1 g mL-1) EDC mixture spiked in extracted Lake Apopka water................................................................................................................... 157 5-6 Chromatogram of several potential ED Cs found in Lake A popka water (FWL1874).........158 6-1 Calibration curve for E1 using the SPME-OFD method with GC/MS.................................170 6-2 Calibration curve for 17 -E2 using the SPME-OFD method with GC/MS.........................171 6-3 Chromatogram of endogenous steroids (a nd vitamin E) in the SPME-OFD extract of E2injected alligator plasma..................................................................................................172 6-4 LC/MS/MS chromatogram and mass spectrum of E2 detected in E2-injected alligator plasma......................................................................................................................... .....173 6-5 Chromatogram and mass spectra for the two di-TMS-E2 peaks ( and -E2). B...............174
14 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy NOVEL ANALYTICAL STRATEGIES FOR THE CHARACTERIZATION OF STEROIDS AND ENDOCRINE DISRUPTING COMPOUNDS (EDCS) IN BIOLOGICAL AND ENVIRONMENTAL SAMPLES By John A. Bowden May 2009 Chair: Richard A. Yost Major: Chemistry Ongoing research has demonstrat ed that specific environmental contaminants, commonly referred to as endocrine disrupt ing compounds (EDCs), can alter the functioning of the endocrine system by affecting normal growth, development and reproduction. In cons equence, there is a fundamental need for innovative methods to improve the characterization of EDCs and their role in EDC-induced alterations in wildlife, specifically the change s in endogenous steroid profiles. The diverse physiochemical properties of ster oids and EDCs impose inherent analytical limitations and thus current an alytical methodologies for mon itoring these compounds are often limited in scope and scale, limiting the overall u nderstanding of contaminant-steroidogenesis interactions. New strategies that can simultaneously analy ze a wide range of endoge nous steroids in one chromatographic analysis would be a signifi cant improvement over current methods. In the research reported here, application of chroma tographic techniques such as gas and liquid chromatography, coupled with mass spectrome try (GC/MS, LC/MS), are shown to offer promising profiling capabilities in increasing the number of co mpounds detected and quantified. Several innovative sample prepar ation strategies, i nvolving extraction and derivatization, were
15 optimized and tailored to expand the dynamic range of compounds capable of being analyzed in a single analysis. In particular, novel extracti on (solid-phase extraction (SPE) with several sorbents and solid-phase microextraction (SPME) with on-fiber derivatization) and derivatization (including microwave and solvent enhancement) strategies were explored and optimized. Sample preparation techniques were evaluated and comp ared in terms of yield, reproducibility, and overall analysis time. The overall profiling schemes were effective fo r the characterization of a diverse suite of environmental EDCs in surface water samples and st eroids in alligator plasma. The analysis of endogenous steroids in the American alligator, in concert with th e detection of EDCs from its natural (indigenous) water environment, provides a unique opportunity to correlate changes in the normal endocrine milieu in response to possible EDC exposure.
16 CHAPTER 1 INTRODUCTION The American alligator ( Alligator mississippiensis ) has exhibited altered steroidogenesis when exposed to certain anthropogenic com pounds, commonly referred to as endocrine disrupting compounds (EDCs). Present research is aimed to help understand the contaminantsteroidogenesis interactions and the physiological level at whic h these contaminants can alter the functioning of the endocrine system. Current resear ch strategies to monito r both the presence (in the environment) and activity (in an organism) of EDCs often focus on the analysis of small groups of chemically similar compounds. Thus, new techniques capable of characterizing on a broader scope the EDCs in the environment and the EDC-induced impacts on wildlife would be an improvement over current methods. In this study, the application of chromatographic techniques, coupled with mass spectrometry, offe r promising capabilities for increasing the number of compounds (EDCs and steroids) analyzed in a single analysis In addition, innovative sample preparation strategies were optimized and tailored to increase the selectivity and reduce the analysis time of the overall methods. The developed methods were used in the characterization of steroids and EDCs in biol ogical and environmental samples, respectively. Overview of Endocrine Disruption Background on Endocrine Disruption The basic role of the endocrin e system is to regulate physio logical processes in the body including growth, development, and reproduc tion. The mechanism by which the endocrine system regulates and maintains these processes is by releasing chemical messengers, also called hormones.1, 2 As they are secreted, these hormones ar e transported (typica lly protein-bound) to various destinations throughout the body. At the ta rget cell, hormones bind to specific receptors as they diffuse through the cell membrane.2-4 The hormone-receptor complex signals specific
17 biological actions, including the growth and development of specific tissues.4-6 Disruption in normal endocrine activity can occur with exposur e to certain environmental chemicals, often resulting in atypical growth, development, and reproduction.4, 7-9 These contaminants, often referred to as endocrine di srupting compounds (EDCs), have been defined by the US Environmental Protection Agency (EPA) as, Â“exogenous agents that interfer e with the synthesis, secretion, transport, binding act ion, or elimination of natural hormones in the body that are responsible for the maintenance of homeostasis, reproduction, development, and/or behaviorÂ”.10 As early as the 1950s, evidence of endocrine di sruption originated from observations of reproductive failure, inappropriate behavior, and population declin e in wildlife including fish, birds, reptiles, and mammals.11-15 Through the progression of applied science, examples of endocrine disruption have often been linked to the effects of anthropogenic compounds. A noteworthy example of the disrup tive effects of anthropogenic compounds on wildlife has been well-documented with the pesticide, dichlorodiph enyltrichloroethane (DDT). The benefits of DDT, as an agent to avoid disease-carrying insect s, were first realized by Paul Muller in 1939.11, 16, 17 Subsequently, DDT was heavily applied as a pesticide in an effort to increase the productivity of agricultural practices; upwards of two million tons globa lly since its inception.16 Numerous examples of endocrine disruption have been associated with DDT (and its analogs), including abnormal reproductive developm ent and several types of cancers.18, 19 Coincidentally, at same time that DDT was first being marketed as a pesticide, another chemical named diethylstilbestrol (DES) was bei ng proposed as a promising pharmaceutical. In contrast to DDT, DES was developed for dire ct human treatment. First introduced in 1938 by Dodds and coworkers, DES was prescribed to ai d in the prevention of pregnancy complications, most notably spontaneous abortions.20-23 During its tenure, DES was prescribed to over five
18 million women, many of whom took the drug as a precautionary measure.11, 21 It was later discovered that not only was DES ineffectiv e in avoiding pregnancy complications, but it increased cancer-related occurrences in the offspring of medicated mothers.20-24 Laboratorybased studies have substantiated the effects of human exposure to DES, as the offspring of mice exposed to DES exhibited an increase in reproductive abnormalities and tumors.8, 25-27 To date, endocrine disruption by DES is the clearest ex ample of the human consequences sustained through to exposure to hormone-like compounds. Several books7, 11, 24, 28 and workshops/seminars10, 29, 30 have focused on bringing the issue of endocrine disruption to the forefront of di scussion on environmental and human health. These discussions illustrate a fundamental need for a better understanding of the cause/effect relationship in which EDCs exert disruptive effects on both wildlife and humans. Initially, endocrine disruptors were thought to interfere solely with estrogens (ex. 17 -estradiol, shown in Figure 1-1), by impeding normal estrogen upkeep or binding to and limiting the activity of estrogen receptors (ER or ER ). However, recent studies have shown that EDCs can also have a deleterious effect on other mechanisms of th e body, including the disrup tion of the thyroid, the immune and nervous systems, and other horm ones such as androgens and progestins (e.g. testosterone and progesterone, re spectively, shown in Figure 1-1).3, 5, 10, 24, 31, 32 Mechanisms of Endocrine Disruption The preliminary concepts regarding the mech anisms by which EDCs can disrupt normal hormonal activity were first realized from re search done by Soto and Sonnenschein in the 1980s.11, 32-34 During their research, an unknown contam inant was interfering by proliferating breast cancer cells, a normal activ ity of endogenous estrogens. It was later discovered that this contaminant was p -nonylphenol, an ingredient of plastic labware.11, 32-34 A similar result was
19 obtained in the work done by Feldman, in which th e discovery of another plastic constituent, bisphenol A, had interfered w ith the normal binding of estrog ens to a receptor in yeast.11, 32, 34, 35 The important implications from these studies gave rise to a new lin e of thinking about the mechanisms by which environmental contamin ants can interfere with normal hormonal processes. It should also be not ed that these studies also serv ed notice to the increasingly unexpected origins of EDC exposure. Since these early studies, several mechanisms of endocrine disruption have been proposed, most notably in the alteration of 1) steroidogeni c enzymes, 2) normal hormone homeostasis, 3) and the binding of hormones to receptors.4, 10, 20, 32, 36, 37 Of particular interest are the EDCs that disrupt normal reproduction by competing with e ndogenous sex steroids for receptors, by either agonistic or antagonistic actions.4, 9, 20 An example is DDT, which exerts disruptive effects of on the endocrine system through its ability to act as an estrogen agonist by binding to and altering the activity of the estrogen receptor.11, 14, 18 Antagonists, such as the DDT degradation product dichlorodiphenyldichloro ethylene (DDE), can act by blocking or inhibiting the receptors.11, 14, 18, 38, 39 In addition, recent research has also demonstrated that some EDCs have the capability to be active through multiple mechanisms.5, 9, 32, 40 Types of Endocrine Disrupting Compounds (EDCs) The ubiquitous nature of EDCs in a variety of matrices, including water, soil, air, and food,24, 36, 41 is of direct consequence to the varied applications of intended usage. EDC usage ranges from agricultural and industrial practices, to an assortment of chemicals found in many household and personal care products.42-45 The agricultural industry, a major source of EDC pollution, is responsible for the in troduction of several pesticides, herbicides, and fungicides into the environment in several cases from thousands to millions of tons per year.46 Moreover, DDT and many other organochlorine pesticides (i ncluding dicofol, endosulfan, and several DDT
20 byproducts) have since shown th e potential to disrupt norma l estrogenic, androgenic and thyroidal functions.3, 31, 38, 39, 47, 48 Several other widely used compounds, including herbicides (atrazine), fungicides (vinclozo lin), flame retardants (polychl orinated biphenyls, PCBs), plasticizers (phthalates, bisphe nol A), surfactants (nonylphenol), and natural/synthetic hormones (ethinylestradiol), are also suspected endocri ne disruptors; these are shown in Figure 1-2.3, 10, 24, 31, 47, 48 It is troublesome that only a few of the 80,000 registered chemicals for practical usage have been tested as endocrine disruptors.30 EDC determination is complicated by the fact that there is little evidence correla ting EDC-induced actions to a corresponding chemical property. Furthermore, as shown in Figure 1-2, EDCs typically exhibit little resemblance to the endogenous sex steroi ds they disrupt. Endocrine Disruption Effects Consequences of Environmental Exposure Studies have demonstrated reproductive dysfunc tion in wildlife on several tiers of the food chain. Case studies of fish expos ed to EDCs have been presen ted, including turbots exhibiting altered hormone levels when exposed to ppt (pg mL-1) levels of bisphenol A and nonylphenol,49 numerous fish exhibiting incr eased neoplasia when exposed to PCBs and pesticides,13 carp exhibiting altered hormone levels when ex amined near a sewage treatment plant,50 and mosquitofish exhibiting decreased sperm and horm one levels when exposed to organochlorine pesticides.51, 52 It should be noted that even though lowe r level aquatic organisms, such as fish, routinely have to deal with contaminants di rectly from sewage treatment plant effluent, agricultural runoffs, and industria l waste, organisms which consum e these fish (birds, mammals, and humans) often encounter these same contaminan ts in a larger dosage due to bioaccumulation and biomagnification (upwards of a million times or more of the original exposure11, 12, 15, 24).
21 Accordingly, many studies have attempted to de termine the endocrine disruption effects on these organisms at the higher end of the food chain, including case studies i nvolving the population decline and abnormal sexual develo pment of several avian species,11, 53 altered sexual development and sex reversal fo r several reptiles and amphibians,54-57 and the population decline and suppressed immune and reproductive systems of several marine mammals.12, 13 Furthermore, case studies with rodents and pigs exposed to DES and nonyphenol, respectively, have illustrated that many of these EDCs can be carcinogenic a nd have the capability of creating abnormal effects over several generations.58-60 The Alligator Model Ongoing research in the Guillette research gr oup at the University of Florida has been directed towards understanding the mechanis ms by which the American alligator ( Alligator mississippiensis ) manages exposure to EDCs and the consequential EDC-induced ramifications. The American alligator, along with Lake Apopka (FL), have been ideal models for examining the relationship between wildlif e and EDC exposure. Lake Apopka a large lake in central Florida, has experienced contamination thr ough various entry points, including extensive agricultural and industrial waste. Lake Apopka also suffered a large chemical spill of dicofol and DDT (and its analogs) in 1980,52, 61, 62 which may have been a factor in the population decline of alligators in the 1980s.62, 63 The American alligator is an important sp ecies for accessing both environmental exposure and the resulting consequences to wildlife and ul timately humans. On the surface, alligators, like humans, are top-of-the-food chai n predators and experience si milar consequences to the bioaccumulation and biomagnification of EDCs in various fatty tissues and food webs.11, 48, 64 Physiologically, alligators are unique in that th ey are temperature sex dependent (TSD) animals, thus the sex of the offspring is determined by the temperature in which the eggs are incubated.65
22 The theory of TSD also includes the concept that other environmental f actors (such as exposure to EDCs) can exert an effect on sex determina tion. Thus, the consequential outcome of EDC exposure can be readily monito red. Studies have shown that exposure of estrogenic compounds to alligator eggs at key intervals can result in sex-reversed alligators.65, 66 Furthermore, the endocrine system has been consider ed to be evolutionarily static;11 thus, as the same steroids are responsible for similar endocrine actions acro ss animal species, so should the mechanisms by which normal endocrine processes be altered. Several key perspectives in the understa nding of endocrine disruption have been introduced through examinations of the American alligator, including th e significance of timing in endocrine exposure, the mechanisms by which endocrine exposure occurs, and the consequential endpoint results from endocrine e xposure. Through several case studies, timing of the exposure to EDCs has shown to have a pr ofound effect. Exposure to EDCs during embryonic and neonatal stages has been shown to result in permanent deviation from normal reproductive growth and developmental processes.65, 67-70 Furthermore, the mechanisms by which EDCs affect alligators have been pinpointed towards two central themes, the disr uption of steroidogenic enzymes and the alteration of hormone/receptor binding.66, 71-73 In comparison studies to alligators sampled from contro lled environments (low contamination), several reproductive anomalies have been demonstrated in alligators examined from contaminated lakes, including altered sex steroid concentrations,67, 68, 70, 73-75 morphological irregularities,67, 68, 70, 73-75 and low clutch viability.76 Moreover, EDCs have been shown to have an effect in other aspects, including the skeletal,77 immune,78 and nervous system.69 The Human Impact The number of human reproductive disord ers has dramatically risen since 1970,11, 21, 79 prompting the scientific community to suggest the possible activity of EDCs to be the
23 explanation. A decline in sperm quality and qua ntity has been the most significant result and increases in testicular, prostate ovarian, and breast cancers have also been observed and may be linked to EDC exposure.10, 20, 21, 24, 80, 81 Recent data suggest that because of their ubiquitous nature, the general public carries and is exposed to physiologically signi ficant levels of EDCs.30, 81 Moreover, several reports have documented the potential of EDCs to negatively affect human populations, including an increase in occurrences of preterm birth, and behavioral, reproductive, and immune dysfunction.20, 24, 30, 82 Human exposure to EDCs can occur directly or indirectly. Direct exposure is typically mediated by a compound such as anabolic or phar maceutical steroids designed to elicit a specific biological response. Anabolic st eroids have been used to e nhance athletic performance by increasing muscle strength and size; however, th ey have also been f ound to disrupt a host of other physiological processes.2, 83-85 Pharmaceutical steroids, such as the active ingredient in contraceptive pills (ethinylestradiol Â– EE2), have been used to prevent pregnancy by inhibiting the normal pregnancy processes.86 Pregnancy-related steroids (such as EE2) have been shown to have many beneficial qualities; however, there have been instances in which these compounds have promoted endocrine disruptive actions, most notably, in the case of diethylstilbestrol (DES). Although there is still mu ch to be learned about the illeffects of direct EDC exposure, indirect exposure is the more challenging concern to humans. I ndirect exposure can occur from unintended contact with many of the pesticides, PCBs, phenols, and other EDCs previously mentioned.20, 24, 30, 82 Timing of exposure is widely viewed as the mo st important factor in terms of endocrine disruption, as certain developmental stages (prenatal, neonatal, and pubertal) are more susceptible to disruption.10, 21, 24, 30, 36, 59, 80 During these developmental periods, permanent
24 organizational and developmental processes ar e undertaking and can be altered through the disruptive actions of EDCs.63 In addition, specific requirement s of hormone levels and actions are required at specific intervals for normal growth and development.22, 24, 59, 80, 87 Young women in Mexico living near highly mode rnized agricultural practices had altered breast development when compared to women living in rural areas.88 Studies have also show n that children can be exposed to other EDCs through dental sealants (bisphenol A),24, 89 baby products (phthalates),90 and even through human breast milk (DDT, PCBs).24, 91 Common links of endocrine disruption has been conceptualized in humans; however, li ttle evidence has been presented which fully explains these hypotheses. Controversy of Endocrine Disruption The theory behind human endocrine disruption is currently incomplete, mostly due to the complicated nature of EDCs and the limited amount of definitive information regarding their action. These factors have lead to speculation abou t the capability of EDCs to exert a destructive influence on humans. Skeptics point to several inherent factors of most EDCs, such as the low potency in comparison to natura l sex steroids, the environmenta lly low doses exhibited from natural occurring exposure, and the fact that most EDCs are found below the No Observed Adverse Effect Level (NOAEL).11, 34, 92-94 The low measured potency of some highly publicized EDCs has created some skepticism as to wh ether these EDCs are potent enough to exert a response over the much more potent endogenous estrogens.11, 34, 93, 95 The estrogenic potency of many EDCs, like DDT and nonylphenol, in some cases has revealed estrogenic potency levels close to a million times lower than that of natural estrogens.20 Furthermore, many laboratorybased studies that measured an endocrine disr uptive response were performed at high dosages, leading skeptics to argue the realistic probability of endocrine disrupti on at environmentally
25 active concentrations was minimal.92 Even further, exposure to these low environmental levels should not reflect a significant response, accord ing to current toxicology principles. Although there is validity to many of these st atements, several relevant factors are not considered. The ability of an EDC to affect an organism can be influenced by several factors, such as age, sex, species, dosage, and duration of exposure.4, 20 Although the potency of some EDCs are low in comparison to natural sex steroi ds, several field and la boratory-based studies have shown that the seemingly low potent EDCs can still exert a dramatic endocrine response.11, 92 Other influential factors, such as bioaccumula tion, timing of exposure, and concentration/halflife of the EDCs can also increase the potency of a compound.11, 38, 46, 64, 82 Furthermore, the consequences from EDC exposure often under go long latency periods and exposure is not reported until later stages in life.10, 20, 36 In combination with the fact that EDCs can bioaccumulate and biomagnify, an underlying proposition is that within the framework of an industrialized country, EDC exposure is chronic and is never isolated.36 Thus, as skeptics have assert ed that environmentally active concentrations may not be significant; mixtures of EDCs may impose a deeper threat. Several studies have examined the additive and synerg istic nature of EDCs and demonstrated that mixtures of EDCs can have unpredictable consequences.10, 20, 36 Furthermore, many degradation products (ex. DDE) of known EDCs (DDT) ha ve been shown to be as potent and environmentally persistent as the parent compound.3, 38, 39 The toxicology profiles of many EDCs do not follow standard protocols. It has been shown that many EDCs actually exert a greater effect at smaller concentrations than at the higher concentrations that are tested in laboratory-based studies.92, 94 Since most sex steroids are protein-bound until they are released for activity, the active and free steroids are typically present
26 in low concentrations. Moreove r, the EDC compounds, which te nd to bioaccumulate, are often not protein bound, and thus are frequently more available to cause disruption.15, 46, 82 Indeed, in many cases, any exposure to EDCs can be consid ered enough to exert a biological response. The lack of a well-defined cause/effect re lationship between EDC exposure and the EDCmediated responses has been a major driving force for continued research. Moreover, to understand EDC effects, bette r associations between obs ervations in wildlife ( in vivo ) and in research laboratories ( in vitro ) need to be made.3, 20 Nevertheless, the persistent and ubiquitous nature of EDCs illustrates the eminent need to address the specific human impacts. Strategies for Endocrine Disrupting Characterization Issues Regarding Endocrine Activity Of the thousands of chemicals used for da ily household, industri al, and agricultural practices, only a miniscule fraction have been test ed for estrogenicity (or capability to exert an estrogen action).11, 96 Although not a focal point of this i nvestigation, research involved in measuring endocrine activity (est rogenicity, androgenicity) must be addressed. At the forefront of any discussion regarding whether a speci fic compound possesses e ndocrine disruptive capabilities lie examinations from specific assays designed to di scriminate compounds that have either estrogenic or androgeni c activity. The E-SCREEN is one of the more popular methods to determine the estrogenic ity or estrogenic potency of suspect compounds.79, 97 These methods use estrogen-sensitive cells in vitro to measure the consequential cell proliferation (a role of natural estrogens) achieved when in the presence of estrogen-like compounds.20, 79, 98, 99 The estrogenic (or androgenic) property of compounds is only one link in the characterization of endocrine disruption.
27 Detection of Environmental EDCs Connected to the debate of whether a compound possesses estrogenic (or androgenic) capabilities lies the determinati on of where these potential EDCs end up in the environment. It has been previously stated that these environm ental EDCs have dramatic effects on wildlife, since animals (fish, birds, reptiles, mammals, and ultimately humans) rely on the environment for food and water. Accordingly, th e analysis of water has been th e most widely studied matrix. These studies involve measuring both endoge nous (estradiol and es trone) and exogenous hormones (DES and EE2), alkylphenols (nonylphenol and octylphenol) and bisphenol A, in effluent derived from sewage (STPs) and wastew ater treatment plants (WWTPs) that end up at physiologically high levels (ng Â– g L-1) in nearby rivers, ground and surface water, sediment and sewage sludge due to ineffective treatment.100-124 The other large source of water contamination is from the excretion of hormone s from concentrated animal feeding operations (CAFOs).125-127 It has been stated that about 90% of the current endogenous hormonal contamination is from livestock excretion, as the excrement is often flushed into nearby waterways or made into manure for agricultural purposes, which eventually integrates into agricultural runoffs.125, 127 Concerns over water cont amination also arise from the fact that EDCs can be persistent and degrade into othe r products that may be unknown and harmful.46, 127 Furthermore, a direct impact to humans is realiz ed by the fact that some of the contaminated water is destined for human consumption.127 Finally, water is not the only route of EDC exposure. Several studies have shown EDCs can be found in relevant concentrations in soil,44, 128 indoor air and dust,44, 129 and even food.130-134 The recent interest for analyzing environmenta l EDCs has been to correlate results from chromatographic techniques (detection of EDCs ) with bioassays (measure hormonal activity).103,
28 111, 113, 135 The most common strategies involve coupli ng the chromatographic techniques (liquid chromatography (LC) and gas chromatography (G C)) with mass spectrometry (MS) and tandem mass spectrometry (MS/MS). Gas chromatography/mass spectrometry (GC/MS) has been a powerful technique for analyzing EDCs for many years due to its ability to provide excellent chromatographic resolution and the fact that ma ny potential EDCs are GC -ready (that is, volatile and thermally stable). GC/MS, with electron io nization (EI), also has extensive mass spectral libraries that aid in the id entification of known and unknown ED Cs. GC/MS has been routinely used for the analysis of many organochlori nes (DDT, DDE, methoxychlor, endosulfan, etc.),101, 116, 129-131, 136, 137 PCBs,129 phthalates,129 and other EDCs (atrazine).101, 131, 137, 138 However, the analysis of polar, non-volatile EDCs, typically require deriva tization for GC analysis. The application of derivatization for the analysis of polar EDCs has been performed for the steroidal hormones,103, 104, 106, 108, 110, 115, 119-126, 135, 139, 140 alkylphenols,103, 108, 110, 114, 118, 119, 122, 124, 129, 135, 139, 140 and bisphenol A,103, 108, 110, 114, 118, 121, 122, 135, 139, 140 but these methods of ten require extensive sample preparation procedures. Due to the extra sample preparation steps re quired for polar EDC analysis by GC/MS, an impetus towards employing LC/MS has surfaced. LC/MS, employing both positive and negative mode electrospray (ESI) and atmospheric pressu re chemical ionization (APCI) techniques, has been used to analyze polar EDCs including the steroidal hormones,105, 107, 111-113, 116, 117, 139, 141 alkylphenols,100, 111-113, 117, 141 bisphenol A,100, 112, 113, 117, 139, 141 and the polar degradation products of many of the non-polar EDCs.116, 129, 136 The detection of polar degr adation products of many of the non-polar EDCs has gained recen t attention, most likely due to th e lack of char acterization of these compounds. In addition, a few reports have used LC/MS for non-polar EDCs.102, 112, 142, 143
29 Detection of Steroids and EDC-Induced Responses The analysis of steroids with in an organism can be define d by two disciplines: 1) the detection of endogenous or exogenou s anabolic steroids for enhan cement in sports and livestock production, and 2) the clinical screening of endogenous steroids for the charac terization of endocrine disorders. Both of these applications will be discussed in detail. At the 1976 Olympics in Montreal, the Inte rnational Olympic Committee (IOC) banned the use of anabolic steroids prior to competitions.144-147 Furthermore, several regulatory organizations have identified and banned several growth promoters in the production of livestock destined for human consumption.148 The application of anabolic st eroids has been shown to have several endocrine disrupting impacts, even though the mechanisms by which these compounds work are still not completely understood.84, 149 In fact, many of the currently used compounds for illicit purposes were initially designed for medicinal use of se veral hormone-related disorders.83, 150 Initial detection of anabolic compounds by the IOC was done by immunoassay-based techniques (IA).145, 147 Since then, GC/MS, capable of multi-residue analysis, has become the method of choice for the detection of steroids of abuse for human competition,144, 146, 147, 151-155 equine racing and breeding,156-158 and livestock production.159-165 These strategies are centered on the analysis of urine, and often require en zymatic hydrolysis, extrac tion, and derivatization procedures. These extensive sample preparation techniques, along with th e increasing synthesis and discovery of new anabolic agents, have cr eated an impetus toward utilizing methods that require less sample preparation, such as LC/MS (and LC/MS/MS).85, 150, 166-171 Due to the numerous ways an EDC can enter an d affect an organism, interest in methods capable of routinely characterizing endogenous ster oids are needed and would be an important tool in understanding both metabolic response and the mechanisms by which normal endocrine
30 function can be altered. Several strategies exist for the determ ination of endogenous steroids, including target analysis (one analyte), metabolic profiling (one class or several closely related analytes), and fingerprinting (trends de rived amongst all analytes separated).172 A commonly employed target method for the analys is of endogenous steroids is the use of IA techniques. IA-based techni ques, such as radio immunoassay (RIA), have frequently been used to detect specific steroids in several animal61, 173-177 and clinical studies.6, 178-182 IA-based techniques are based on the determination of st eroid concentrations by measuring the binding of specific antigens (steroid) to specific antibodies (protein).181 The high specificity and sensitivity of these techniques (detection at pg mL-1),174 has made them very popular among biologists and endocrinologists; however, these techniques often suffer from fa lse-positives (due to crossreactivity),6, 182-185 variability in commercial IA kits,185, 186 and the limited availability of antibody types that hampers unknown analysis.184 Furthermore, IA techniques are predominantly target-based methods, thus limiting the ability to study steroids on a comprehensive level. Despite its popularity and routin e use for the determination of physiological alterations of steroids, chemists and clinicians typically em ploy both GC/MS and LC/MS techniques due to the expansion of components that can be analyzed in a single inje ction. GC/MS for many years has been considered the preferred method for the detection of steroid profiles. Although derivatization is of ten required, which can be time-consuming, GC/MS has unsurpassed separation capability for analyzing a wide range of analytes and steroid classes.154, 187-192 LC/MS has more recently been used in an effort to avoid the tedious GC-required sample preparation techniques. Although in the screen ing of endocrine disorders, LC /MS has often been considered more of a complimentary technique.6 The difficulty for steroid prof iling with LC/MS is two-fold, 1) LC/MS does not provide the same chromatogr aphic resolving power as GC/MS, and 2) the
31 predominant ionization mechanisms target steroi ds of different polarities (APCI, nonto midpolar, and ESI, polar).6, 185, 193, 194 The use of steroid profiles for the diagnosis of endocrine-related prob lems has created an overlap of both GC and LC analysis. Both tec hniques have been utilized for the diagnosis of several endocrine-related disorders or dis eases, including congenita l adrenal hyperplasia (CAH),6, 154, 179, 187, 188, 195, 196 CushingÂ’s syndrome,6, 182, 187, 188, 195, 197 adrenal/gonadal tumors,182, 187, 195, 198 and several other androgenic a nd estrogenic related disorders.6, 154, 182, 187, 188, 190, 191, 195, 199-204 Of particular interest are those methods that are designed for newborn screening, since endocrine disruption is most prev alent during developmental stages.179, 187, 191, 204 Steroid profiles have also been used to monitor stress levels,182, 205 aging,201 and the influence of drugs, such as antibiotics, on an individual.206, 207 The monitoring of endocrine-related disorders is predicated on the fact that endocrine disruption is a physiological proces s that affects specific hormones and tissues. Several theories on breast and prostate cancer focus on the over-exposure of hormones to these tissues.24, 198, 202, 208, 209 Accordingly, hormonal therapeutic treatments have created a need for monitoring the physiological ramifications of these actions. Although the ramifications of EDC exposure to wildlife have shown the potential to disrupt normal steroidogene sis, few cause/effect relationships have been established. Furtherm ore, human examinations of the effect of environmental EDC exposure on endoge nous steroid profiles are limited. For the effective detection and identification of both EDCs and steroids in environmental samples, purification by extrac tion is often required. Sample purification, also known as extraction, is utilized to pre-concentrate the desired analytes while removing unwanted contaminants from the sample. Since most EDC analysis is based on the analysis of water,
32 simple solid-phase extraction ( SPE) techniques are often empl oyed (except for cases involving the analysis of sludge or sedi ment that employ extra sonicati on steps). SPE procedures are usually performed by employing specific el ution strategies with either C18102, 105, 106, 112, 120, 124, 210-214 or polymeric based cartridges.106, 116, 119, 122, 124, 140, 212, 213, 215-221 Other methods, including solid-phase microextraction (SPME)104, 222 and stir-bar sorpti ve extraction (SBSE)223 have also been used. Extraction of free steroids (endogenous or exogenous) from biological samples (urine, plasma) is more complicated. The most simplistic extraction of steroids from biological samples is liquid-liquid extraction (LLE). LLE methods typically employ diet hyl ether, dichloromethane, or methyl tert -butyl ether as extraction solvents.113, 147, 150, 156, 168, 197, 198, 202, 224-228 Despite the simplistic and relatively efficient extraction met hods of LLE, the application of SPE for steroid extraction in biological samples ha s dominated. Application of C18164, 229-237 and polymeric based cartridges238-240 have been widely used and have allowe d for the investigation of steroids at lower physiological concentrations. As with ED Cs, several other extraction mechanisms have been investigation, including SPME222, 241, 242 and SBSE.223 Other matrices, including meat, feces, and hair have also been analyzed a nd involve various combinations of extraction procedures.153, 177 Ultimately, for the effective comprehe nsive characterization of steroids and EDCs in biological and environmental samples, sample preparation strategies need to be optimized. Background on EDC/Steroid Ch aracterization Strategies Analytical Methodology In this study, analysis of biological (alligat or plasma) and environmental (surface water) samples follows the methodology as described in Figure 1-3. A diverse suite of steroids, differing in chemical structur e and physiological function, is analyzed by both GC/MS and
33 LC/MS techniques. GC was first intr oduced in 1952 by James and Martin.188, 243 LC was first successfully implemented by several independe nt parties in the late 1960s/early 1970s.244 The first and very critical step s in the analysis of steroids by chromatographic techniques is the isolation of steroids from sample matri ces by extraction, which is needed for both GC/MS and LC/MS. LLE, SPE, and SPME techniques were ex amined and optimized for the isolation of steroids from both plasma and water. The second step, for GC/MS analysis only, required derivatization procedures to increase their chromatographic and detection properties. To minimize the drawbacks of derivatization, seve ral methods to improve and/or enhance the derivatization process were explored, including: 1) the optimization of conditions for multi-class derivatization, 2) employment of on-fiber derivatization for SPME, and 3) the investigation of microwave heating to increase derivatization reac tion rates. The final two steps involved the analysis by both GC/MS and LC/MS. Comparisons between GC/MS and LC/MS analyses were addressed in the detectio n and quantification of steroids in pl asma. A list of the chosen steroids (both endogenous and exogenous) used for method development can be seen in Figure 1-4. In addition to developing a steroid profile, the development of a comprehensive method for the analysis of EDCs only by GC/MS is also described. Both polar a nd non-polar EDCs are separated and detected in a single chromatogr aphic analysis by employing novel extraction and derivatization procedures. EDC development fo llows the same analytical methodology as the steroids (Figure 1-3), for the determination of EDCs in water by GC/MS. The EDCs analyzed for the comprehensive profile are shown in Figure 1-5. Finally, interpretation of data will include accessing correlat ions between EDC content in water to the corresponding steroid levels found in plasma from alligators living in contaminated water.
34 Gas Chromatography Gas chromatography is a chromatographic tec hnique that is based on the separation of analytes in the gas phase using a packed capillary column. Thus, for an analyte to be analyzed by GC, it is critical for the analytes to possess su fficient volatility and thermal stability. As the sample is introduced and volatilized in the high-t emperature injection por t (typically 200 C or higher), the sample is focused onto the head of the column by a flow of helium. The introduction of the sample onto the GC column is critical, as there is a need to maximize the amount of analyte onto the column while minimizing sample loss.188, 245 Specific injection types, such as splitless injection, can be implemented to allow imp roved trace analysis by directing the bulk of the sample onto the column.188, 245 As the sample enters the column, the analyt es within the sample experience specific interactions with the stationary phase coated on the inside of the GC column. A typical GC column is coated with 5% phenyl and 95% dime thylpolysiloxane. As the analytes are pushed through the column by the mobile phase (helium) and as the oven temperature is ramped, the stationary phase holds the unique ca pability to interact with the an alytes. Over the length of the column (30 m), each analyte is retained based on their individual affinities to the stationary phase. Components with weak affinities for the stationary phase elute more rapidly, followed by those components which have greater affinity to th e stationary phase. The rate at which analytes are separated and eluted can also be influen ced by mobile phase velocity and temperature programming. Since the column is housed in an oven, a temperature program (temperature ramp) can be implemented to effectively speed up or sl ow down the rate by which the analytes interact with the stationary phase.188, 243, 245 Since the temperature of the column is much lower than the injection port, a temperature progr am (increase in oven temperature) can be used to re-evaporate the analytes as they are introduced and tr avel through the capillar y column. Temperature
35 programming can be used to improve separation and reduce overall analysis time for a diverse mixture of compounds. In the present study, the m obile phase velocity was held constant. The time that it takes for an analyte to elute from the column is commonly referred to as the retention time (tr). Using the same instrument method, retention time is highly reproducible with GC and is often regarded as a tool of identification. As th e components elute, a chromatogram is created, displaying the eluted anal ytes as peaks in relation to time. Once eluted, the analytes exit the column and enter the detector, which in this case is a mass spectrometer. As the ions enter the detector, the analyt es are ionized by a process called electron ionization (EI). In this ion sour ce, the analyte vapor is bombarded by electrons created by a wire filament.188, 246, 247 The mechanism by which this process occurs can be seen below in Equation 1-1, where M is the analyte in the gas phase, eis an electron, and M+ is the molecular ion. M + eM+ + 2e(1-1) A schematic of the EI process is shown in Figure 1-6. The molecula r ion represents the nominal mass of the analyte and is created by th e removal of an electron. The removal of the electron forms a radical, which is generally uns table and consequently breaks into corresponding mass fragments.188, 246, 247 Since most analyses are performe d with the same energy (70 eV), reproducible fragmentation patterns are obtained, allowing the creation of searchable spectral libraries.172, 188 These libraries, along with retention time, greatly aid in the positive identification of known and unknown analytes. Liquid Chromatography Biological compounds, such as steroids, ar e commonly analyzed by LC to avoid the tedious and laborious sample pr eparation procedures often required by GC to impart sufficient volatility and thermal stability. The analysis of compounds by LC follows a similar mechanism to GC in that the separation mechanism involve s both a mobile and stationary phase. However,
36 unlike in GC analysis, the mobile phase plays an active role with the stationary phase in the separation process. Here, analytes are separated in liquid phase based on each analyteÂ’s ability to dissolve into the stationary phase.245, 248, 249 The stationary phase inside the column is typically a C-18 packed silica (with a specified length, diameter and particle size), and are most retentive for non-polar neutral analytes. Th e inclusion of a polar mobile phase creates what is called a reverse-phase separation. In reverse-phase LC, elution order occurs in order of decreasing polarity and can be manipulated by the al teration of the mobile phase composition.245, 248, 249 The different interactions of each analyte with the mobile and stationary phase, causes the retention and consequential separation over the duration of the column with a specified flow rate. Separation parameters, such as mobile phase co mposition, flow rate, temperature, and mobile phase gradient, can be altered to effectively increase separation and shorten analysis time.245, 248, 249 As components elute from the column, the liquid mixture of mobile phase and target analytes must then be ionized for detection by MS. The two most commonly employed ionization mechanisms for LC/MS are electrospray ionization (ESI) and atmospheric pressure chemical ionization (APCI). Although several papers employ ESI for the analysis of steroids, ionizatio n by ESI is most effective for the analysis of steroids exhibiting highly polar character.6, 185, 193, 194 APCI, on the other hand, is most useful for the analysis of analytes whic h exhibit lowto mid-polarity.6, 185, 193, 194, 250 Since most steroids fall within the low to mid-polarity range, APCI is better suited for the ionizatio n of steroids separated by LC.185, 193 Also, APCI is capable of handling hi gh mobile phase flow rates (1 mL or higher),193 which is of particular interest for the employment of monolithic columns. Monolithic columns optimally operate at higher flow rates, th us making the use of ESI (typical mobile phase flow rates of < 1 mL) not as ideal. Due to the macroand mesopores inside of monolithic
37 columns, faster and improved separation can be obtained;251 thus, the application of APCI is well suited for the analysis of steroids with the use of either monolithic or particulate columns. Ionization by APCI occurs in the gas pha se. A schematic of the mechanism by which APCI occurs is shown in Figure 1-7. After the an alytes (and mobile phase) leave the column into a small capillary, they are heated with a ne bulizer in a high temp erature region (550 C).247, 250 As the analytes leave the heated capillary, the an alytes react with the gas phase ions produced by the discharge from a corona needle. At atmosphe ric pressure, a high number of collisions occur between the analytes and the reagent molecules (H2O, O2, NH3), which subsequently produce reagent ions (H3O+, NH3 +, etc.). APCI is generally similar to chemical ionization (CI), in that ionization of the analyte is dependent upon the prot on affinity of the analyte being higher than the proton affinity of the reagent ions created by the corona needle.247 Post-collision, the analytes obtain a proton from the reagent ions and form [M+H]+ molecular ions, and sometimes fragment into smaller ions.247, 252 Mass Spectrometry (MS) and Tandem Mass Spectrometry (MS/MS) The application of MS and ta ndem MS with ion trapping is critical to some work performed here, and thus a brief background of this technique is de scribed below. As the ions are formed using either GC or LC, they are focused into an ion trap by a series of electrostatic lenses. The quadrupole ion trap (QIT ) consists of three radially symmetric hyperbolic electrodes: two endcaps and one ring electrode.188, 246, 253 The pre-formed ions are trapped in the ion trap with the application of a field and the aid of helium, which collisionally cools and reduces the kinetic energies of the ions. As the ions ar e centered in the trap, a radio frequency is subsequently applied to the ri ng electrode and a quadr upolar field is created, which keeps the ions trapped with stable trajectories.246, 253 Full mass spectra can be generated by ramping the radio frequency amplitude and applying a supplemental AC voltage.188, 246, 253 As the amplitude
38 is increased, ions with increasingly larger masses become unstable in the ion trap and are ejected, and the mass-to-charge ( m/z ) ratio of the ejected ions is proportional to the applied radio frequency amplitude. The ejected ions leave the ion trap through holes in the endcaps and are subsequently detected by an electron multiplier after conversion using a dynode. For the comprehensive analysis of both st eroids and EDCs, total ion current (TIC) chromatograms were obtained. The TIC chromatogram s display all of the separated analytes and the corresponding mass spectral information. The ch romatographic peaks in the TIC are based on the amount of full MS signal acquired during specific elution times. Each peak is comprised of a single analyte or several analytes that share similar retention time s. Identification of each analyte peak can be made by analyzing the mass spectr um of each individual chromatographic peak. However, if the characteristic ions are know n for each particular analyte of interest, reconstructed ion chromatograms (RIC) can be produced. The application of RIC can further increase the chromatographic resolution obtai ned and aid in improved identification and quantification.247 For the comprehensive analysis of both steroids and EDCs, RICs were obtained using instrument software for each specific anal yte based on two known characteristic ions. The characteristic ions were experimentally obtained by processing standards. The application of tandem spect rometric techniques (MS/MS) following GC or LC stems from improving the ability to iden tify and quantify at trace levels.254 During MS/MS, all of the ions in the ion trap (including potential interferences) are ejec ted using an applied waveform except for a pre-selected ion of in terest, termed the precursor ion.246, 247, 253 The precursor ion (specific m/z ) is held within the trap by the creation of a notch in the applied waveform. As the ion of interest is held within the trap, a supplemental AC wave form is applied with a specific amplitude that causes the precursor ion to collide with helium and is subsequently cleaved into
39 characteristic mass fragments, termed product ions.246, 247, 253 Once the product ions (different m/z from parent ion) are created, th ey are ejected to form a produ ct ion spectrum. A product ion spectrum is the corresponding mass fragments of a specific precursor ion. Product ion spectra, along with retention time, characte ristic MS identifica tion ions (RIC), and MS/MS, aid in the identification and quantification of ster oids and EDCs in complex matrices. Sample Extraction The mechanism of liquid-liquid extraction (LLE) is based on the relative solubility of an analyte in immiscible solvents (shown in Figure 1-8). Upon the mixing of the two solvents (step 1, one solvent may be the biological sample), th e analytes of interest are separated into the partition by which they are most soluble. Post -mixing (step 2), the analytes of interest (supernatant in organic layer) can be obtained from fractioning o ff the raffinate (aqueous) layer (step 3). LLE is considered to be a relatively efficient extraction t echnique; however, it is commonly regarded as a time-consuming technique that often suffers from incomplete removal of all matrix interferences.255-258 To date, solid-phase extraction (SPE) has genera lly replaced LLE due to its quicker sample processing when performed using a vacuum mani fold, higher specificity towards analytes, and more economical use of solvent.255, 259, 260 SPE was first introduced by Rohm and Haas in the early 1970s, and since then, the most common SP E phase has become octadecylsilane silica (C18).260 Extractions normally empl oy reverse phase C18 mechanis ms and typically involve a polar sample matrix (water, plasma) and a non-po lar stationary phase (silica based). Extraction strategies, similar to chromatographic techniques, center on the retention of select analytes on a SPE cartridge phase until an appropriate solvent is added to promote elution. SPE methods entail five steps (shown in Figure 1-9) 1) conditioning of the statio nary phase in the extraction cartridge, 2) loading of sample onto the cartridge 3) washing of cartri dge to remove undesired
40 matrix interferences, 4) drying of the cartridge, and 5) eluting the analyt e to unbind the analytes from the solid phase by applying the appropriate elution solvent.260 Several advances have been made to improve solid phase extraction. Polymeric cartridges operate with similar extracti on steps, although some extractio n steps are often eliminated.255, 259, 260 Polymeric cartridges often do not require prec onditioning steps, which can often lead to the unwanted removal of phase from the stationary phase.255, 259, 260 Also, polymeric cartridges often have better recoveries for a wider range of compounds due to th eir greater surface area.260 An alternative extraction technique, SPME, was first introduced by Pawliszyn in 1990.261, 262 SPME employs a phase-coated fiber that absorbs analytes in one of two modes, 1) headspace or 2) direct immersion. In head space extraction, analytes are made volatile and absorb onto the fiber in the gas phase. Direct immersion occurs when the fiber is immersed into a solution containing the analytes of interest. Direct im mersion is the most applicable mechanism for extracting steroids out of biological matrices due to the lack of volatility of most steroids. The SPME procedure is shown in Figure 1-10. Generall y, the fiber is immersed into the biological sample and extraction occurs by the fiberÂ’s ability to equilibra te between the analyte and the fiber (step 1).172, 256, 262-265 Equilibrium can be more efficien tly reached through the addition of salts, adjusting the pH change, and stirring.241, 256, 262, 266-268 As with SPE cartridges, efficient extraction of analytes is made possible by the application of the corr ect extraction phase. A common SPME phase type is polyacrylate (PA), which has been shown to be successful for several polarity ranges.241, 269 However, for the analysis of polar compounds with a PA fiber, derivatization is often needed to enhance GC analysis. For on-fiber derivatization (OFD),104, 222, 241, 242, 270 the fiber is exposed to deri vatization reagent vapor and the extracted analytes are headspace derivatized on the fiber (step 2). SPME-OFD is efficient by eliminating sample loss, which
41 commonly occurs during the evaporation and reconstitution steps employed by traditional derivatization procedures. Post-ext raction (and derivatization), the an alytes are desorbed off from the fiber when they are directly injected into the GC inlet (step 3). SPME is often noted for its simplicity, low detection limits, and low consumption of solvents; 256, 264, 267-269, 271, 272 however, SPME is not considered as a high-throughput method and method development can be tedious.257, 264 In addition, proteins have been shown to complicate extraction with certain fiber phases.267, 268 Analytical Derivatization Analytical derivatization (post-extraction) has of ten been applied to alter analytes, which in their natural or normal form, are not amenable to analysis. GC derivatization is commonly applied to increase the volatility, thermal stabil ity, separation, and detectability of an analyte.273275 Although derivatization ha s been noted to be both tedious and time-consuming,119 it can provide an opportunity allow for the profiling of more compounds by expanding the polarity range of amenability. Functiona l groups, like hydroxyls, amines and thiols, contain active hydrogens, which can be replaced by functional groups that are amenable to GC analysis. With derivatization, both polar and nonpolar compounds can be successfully separated, ionized, and detected in a single analysis. The most effective strategies for steroid derivatization are acylation and silylation. Acylation, the replacement of active hydrogen with perfluoroacyl groups (PFA), is often employed to increase detection w ith an electron capture detector (ECD).273-275 Lastly, silylation, the most commonly used method, involves the replacement of the active hydrogen with a trimethylsilyl (TMS) group to increase the volatility and ther mal stability of the analyte.273, 274, 276 Derivatization strategies for polar EDCs also typically employ silylation techniques. The reaction mechanism for both acy lation and silylation is shown in Figure 1-11, for the steroid 17 -estradiol.
42 In the silyl derivatization of steroids a nd polar (phenolic) EDCs active hydrogens are replaced with TMS groups by the applic ation of silyl reagents, such as N,O -bis-(trimethylsilyl)trifluoroacetamide (BSTFA) with 1% trimethylchlorosilane (TMCS) and N -methylN trimethylsilyl-trifluoroacetamide (MSTFA). Th ese silylating reagents are often considered comparable in silylating pr operties and are thoroughly de scribed in several reviews.273-276 Reaction time and temperature are the most cr ucial parameters for optimal derivatization strategies. Typical reaction conditi ons for the derivatization of steroids and EDCs with silylation strategies are 60 Â– 70 C for 30 Â– 180 minutes (1 59). Common methods for the derivatization of these groups typically require 45 Â– 90 C and 30 Â– 180 minutes.124, 215, 220, 226, 277-279 Steroid derivatization can often be complicated by the fact that steroids typically contain some combination of two active carbonyl a nd hydroxyl functional gr oups. Although steroids with only carbonyl groups (progesterone, androstene dione, etc.) present no analytical challenge to GC analysis, modification is required for those that contain one or more hydroxyl groups. Since steroids contain various combinations (location and number) of carbonyl and hydroxyl groups, the successful derivatization is depende nt upon the selective de rivatization of only hydroxyl groups, in an effort to minimize the formation of unwanted derivatives.280 This is an easy task for the analysis of small analyt e sets, but becomes increasingly difficult for comprehensive analyses. The conditions employed mu st be carefully made to effectively allow selective derivatization. The importance of selectiv e derivatization, or targeting specific analytes while avoiding others, is a key factor to succes sful derivatization on a comprehensive level. Methods are needed for the comprehensive derivatiz ation of steroids that effectively derivatize the hydroxyl groups while minimizing unwanted carbonyl derivatization.
43 The drawbacks of derivatization have typica lly been the reason for the impetus towards using alternative methods, such as LC/MS, for st eroid analysis. Derivati zation, albeit an organic reaction, has never been regarded as a reprodu cible technique, and often suffers from long reaction times, sample loss, and artifact formation.119, 172, 193, 280 Furthermore, most silylation reactions are moisture sensitive, making long term stability a concern. However, several advances towards improving deriva tization have been made to ta ke advantage of the excellent chromatographic power of GC. The enhancemen t strategies for derivatization include the implementation of solvent-based reactions to increase yield and reduce artifact formation,273, 274, 276, 281-283 application of new reagents for more highly specified derivatization products,273-276, 282, 284, 285 and the application of catalysts, such as TMCS, to aid in the derivatization of steroids which have multiple or hindered hydroxyl groups.273-276, 286 Furthermore, alternative heating mechanisms, such as microwave-accelerate d derivatization (MAD), have recently been introduced as a time-saving derivatization mechanism233, 287-289 and will be discussed in more detail in Chapter 4. Executive Summary The dissertation is composed of seven chapte rs. The underlying theme of the dissertation is to improve upon existing methodologies for the ch aracterization of steroids in the American alligator in an effort to elucidate the potentia l steroid responses in relation to EDC exposure. Chapter 1 includes an introduction of EDCs and several methods by which EDCs and EDCinduced steroid concentrations can be measure d, including GC/MS and LC/MS. Chapter 2 details an investigation of the best sample extraction st rategies for analysis of both EDCs in water and steroids in plasma using LLE, SPE, and SPM E methods. For the GC/MS analysis of a comprehensive mixture steroids, effective derivatizat ion strategies are required and needed to be optimized, which is the topic of chapter 3. In this chapter, a series of experiments to optimize and
44 enhance GC derivatization, util izing different derivatizing re agents, solvents, and heating mechanisms, were performed. Furthermore, chapte r 4 discusses and demons trates the merits of microwave-accelerated derivatizatio n of steroids using an organic synthesis microwave. Chapter 5 focuses on the development of a comprehensiv e method for the detection of both polar and non-polar (EDCs) in a single analysis by GC/MS. The method was subsequently applied to detect EDCs in water samples from Lake Apopka. Chapter 6 contains seve ral analyses involving the characterization of endogenous steroids in plasma using LC/MS, GC/MS and tandem mass spectrometric methods. Characterizations included comparisons of chromatographic techniques to IA-based techniques and the elucidation of multiple endogenous steroids. Chapter 7 contains conclusions and suggestions for future work.
45 Figure 1-1. Examples of endogenous sex steroids.
46 Figure 1-2. Examples of endocri ne disrupting compounds (EDCs).
47 Figure 1-3. Analytical methodology for the determination of steroids and EDCs from plasma and water samples.
48 Figure 1-4. Structures of the ster oids investigated in this study.
49 Figure 1-5. Structures of the ED Cs investigated in this study.
50 Figure 1-6. Electron ioni zation (EI) process.
51 Figure 1-7. Atmospheric pressure ch emical ionization (APCI) process.
52 Figure 1-8. Liquid-liqui d extraction steps.
53 Figure 1-9. Solid-phase extraction steps.
54 Figure 1-10. Solid-phase microextracti on steps (with on-fiber derivatization).
55 Figure 1-11. Acylation (PFA) and sily lation (TMS) reaction mechanisms for 17 -estradiol.
56 CHAPTER 2 EVALUATION OF EXTRACTION STRATEGI ES FOR THE ANALYSIS OF STEROIDS AND ENDOCRINE DISRUPTING COMPOUNDS (EDCS) IN ENVIRONMENTAL AND BIOLOGICAL SAMPLES USING GAS CH ROMATOGRAPHY/MASS SPECTROMETRY (GC/MS) Introduction Several field and laboratory based studies have shown that exposure to endocrine disrupting compounds (EDCs) can have a delete rious effect on wildlife and human health.7, 8, 11, 14, 290 Several of the observed effect s in these studies include a ltered steroidogenesis, abnormal reproductive development, and in appropriate sexual behavior.13, 14, 29, 290 In principle, EDCs can enter the water environment in effluent from sewage treatment plants,96, 100-124, 135, 143 excreta from livestock production systems,125-127 agricultural and indu strial runoff and waste,46, 96, 127 and the use and disposal of many pe rsonal care household products.42-45, 96, 291 As a result of the ubiquitous nature of these anthropogenic co mpounds, several reports have detected their presence in other matrices, including air,44, 129 soil,44, 128 food,44, 130-134 and wildlife.12, 14, 70, 292 In consequence to this and the ever-expanding li st of potential EDCs, innovative methods capable of detecting changes in their pr esence and activity are needed. Analytical methods designed to monitor envi ronmental EDCs and endogenous steroids are typically based on the applic ation of gas or liquid chro matography coupled with mass spectrometry (GC/MS and LC/MS). However, the e ffective analysis of environmental EDCs and endogenous steroids in co mplex matrices is largely dictated by the efficiency of sample preparation procedures employed pr ior to chromatographic methods.172 In particular, optimal extraction techniques are required to isolate and pre-concentrate chemically similar analytes by a procedure which exploits the specific differenc es between the analytes of interest and the interfering sample matrix. Offline/online extr action techniques have been employed for the
57 analysis of EDCs and steroids, including liqui d-liquid extraction (LLE), solid-phase extraction (SPE) and solid-phase mi croextraction (SPME). GC/MS Analysis of EDCs and Steroids GC/MS, with offline extraction techniques, has been successfully implemented for the analysis of many non-polar environmental EDCs including organochlorine pesticides and several other polychlorinated compounds, phthala tes, polychlorinated bromides (PCBs), and several other anthropogenic compounds.101, 116, 129-131, 136-138 For more polar analytes, such as the alkyphenols, bisphenol A, natural and syntheti c steroids, and polar degradation products, derivatization procedures are also required for traditional GC/MS.106, 108, 110, 114, 115, 118-124, 126, 135, 139, 214, 216, 220, 285 Thus, the current trend is to employ LC /MS for the analysis of polar EDCs and steroids to avoid deri vatization procedures;105, 107, 111-113, 116, 117, 133, 136, 139, 141, 213 however, in combination with the superior chromatogra phic resolution of GC, the application of derivatization allows co mpounds from both ends of the polarity spectrum to be separated and detected in a single analysis. Therefore, techni ques which enable the extraction of chemically diverse analytes in environmental and bi ological samples would permit GC/MS with derivatization to ha ndle analytes over a wide polarity range. Liquid-Liquid Extr action (LLE) A classical extraction method is liquid-liq uid extraction (LLE), which involves the partitioning of analytes between two immiscible so lvents. Solvents such as dichloromethane, diethyl ether, and methyl tert -butyl ether have been utilized for the extraction of analytes in plasma and urine.113, 147, 150, 156, 168, 197, 202, 224-228, 293 The frequent use of LLE arises from its simplicity; however, it commonly suffers from an excessive use of solvents, the creation of emulsions, and insufficient analyte specificity.256, 258, 268, 294
58 Solid-Phase Extraction (SPE) In response to the incomplete removal of interferences using LLE, several extraction methods with increased selectivity have been developed, most notably solid-phase extraction (SPE). SPE is characterized by th e retention of analytes on a so lid sorbent packing inside an extraction cartridge and is the most commonly employed method for the extraction of analytes in environmental and biological samples.172, 255, 256, 258 The most common sorbents used include C18 and polymeric-based cartridges (such as Oasis HLB). A recently developed sorbent strategy is the implementation of liquid-liqui d cartridge extraction (LLCE), which employs the fundamental principles of both LLE and SPE methods. LLCE is performed by the rete ntion of the sample matrix (with analytes) in the sorbent packing, followed by the removal of analytes of interest by the addition of an immiscible solvent. One adva ntage of this extraction method over SPE is that the cartridge experiences less sa mple plugging; another advantage is that the extraction can be performed in just two steps. A schematic of the LLCE process can be seen in Figure 2-1. Although SPE has been shown to be advantageous for the extraction of a variety of select steroids106, 112, 116, 119, 120, 122, 124, 140, 164, 169, 211, 212, 214, 215, 217, 218, 220, 221, 229, 230, 234, 239, 240, 295 and EDCs,102, 106, 112, 116, 119, 122, 124, 140, 210, 214, 216-220, 240 it is not without disadv antages. Most cartridges suffer from sample loss, tedious extrac tion procedures, and cartridge plugging.172, 256, 268, 269 SPE strategies remain an excellent mechanism for the extraction of a set of chemically similar analytes; however, for the extract ion of a multi-class mixture, me thods typically require the use of separate aliquots, extra elut ion steps, or multiple cartridges.116, 214-216, 229, 240, 296 Solid-Phase Microextraction (SPME) An alternative in the extraction of environmen tal and biological samp les is the application of solid-phase microextraction (SPME),266, 268, 297-302 which was first introduced in 1990 by Pawliszyn and coworkers.261, 262 SPME employs a phase-coated fiber with the ability to extract
59 analytes out of the sample matrix by either direct immersion (DI) or headspace (HS) methods. Although HS-SPME is the most widely used SPME strategy, the low volatility of most polar EDCs and steroids make DI-SPME the more a ppropriate choice for extr action. Effective SPME analysis is predicated upon reaching a partitioning equilibrium between the analytes of interest in solution or the gas phase and the phase-coated fiber.256, 263-265, 271 Similarly to SPE, success of SPME is based on the selection of the appropriate extraction phase. The two most common fiber phases are the non-polar polydimethylsiloxane (P DMS) and the more polar polyacrylate (PA) fiber. Also like SPE, SPME is typically limite d by the capabilities of the fiber phases, although PA fibers have been considered to be capab le of extracting a wide r range of analytes.241, 269 Several examples exist for the SPME analysis of nonpolar GC-ready EDCs;222, 269, 303 however, for the GC analysis of polar EDCs and steroids, derivatiza tion is often required.104, 222, 241, 242, 270 For the extraction of polar analytes with th e PA fiber with GC/MS, one aspect needs to be considered. SPME is capable of pre-concentra ting extracted analytes on the fiber due to its ability to absorb analytes from samples ont o a relatively small su rface area on the fiber.262, 299 Thus, analytes are typically de sorbed into the GC/MS injection port at a higher concentration than observed by traditional SPE extraction. As a result, SPME is capable of detecting several steroids and polar EDCs without derivatization, a result atypical of tradi tional GC/MS analysis with other extraction techniques. Solid-Phase Microextraction with On-Fiber Derivatization (SPME-OFD) In contrast to standard SPME, the implemen tation of on-fiber deri vatization (OFD) is a new methodology designed to improve the analysis of polar analytes by SPME.104, 222, 241, 268 The implementation of SPME-OFD not only enables the analysis of a wider range of polarity for GC/MS, but also increases the detectability of thos e polar compounds whic h are not directly amenable to GC/MS analysis. SPME-OFD has previ ously been implemented for the analysis of
60 small sets of polar compounds by first DI-SPME, followed by the headspace (HS) derivatization of the extracted analytes on the fibe r with derivatization reagent vapor.104, 222, 241, 242, 265, 270 It has been shown that derivatizing onfiber leads to single derivati zation products and eliminates evaporation steps often required by othe r extraction/derivati zation strategies. In this study, the objective was to examin e innovative strategies for the simultaneous extraction of a large suite of chemically diverse steroids and EDCs in a single GC/MS analysis. Several extraction methods were evaluate d, including LLE and SPE (with offline silyl derivatization), and SPME (with a nd without OFD). Each method was evaluated in its ability to extract a diverse mixture of st eroids and EDCs in spiked water and human plasma samples. Experimental Section Steroids and Endocrine Di srupting Compounds (EDCs) The natural and synthetic st eroids examined were 17-estradiol (E2), estrone (E1), diethylstilbestrol (DES), ethinylestradiol (EE2), progesterone (PROG), testosterone (T), androstenedione (AE), pregnenol one (PREG), and cholesterol (C HOL) and were purchased from Sigma Aldrich (St. Louis, MO). The potential EDCs examined, 4-octylphenol, 4-nonylphenol, vinclozolin, bisphenol A, tric losan and 1,1,1-trichloro-2,2-bis (4-hydroxyphenyl)ethane (HPTE), were also acquired from Sigm a Aldrich. The other EDCs examined, p,pÂ’-DDT, p,pÂ’-DDD, p,pÂ’DDE, p,pÂ’-DDA, p,pÂ’-methoxychlor, dicofol, dieldri n, endrin, heptachlor, heptachlor epoxide, lindane, dibutyl phthalate, -endosulfan, trifluralin, and atrazi ne, were acquired from the EPA (Research Triangle Park, NC). Purity grades we re labeled at 98% or higher, except for endrin (83%) and dibutyl phthalate (80%). Internal standards, anthracene, 7,8-dimethoxyflavone and 17 -estradiol-acetate, were obtained from Sigma Aldrich. A spiked mixture (1000 ng mL-1) of all of the steroids and EDCs was prepared in an alytical grade methanol (Fisher Scientific, Fair
61 Lawn, NJ). For the determination of relative extraction efficiency of each analyte in the plasma and water samples, the steroid/EDC mixture was adde d to each matrix at a final concentration of 200 ng mL-1. The ions ( m/z ) used for quantification for each ex pected steroid and EDC product (with TMS-derivatization) were di-TMS-17-estradiol (285, 416), mono-TM S-estrone (257, 342), diTMS-diethylstilbestrol (217, 412), di-TMS-ethinyles tradiol (425, 440), progesterone (124, 314), mono-TMS-testosterone (226, 360) androstenedione (148, 286), mono-TMS-pregnenolone (298, 388), mono-TMS-cholesterol (368, 458), monoTMS-4-octylphenol (179, 278), mono-TMS-4nonylphenol (179, 292), vinclozolin (178, 286), di -TMS-bisphenol A (357, 372), mono-TMStriclosan (200, 362), mono-TMS-HPTE ( 343, 390), p,pÂ’-DDT (165, 235), p,pÂ’-DDD (165, 235), p,pÂ’-DDE (246, 318), mono-TMS-p,pÂ’-DDA (200, 337) p,pÂ’-methoxychlor (227, 228), dicofol (139, 251), dieldrin (79, 263), e ndrin (281, 317), heptachlor (272, 371) heptachlor epoxide (353, 355), lindane (181, 219), dibut yl phthalate (149, 278), -endosulfan (195, 241) trifluralin (264, 306), and atrazine (200, 215). The compounds listed without the TMS prefix were not derivatized upon the addition of de rivatizing reagent. The ions ( m/z ) used for the analysis of several of the polar steroids /EDCs (underivatized, no derivatization reagent added) were 17estradiol (213, 272), estrone ( 185, 270), diethylstilbestrol (145 268), testosterone (124, 288), pregnenolone (231, 316), 4-octylphenol (107, 206), 4-nonylphenol (107, 220), bisphenol A (213, 228), and triclosan (218, 289). Liquid-Liquid Extrac tion (LLE) Method The liquid-liquid extraction was carried out by the addition of 2 mL of methyl tert -butyl ether (MTBE, Fisher Scientific) to 1 mL of spiked plasma. The sample was then vortexed for one
62 minute and stored in a Â–20 C freezer for one hour. The supernatant was removed by decanting off the organic layer. Solid-Phase Extraction (SPE) Methods The solid-phase extraction of spiked plas ma and water samples was carried out on a Prepsep vacuum manifold (Fishe r Scientific). MTBE, ethyl a cetate (EA), acetonitrile (ACN), methanol (MeOH), water (H2O), and ammonium hydroxide (NH4OH) were acquired from Fisher Scientific and were used duri ng the extraction procedures. The cartridges examined were Oasis HLB (30 mg, 1 mL, Waters, Milf ord, MA), Generik Â– C18 (50 mg, 1 mL, Sepax Technologies, Newark, DE), and HM-N Isolute LLCE cartrid ges (1 mL, Biotage, Charlottesville, VA). The Oasis HLB extraction was performed by preconditioning the cartridge with 1 mL of MTBE, 1 mL MeOH, and 1 mL H2O. Following sample loading (1 mL), the analytes were eluted from the cartridge with four steps: 1 mL 40% MeOH:H2O, 1 mL H2O, 1 mL 10% MeOH: 2% NH4OH in H2O, and 1 mL 10% MeOH:MTBE. The four elution volumes were combined. The C18 extraction was performed by precondi tioning the cartridge with 1 mL MeOH and 1 mL 10% ACN:H2O. After 1 mL of sample was loaded, the cartridge was washed with 0.5 mL of H2O. The analytes were eluted with 1 mL of MeOH. The LLCE was performed by adding the sample (1 mL) directly to the cartridge, followed by elution with 3 mL of MTBE. Offline Derivatization Strategy for LLE and SPE Extracts Post-LLE and -SPE, each elution volume was ev aporated to dryness with heated (40 C) ultra-high-purity (UHP) nitrog en. The extract was subsequent ly reconstituted with 500 L of analytical grade methanol. A volume of 200 L (of the reconstituted residues) were added to 4 mL glass vials and evaporated to dryness with UHP nitrogen. The deriva tization reactions were
63 carried out using N -methylN -trimethylsilyl-trifluoroacetamide (MSTFA) (Pierce, Rockford, IL). MSTFA was added (200 L) to each vial and reaction was performed at 55 C for 30 minutes using a Thermolyne Type 16500 Dri-Bath heater. Po st-derivatization, the sample was evaporated to dryness with UHP nitrogen and reconstituted with isooctane (Fisher Scientific). Relative Response Factor (RRF) and Relative Extraction Efficiency (REE) Values for the LLE and SPE Methods The relative response factors (RRFs) using both LLE and SPE methods were determined by adding the steroid/EDC mixture to deionized water and plasma samples prior to (B) and after (A) extraction (each in triplicate). For each samp le (both A and B), RRF values were generated by dividing the peak area of each analyte by the peak area of the surrogate, anthracene (added post-extraction to a final concentration of 100 ng mL-1). The relative extraction efficiency (REE) for each analyte was determined by dividing th e RRF value (B) by the RRF value (A). In addition, the steroid/EDC mixtur e was also spiked at severa l concentrations: 2000, 200, 20, 2, and 0.20 ng mL-1 in deionized water and extracted by the LLCE method (in triplicate). The extracts were evaporated and derivatized according to derivatization procedures. Solid-Phase Microextraction (SPME) Method SPME analysis was performed using a fiber coated with an 85 m polyacrylate (PA) phase (Supelco). Traditional DI-SPME was performed by im mersing the PA fiber into the spiked water samples for 30 minutes at 25 C. Extraction e quilibrium was expedited with the aid of a magnetic stir bar. Post-extrac tion, the fiber was injected into the GC injection port and the analytes were desorbed from the fiber for five minutes. The concentration levels of the spiked steroid/EDC mixture were at 0.2, 0.7, 2, 20, and 200 (and 500 ng mL-1 for the steroids). The internal standard for the SPME analysis, 7,8-dime thoxyflavone, was added to the spiked samples at a concentration of 100 ng mL-1 prior to extraction.
64 Solid-Phase Microextraction On-Fib er Derivatization (SPME-OFD) Method The extraction procedure for SPME-OFD method was followed as previously described, but was followed by an additional on-fiber de rivatization step. The on-fiber derivatization occurred in a 4 mL vial. A total of 50 L of MSTFA was added and placed into a heated water bath at 55 C. The PA fiber was exposed to th e derivatization reagent vapor (for 30 minutes) by piercing the Teflon-coated silicone septum with the SPME holder. 17 -estradiol-acetate was utilized as the internal standard for the SPME-OFD analysis and was derivatized. SPME Fiber Monitoring A simple method was created to monitor the condition of the SPME fiber by DI-extraction of 2,6-dichlorophenol (2,6-DCP, Sigma Aldric h) in methanol (1 ppm). The 2,6-DCP was extracted from the methanol for 10 minutes and in jected/desorbed into the GC injection port. The procedure was performed after every extraction (w ith or without OFD) and the percent RSD of the 2,6-DCP peak area was recorded. Potential carryover peaks were also monitored with the method. Spiked Human Plasma (SPME with and without OFD) The spiked steroid/EDC mixture (a t a final concentration of 200 ng mL-1) was added to each vial and evaporated to dryness. To each vi al, 1 mL of expired human plasma was added. Prior to SPME, 2 mL of MTBE was added to the plasma and vortexed for 45 seconds. The sample was then put into a freezer (20 C), and d ecanted off into another vial after 1 hour. The extract was evaporated with UHP-nitrogen and rec onstituted with analyti cal grade water (Fisher Scientific). The sample was then sonicated for 1 minute and extracted by DI-SPME. Postextraction, the fiber was injected into the GC (except for when OFD was required prior to injection).
65 Gas Chromatography/Mass Spectrometry (GC/MS) GC/MS analysis of water and plasma extrac ts was achieved using a ThermoFinnigan Trace GC 2000 gas chromatograph/quadrupole ion trap mass spectrometer, equipped with an AS3000 Autosampler (San Jose, CA) and Xcalibur 1.4 da ta acquisition software. An SLB-5ms capillary column (Supelco, Bellefonte, PA) w ith dimensions of 30 m x 0.25 mm x 0.25 m (film thickness), was utilized for the separation of components. The ion source and transfer line temperatures were set to 200 C and 300 C, resp ectively. Splitless inject ion was employed at a temperature of 280 C and a split flow of 50 mL min-1. The carrier gas, UHP-helium (99.99%), was used at a flow rate of 1 mL min-1. A temperature program was employed at an in itial temperature of 120 C and held for 2 minutes. The temperature ramp increased to 250 C at 15 C min-1. After 250 C was reached, the ramp increased to 300 C at 5 C min-1 and was held for 5 minutes. Total GC analysis time for the separation of the underivatized and de rivatized steroid/EDC species was 34.65 minutes. For SPME fiber monitoring, the same temperature program was employed. 2,6-DCP had a retention time of 7.67 minutes. Results and Discussion Method Development The steroids and EDCs selected for the comprehensive extraction encompassed a wide variety of compounds differing in physiochemical properties and intended applications. Several of the EDCs chosen represent anthropogenic compounds th at are heavily used in agricultural and industrial practices, and have s ubsequently been found in the e nvironment at relatively high concentrations (ng mL-1 Â– L-1 range).46, 55, 301 The chemically diverse EDCs examined include organochlorines (DDT, DDE, DDD, DDA, linda ne, methoxychlor, endosulfan, dicofol and HPTE) and several other chlorinated compounds (triclosan, dieldrin, endrin, heptachlor,
66 heptachlor epoxide, and kepone), alkylphenols (n onylphenol and octyphenol), phthalates (dibutyl phthalate), triazines (atrazine), bisphenol A, and several other su spected EDCs (trifluralin and vinclozolin). The natura l and synthetic hormones investigated were E2, E1, DES, EE2, PROG, T, AE, PREG, and CHOL. Even amongst chemically similar EDCs, such as the organochlorine compounds, degradation products (DDA, HPTE) can present chromatographic challenges by requiring additional methods for their characteriz ation, along with the char acterization of their non-polar precursor.116, 129, 136 Thus, methods capable of extr acting analytes with varying chemical properties would be useful. The develo pment of a method capable of extracting a wide range of compounds involved c hoosing several common extract ion cartridges and elution strategies. All of the extracts were offline derivatized with the sa me procedure, thus any errors determined in the extraction possess some level of error associated wi th the derivatization. The RRF Values and Extraction Effi ciency for Spiked Water Using SPE A comparison of RRF values obtained after extraction (B) presente d a comparison between different cartridges of different so rbent types. It is difficult to co mpare extraction capabilities of cartridges with different sorbent t ypes, due to the difficulty in se lecting an appropriate internal standard capable of being used across cartridges; thus, a surrog ate (anthracene) was added postextraction to allow a comparison between cartridges. The calculat ed RRF values are shown in Figure 2-2 and represent the peak area of each an alyte after extraction divided by the peak area of the surrogate. Since all components were added at the same concentration, a determination of the optimal extraction cartridge was obtained. The three main sorbent types examined were an Oasis HLB, C18, and LLCE cartridge. Data in Figure 2-2 show the RRF values obt ained for the chlorinated EDCs. As shown in every case, the LLCE cartridge gave the highest RRF values post-extrac tion, indicating that it extracted the highest amount of analyte. Th e Oasis HLB cartridge had lower RRF values
67 compared to the LLCE cartridge; however, the Oasis HLB extractions had a lower standard deviation in comparison to the LLCE cartridge. The C18 cartridge had the lowest RRF values and in some cases was not capable of extracti ng some of the chlorinated EDCs. (endosulfan and HPTE). DDA, was not extracted by any of the three cartridges using the selected elution strategies. The data in Figure 2-3 display the RRF values obtained for the hydroxylated EDCs. In all cases, the RRF values obtained using the LLCE cartr idges were either highest or comparable to those obtained from the other two cartridges, although the RRF values obtained for the Oasis HLB cartridges were comparable for TMS-BP A and TMS-HPTE. In addition, the Oasis HLB cartridge had dramatically lower standard devi ation when compared to the LLCE cartridge. The RRF values shown in Figure 2-4 represent the ability to extract the other remaining EDCs, trifluralin, atrazine, vinclozolin and dibut yl phthalate. The RRF values were typically highest while using the LLCE cartr idge (except for atrazine). In Figure 2-5, the RRF values obtained fo r the Oasis HLB and LLCE cartridges were comparable for all steroids (underivatized and TMS-derivatized), although the Oasis HLB cartridge had the lowest standa rd deviation. The C18 cartridge was capable of extracting the steroids, but for most of the an alytes the RRF values were lowe r then those obtained with the other two cartridges. The values in Table 2-1 repr esent the REE values from wa ter (RRF obtained from spike before extraction (B)/RRF obtained from spike postextraction (A)) of all th e analytes studied in this experiment for all three cartridges. The REE values illustrate the ability of each cartridge to retain the total amount of analyte. The REE values were the highest using for the LLCE cartridge. The average REE for all the analytes using the LLCE cartridge was 76% ( 12%). The
68 Oasis HLB and C18 cartridges had an averag e REE of 46% ( 17%) and 63% ( 9%), respectively. Although the extraction of water using the LLCE car tridge typically has the highest standard deviation, the RRF values and analyt e coverage was better with the LLCE cartridge than the Oasis HLB and C18 cartridges. The RRF Values and Extraction Efficiency for Spiked Plasma Using SPE For the extraction of plasma samples, the main emphasis was on the extraction of the steroids, although the ability to extract EDCs fr om plasma was also included. The RRF values obtained from extracting the spiked steroids fr om plasma samples is shown in Figure 2-6. The extraction with the Oasis HLB and C18 cartridges ha d the highest RRF values for all the steroids examined (except for TMS-E1). For the spiked plasma, the REE values of each extraction type (Oasis HLB, C18, LLCE, and LLE) are shown in Table 2-2. Since the extraction of plasma is most applicable for the analysis of endogenous steroids, the optimal extraction method was determined to be the Oasis HLB and C18 cartrid ges, although the C18 cartr idge typically had a greater propagation of error. The LLCE cartridge was not comparable to the Oasis HLB and C18 cartridges. The LLE extraction had relatively hi gh REE values for the estrogens; however, it typically had a high propagation of error and wa s not optimal for the other steroids examined. Analysis of Spiked Water Us ing the LLCE Cartridges The extraction of spiked water was perf ormed using the LLCE method at several concentrations. The LLCE method wa s capable of extracting most of the steroid/EDC mixture, but was not able to extract most analytes efficiently at concen trations below 2 ng mL-1. In addition, the RSD values were typi cally high at each c oncentration, likely due to the selection of an inefficient internal standard (7,8-dime thoxyflavone). The LLCE method does; however, exhibit the capacity to extract se veral steroids and EDCs in wate r. Future analyses will involve an investigation of a be tter internal standard.
69 SPME Analysis of Water The utilization of SPME (with a PA fiber) was ba sed on its ability to preferentially extract polar compounds, although the PA fiber has been shown to possess some capabilities for the extraction of non-polar compounds. Du e to its ability to preconcen trate analytes on the fiber, several polar EDCs and steroids were detected with GC/MS without deri vatization (as low as 0.2 ng mL-1). The SPME fiber was also able to extract several non-polar EDCs from the water. The R2 values and concentration range (in which th e analyte was detected) for each non-polar and polar steroid/EDC is shown in Table 2-3. Concentration ra nges were between 0.2 and 200 ppb (ng mL-1), except for the steroids, which extended to 500 ppb. Concentration range was evaluated by including th e levels that had peaks observed wi th a signal-to-noise (S/N) value > 10. Several polar steroids and ED Cs (triclosan, octylphenol, bi sphenol A, dicofol, DES, E1 and E2) were detected underivatized in the spiked water at sub ng mL-1 levels with SPME. Likewise, several non-polar EDCs (DDE, DDD, lindane heptachlor, DBP, heptachlor epoxide, methoxychlor, and dieldrin) were detected below ng mL-1. The analysis of SPME without derivatization was not possibl e for testosterone, choleste rol, androstenedione and ethinylestradiol. SPME Analysis of Plasma The GC/MS analysis of spiked plas ma (at a concentration of 200 ng mL-1) using the SPME-OFD method exhibited an increase in S/ N values over the traditional SPME method (without derivatization, as s hown in Table 2-4, except for TMS-T). The SPME-OFD of derivatized steroids and EDCs were capab le of extraction to as low at 0.2 ng mL-1 in plasma. Three steroids (testosterone, chol esterol, and ethinylestradiol) which were not detected with traditional SPME were now detected using the SPME-OFD method.
70 Monitoring Fiber Condition To date, there is no method available to m onitor the condition of the SPME fiber. A monitoring method would be advantageous when performing SPME strategies that have the potential to deteriorate the ex traction phase on the fiber. The application of SPME to extract components from plasma can lead to complica tions with the extrac tion phase due to the interaction of the fiber with proteins in the plasma. Furthermore, the employment of silyl reagents, such as MSTFA at high temperatures, can degrade the phase on the fiber. Direct contact of the fiber to the derivatizing reagent can be detrimental to the fiber (as shown in Figure 2-7); and thus, headspace derivatization was ut ilized, although the headspace vapor can condense onto the fiber and deteriorate the extraction phase. In addition, the derivatizing reagent has been shown to exist after desorption of the fiber in the GC injection port. A simple method to monitor these effects was performed using an extraction step after each SPME analysis. The analysis involved the extraction of 2,6-DCP in water and the analysis of its peak area. The remaining MSTFA and other derivatized co mponents were eliminated af ter the 2,6-DCP extraction. The RSD of the 2,6-DCP peak area for the derivatize d and non derivatized experiments was less than 10%. Furthermore, the PA fiber was capable of performing at least fifty SPME-OFD analyses before requiring a replacement fiber. Conclusion Most SPE cartridges have not been fully expl ored in terms of their ability to extract analytes over a wide polarity range in a single extraction. Furthermore, the capabilities of LLCEbased cartridges have not been examined in their ability to extract steroids and EDCs. The results presented in this study illustrate that th e Oasis HLB cartridges typically had better reproducibility when extracting steroids and ED Cs from water. However, the LLCE cartridges typically extract the most amount of analyte and have higher rela tive extraction efficiencies. In
71 addition, the LLCE extraction was simple and was e ffective in the extraction of a diverse mixture of analytes in a single analysis. For the extract ion of plasma, Oasis HLB and C18 extraction were the best extraction strategies, a lthough the methods examined in this study were not effective below ng mL-1. The application of SPME for the analysis of st eroids and EDCs exhibi ted great potential or the simultaneous extraction of both non-polar and polar compounds. SPME, with GC/MS, was capable of detecting several polar EDCs at concentrations below ng mL-1. Extraction of polar compounds in plasma by SPME-OFD provided enhan ced GC/MS detectability for most of the polar steroids and EDCs. Although SPME-OFD requi res an additional sample preparation step prior to analysis, it is clear that is has great potential for the an alysis of endogenous steroids in plasma at relatively low concentrations.
72 Table 2-1. Relative extraction efficiencies (REE ) for steroid/EDC mixture in water using the selected extraction strategies Oasis (%) C18 (%) LLCE (%) DDT 28 8 21 12 63 7 DDE 25 3 30 6 68 5 DDD 29 5 33 9 68 6 Trifluralin 28 25 38 8 Atrazine 73 43 75 10 68 31 Lindane 42 5 100 19 76 4 Vinclozolin 35 6 83 12 56 7 Heptachlor 32 4 DBP 39 5 70 10 81 8 Heptachlor epoxide 34 11 67 12 Endosulfan 30 8 71 10 Methoxychlor 31 3 39 4 76 8 Dieldrin TMS-Triclosan 36 10 55 9 92 21 TMS-Octylphenol 35 11 69 4 75 14 TMS-Nonylphenol 27 7 39 5 63 10 TMS-HPTE 47 26 103 21 109 39 TMS-BPA 46 11 82 3 90 13 TMS-DDA 28 39 TMS-Dicofol 24 7 37 16 61 5 TMS-DES 46 10 68 11 90 14 TMS-Estrone 46 8 72 7 92 16 TMS-Ethinylestradiol 42 6 68 8 92 14 TMS-Estradiol 49 7 70 6 92 11 TMS-Testosterone 43 6 77 5 91 12 TMS-Pregnenolone 41 4 53 4 91 14 Androstenedione 43 10 71 6 94 11 Progesterone 37 6 67 0 90 9 Error values represent the error propagated through the extraction efficiency calculation (B/A calculation). TMS indicates that the steroid/EDC was trimethylsilyl derivatized. Missing values indicate that the compound was not detected.
73 Table 2-2. Relative extraction efficiencies (REE ) for steroid/EDC mixtur e in plasma using the selected extraction strategies Oasis HLB (%) C18 (%) LLCE (%) LLE (%) DDT 11 5 11 4 12 1 31 13 DDE 2 0 10 3 14 1 28 11 DDD 8 2 24 5 13 1 32 13 Trifluralin 15 4 44 22 17 6 Atrazine 77 11 110 35 Lindane 74 10 82 18 42 6 89 33 Vinclozolin 73 13 31 6 Heptachlor 19 6 DBP 24 3 92 37 5 3 20 6 Heptachlor epoxide 26 7 65 18 28 1 68 30 Endosulfan 23 25 23 4 Dieldrin 81 19 Methoxychlor 16 3 51 13 21 1 53 21 TMS-Triclosan 37 8 57 15 30 3 59 23 TMS-Octylphenol 58 8 48 13 31 2 64 20 TMS-Nonylphenol 21 3 22 6 17 2 32 14 TMS-HPTE 34 16 41 12 46 4 52 18 TMS-BPA 63 17 66 19 115 11 112 39 TMS-DDA 80 27 8 2 TMS-Dicofol 8 8 37 4 14 1 26 11 TMS-DES 34 6 32 9 36 3 61 14 TMS-Estrone 91 13 85 19 88 28 TMSEthinylestradiol 89 7 76 13 52 3 99 32 TMS-Estradiol 83 6 76 14 63 3 98 32 TMS-Testosterone 88 12 88 20 78 8 117 41 TMS-Pregnenolone 58 7 67 15 31 4 71 27 Androstenedione 72 10 98 25 82 11 Progesterone 49 7 94 27 65 11 Error values represent the error propagated through the extraction efficiency calculation (B/A calculation). TMS indicates that the steroid/EDC was trimethylsilyl derivatized. Missing values indicate that the compound was not detected.
74 Table 2-3. The R2 values and linear concentration range (in ppb) obtained from SPME analysis of spiked water (without OFD). Range (in ppb) R2 Value DDT 2-200 0.9917 DDE 0.2-200 0.9921 DDD 0.2-200 0.9974 Trifluralin 0.2-200 0.9693 Atrazine 2-200 0.9956 Lindane 0.2-200 0.9986 Vinclozolin 2-200 0.9832 Heptachlor 0.7-200 0.9962 DBP 0.2-200 0.9981 Heptachlor epoxide 0.2-200 0.9976 Endosulfan 2-200 0.9971 Methoxychlor 0.7-200 0.9914 Dieldrin 0.7-200 0.9981 Triclosan 0.2-200 0.9979 Octylphenol 0.2-200 0.9621 Nonylphenol 2-200 0.9981 HPTE 2-200 0.9981 BPA 0.2-200 0.9929 Dicofol 0.2-200 0.9962 DES 0.2-500 0.9983 Estrone 0.7-500 0.9993 Estradiol 0.7-500 0.9976 Pregnenolone 2-500 0.9989 Progesterone 2-500 0.9957 The steroids (androstenedione, te stosterone, cholesterol, and ethinylestradiol) and the EDC (DDA) were not detected with SPME without de rivatization. Concentration ranges were between 0.2 and 200 ppb (ng mL-1), except for the steroids, which extended to 500 ppb. Concentration range was evaluated by including the levels that had peaks observed with a signal-to-noise (S/N) value > 10.
75 Table 2-4. Signal-to-noise (S/N) values of the pol ar steroids/EDCs extracte d from spiked plasma (200 ng mL-1) using SPME (with and without OFD). Polar EDCs/Steroids S/N (no OFD) S/N (with OFD) Octylphenol 1900 2400 Nonyphenol 270 850 Triclosan 1100 1400 Bisphenol A 3100 29000 DES 1500 5900 Testosterone 360 180 Estrone 2800 6700 Estradiol 760 5200 Pregnenolone 170 670 Cholesterol ND 360 HPTE ND 2500 Ethinylestradiol ND 1000 ND = not detected.
76 Figure 2-1. Liquid-liquid cartridge extraction (LLCE) procedure.
77 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 DDTDDEDDDLindaneEndosulfanHeptachlorHeptachlor EpoxideMethoxychlorPeak Area Ratio Oasis HLB C18 LLCE Figure 2-2. Peak area ratios of th e non-polar chlorinated EDCs in spiked water were obtained by dividing the peak area of the analyte by the peak area of the surrogate, anthracene. Error bars correspond to the standard deviation of the mean (n=3).
78 0.000 0.200 0.400 0.600 0.800 1.000 1.200 1.400 TMS-OPTMS-NPTMS-TriclosanTMS-BPATMS-DESTMS-HPTETMS-DicofolPeak Area Ratio Oasis HLB C18 LLCE Figure 2-3. Peak area ratios of the polar ED Cs in spiked water were obtained by divi ding the peak area of the analyte by the p eak area of the surrogate, anthracene. Error bars correspond to the standard deviation of the mean (n=3).
79 0.000 0.200 0.400 0.600 0.800 1.000 1.200 TrifluralinAtrazineVinclozolinDBPPeak Area Ratio Oasis HLB C18 LLCE Figure 2-4. Peak area ratios of th e remaining EDCs in spiked water were obtaine d by dividing the peak ar ea of the analyte by t he peak area of the surrogate, anthracene. Error bars corres pond to the standard deviation of the mean (n=3).
80 0.000 0.100 0.200 0.300 0.400 0.500 0.600 0.700 0.800 AdrostenedioneProgesteroneTMS-TestosteroneTMS-PregeneloneTMSEthinylestradiol TMS-EstroneTMS-EstradiolPeak Area Ratio Oasis HLB C18 LLCE Figure 2-5. Peak area ratios of th e steroids in spiked water were obtained by di viding the peak area of the analyte by the pea k area of the surrogate, anthracene. Error bars correspond to the standard deviati on of the mean (n=3).
81 0.000 0.100 0.200 0.300 0.400 0.500 0.600 0.700 AdrostenedioneProgesteroneTMS-TestosteroneTMS-PregeneloneTMSEthinylestradiol TMS-EstroneTMS-EstradiolPeak Area Ratio Oasis HLB C18 LLCE LLE Figure 2-6. Peak area ratios of the polar ED Cs in spiked plasma were obtained by divi ding the peak area of the analyte by the peak area of the surrogate, anthracene. Error bars corres pond to the standard deviation of the mean (n=3).
82 Deterioration of Fiber Deterioration of Fiber Figure 2-7. Deteriorated SPME fiber. The extraction phase on the section indi cated deteriorated over several extractions (with OFD).
83 CHAPTER 3 ENHANCEMENT OF CHEMICAL DERIVATIZATION OF STEROIDS BY GAS CHROMATOGRAPHY/MASS SPECTROMETRY (GC/MS) Introduction Metabolite profiling, particularly in the form of urinary steroi d profiles, has proven useful in a variety of fields, including the clinical diagnosis and etiology of endocrine disorders,182, 187, 188, 195, 200 the detection of steroid doping involved in competition146, 154 and livestock,139, 161, 162, 165 and the monitoring of steroids as environmental contaminants.215, 229, 304 The development of a comprehensive steroid profile method using gas chromatography/mass spectrometry (GC/MS) could provide an excellent mean s to gather more complete metabolic information, providing a better understanding of the mechanisms by which steroid pathways are altered. GC/MS and Derivatization GC/MS is well-suited for the identification of a large number of potential steroids and metabolites due to its high chromatographic re solution capacity and reproducible ionization efficiency;182, 188, 190 however, its success is dictated primar ily by the efficiency of the extraction and derivatization procedures empl oyed prior to injection of the sample. Derivatization replaces functional groups in an effort to make th e compound more amenable to standard GC/MS analysis, by increasing the volatility and/or thermal stability of a compound. Although derivatization can be time-consum ing, it allows for the profili ng of more compounds by allowing both polar and non-polar steroids to be successfully separated. Although carbonyl groups present no analytical challenge to GC/MS analys is, hydroxyl groups require modification.273 Since steroids may contain various combinations of carbonyl and hydroxyl grou ps, the comprehensive analysis of steroids is most successful with the selective deri vatization of only hydroxyl groups to minimize the formation of artifacts or undesirable derivatization products.280
84 Silylation Strategies The most commonly used derivatization me thodology for steroids is silylation, where active hydrogens on hydroxyl groups are replac ed with trimethylsilyl (TMS) groups.193, 241, 280 The important conditions to be optimized for the silylation of steroids are the reaction time and temperature,215, 277, 279 providing conditions that do not promote undesired derivatization,280 yet are still capable of driving the reaction to completion (typically 60 Â– 70 C for 30 Â– 90 minutes).103, 124, 215, 220, 226, 277, 278 In addition, there have b een several advances towards improving the derivatization process, including solvent enhancement273, 274, 276, 281-283 and alternative heating methods, such as mi crowave-accelerated de rivatization (MAD).196, 233, 287, 288 MAD has been shown to be an effective h eating method, producing derivatization conditions comparable to traditional methods while requi ring less reaction time. The use of sonicationassisted derivatization (SAD) has also been discussed190 as a method to improve derivatization efficiency, but there are no studies that explicitly investigate its potential. It is clear that for any effective derivatization, especially for comprehe nsive analysis, enhanci ng techniques should be considered and optimized. Current research in steroid analysis by GC/MS focuses on multi-analyte detection, targeting either small groups of related ster oids or steroids within specific classes.139, 161, 162, 203, 215, 229, 233, 279, 282, 287, 304 However, the lack of established protocols or standard methods for reliable steroid derivati zation is a deterrent to steroid analysis by GC/MS.305 Most published methods employing GC/MS for steroid analysis om it a discussion of the determination of the reaction settings, and often the r eaction conditions are not included in enough detail to determine whether or not optimal conditions were employe d for all analytes. To our knowledge, only a few research studies have investigated the op timization of the derivatization conditions;215, 233, 277, 279,
85 282, 304 however, these examinations were restricted to investigating only a small set of steroids rather than a comprehensive mixture. In this study, derivatization of a large suite of diverse steroids was systematically optimized using GC/MS by performing a detailed inve stigation of three sily lating reagents over a series of reaction times and temperatures. Th e role of derivatizati on enhancers was also examined, specifically the enhancing effects of fered by the use of solvents and non-traditional heating methods, such as MAD and SAD. The procedures developed not only describe the ideal derivatization strategi es for investigating steroids comprehens ively, but also give insight into the ideal reaction conditions required for steroids on an individual level. Experimental Section Chemicals and Reagents The natural and synthetic ster oids testosterone (T), 17 -estradiol (E2), estrone (E1), androstenedione (AE), ethinylestradiol (EE2), 17-methyltestosterone ( 17-MT), progesterone (P), pregnenolone (PREG), cholesterol (CHOL), cor ticosterone (CORT), and dihydrotestosterone (DHT), and the non-steroidal synthetic estrogen, diethylstilbestrol (DES) were acquired from Sigma (St. Louis, MO). The surrogate used for the evaluation of the deri vatization reactions was dichlorodiphenyldichloro ethylene-p,p (DDE, EPA Research Tr iangle Park, NC). Stock solutions of each steroid and the steroid mixture were made to 100 g mL-1 in analytical grade methanol (Fisher Scientific, Fair Lawn, NJ) and stored at Â–20 C. The reagents used were derivatization grade N,O -bis-(trimethylsilyl)-trifluoroacetamide (BSTFA) with 1% trimethylchlorosilane (TMC S) (BSTFA/TMCS, Supelco, Bellefonte, PA), N methylN -trimethylsilyl-trifluoroacetamide (M STFA) (Pierce, Rockford, IL), and N,O -bis(trimethylsilyl)-acetamide (BSA) (Pierce). The so lvents used in the r eactions were anhydrous
86 dimethylformamide (DMF, Sigma), anhydrous acetonitrile (ACN, Si gma), and extra-dry pyridine (PYR, Acros Organics, Morris Plains, NJ). GC/MS Analysis The GC/MS analysis was performed usi ng a Trace GC 2000 gas chromatograph/ quadrupole ion trap mass spectrometer with an AS3000 autosampler (Thermo Finnigan, San Jose, CA). The column used was a Zebron ZB-5 30 m x 0.25 mm capillary column with a film thickness of 0.25 m (Phenomenex, Torrance, CA). The ion source and transfer line temperatures were set to 200 C and 300 C, re spectively. The injector was set to splitless injection at a temperature of 280 C with a split fl ow of 50 mL min-1. The temperature program was set to begin at 120 C for 2 minutes, elevated at 15 C min-1 to 250 C, and finally increased by 5 C min-1 to 300 C (maintained for 5 minutes). Th e carrier gas used was ultra-high-purity helium (99.999%) at a flow rate of 1 mL min-1. The data acquisition software used was Xcalibur 1.4. Electron ionization was used at 70 eV and the mass spectrometer was set to full scan, m/z 50 Â– 600. Derivatization Experiments Derivatization overview Each sample contained the twelve selected st eroids and the surrogate and was dried down using ultra-high-purity nitrogen to represent sample conditions th at normally exist following solid-phase extraction. The residue was reconstituted with a volume of 200 L of derivatizing reagent (with surrogate to a final concentration of 5 g mL-1) and vortexed. The block and water bath reactions took place in a Thermolyne Type 16500 Dri-Bath and a Fisher Scientific Isotemp Immersion Circulator (Model 70, Pittsburgh, PA), respectively.
87 The characteristic ions selected for peak ar ea identification, shown in Table 3-1, consisted of the base peak and the molecular ion for each steroid and the surroga te. The relative response factor (RRF) for each of the products formed wa s calculated by dividing the peak area of the steroid product by that of the peak area of the surrogate (derivatization inactive). Derivatization time and temperature Each derivatization parameter was exam ined by running triplicate samples (200 L each) and pooling them post-heati ng into a single vial (600 L total). The pooled sample was then analyzed in triplicate using GC/MS. Optimizatio n of the derivatization procedure for the steroid standard mix consisted of examining reaction times and temperatures using three silyl reagents: BSTFA/TMCS, MSTFA, and BSA. The derivatizati on times and temperatures investigated were 15, 30, 45, and 60 minutes, and 40, 55, 70, and 90 C. The optimal time and temperature combination for each reagent was then used for the subsequent enhancement comparisons. Derivatization enhancing experiments Solvent enhancement was examined by adding so lvent and derivatizing reagent (1:1) to a total volume of 200 L for derivatization. MAD was evaluated by comparing the RRF va lues obtained to those obtained using traditional thermal (block) heating. The M AD was performed using a 1000-W Half-Time convection/microwave oven (Apollo Worldwide, Pa lm Beach, FL). The microwave settings for each reagent were: 900 W for 0.5 and 1 minut e (BSTFA/TMCS and MSTFA) and 900 and 1000 W for 1 min (BSA). SAD was examined by comparing the best overa ll method for each reagent using the block heater to the RRFs obtained using a heated sonicated water bath. The derivatization reactions in the water bath were run both with and without sonication to eval uate the potential improvements
88 achieved using sonication. The only deviation was that the water bath temperature for the BSA reaction was at 80 C instead of 90 C, due to the temperature constrai nts of the water bath heater. Results and Discussion Derivatization Optimization Derivatization time and temperature The steroid mixture was selected to evaluate the effects of the possi ble combinations of functional groups (carbonyl and hydroxyl) at various locations on the steroid structure. Even though the steroids analyzed were all at the same concentration, they exhibited a wide range of RRFs. Due to keto-enol tautomerism, derivatiz ation efficiency, and the tendency of carbonyl groups to derivatize under harsh conditions, a ba lance in reaction time and temperature was required. The selection of target derivatives wa s based on the optimal product (underivatized or derivatized) for each steroid that would be the best compromise for allowing comprehensive profiling for the wide polarity range of steroi ds examined. Two steroids, PROG and AE, contain only carbonyl groups and did not require deriva tization, while the predominant product observed for 17-MT was also underivatized due to the steric hindrance of the hydroxyl group at the 17 position. The remaining hydroxylated steroids we re sufficiently deriva tized; only di-TMS-EE2 was difficult to derivatize, most likely due to the triple bond near the hydroxyl group. Shareef et al. concluded that this may be due to a breakdown of the EE2-derivative;282 however, this may also be due to more favorable condi tions for the production of mono-TMS-EE2. Di-TMS-DES had two GC peaks (two isomers); only the dominant peak (tr = 12.12 min) was used for the RRF determination. In general, the derivatization conditions beyond 60 minutes and above 70 C (90 C for BSA) did not provide any added advantage in the RRFs obtained. In addition, temperatures of 40
89 C or below were not adequate for efficient deri vatization at the reaction times tested in this study. The optimal conditions (reagent, reaction time and temperature) for obtaining the greatest RRF value for each individual steroid are displayed in Table 3-2. The derivatization of the steroid mixture us ing BSTFA/TMCS was most successful at producing the highest RRF values in the range of 55 Â– 70 C fo r 15 Â– 30 minutes. As shown in Table 3-2, seven of the twelve steroids analyzed displayed the highest RRF using BSTFA as the reagent. This reagent was the most successful in creating conditions that allowed the highest RRF values of PROG, AE and 17-MT. This is a key result in the effort to reduce sample complexity by avoiding the derivatization of compounds which are amenable to GC/MS. Conditions beyond 30 minutes using BSTFA/TMCS for the three underivatized compounds (PROG, AE, and 17-MT) exhibited a reduced RRF valu e, likely due to the formation of artifacts (data not shown). Furthermore, BSTFA/TMCS was also ideal for the deri vatization of most of the mono-derivatives, including CHOL, E1 and PREG. The best se t of conditions using BSTFA/TMCS for profiling the multi-class mixture was determined to be 70 C for 30 minutes. For the derivatization of the steroid mixture using MSTFA, temperatur es higher than 55 C and reaction times longer than 30 minutes exhibited no increase in RRF values. As shown in Table 3-2, four of the twelve targeted derivativ es had the highest RRF values using the MSTFA reagent: di-TMS-EE2, mono-TMS-E1, mono-TMS-T, and mono-TMS-DHT. However, the RRF values for the underivatized steroids PROG, AE, and 17-MT, were much lower than the levels achieved using BSTFA/TMCS. The best overall reaction condition for using MSTFA reagent was 55 C for 30 minutes. The RRF values obtained with the BSA reagent we re in almost all cases lower compared to those obtained with the other two silyl reagents, particularly fo r the underivatized steroids. As
90 shown in Table 3-2, the RRF values generated using BSA for two of the eleven derivatized steroids examined were lower (typically by a fact or of 10) than the RRF values achieved using the other reagents; the elevent h, di-TMS-DES, the most easily de rivatized species of the study, yielded a 2x higher RRF with BSA. The best reaction condition for the multi-class steroid analysis using BSA was 90 C for 30 minutes. Derivatization reagent For the derivatization of steroids with tw o or more hydroxylated groups, MSTFA is the best reagent choice. BSTFA is the more approp riate reagent choice for the derivatization of mono-hydroxylated compounds. Both MSTFA and BSTF A were comparable in their ability to derivatize the multi-class steroid mixture; however, the optimal reaction condition for BSTFA/TMCS (at 70 C for 30 min) offers the best compromise, with the ability to derivatize a wide variety of steroi ds without the derivatiz ation of carbonyl groups. Derivatization Enhancers Solvent enhancement The RRF values for the steroid products obt ained by adding a solvent (1:1 to each silyl reagent) during the derivatizati on reaction were evaluated agains t the RRF values generated for derivatization reactions without solvent. Tables 3-5 to 35 present the change in RRF for enhanced derivatization compared to those obt ained with the traditi on block derivatization method. BSTFA/TMCS, with the solvents DMF, ACN, and PYR, generally showed an increase in RRF values over those generated without the ai d of solvents (Table 3-3). The underivatized steroids, AE and PROG, had higher RRF values with the addition of solvent (except for AE with DMF). However, underivatized 17-MT exhibite d low RRF values (using DMF and PYR), implying that its derivatization was facilitated. ACN was the most effective solvent in preventing
91 the derivatization of carbonyl st eroids, and thus had the largest RRF values for these compounds. The RRF values for di-TMS-EE2 exhibited dramatic increases with the addition of DMF and PYR. It has been suggested that those so lvents prevent the br eakdown of di-TMS-EE2.277, 281, 282 MSTFA reactions employing solvents (Table 3-4) resulted in essentia lly the opposite effect in RRF values when compared to BSTFA/TMCS. A comparison of the RRF values showed no benefit when compared to the conventional solvent-less block heating method. No solvent enhancement was observed with MSTFA. The RRF values obtained using BSA with solven ts generally exhibited either an increase or no change (Table 3-5). Pyridine provided s lightly better RRFs for th e derivatizat ion of monoTMS steroids, while DMF was the best at protecting the underivatized steroids. Overall, the addition of solvent during th e derivatization process was shown to be beneficial in creating a more effective reaction for both BSA and BSTFA/TMCS. For BSTFA/TMCS, the use of PYR, ACN, and to a le sser extent DMF, genera ted higher RRF values indicating an improved derivatiz ation reaction. In addition to increased RRF values, solvent usage during the derivatization re action reduced the amount of reag ent needed, thus reducing the cost per sample. Microwave-accelerated derivatization (MAD) The RRF values generated in the microwave are compared to the RRF values obtained using the traditional block heater in Tables 33 to 3-5. The microwave experiment was designed solely to analyze the change in RRF values a nd not to compare RRF values at similar reaction temperatures. MAD with BSTFA/TMCS, MSTFA and BSA at 900 W for at least one minute provided higher RRF values than those generated using th e traditional block derivatization for all eleven steroids except for di-TMS-EE2 with BSA. However, the MAD reaction at 900 W for 30 seconds
92 was not as effective as the 1-minute microwave reaction for both MSTFA and BSTFA/TMCS (except for mono-TMS-T with BSTFA/TMCS). The underivatized steroids had higher RRF values with use of microwave heating, highlighti ng the unique potential of microwave heating to provide strong enough heating conditi ons for the effective derivati zation of steroids, yet still providing an environment which does not for ce the derivatization of carbonyl groups. Silyl reagents combined with microwave heat ing can be a successful technique for the rapid derivatization of a variety of different steroids. Furthermore, as shown here, the application of MAD often provided higher RRF values when co mpared to traditional heating methods with a drastic reduction in derivatization time (1 minute to 30 minutes). Sonication-assisted derivatization (SAD) The concept of SAD is based on the promoted agitation of compounds in a heated solution, thus potentially increasing the frequency of interaction between the steroids and the derivatization reagent. Sonication enhancemen t was examined by comparing the RRF values generated in a sonicated water bath to those obtained using th e same time and temperature on a block heater. The RRF values employing SAD for all three silylating reagents exhibited an increase when compared to those generate d using a traditional block heater (Tables 3-3 to 3-5). However, SAD yielded very similar RRF values to those obta ined for derivatization in a water bath without sonication, indicating little or no be nefit of sonication for the deriva tization of steroids. The data do, however, highlight the potential of water bath heating for more effici ent derivatization than block heating. Conclusion This investigation was focused on the optimi zation and enhancement of methodologies for the derivatization of a wide polar ity range of steroids in a si ngle chromatographic analysis. The
93 results have defined more accurately the optim al derivatization conditions needed on three levels: 1) for each individual steroid, 2) fo r groups of related steroids, and 3) for the comprehensive profiling of a suite of unrelated steroids. In addition, the resulting information serves as a tool for future reference in the pr ediction of steroid deriva tization for steroids for which there is no analytical deri vatization protocol avai lable. The application of solvent and the use of microwave heating were f ound, in most instances, to be more efficient than the optimized traditional heating methods. The use of solv ent was found to be most effective with BSTFA/TMCS, resulting in a general increase in RRF values with a reduced amount of reagent needed per sample. Microwave-accelerated deri vatization at 900 W for 1 minute provided an increase in RRF values for all the steroids ex amined in the study and reduced the derivatization time (1 minute to 30 minutes). Sonication-as sisted derivatization did not enhance the derivatization of the st eroids, but did highlight the potential of water bath heating for derivatization. Future investigations will include a rigorous examination of the combinatorial effects of the derivatization enhancing techniqu es and its application to biological samples.
94 Table 3-1. Targeted deriva tives and their characteristic ions and retention times Steroid Abbreviation Functional Groups Molecular Weight Target Derivative Characteristic Ions Retention Time (min) Surrogate DDE none 318 Underivatized 246, 318 10.93 Diethylstilbestrol DES 2 C-OH, C-OH 268 Di-TMS 412 217 12.12 Dihydrotestosterone DHT C=O, C-OH 290 Mono-TMS 362 347 14.36 Estrone E1 C=O, C-OH 270 Mono-TMS 342 257 14.49 Androstenedione AE 2 C-OH, C-OH 286 Underivatized 148, 286 14.68 17 -Estradiol E2 2 C-OH, C-OH 272 Di-TMS 416 285 14.90 Testosterone T C=O, C-OH 288 Mono-TMS 360 226 15.02 17-Methyltestosterone 17-MT C= O, C-OH 302 Underivatized 229, 302 15.06 Pregnenolone PREG C=O, C-OH 316 Mono-TMS 298, 388 15.63 Ethinylestradiol EE2 2 C-OH, C-OH 296 Di-TMS 425, 440 16.02 Progesterone P 2 C-OH, C-OH 314 Underivatized 124, 314 16.42 Cholesterol CHOL C-OH 386 Mono-TMS 458 368 19.46 Corticosterone CORT 2 C=O, 2 C-OH 346 Di-TMS 490 475 21.57 Italics indicate molecular ion ( m/z ), while the other ion listed is the base peak. The surrogate chosen was DDE due to its de rivatization inactivity.
95 Table 3-2. The RRF values obtained using various time, temperature and derivatiz ing reagent combinations MSTFA BSTFA with 1% TMCS BSA 55C 55C 70C 55C 90C 15 min 30 min 45 min 15 min 30 min 15 min 30 min 60 min 45 min 30 min AE 0.12 0.22 0.16 0.26 0.26 0.25 0.24 0.16 0.04 0.05 17-MT 0.08 0.16 0.21 0.27 0.29 0.26 0.27 0.14 0.05 0.05 PROG 0.13 0.18 0.15 0.25 0.25 0.24 0.24 0.12 0.03 0.03 Mono-TMS-CHOL 0.50 0.49 0.50 0.53 0.53 0.57 0.48 0.52 0.18 0.27 Mono-TMS-DHT 0.15 0.16 0.13 0.13 0.13 0.13 0.13 0.13 0.04 0.06 Mono-TMS-E1 2.24 2.59 2.20 2.60 2.40 2.85 2.88 2.68 0.99 1.30 Mono-TMS-PREG 0.43 0.49 0.44 0.53 0.56 0.57 0.53 0.55 0.12 0.18 Mono-TMS-T 0.40 0.46 0.44 0.34 0.32 0.31 0.31 0.18 0.12 0.14 Di-TMS-CORT ND ND 0.06 0.06 0.03 0.03 0.04 0.09 0.03 0.05 Di-TMS-DES 2.04 2.39 2.43 1.89 1.78 1.68 1.91 2.73 3.44 3.08 Di-TMS-E2 2.18 2.13 1.88 1.65 1.57 1.57 1.84 1.56 0.97 1.22 Di-TMS-EE2 0.67 0.68 0.69 0.07 0.06 0.08 0.12 0.09 0.14 0.20 ND = not detected.
96 Table 3-3. The RRF changes due to various de rivatization enhancemen ts with BSTFA/TMCS BSTFA + 1% TMCS DMF Pyridine Acetonitrile Microwave 30s Microwave 1min Water Bath Sonication AE 0.8 0.2 1.3 0.2 1.6 0.2 1.6 0.0 2.3 0.0 1.1 0.1 1.0 0.0 17-MT 0.3 0.0 0.5 0.0 1.0 0.1 1.6 0.0 2.3 0.1 1.1 0.1 0.9 0.1 PROG 1.1 0.1 1.4 0.1 1.7 0.1 1.4 0.1 2.2 0.0 1.2 0.0 1.1 0.0 Mono-TMS-CHOL 1.5 0.2 1.4 0.0 1.3 0.1 0.7 0.0 1.8 0.2 0.9 0.1 0.9 0.1 Mono-TMS-DHT 1.8 0.1 2.0 0.0 2.3 0.0 1.0 0.1 1.4 0.1 1.0 0.0 1.1 0.1 Mono-TMS-E1 0.9 0.1 1.0 0.0 1.1 0.0 0.8 0.0 1.4 0.0 0.9 0.1 0.9 0.0 Mono-TMS-PREG 1.1 0.1 1.2 0.1 1.1 0.1 0.8 0.0 1.6 0.2 1.0 0.0 1.0 0.1 Mono-TMS-T 1.7 0.1 1.8 0.0 1.9 0.1 4.5 0.0 2.8 0.3 1.2 0.1 1.2 0.0 Di-TMS-DES 1.7 0.1 1.6 0.0 1.4 0.0 0.9 0.1 1.0 0.0 0.9 0.0 0.9 0.0 Di-TMS-E2 1.6 0.0 1.5 0.2 1.5 0.2 0.7 0.1 1.4 0.1 1.0 0.0 1.1 0.0 Di-TMS-EE2 12.0 1.4 9.1 1.6 2.8 1.5 0.7 0.1 2.0 0.1 1.1 0.0 1.3 0.0 Values are averages of three in jections, normalized to the block heating method w ithout enhancement. Highlighted values indica te the highest RRF value obtained for each steroid. Microwave power was se t to 900 W. Values greater than 1 indicate an increase in RR F over the block method; values less than 1 i ndicate a decrease in RRF. The averages are shown the standard deviation of the me an (also normalized). Error of 0.0 indicates an error below the significant figures shown. Di-TMS-C ORT was not effectively deriva tized with any of the enhancing experiment s and was not included in the tables.
97 Table 3-4. The RRF changes due to various derivatization enhan cements with MSTFA MSTFA DMF Pyridine Acetonitrile Microwave 30s Microwave 1min Water Bath Sonication AE 0.7 0.0 0.4 0.0 0.5 0.0 1.1 0.0 1.5 0.1 1.2 0.1 1.0 0.0 17-MT 0.3 0.0 0.5 0.0 0.8 0.0 1.1 0.0 1.8 0.6 1.1 0.0 1.1 0.0 PROG 0.7 0.0 0.3 0.0 0.5 0.0 1.0 0.1 1.5 0.0 1.3 0.1 1.2 0.1 Mono-TMS-CHOL 1.1 0.0 0.8 0.1 1.1 0.1 1.2 0.1 1.4 0.1 1.5 0.1 1.4 0.0 Mono-TMS-DHT 0.8 0.1 0.8 0.0 0.9 0.0 1.0 0.1 1.0 0.1 1.1 0.0 1.1 0.0 Mono-TMS-E1 1.0 0.1 0.9 0.0 1.0 0.0 1.1 0.0 1.1 0.0 1.6 0.0 1.5 0.1 Mono-TMS-PREG 0.9 0.1 0.8 0.0 1.0 0.0 1.1 0.1 1.1 0.1 1.3 0.1 1.2 0.0 Mono-TMS-T 0.8 0.0 0.7 0.0 1.0 0.0 1.0 0.0 1.2 0.1 1.2 0.0 1.2 0.1 Di-TMS-DES 1.0 0.0 1.0 0.0 1.0 0.1 1.1 0.1 1.1 0.1 1.1 0.0 1.1 0.0 Di-TMS-E2 0.9 0.0 0.9 0.1 1.0 0.0 0.7 0.0 1.1 0.1 1.2 0.0 1.1 0.0 Di-TMS-EE2 1.3 0.2 1.1 0.1 1.2 0.2 0.8 0.1 1.2 0.1 1.7 0.0 1.6 0.1 Values are averages of three in jections, normalized to the block heating method w ithout enhancement. Highlighted values indicat e the highest RRF value obtained for each steroid. Microwave power was se t to 900 W. Values greater than 1 indicate an increase in RR F over the block method; values less than 1 i ndicate a decrease in RRF. The averages are shown the standard deviation of the me an (also normalized). Error of 0.0 indicates an error below the significant figures shown. Di-TMS-C ORT was not effectively deriva tized with any of the enhancing experiment s and was not included in the tables.
98 Table 3-5. The RRF changes due to various derivatization enha ncements with BSA BSA DMF Pyridine Acetonitrile Microwave 1min Microwave 1min1 Water Bath Sonication AE 1.5 0.0 1.4 0.1 1.3 0.0 1.5 0.3 1.6 0.2 1.2 0.0 0.9 0.1 17-MT 1.7 0.1 1.3 0.3 1.3 0.1 1.5 0.1 1.6 0.1 1.3 0.0 1.0 0.1 PROG 1.6 0.0 1.5 0.1 1.2 0.1 1.5 0.2 1.5 0.2 1.2 0.0 1.2 0.1 Mono-TMS-CHOL 1.0 0.0 1.1 0.0 0.9 0.0 1.1 0.0 1.1 0.0 1.3 0.1 1.2 0.1 Mono-TMS-DHT 1.6 0.2 1.9 0.1 1.0 0.1 1.5 0.1 1.6 0.2 1.3 0.2 0.2 0.0 Mono-TMS-E1 1.0 0.0 1.3 0.0 1.0 0.1 1.2 0.1 1.3 0.1 1.1 0.0 1.0 0.0 Mono-TMS-PREG 1.0 0.0 1.1 0.1 1.0 0.0 1.1 0.0 1.2 0.0 1.1 0.0 1.2 0.1 Mono-TMS-T 1.3 0.1 1.4 0.2 1.1 0.1 1.4 0.1 1.4 0.1 1.2 0.0 1.1 0.1 Di-TMS-DES 1.0 0.0 1.0 0.0 1.0 0.1 1.1 0.0 1.2 0.0 1.1 0.1 0.9 0.0 Di-TMS-E2 1.0 0.0 1.0 0.1 1.0 0.1 1.2 0.0 1.3 0.1 1.1 0.0 1.1 0.0 Di-TMS-EE2 0.9 0.1 0.8 0.0 0.7 0.0 1.6 0.1 1.4 0.0 1.1 0.3 0.7 0.1 1Microwave at 1000W, rather than 900W. Values are averages of three inj ections, normalized to the block heating method without enhancement. Highlighted values indicate the highest RRF value obt ained for each steroid. Values gr eater than 1 indicate an inc rease in RRF over the block method; values less th an 1 indicate a decrease in RRF. The averag es are shown the standard deviation of the mean (also normalized). Error of 0.0 indicates an error below the significant figures shown. Di -TMS-CORT was not effectively derivatized with any of the enhancing experiments and was not included in the tables.
99 CHAPTER 4 ENHANCED ANALYSIS OF STEROIDS BY GAS CHROMATOGRAPHY/ MASS SPECTROMETRY (GC/MS) USING MICROWAVE-ACCELERATED DERIVATIZATION (MAD) Introduction The unrivaled chromatographic resolving power of capillary gas chromatography/mass spectrometry (GC/MS) has made it an indispensa ble tool for the characterization of steroid trends associated with metabolic disorders,154, 182, 187, 188, 195 doping in sports,146, 162, 165 and the monitoring of steroids as chemical pollutants;106, 121, 124, 125 however, the time-consuming extraction and derivatization proced ures needed for steroid analys is by GC/MS have traditionally impeded further advancement in the field. Sample derivatization is required to improve the volatility and thermal stability of steroids containing hydroxyl groups, yet the difficulty of optimizing sample preparation techniques for a la rge suite of varying components has led most steroid analysis to be performed in smaller subsets by liquid chromatography/mass spectrometry (LC/MS),85, 107, 168, 185, 203, 306 liquid chromatography/tandem ma ss spectrometry (LC/MS/MS),202, 209, 307-309 and gas chromatography/tandem mass spectrometry (GC/MS/MS).119, 120 In an effort to take advantage of the profiling capabilities of GC /MS for steroid analysis, the search for more efficient derivatization techniques has predom inantly focused on increasing yield, stability, selectivity, and reducing the formati on of undesirable products or artifacts.273, 280 For steroid analyses, there are several common derivatization strategi es for GC/MS, including the optimization of reaction times (30 Â– 180 minutes) and temperatures (60 Â– 70 C) with various trimethylsilyl reagents (TMS),215, 273, 274, 276, 277, 279, 282 the application of N -methylN -[tertbutyldimethyl-silyl]trifluoroacetamide (MTB STFA) for increased derivative stability,284, 285 the application of the two-step methoxime derivatization,164, 184, 207, 310, 311 and solvent enhancement as a means to reduce artifact formation.274, 276, 281, 282 Traditional thermal derivatization methods
100 for steroid analysis employ block heaters or water baths, which of ten require considerable time for the bulk solution to reach conditions capable of comple ting the reaction. High reaction temperatures and extended reaction times can be a problem for the derivatization of a comprehensive mixture; steroids typically c ontain some combination of hydroxyl and ketone functional groups; while ketone groups pose no diffi culty to GC/MS analysis, they are prone to unwanted derivatization under certain reaction conditions.280 Thus, the development of a selective derivatization method wh ich specifically targets hydroxyl groups and minimizes ketone derivatization would provide the best derivatization stra tegy. Innovative methods which minimize derivatization conditions (reaction time and temperature) while still providing the capability of comprehensive derivatization have not been actively pursued. Microwave-Accelerated Derivatization (MAD) Over the last two decades, microwave techno logy has become an ideal alternative for applications which require heating of solutions The application of mi crowave technology has been successfully adapted in the areas of sample drying,312-314 digestion,315-317 extraction,318, 319 organic synthesis,320-324 and has begun to cross over in to other techniques, including derivatization. Microwave-accelerat ed derivatization (MAD) has b een successfully implemented in the derivatization of sugars,325 amino acids,289 fatty acids,288 aliphatic alcohols,326 select steroids,196, 233, 287, 288 and a few other compounds,319, 327, 328 although these studies often employ domestic microwave ovens and typically omit disc ussions regarding the optimization of reaction conditions. Organic and derivatization r eactions have employed microw ave heating not only because of its ability to reduce analysis time, but also because it ofte n produces comparable or higher yields to traditional methods.287, 289, 321, 325, 327, 328 Traditional thermal derivatization methods conductively heat the sample, thus prolonged start up conditions are often ne cessary to facilitate
101 the heating of the entire bulk solution to a certain reaction temperature. In comparison, microwave systems heat volumetrically (entire sample all at once), resulting in a more efficient way to bring the bulk solution to the required r eaction temperature. The microwave heating of a solution is strongly dependent upon, among other things, the presence of polar compounds (reactants, reagents, solvents) in the solution th at will absorb the micr owave energy. The ability of a compound in solution to heat a solution inside a micr owave oven and provide a rate enhancement is directly related to two intrin sic properties: the com poundÂ’s dielectric constant ( Â’) and dielectric loss ( Â”). The dielectric constant of a co mponent, or the magnitude of the permanent dipole, is a measure of the compone ntÂ’s ability to be polarized by microwave energy.322-324, 329, 330 Components with large dielectric consta nts (or strong dipoles) try to align with the field, creating heat by ag itation and intermolecular friction.321-324, 330 The dielectric loss of a component describes how e fficiently this energy is conv erted to heat (compounds with dielectric loss values between 1 and 10 are consider ed to dissipate heat well to the solution). The resulting heat (thermal effects) dissipates and directly heats the solution, and is one of the explanations for the enhanced reaction rates. Ot her explanations for the resulting rate increase include specific and non-thermal microwave eff ects. Specific microwav e effects, which are essentially specific thermal effects characteri stic of only microwav e heating, include the superheating of solvents, generation of hot s pots due to selective heating of individual components, and volumetric heating.322-324, 329-331 Non-thermal effects are generally microwave phenomena which are not thermal in nature.329, 330 The relative contribution of the various microwave effects are largely unknown, leading to differing opini ons regarding the fundamentals of this technique.321, 329, 332 Clearly, some combination, if not all of the effects, contribute to the enhancements in derivatization achieved over traditional methods.
102 MAD with Polar Organic Solvents Combining polar organic solvents with varyi ng dipole moments with silyl reagents offer differing capabilities to absorb microwave energy and transfer the generated heat to rest of the solution. For traditional thermal derivatization methods, the application of solvents with silylating reagents is typically employed fo r solubility purposes. In addition, the added advantages of using solvents for steroid derivatization often include an inherent catalytic capability276 and an ability to decrease the produc tion of partial or unwanted derivatives.277, 281, 282 The many benefits of using certain solvents for traditional thermal derivatization in combination with the microwave heating benefits derived from using polar solvents should be present, and these effects we re examined in this study. MAD with Several Derivatization Strategies Trimethylsilyl reagents such as N,O -bis-(trimethylsilyl)-trifluoroacetamide (BSTFA) and N -methylN -trimethylsilyl-trifluoroacetamide (MSTFA ) are the most common reagents for the derivatization of steroids by GC/MS.172, 273, 276 The application of MAD and silyl reagents can offer the ability to improve th e derivatization response while lowering overall analysis time compared to traditional thermal methods at longer analysis times. Howeve r, silyl derivatization still has inherent flaws not associated with the time-consuming reaction time, in particular, irreproducibility and the inability to avoid unwanted product formation and the decomposition of desired products.280 These flaws have led to the creation of alternative deriva tization strategies, most notably the introduction of methoxime (MO) and tert-butyldimethylsilyl (TBDMS) derivatives. Methoxime derivatiz ation, a two-step strategy (s hown in Figure 4-1), protects derivatization-sus ceptible ketone groups by converting them to methoximes prior to the trimethylsilyl derivatizati on of the hydroxyl groups.172, 193, 273 N -methylN -[tert-butyldimethylsilyl]trifluoroacetamide (MTBSTFA), which fo rms TBDMS derivatives (shown in Figure 4-2),
103 was created to solve the stability concerns associated with TMS-derivatives, as TBDMSderivatives are more stable.284, 285 To date, MAD has not been implemented to eliminate the timeconsuming reaction conditions commonly a ssociated with these techniques. In this study, MAD was examined as an al ternative heating approach for the rapid derivatization of steroids using a synthesis microwave system First, the maximum temperatures achieved in one minute at various power levels were recorded and aided in the understanding of the microwave heating contributions of the derivatizing reagent, solvent, and combinations of reagent-solvent. Secondly, optimal microwav e derivatization condi tions (reaction time, microwave power, and derivatizat ion reagent) were systematically optimized. Unlike current examinations with domestic microwave ovens, MAD with the synthesis microwave system allowed an evaluation on two levels: 1) compar ing relative response factors (RRFs) generated between MAD and traditional thermal methods at the same reaction temperature (made possible by the ability of the synthesis microwave system to explicitly control microwave power) and 2) comparing traditional thermal methods to MAD at different radiation powers regardless of temperature. The absolute RRF values generated with the synthesis microwave system were also compared to those generated with a domestic microwave oven. In additio n, this investigation studied the use of various organi c solvents during the MAD reaction and the resulting increase in RRF obtained when compared to those generated without the use of solvent. Furthermore, two other widely used but tedious de rivatization strategies (methoxime and tert-butyldimethylsilyl) for steroid analysis were also examined and were implemented with shorter reactions times using MAD. Concomitantly, this examination also exam ined the efficacy of using MAD in obtaining a comparable derivatization for a multi-class ster oid mixture to traditional thermal techniques. Finally, MAD was compared to th ermal derivatization in analyzi ng steroids spiked in plasma.
104 Experimental Section Steroids, Solvents, and Reagents The multi-class mixture of steroid and ster oid-related compounds, testosterone (T), 17 estradiol (E2), estrone (E1), androstenedione (AE) ethinylestradiol (EE2), 17-methyltestosterone (17-MT), progesterone (P), pregnenolone (PREG) cholesterol (CHOL), corticosterone (CORT), diethylstilbestrol (DES), and dihydrotestosterone (DHT), were all acquired from Sigma (St. Louis, MO). The surrogate chos en for the determination of th e relative response factors was dichlorodiphenyldichlo roethylene-p,p (DDE, EPA Research Triangle Park, NC). DDE was chosen due its inactivity during th e derivatization process. All components in the standard steroid mixture and the surrogate were made to 5 ppm (5 g mL-1) in analytical grade methanol (Fisher Scientific, Fair Lawn, NJ). The methanol was blown down using ultra-high-purity (UHP) nitrogen. The resulting residue (comparative to post-extraction conditions) was derivatized by adding derivatizing reagent (and solvent wh en applicable) to a total volume of 200 L. The reagents used for the derivatization analyses were derivatization grade N,O -bis(trimethylsilyl)-trifluoroacetamide (BSTFA) with 1% trimethylchlorosilane (TMCS) (BSTFA/TMCS, Supelco, Bellefonte, PA), N,O -bis-(trimethylsilyl)-trifluoroacetamide (BSTFA) (Supelco), N -methylN -trimethylsilyl-trifluoroacetamide (MSTFA) (Pierce, Rockford, IL), N,O bis-(trimethylsilyl)-acetamide (BSA) (Pie rce), MOX reagent (2% methoxyamineHCl in pyridine) (Pierce), and N -methylN -[tert-butyldimethyl-silyl]t rifluoroacetamide (MTBSTFA) (Pierce). The solvents examined for the reagent:solve nt derivatization reac tions were anhydrous dimethylformamide (DMF, Sigma), anhydrous acet onitrile (ACN, Sigma), hexane (HEX, Fisher
105 Scientific) and extra-dry pyridine (PYR, Ac ros Organics, Morris Plains, NJ). Methyl tert -butyl ether (MTBE) was acquired from Fisher Scientific. Instruments and Apparatus Traditional thermal derivatization A Thermolyne Type 16500 Dri-Bath heater was obtained from Sybron Corporation (Dubuque, IA). The temperature was monitored using a thermometer and reactions were performed in 4 mL glass vials with screw caps. Synthesis microwave system A Discover S-Class synthesis microwave system (2455 MHz, 300 W max power) with Synergy data acquisition soft wate (version 1.34) was obtained from CEM (Mathews, NC). The microwave system was equipped with simulta neous control and real-time monitoring of temperature, power and pressure. The temperat ure of the microwave reaction was monitored with an infrared sensor inside the microwave vessel. Reactions were performed in 10 mL CEM glass microwave vials equipped wi th specialized snap caps. An overhead view of the Discover SClass synthesis microwave system is shown in Figure 4-3. Domestic microwave oven A Half-Time convection/microwave oven (2450 MHz, 1000 W max power) was obtained from Apollo Worldwide, Inc. (P alm Beach, FL). It was equipped with power cont rol (10 levels) and had no real-time monitoring cap abilities. Reactions were perfor med in 4 mL glass vials with screw caps. Gas chromatography/mass spectrometry (GC/MS) GC/MS was performed using a ThermoFi nnigan Trace GC 2000 gas chromatograph/ quadrupole ion trap mass spectrometer with an AS3000 Autosampler (San Jose, CA). Xcalibur 1.4 was the data acquisition software employe d. The Trace GC was fitted with an SLB-5ms
106 capillary column with dimensions of 30 m x 0.25 mm and a f ilm thickness of 0.25 m (Supelco). Splitless injection was employed at a temperature of 280 C with a split flow of 50 mL min-1. The temperature program was set initially at 120 C (held for 2 minutes), then elevated 15 C min-1 to 250 C, and finally increased 5 C min-1 to 300 C (held for 5 minutes). The ion source and transfer line temperatures were set to 200 C and 300 C, respectivel y. The carrier gas was ultra-high-purity (UHP) he lium (99.99%) and used at a flow rate of 1 mL min-1. Derivatization Procedures Maximum temperatures Maximum temperatures (Table 4-1) were obt ained individually for each of the reagents used (BSTFA/TMCS, BSTFA, MSTFA, BSA, MOX, and MTBSTFA), at three power levels for each of the solvents used (ACN, DMF, PYR, and HEX), and for 1:1 combinations of BSTFA/TMCS with each of the solvents. Maximu m temperatures were obtained by loading a total volume of 200 L of a sample into a 10 mL microwave vial and heating at different power levels (100, 200, and 300 W) for 1 minute. Each sample was run at each power level in triplicate, and allowed the prediction of maximum temperatures for all subsequent derivatization reaction conditions within the microwave. Derivatization overview In each derivatization experiment the mixture of all twelve steroids was added to either a 10 mL microwave vial or to a 4 mL block-heat er vial and evaporated under a flow of UHPnitrogen gas. The steroids were reconstituted using 200 L of the derivatizing reagent or reagent-solvent mixture to a concentration of 5 ppm (5 g mL-1) and were heated according to the derivatization procedures (in triplicate). The solutions were then evaporated under UHPnitrogen and reconstituted to the same concentrat ion using isooctane (Fisher Scientific). Relative response factors (RRFs) were dete rmined by dividing the peak area of the steroids (for target
107 underivatized or derivatized st eroid products) by the peak area of the surrogate, DDE. The characteristic ions (molecular ion and base peak ) summed to calculate the peak area for each of the expected steroid produc ts are shown in Table 4-2. Di-TMS-CORT was not effectively derivatized with any of the enhancing experiments (except under certain conditions with MSTFA) and is not included in the tables. Tables 4-3, 4-5, 4-6, and 4-7 show RRF values for eleven steroids normalized to those for the thermal heating method either at the same temperature or at set powers regardless of temperature. Table 4-8 shows these RRF values normalized to a solvent-less method. Values greater than one indicate an increase in RRF over the thermal method; values less than one indicate a decrease in RRF. Absolute RRF valu es (shown in Figures) represent the actual RRF values calculated for each steroid. Traditional methods For each reagent, the thermal derivatizati on method (reaction time and temperature), was optimized (from Chapter 3); the data from these op timized conditions were used as a reference to measure the enhancement achieved by MAD. The BSTFA/TMCS reaction was performed at 70 C for 30 min, MSTFA at 55 C for 30 min, BS A at 90 C for 30 min, MOX at 60 C for 180 min and at 60 C for 30 min (both followed by BSTFA/TMCS at 70 C for 30 min), and MTBSTFA at 60 C for 120 min. Synthesis microwave system vs traditional heating methods The first experiment directly compared MAD with the synthesis microwave system to the traditional block-heating method. Each silyl re agent (BSTFA/TMCS, MSTFA, and BSA) was run in three separate segments: (1) using the traditional thermal derivatization (as previously described), (2) using the synthesi s microwave system derivatizati on at various wattages for one minute (BSTFA/TMCS Â– 200 and 300 W, MSTFA Â– 100 and 300 W, and BSA Â– 300W), and (3)
108 using the synthesis microwave system at the traditional thermal derivatization temperature for one minute. The two thermal MOX reagent met hods were performed by first employing thermal derivatization methods for both the methoxime and TMS reactions (using BSTFA/TMCS) and comparing it to the same MOX/BSTFA/TMCS appr oach but with each step performed with the synthesis microwave system at 300 W for one minute. Derivatizati on with MTBSTFA was performed with the synthesis microwave system at 300 W for one minut e; the results were compared to the previously described thermal derivatization method. Fina lly, derivatization with BSTFA/TMCS was performed with the synthesis mi crowave system at different irradiation times (0.5, 1, and 2 minutes). Synthesis microwave system vs. domestic microwave oven The synthesis and domestic microwave ovens were directly compared using both BSTFA/TMCS and MSTFA. Each reagent was run separately for one minute using the synthesis microwave system at 300 W, the domestic microwave oven at 300 W, and the domestic microwave oven at full power (1000 W). In addi tion, percent RSD of the RRF values obtained using MSTFA and BSTFA/TMCS were also compar ed between the synthesis microwave system and the domestic microwave oven (shown in Table 4-4). Solvent enhancement with MAD The ability of each solvent to absorb micr owave energy and enhan ce the derivatization reaction was analyzed. MAD with BSTFA/TMCS was analyzed us ing the synthesis microwave system for one minute at 300 W with each of the four organic solvents (ACN, PYR, DMF, and HEX) at a ratio of 1:1, and se parately without solvent. A fo llow-up experiment was performed with ACN to determine the approximate composition (0:1, 1:3, 1:1, and 3:1) of solvent needed to achieve enhancement during derivatization.
109 Microwave heating of plasma extract The steroid mixture was adde d to six vials and subseque ntly blown down with UHPnitrogen. One milliliter of expired human plasma (past medicinal use date, from local blood bank) was added to all of the vials and vortexed. The six samples were then loaded onto 1 mL HM-N Isolute cartridges (liquidliquid extraction cartridges, Biot age, Charlottesville, VA). The steroids were eluted with 3 x 1 mL additions of MTBE, blown down, and reconstituted with MSTFA to a final concentration of 200 ng mL-1. Two separate derivatization methods were utilized (in triplicate): derivatized either with the synthesis microwave system (one minute) or block heater (thirty minutes) at 55 C. Results and Discussion Maximum Temperatures of Reagent, Solven t, and Reagent:Solvent Combinations The microwave properties of most organic solvents are known;333 unfortunately, the microwave properties of the derivatization reagents are currently unknown and to this date it has been difficult to determine their heating cont ributions to microwave derivatization reactions. Initial experiments focused on determining th e microwave compatibility of the various derivatization components analyzed in this st udy. The maximum temperatures reached for each reagent (BSTFA/TMCS, BSTFA, BSA, MSTFA, MOX and MTBSTFA) at three power levels (100, 200 and 300 W) for one minute were obtained using real-time temperature monitoring with the synthesis microwave system. The average (n=3) maximum temperature achieved for each power level is shown in Table 4-1. MTBS TFA, MOX, and MSTFA coupled well with microwave energy (all over 100 C at 100 W fo r only one minute). BSTFA/TMCS and to a larger extent BSA, did not couple well with th e microwave energy. However, as the power level was increased from 100 to 300 W, the differe nce in maximum temperature obtained became smaller between MSTFA and BSTFA/TMCS (36 C to 21 C). BSA had the lowest maximum
110 temperatures, although it still reached 90 C in one minute at 300 W. The maximum temperatures generated for each reagent ove r all three powers (100, 200 and 300 W) were determined in triplicate and had excellent repr oducibility (typically 1%). The four organic solvents (DMF, ACN, PYR and HEX) were exam ined in a similar fashion, as well as the maximum temperature obtained using mixtures of reagents (BSTFA/TMCS and MSTFA) and solvents (Table 4-1). The addition of ACN, PYR and HEX all lowered the final temperature achieved when using MSTFA (in comparison to us ing MSTFA individually), likely due to the low boiling points of the three solvents (82, 115, and 69 C, respectively) in comparison to the boiling point of MSTFA (132 C). The same eff ect is observed with the use of BSTFA/TMCS (boiling point 45 Â– 55 C). The addition of DM F (boiling point 153 C) increased the final temperature for both reagents but more dram atically for BSTFA/TMCS. Furthermore, the catalyst TMCS was shown to actually lower the fi nal temperature for BSTFA by almost 20 C at 300 W. The addition of solvent to BSTFA/TM CS and MSTFA also had very reproducible maximum temperatures (Table 4-1), which allo wed for reproducible de rivatization reactions using the synthesis microwave. Overview of MAD Analysis Evaluation of MAD began by first isolating th e contributions of mi crowave heating in comparison to thermal heating. A common pract ice with current micr owave derivatization methods is to disregard or ignore the temperature of the MAD reaction,196, 233, 287-289, 326-328 often due to a lack of means to monitor temperature with domestic microwave ovens. In this study, in contrast, the heating differenc es and RRF values were determined by performing both the synthesis microwave and thermal reactions at th e same reaction temperature. Further evaluation focused on comparing RRFs generated using the synthesis microwave system at different wattages (regardless of temperature achieved) an d different radiation times to those obtained
111 using the thermal heating methods. Also, since most MAD analyses to date have been performed using a domestic microwave oven, a comparison between a domestic microwave oven and a synthesis microwave system was performed. MAD with MSTFA For MSTFA, when comparing RRF values us ing the synthesis microwave system (one minute) to those obtained by a traditional therma l heating method (30 minutes), the RRFs for all steroids were equal to or slightly improved at the same reaction temperat ure of 55 C (Table 43). MSTFA coupled so well with the microwave energy, it only required an average of 5 W to maintain a temperature of 55 C inside the reacti on cell. A comparison of RRF values at different wattages (100 and 300 W) to the thermal deriva tization showed that the RRFs obtained at 100 W for one minute were in accordance with the ther mal derivatization RRF values (as shown in Table 4-3). However, since MSTFA coupled so well with the microwave energy, all of underivatized and mono-TMS derivative species e xperienced a decrease in RRF values due to the harsher reaction conditions at 300 W for one minute. The di-TMS derivatives were not negatively affected at 300 W, most likely due to the ha rsher conditions required for the derivatization of multiple hydroxyl groups. Also, the only reaction conditions in which di-TMSCORT derivatized effectively was at 300 W fo r one minute using MSTFA (data not shown). A comparison of absolute RRFs obtained with th e synthesis microwave oven at 300 W was also made to those obtained from a domestic micr owave oven at 300 W a nd 1000 W (Figure 4-4). Due to the high reactivity and ability of MSTF A to couple with microwave energy, there was little difference between the two microwave systems for almost ev ery derivatized species at 300 W (except for mono-TMS-DHT). A compar ison at 300 W (and 1000 W) for the three underivatized species (17-MT, PROG, and AE) s howed higher absolute RRF values in the domestic microwave oven, likely due to the sy nthesis microwave system providing continuous
112 power in contrast to the domestic microwave oven which operates with an on/off duty cycle. Thus, part of the reason why the domestic microw ave oven provides higher absolute RRF values than the synthesis microwave system was becau se the domestic microwave oven was compared at 300 W (a duty cycle of 30% on time, 70% off time, which may lead to a more favorable reaction than with continuous power). The micr owave heating on a duty cycle (domestic oven) may lead to an increase in superheating and hi gher peak temperatures of the bulk solution; however, the applicatio n of the continual power (using the synthesis microwave system) provides more consistant reaction temperatures and thus, more reproducible reactions. The percent relative standard devi ation (RSD) associated with th e RRF values obtained with the synthesis microwave system and the domestic microwave oven are shown in Table 4-4. As shown, for reactions performed with BSTFA/ TMCS, the synthesis microwave system has dramatically lower RSD values (except for unde rivatized-17-MT). For MSTFA, the percent RSD values of the synthesis microwave system were approximately equal to or lower than the domestic microwave oven. The RSD values obta ined using MSTFA were comparable between the two microwaves likely due to two factors: 1) the high reactivity of MSTFA and the ability of MSTFA to couple well with microwave heating (by effectively absorbing microwave energy). BSTFA/TMCS, on the other hand, is a less reactive reagent than MS TFA. Due to the fact that BSTFA/TMCS derivatization typi cally requires harsher reacti on conditions, the continual microwave heating offered by the synthesis mi crowave system provided a more efficient derivatization medium and resulted in RRF values with low RSD (in comparison to the domestic microwave oven). Comparisons of the absolute RRF values from the domestic microwave (at 300 W and 1000W) to those obtained with the synthesis microwave at 100 W were similar (data not shown). Due to the high reactivity of MSTFA, the predominant product of EE2 was the
113 completed derivatization product, di-TMS-EE2 (as shown in Figure 4-4). The standard deviation of the absolute RRF values for both microwave ove ns was similar for all the steroids analyzed (except for mono-TMS-EE2). MAD with BSTFA/TMCS BSTFA/TMCS did not absorb the microwave energy as well as MSTFA, and required an average of 50 W to maintain 70 C for one minute. As shown in Table 3, the RRF values achieved for BSTFA/TMCS by MAD (one minute) were significantly lower for the estrogens (E2, E1, EE2, and DES) than those obtained with the thermal derivatization method (thirty minutes). The other steroid RRF va lues at 70 C were similar to those obtained with the thermal method. RRF values at different wattages fo r one minute (200, and 300 W, regardless of temperature, Table 4-5), genera lly showed similar RRF values when compared to the thermal method. Furthermore, the underivatized steroids had higher RRF values for all microwave comparisons when compared to thermal derivati zation, suggesting that mi crowave heating does not force their derivatization to form artifact s. The absolute RRF values obtained for the estrogens (E2, E1, EE2, and DES) with the domestic mi crowave oven (at 300 and 1000 W) exhibited a dramatic decrease in absolute RRF va lues in comparison to the synthesis microwave system (Figure 4-4), along with a greater st andard deviation. The predominant product of EE2 using BSTFA/TMCS was the mono-derivative. It should be noted that even though, in some cases, the synthesis microwave system had abso lute RRF values inferior to the domestic microwave oven (mono-TMS-CHOL, mono-TMST, mono-TMS-DHT and mono-TMS-PREG), the standard deviation was smaller using the synthesis microwave. The optimal irradiation time for the deriva tization of steroids using the synthesis microwave system and BSTFA/TMCS (and MSTFA, data not shown) was found to be one minute (as shown in Figure 4-5). Irradiation times of 30 seconds had lower absolute RRF values,
114 while irradiation times longer than one minute (2 min) typically had higher RRF values but with a significant decrease in reproducibility (d ue to possible promotion of artifacts). MAD with BSA The RRF values generated by the synthesis mi crowave system using BSA were about the same when compared to the traditional therma l heating method at 90 C (except for mono-TMSEE2 and di-TMS-DES, as shown in Table 4-6). However, the absolute RRF values were generally much lower in comparison to those generated by both MSTFA and BSTFA/TMCS (data not shown). Due to the weak ability of BSA to absorb microwave energy, only 300 W was compared to the thermal deriva tization method. For every derivatiz ed species, the RRFs obtained with the synthesis microwave system were equal to or larger than the ther mal derivatization at 90 C (with dramatic increases for mono-TMS-EE2 and di-TMS-DES). MAD with Methoxylamine/BSTFA/TMCS Strategy The potential of using two other steroid de rivatization strategies with MAD was also investigated; the data are shown in Table 4-7. The popular two-step deri vatization strategy (MOTMS) usually requires some combination of l ong reaction times (15 minute to 100 hrs) or high reaction high temperatures (room temperatur e to 100 C) for the methoxime derivatization followed by 30 to 180 minutes at 60 to 70 C for the subsequent trimethylsilyl derivatization.164, 184, 207, 310, 311 It is common to require extensive reaction conditions, es pecially for the comprehensive derivatizatioin of a chemically diverse mixture of anal ytes, and thus the timeconsuming aspect of this method has beco me a major drawback. Two thermal MO-TMS procedures were performed at 60 C, (1) fo r 30 minutes, and (2) fo r 180 minutes, both subsequently followed by a silylation procedur e using BSTFA/TMCS (70 C for 30 minutes). Although all steroids were deriva tized (as MO, MO-TMS or TMS de rivatives), in almost every case the RRF values using the MAD method were slightly less than the two-step thermal
115 derivatization method. However, the slightly lower RRFs achieved by MAD were obtained using a two-step microwave procedure, which had a to tal analysis time of two minutes (methoxime Â– one minute, silylation Â– one minute). The two ther mal derivatization methods had a total analysis time of 60 and 210 minutes at 60 C. Furthermore, MO derivatives are often susceptible to forming isomeric peaks of some derivatized species,310 as seen most notably here with MOTMS-T, bis-MO-PROG and MO-17-MT. MAD with MTBSTFA The other reagent, MTBSTFA, is very regiospecific in its capability to derivatize steroids;193 thus, only mono-derivatives are formed with the hydroxyl groups at specific locations due the bulky nature of its derivative attach ment (TBDMS). MTBSTFA was effective for the derivatization of all of the estrogens (E2, E1, EE2, and DES), due to the lack of steric hindrance and high acidity of the hydroxyl group at the 3position. The other hydroxylated steroids were not successfully derivatized with MTBSTFA. Reaction conditions for derivatization using MTBSTFA are typically 60 Â– 75 C for 90 minutes to overnight.273, 284, 285 Here, it was demonstrated that estrogens can be effectivel y derivatized using MTBSTFA with MAD in one minute (at 300 W) with improved reproducibility when compared to the thermal method (Table 4-7). MAD with Polar Organic Solvents Several methods exist to enhance or improve thermal derivatization reactions, but few studies exist demonstrating method to improve derivatization results by MAD. Based on our previous work (Chapter 3), it was established th at during thermal deriva tization, the addition of solvent with BSTFA/TMCS does indeed improve RRF values when compared to solvent-less derivatization. However, thermal derivatization with solvent for MSTFA does not generate the same increase in RRF. A benefit of solvent us e is the reduced volume of reagent used during
116 each reaction. Typical derivatizat ion methods use the derivatizing reagent exclusively as the solvent; thus, the addition of so lvent can decrease the amount of reagent required and thereby the cost per sample without sacrificing results. The solvents examined were chosen based on relevant microwave heating properties, including dielectric constant ( Â’), dielectric loss ( Â”) and boiling point. DMF and ACN both have high di electric constants (36.7 and 37.5, at 20 C, respectively), and thus does not absorb microwave energy as well; however, ACN has a much lower boiling point (82 C) than does DMF (153 C). Superheating is a microwave phenomenon wherein the entire bulk solvent can reach up to 15 Â– 20 C above their normal boiling point. Hence, the lower boiling point of ACN may affect the extent of the rate enhancement in two ways: 1) a limitation of the temperature ach ievable in the microwave reaction and 2) superheating of the ACN. Pyridine has a modera te dielectric constant (12.3 at 25 C) and a boiling point of 115 C. The final solvent examined was the non-polar hexane (HEX), which has an extremely low dielectric constant (1.88 at 25 C) and a low boiling point (69 C). HEX was examined to illustrate that there was little or no enhancement for steroi d derivatization with nonpolar solvents. Finally, the dielect ric loss, or the ability of co mponents to transform microwave energy into heat for the solution, al so plays a pivotal role. Solvents with dielectric losses in the range of 1 to 10 are considered to be moderate in their capability to transform the microwave energy that is absorbed into heat,323 while values below one are poor at transforming microwave energy into heat. The dielec tric losses of DMF, ACN, and HEX are 6.070, 2.325, and 0.038, respectively. The dielectric loss for PYR is not available. An enhancement was clearly evident when th e synthesis microwave system was compared to the solvent-less derivatization method for ACN, DMF, and PYR (as shown in Table 4-8). For all the derivatized species, these three solvents (ACN, DMF, and PY R) all showed an increase in
117 RRF values over those obtained without the us e of solvent. Among the solvent enhancements, ACN was slightly better for all th e steroid derivatives monitored (e xcept for some derivatives of EE2). Both DMF and PYR favored the formation di-TMS-EE2, but with considerable irreproducibility, while ACN favored the formation of mono-TMS-EE2 with improved reproducibility. Furthermore, while all the underivatized species (17-MT, PROG and AE) experienced lower RRF values when solvent wa s added, the RRF values were highest using ACN. The high temperatures, and thus the hars h reaction conditions, achieved with the addition of DMF may explain why many of the mono-TMSderivatives had smaller increases in RRF values. As expected, HEX was not beneficial in enhancing the derivati zation and actually had values dramatically lower than the solventless heating method. Furthermore, ACN was the fastest to evaporate after deriva tization and prior to reconstituti on with isooctane (on average, blow down times for PYR and DMF were a pproximately: 3x to 10x longer than ACN, respectively). The solvent effect on RRF values using AC N was examined at different composition percentages (0:1, 1:3, 1:1 and 3:1 ACN), as shown in Figure 4-6. BSTFA/TMCS, with any addition of ACN resulted in an increase in RRF values when compared to those obtained without ACN (DMF and PYR also produced similar result s, data not shown). The domestic microwave oven yielded similar increases in RRF values with the addition of solvent; although with a considerable increase in standa rd deviation (data not shown). MAD of Spiked Plasma Extracts Spiked plasma extracts were examined to determine the efficacy of MAD at derivatizing steroids at low concentrations (200 ng mL-1) in a complex matrix. As shown in Table 4-9, the RRF values illustrate an increase for all deri vatized and underivatized species when using MAD
118 (one min) in comparison to those obtained by the thermal derivatizati on method (thirty minutes with MSTFA) at the same temperature of 55 C. Conclusion These results show that MAD can perform de rivatization strategies more rapidly than traditional methods. There are also other aspects to be considered. In addition to derivatization time (typically in one minute), additional times need to be factored in when considering overall analysis time, as the synthesis microwave system requires extra cool down time before the vial is released from the reaction vessel. For reactions at higher powers and te mperatures, upwards of five minutes was sometimes required to cool down before further analys es could be performed (dictated by a specific releasing temperature). Also, since it is preferable to analyze the derivatized steroids immediately after derivatization, analyses we re performed one-at-a-time and the overall analysis time was ultimately limited by the length of the GC temperature program. Furthermore, with the added c ontrol of temperature and power in the synthesis microwave reaction cell, sample derivatization time was decreased without a lo ss in reproducibility. This study has focused on evaluating the poten tial advantages of MAD (using a synthesis microwave system designed for organic synthe ses) in comparison to traditional thermal derivatization methods. Th e ability to monitor an d control temperature permits the synthesis microwave system to achieve reproducible r eaction temperatures, a nd thus reproducible reactions. MSTFA was the most effective reagent for heating in the microwave, resulting in larger absolute RRF values and comparable re producibility to other reagents. Due to the high reactivity and ability to absorb microwave ener gy, there was little difference for MSTFA in the absolute RRF and standard deviation values between the domestic and synthesis microwave system. BSTFA/TMCS, on the other hand, generally showed improved RRF values for the estrogens and a decrease in the standard deviatio n for the rest of the derivatized species achieved
119 with the synthesis microwave system in comparison to those obtained with the domestic microwave oven. In general, however, MAD with BSTFA/TMCS and BSA was not as successful as MAD with MSTFA in providing increased RRF values over the thermal methods. Results for MOX and MTBSTFA derivatizatio n reactions using MAD were comparable to traditional thermal derivatization methods, but in significantly less time. The application of polar solvents with MAD yielded higher RRF values and reduced the amount of reagent needed per sample. ACN was found to be the most effective so lvent for derivatizati on reactions employing BSTFA/TMCS. MAD, using the aforementioned de rivatization strategies provided rapid (1 minute) reactions with comparable or slightly in creased yields and reproducibility. Furthermore, the proposed methods can be used to derivatize a large suite or smaller subsets of steroids. The analysis of spiked plasma revealed that MAD was slightly more effective in derivatizing the steroid mixture in a complex matrix than therma l heating. Future studies will aim to improve the quantitative ability of the proposed met hod for endogenous steroids in plasma.
120 Table 4-1. Maximum temperatures (in C ) achieved with the synthesi s microwave system for re agent and reagent:solvent combinations Power, Solvent MSTFA BSTFA/TMCS BSTFA BSA MOX MTBSTFA 100 W 128 1.0 92 0.0 95 0.4 56 0.6 120 2.1 127 0.6 200 W 144 0.6 117 1.0 125 0.7 78 4.0 138 2.5 157 0.6 300 W 153 0.6 132 0.6 150 0.6 91 2.0 142 1.5 169 0.6 100 W ACN 94 1.0 82 1.0 200 W ACN 114 1.2 101 1.5 300 W ACN 124 1.0 111 1.0 100 W DMF 130 0.6 125 2.3 200 W DMF 146 1.2 143 1.5 300 W DMF 156 2.0 152 2.6 100 W HEX 100 2.5 74 0.6 200 W HEX 121 1.5 97 0.6 300 W HEX 129 1.0 109 1.0 100 W PYR 113 0.6 92 0.0 200 W PYR 134 0.6 116 0.0 300 W PYR 142 1.2 126 0.6 The addition of solvent to each of the reagent:solvent mixtures was 1:1. Each of the microwave heating reactions was performed for 1 minute at the given power levels. The standard devi ation (n=3) of each temper ature is also shown.
121 Table 4-2. Targeted derivative s and ions summed to measure peaks areas fo r all the derivatization reagent strategies Steroid Abbreviation Molecular Weight TMS Derivative TMS Ions MO/MOTMS Derivative MO/MOTMS Ions TBDMS Derivative TBDMS Ions (Surr.) DDE 318 Underivatized 246, 318 Underivatized 246, 318 Underivatized 246, 318 17Methyltestosterone 17-MT 302 C=O, C-OH Underivatized 229, 302 MO 313, 331 ND -Androstenedione AE 286 2 C=O Underivatized 148, 286 Bis-MO 313, 344 ND -Progesterone P 314 2 C=O Underivatized 124, 314 Bis-MO 341, 372 ND -Dihydrotestosterone DHT 290 C=O, C-OH Mono-TMS 347, 362 MO-TMS 286, 391 ND -Estrone E1 270 C=O, C-OH Mono-TMS 257, 342 MO-TMS 340, 371 Mono-TBDMS 327, 384 Pregnenolone PREG 316 C=O, C-OH Mono-TMS 298, 388 MO-TMS 386, 417 ND -Testosterone T 288 C=O, C-OH Mono-TMS 226, 360 MO-TMS 268, 389 ND -Cholesterol CHOL 386 C-OH Mono-TMS 368, 458 Mono-TMS 368, 458 ND -Ethinylestradiol EE2 296 2 C-OH Mono-TMS 425, 440 Mono-TMS 425, 440 Mono-TBDMS 353, 410 17 -Estradiol E2 272 2 C-OH Di-TMS 285, 416 Di-TMS 285, 416 Mono-TBDMS 329, 386 Diethylstilbestrol DES 268 2 C-OH Di-TMS 217, 412 Di-TMS 217, 412 Di-TBDMS 467, 496 Corticosterone CORT 346 2 C=O, 2 C-OH Di-TMS 475, 490 ND -ND -Italics indicate the molecular ion ( m/z ) achieved by trimethylsilyl (T MS), methoxime (MO) and MT BSTFA (TBDMS) derivatization reactions. ND indicates that the steroi d was not derivatized. The surrogate was dichlorodiphenyldichloroethylene-p,p (DDE).
122 Table 4-3. Comparison of the RRF values fo r the synthesis microwave system (MW) nor malized to thermal derivatization methods using MSTFA MSTFA MW 55 CÂ† MW 100 WÂ‡ MW 300 WÂ‡ Underivatized-AE 0.96 0.00 1.00 0.00 0.58 0.00 Underivatized-17-MT 1.06 0.00 0.97 0.00 0.71 0.00 Underivatized-PROG 0.98 0.00 1.03 0.00 0.72 0.00 Mono-TMS-CHOL 1.02 0.00 1.01 0.00 0.98 0.00 Mono-TMS-DHT 1.06 0.00 0.87 0.00 0.32 0.05 Mono-TMS-E1 1.04 0.00 0.99 0.00 0.86 0.00 Mono-TMS-EE2 3.80 1.31 0.10 0.02 0.04 0.00 Mono-TMS-PREG 0.98 0.00 0.98 0.00 0.95 0.00 Mono-TMS-T 0.99 0.00 1.01 0.00 0.96 0.00 Di-TMS-DES 0.99 0.00 1.01 0.00 1.02 0.00 Di-TMS-E2 1.04 0.00 0.97 0.00 0.97 0.00 RRF values are shown normalized to the block heating method at the same temperatureÂ†, and regard less of temperatureÂ‡. Values greater than 1 indicate an increase in RRF over the block method; values less than 1 indicate a decrease in RRF. The standard deviation of the mean is also shown; values of 0.0 indicate a standard deviation less than 0.005.
123 Table 4-4. Percent relative standard devi ation (RSD, in %) of RRF values for each de rivatized steroid using the domestic micro wave oven and the synthesis microwave system BSTFA/TMCS MSTFA Domestic MW Oven Synthesis MW System Domestic MW Oven Synthesis MW System Underivatized-AE 5.4 1.0 1.2 5.1 Underivatized-17-MT 4.6 5.6 9.1 11 Underivatized-PROG 4.0 2.2 0.6 6.3 Mono-TMS-CHOL 22 12 3.9 3.0 Mono-TMS-DHT 45 22 17 3.6 Mono-TMS-E1 31 0.3 2.1 2.5 Mono/Di-TMS-EE2 24 1.0 21 6.1 Mono-TMS-PREG 32 6.0 0.7 2.8 Mono-TMS-T 29 18 1.3 0.5 Di-TMS-DES 34 3.7 0.7 3.7 Di-TMS-E2 50 4.0 2.4 2.3 MAD reactions using BSTFA/TMCS were performed at the optim al conditions of 1000W and 30 0 W (for 1 minute) using the domestic microwave oven and synthesis microwave system, resp ectively. MAD reactions using MSTFA were performed at the optimal conditions of 1000W and 100 W (f or 1 minute) using the domestic microwav e oven and synthesis microwave system, respectively. EE2 was a mono-derivatized product for reactions employing BS TFA/TMCS and di-derivatiz ed for reactions employing MSTFA.
124 Table 4-5. Comparison of the RRF values fo r the synthesis microwave system (MW) nor malized to thermal derivatization methods using BSTFA/TMCS BSTFA/TMCS MW 70 CÂ† MW 200 WÂ‡ MW 300 WÂ‡ Underivatized-AE 1.05 0.00 1.14 0.01 1.09 0.01 Underivatized-17-MT 1.18 0.00 1.19 0.02 1.13 0.01 Underivatized-PROG 1.14 0.00 1.13 0.00 1.06 0.00 Mono-TMS-CHOL 0.91 0.00 0.92 0.01 0.97 0.00 Mono-TMS-DHT 0.94 0.00 1.00 0.00 0.82 0.05 Mono-TMS-E1 0.68 0.00 1.01 0.00 1.07 0.00 Mono-TMS-EE2 0.71 0.00 1.05 0.00 1.11 0.00 Mono-TMS-PREG 0.92 0.00 0.89 0.01 0.91 0.01 Mono-TMS-T 1.03 0.00 1.11 0.00 0.89 0.03 Di-TMS-DES 0.56 0.01 0.93 0.00 1.01 0.00 Di-TMS-E2 0.73 0.00 0.86 0.02 0.85 0.02 RRF values are shown normalized to the block heating method at the same temperatureÂ†, and regard less of temperatureÂ‡. Values greater than 1 indicate an increase in RRF over the block method; values le ss than 1 indicate a decreas e in RRF. The standard deviation of the mean is also shown, values of 0.0 indicate a standard deviation less than 0.005.
125 Table 4-6. Comparison of the RRF values fo r the synthesis microwave system (MW) nor malized to thermal derivatization methods using BSA BSA MW 90 CÂ† MW 300 WÂ‡ Underivatized-AE 1.04 0.00 0.98 0.00 Underivatized-17-MT 1.04 0.00 1.13 0.00 Underivatized-PROG 1.03 0.00 0.96 0.00 Mono-TMS-CHOL 0.97 0.00 1.06 0.00 Mono-TMS-DHT 0.97 0.00 1.01 0.00 Mono-TMS-E1 0.91 0.00 1.44 0.02 Mono-TMS-EE2 0.71 0.01 3.90 1.35 Mono-TMS-PREG 0.94 0.00 1.02 0.00 Mono-TMS-T 0.98 0.00 1.00 0.00 Di-TMS-DES 0.67 0.01 2.96 1.61 Di-TMS-E2 1.00 0.00 1.10 0.00 RRF values are shown normalized to the block heating method at the same temperatureÂ†, and regard less of temperatureÂ‡. Values greater than 1 indicate an increase in RRF over the block method; values less than 1 indicate a decrease in RRF. The standard deviation of the mean is also shown, values of 0.0 indicate a standard deviation less than 0.005.
126 Table 4-7. Comparison of the RRF values fo r the synthesis microwave system normalized to those for the thermal derivatization methods using MOX and MTBSTFA reagent MOX reagent Long Reaction Short Reaction MTBSTFA Long Reaction bis-MO-AE 0.48 0.02 0.63 0.00 di-TBMS-DES 1.01 0.00 bis-MO-PROG 0.74 0.02 0.96 0.00 TBMS-E1 0.98 0.00 Di-TMS-DES 0.95 0.00 0.91 0.00 TBMS-E2 0.68 0.08 Di-TMS-E2 0.98 0.00 0.95 0.00 TBMS-EE2 1.10 0.00 Mono-TMS-CHOL 1.00 0.00 1.11 0.00 Mono-TMS-EE2 1.28 0.06 1.21 0.00 MO-17-MT 1.35 0.00 1.17 0.00 MO-TMS-DHT 0.97 0.00 0.92 0.01 MO-TMS-E1 0.62 0.01 0.78 0.00 MO-TMS-PREG 0.81 0.00 0.88 0.00 MO-TMS-T 0.65 0.01 0.61 0.00 Values are shown normalized to the block hea ting methods regardless of temperature. Valu es greater than 1 indicate an increase in RRF over the block method; values less than 1 indicate a decr ease in RRF. Methoxime/TMS synthesis microwave derivatization method (total reaction time 2 minutes) was normalized to 1) long methoxime/TMS therma l derivatization (total reaction time 210 minutes) and 2) short methoxime/TMS thermal derivatization (tot al reaction time 60 minutes). MTBSTFA reaction normalized to thermal derivatization at 60 C for 120 minut es. The standard deviation of the mean is also shown. St andard deviation in absolu te RRF using MTBSTFA, n=3, synthesis microwave system is: TBDMS-E2 0.01, TBDMS-EE2 0.09, TBDMS-E1 0.04 and di-TBDMS-DES 0.02; thermal method: TBDMS-E2 0.45, TBDMS-EE2 0.26, TBDMS-E1 0.07 and di-TBDMS-DES 0.27).
127 Table 4-8. The RRF change usi ng the synthesis microwave system, derivatization reagent (BSTFA/TMCS), and organic solvent (1:1, v/v) ACN DMF HEX PYR AE 0.88 0.00 0.39 0.05 0.83 0.01 0.86 0.00 17-MT 0.87 0.01 0.01 0.00 0.88 0.00 0.64 0.01 PROG 0.85 0.00 0.42 0.04 0.82 0.00 0.85 0.00 Mono-TMS-CHOL 1.39 0.05 1.24 0.03 0.46 0.12 1.26 0.03 Mono-TMS-DHT 1.71 0.02 1.62 0.04 0.34 0.10 1.69 0.09 Mono-TMS-E1 1.25 0.01 1.12 0.00 0.44 0.07 1.21 0.01 Mono-TMS-EE2 1.15 0.02 0.00 0.00 0.43 0.09 0.87 0.00 Mono-TMS-PREG 1.34 0.04 1.22 0.03 0.43 0.08 1.26 0.04 Mono-TMS-T 1.69 0.19 1.35 0.07 0.35 0.05 1.66 0.18 Di-TMS-DES 1.66 0.01 1.72 0.02 0.26 0.08 1.57 0.04 Di-TMS-E2 1.76 0.03 1.78 0.10 0.24 0.09 1.72 0.04 Values are shown normalized to the solventless microwave heating method. Values greater than 1 indicate an increase in RRF ove r the solvent-less method; values less than 1 indicate a decrease in RRF. The standard deviation of the mean is also shown.
128 Table 4-9. The RRF change in using the synthesi s microwave system for steroids spiked into plasma at 200 ng mL-1 Underivatized-AE 1.1 0.00 Underivatized-17-MT 1.2 0.02 Underivatized-PROG 1.1 0.00 Mono-TMS-CHOL 1.1 0.00 Mono-TMS-DHT 1.1 0.01 Mono-TMS-E1 1.3 0.01 Mono-TMS-EE2 1.8 0.00 Mono-TMS-PREG 1.1 0.00 Mono-TMS-T 1.1 0.00 Di-TMS-DES 1.5 0.04 Di-TMS-E2 1.2 0.00 Microwave derivatization (1 min) normalized to thermal deri vatization (30 min), both were run at 55 C using MSTFA. RRF values are normaliz ed to the thermal heating method. Values greater than 1 indicate an increase in RRF over the block me thod. The standard deviation of the mean (n=3) is also shown
129 Figure 4-1. Strategy for two-step methoxime /trimethylsilyl (MO/TMS) derivatization.
130 Figure 4-2. Strategy for derivatizat ion using the MTBSTFA reagent. .
131 Figure 4-3. Overhead view of the S-Class synthesis microwave system.
132 0 0.5 1 1.5 2 2.5 Di-TMSDES Di-TMS-E2Mono-TMSE1 Mono-TMSEE2 Mono-TMSPREG Mono-TMSCHOL Mono-TMST Mono-TMSDHT PROG17-MTAEDi-TMSCORTRelative Response Factor (RRF) CEM 300W DOM 300W DOM 1000W 0 0.5 1 1.5 2 2.5 3 Di-TMSDES Di-TMS-E2Mono-TMSE1 Mono-TMSEE2 Mono-TMSPREG Mono-TMSCHOL Mono-TMST Mono-TMSDHT PROG17-MTAEDi-TMSCORTRelative Response Factor (RRF) CEM 300W DOM 300W DOM 1000WA B 0.000 0.250 0.500 0.750 1.000 Di-TMS-EthinylestradiolRelative Response Factor 0.000 0.250 0.500 0.750 1.000Di-TMS-EthinylestradiolRelative Response Factor 0 0.5 1 1.5 2 2.5 Di-TMSDES Di-TMS-E2Mono-TMSE1 Mono-TMSEE2 Mono-TMSPREG Mono-TMSCHOL Mono-TMST Mono-TMSDHT PROG17-MTAEDi-TMSCORTRelative Response Factor (RRF) CEM 300W DOM 300W DOM 1000W 0 0.5 1 1.5 2 2.5 3 Di-TMSDES Di-TMS-E2Mono-TMSE1 Mono-TMSEE2 Mono-TMSPREG Mono-TMSCHOL Mono-TMST Mono-TMSDHT PROG17-MTAEDi-TMSCORTRelative Response Factor (RRF) CEM 300W DOM 300W DOM 1000WA B 0 0.5 1 1.5 2 2.5 Di-TMSDES Di-TMS-E2Mono-TMSE1 Mono-TMSEE2 Mono-TMSPREG Mono-TMSCHOL Mono-TMST Mono-TMSDHT PROG17-MTAEDi-TMSCORTRelative Response Factor (RRF) CEM 300W DOM 300W DOM 1000W 0 0.5 1 1.5 2 2.5 3 Di-TMSDES Di-TMS-E2Mono-TMSE1 Mono-TMSEE2 Mono-TMSPREG Mono-TMSCHOL Mono-TMST Mono-TMSDHT PROG17-MTAEDi-TMSCORTRelative Response Factor (RRF) CEM 300W DOM 300W DOM 1000WA B 0.000 0.250 0.500 0.750 1.000 Di-TMS-EthinylestradiolRelative Response Factor 0.000 0.250 0.500 0.750 1.000Di-TMS-EthinylestradiolRelative Response Factor Figure 4-4. Absolute relative response factors (RRFs) for MSTFA (A) and BSTFA/TMCS (B) for both the synthesis microwave system (CEM, at 300 W) and the domestic microwave oven (DOM, at 300 and 1000 W). RRF values for the completely derivatized product of (di-TMS)-EE2 are shown in the inset for both plots (A and B). All the microwave reactions were for one minute. Error bars are shown as the standard deviation of the mean (n=3).
133 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2Di-TMSDES Di-TMS-E2MonoTMS-E1 MonoTMS-EE2 MonoTMS-PREG MonoTMSCHOL MonoTMS-T MonoTMS-DHT PROG17-MTAERelative Response Factor (RRF) 0.5 min 1 min 2 min Figure 4-5. Absolute RRF values for BSTFA/ TMCS at different microwave reaction times using the synthesis microwave system (at 300 W). Error bars correspond to the sta ndard deviation of the mean (n=3).
134 0 0.5 1 1.5 2 2.5Di-TMSDES Di-TMS-E2MonoTMS-E1 MonoTMS-EE2 MonoTMS-PREG MonoTMSCHOL MonoTMS-T MonoTMS-DHT PROG17-MTAERelative Response Factor (RRF) 0% ACN 25% ACN 50% ACN 75% ACN Figure 4-6. Absolute RRF values for BSTFA/TMCS with varying amounts of ACN using the the synt hesis microwave system (at 300 W) for one minute. The total volume was 200 L. Error bars correspond to the standa rd deviation of the mean (n=3).
135 CHAPTER 5 EVALUATION OF DERIVATIZATION STRATEGIES FOR THE COMPREHENSIVE ANALYSIS OF ENDOCRINE DISR UPTING COMPOUNDS USING GAS CHROMATOGRAPHY/MASS SPECTROMETRY (GC/MS) Introduction Over the past twenty years, a renewed inte rest in environmental monitoring has been focused on elucidating the biol ogical role of endocrine disrupting compounds (EDCs).7 Introduced in Our Stolen Future,11 EDCs are usually characte rized as compounds, both natural and synthetic, that possess the pot ential to alter normal endocrine function. An example of the potency of EDCs and their consequential effect on wildlife can be observed at Lake Apopka in central Florida. Along with exte nsive agricultural and municipal pollution, the lake experienced a pesticide spill in 1980, and severa l published reports since have focused on investigating the effect of these compounds on the wildlife.11, 62, 334 One sentinel species us ed to assess the quality of the environment, the American alligator ( Alligator mississippiensis ), has exhibited not only detectable levels of EDCs in eggs292 and serum,70 but also altered sex steroid concentrations,13, 37, 68, 73 reduced clutch viability,76 increased abnormalities in bone density,77 and reproductive morphological abnormalities13, 37, 68, 70, 73 when compared to alligators from reference lakes. The potency of these chemicals is even more problema tic due to their resilience in the environment, their powerful activity at low concentrations, and their ability to accumulate through various food webs.11, 20, 30, 41 Furthermore, many of the metabolite s and degradation products of these suspected EDCs are largely uncharacterized and may possess similar disruptive properties.127 Although there is a lack of defi nitive explanations as to the cause/effect relationship of EDCs to various human disease states (unlik e many wildlife studies), recent human studies examining EDCs suggest not only an obvious burden in the environment, but also a substantial level of health risk to the general public.30, 41 EDC exposure is one suspected cause for the
136 increase in human ailments such as decreased sperm levels,20, 30, 32, 91 preterm birth,20, 30 obesity,30 and breast cancer.20, 30, 32, 91 Furthermore, the list of chemicals known to have endocrine disrupting capabilities is rapidly expanding in both nu mber and variety, raising the need for the development of a comprehensive EDC profiling technique that is capable of analyzing many EDC types in a variety of biological and environmental samples. Current methods to characterize EDCs are ofte n limited in scope, since only compounds of similar chromatographic properties are analyzed. The recent trend in the analysis of EDCs has been to combine chromatographic techniques for detection with bioassays to measure hormonal activity.335 A comprehensive EDC profile, coupled with a bioassay, would provide the ability to examine EDCs and EDC-induced responses on a wider scope. The utility of gas chromatography/mass spectrometry (GC/MS) has been well-reported for th e characterization of EDCs due its amenability to many chemical functionalities, including phthalates,101, 129, 137 pesticides,101, 116, 129, 131, 136, 137 and polycyclic aromatic hydrocarbons.116 However, the analysis of non-amenable GC compounds, such as bisphenol A,106, 114, 121, 122, 124, 139, 214, 216, 220, 285 alkylphenols,106, 114, 121, 122, 124, 139, 214, 216, 220, 285 natural and synthetic steroids,106, 114, 121, 122, 124, 139, 214, 216, 220, 285 and polar byproducts of many non-polar EDCs,129 requires time-consuming derivatization procedures to achieve adequate sensitivity. For the analysis of a mixture of many EDC types, traditional GC/MS analysis often needs to be supplemented by extra derivatization129 or additional liquid chromatography/ mass spectrometry (LC/MS) procedures116, 136 for the inclusion of polar EDCs. LC/MS has been a desi rable alternative for th e detection of many of these polar EDC classes107, 111, 112, 116, 117, 133, 136, 139, 141, 213 because it avoids the tedious derivatization procedures. However, the chromat ographic resolving power of LC is generally inferior to GC, thus the comprehensive profili ng of a wide polarity ra nge of EDCs in one
137 chromatographic run is a challenge. To our knowledge, there are no comprehensive methods which take advantage of the superior chro matographic separation provided by GC for the analysis of both GC-ready and derivatization-required EDCs in a single chromatographic analysis. Although the coupling of GC/MS with deriva tization chemistry can be time-consuming and laborious, it provides additi onal selectivity and expands th e range of compounds that can now be separated and characterized in a si ngle injection. The most common derivatization technique for the analysis of non-amenable EDCs is silylation, which typically employs either N,O-bis-(trimethylsilyl)-trifluoroacetamid e (BSTFA) or N-methyl-N-trimethylsilyltrifluoroacetamide (MSTFA). The silyl deriva tization reaction replaces active hydroxyl groups with a less polar TMS group. In th e literature, most silyl deriva tization methods require reaction conditions ranging between 60 Â– 75 C and 15 Â– 180 minutes.106, 114, 121, 122, 124, 129, 216, 220, 285 The laborious and time-consuming derivatization pr ocedures are often considered a hindrance; however, little effort has been placed on developing methods that reduce the analysis time or improve reaction conditions. The novel use of microwave heating, specifically, microwaveaccelerated derivatization (MAD), has been recently used to derivatize steroids,287, 288 alcohols,326 and amino acids,289 while providing comparable resu lts to block heating methods with a drastic reduction in analysis time. In this study, we report the development of a simple, effective, and rapid derivatization method that provides for the comprehensive pr ofiling of EDCs. Although there exist various methods that can employ GC/MS (without / with chemical derivatization) and LC/MS for the analysis of multiple EDC types, to our knowledg e little effort has been focused on developing a method for the comprehensive analysis of EDCs of a wide range of pol arities in a single
138 chromatographic analysis. This comprehensive EDC method combines the superior profiling capabilities of GC/MS with a standard silyl derivatization reaction. This novel method was developed to be inclusive of a diverse spectrum of endogenous EDCs in environmental matrices. Both thermal and microwave heating deriva tization methods were effective for the characterization of 33 EDCs of various chemi cal and biological properties. The examination investigated the increase in sens itivity achieved by de rivatization, the decreas e in analysis time using microwave heating, and the effectivenes s of the comprehensive method at various concentrations. This comprehensive EDC method wa s then used in a pilotstudy to detect spiked EDCs in surface water from Lake Apopka (Florida ). The pilot-study was followed by an initial analysis of unspiked Lake Apopka water. Experimental Section Chemicals and Solutions The suspected endocrine disrupting chemical s examined were purchased from three sources; p,pÂ’-DDT, p,pÂ’-DDD, p,pÂ’-DDE, p,pÂ’-DDA, p,pÂ’-methoxychlor, dicofol, dieldrin, endrin, heptachlor, heptachlor epoxi de, lindane, dibutyl phthalate, -endosulfan, trifluralin, kepone, atrazine, alachlor, and di ethyl phthalate (acquired from EPA, Research Triangle Park, NC); anthracene, vinclozolin, bisphenol A, 17-estradiol (E2), estrone (E1), diethylstilbestrol (DES), ethinylestradiol (EE2), progesterone, 2,4-dichlorophenol, coumestrol, 4-octylphenol, 4nonylphenol, triclosan and 1,1,1-tr ichloro-2,2-bis(4-hydroxyphenyl )ethane (HPTE) (acquired from Sigma Aldrich, St. Louis, MO); and reso rcinol and benzophenone (acquired from Fisher Scientific, Fair Lawn, NJ). The EDCs were prepar ed individually in anal ytical grade methanol (Fisher Scientific) and added to the compre hensive mixture at a concentration of 5 g mL-1. The solutions were stored at Â–20 C. The purity grad es of all EDCs involved in this study were
139 labeled at 98% or higher, excep t kepone (89%), endrin (83%), and dibutyl phthalate (80%). Chrysene-d12 was added (5 g mL-1, 98% D, Sigma Aldrich) to th e mixture as the surrogate. The surrogate was not added to measure the absolute de rivatization efficiency; rather it was used to measure the derivatization efficiency pertaini ng to various changes in the derivatization parameters. Chrysene-d12 was chosen because it is both deri vatization and microwave inactive. The relative response factor (RRF) values were calculated by dividing the area of each EDC product by the area of chrysene-d12. The comprehensive EDC mixture (all components above) was analyzed with and without derivatization procedures, with the evaluation centered on the enhancement and precision of the data. Derivatization Setup The comprehensive EDC mixture was added (200 L, 5 g mL-1) to standard 4 mL glass vials and blown down with ultra high purity nitrogen. The deriva tization reagents were then added to the EDC residue. The de rivatization reagent used for the analyses was derivatizationgrade N,O-bis-(trimethylsilyl) -trifluoroacetamide (BSTFA) w ith 1% trimethylchlorosilane (TMCS) (Supelco, Bellefonte, PA). The derivatiza tion reactions (described further below) were performed for both the thermal and microwave methods. Post-derivatization, the vials were blown down with ultra high purity nitrogen and the resulting resi dues were reconstituted with isooctane (99%, Fisher Scientific) and subseque ntly injected onto the GC. Reconstitution with isooctane helped eliminate silyl reagent noise at lower temperature levels, allowing improved examination of low-volatile EDCs. The derivatization analyses in this experi ment were all run in triplicate.
140 Block heating Block heating was examined to evaluate deriva tization and its ability to expand the number of compounds separated and detected in compar ison to methods that do not apply derivatization. Block heating also served as a reference for the evaluation of the use of microwave derivatization. The thermal derivatization reacti ons were performed using a Thermolyne Type 16500 Dri-bath block heater. The time and temper ature for the block deri vatization experiments were 30 min and 70 C. Microwave heating Microwave heating was examined as a more ef ficient heating approach to reduce analysis time in comparison to block heating. Microw ave derivatization was performed in a 1000Â–W domestic Half-Time microwave/convection oven (Apollo Worldwide, Palm Beach, Fl). The power levels analyzed were 500 and 900 W. The irradiation time for both power levels was one minute. Gas Chromatography/Mass Spectrometry The GC/MS instrument used was a Ther moFinnigan (San Jose, CA) Trace GC 2000 quadrupole ion trap mass spectrometer with an Autosampler AS3000. The data acquisition software used was Xcalibur 1.4. The column employed was an SLB-5ms capillary column (Supelco) with the dimensions of 30 m x 0.25 mm x 0.25 m film thickness. The temperature program started at 70 C and was held at this temperature for three minutes. The temperature program was then increased to 150 C at a rate of 15 C min-1, followed by an increase to 250 C at a rate of 7 C min-1. The temperature program concluded by ramping at 5 C min-1 to 300 C and subsequently held for two minutes. The carrier gas was ultra high purity helium (99.99%) at a flow rate of 1 mL min-1. The transfer line, ion source and in jection port temperatures were 275, 200 and 280 C, respectively. Splitless injection (2 L) was performed with a split flow of 50
141 mL min-1 (split flow ratio of 10). The MS was turned on at seven minutes and was run in positive full scan mode (approximately 1000 amu sec-1), m/z 50-600, with electron ionization. The retention times (tr) used for identification and the ions used for quantification of each EDC product and the surrogate can be found in Tables 5-1 (underivatized) and 5-2 (derivatized). Semi-Quantitative Calibration The calibration study was performe d to elucidate the effectiveness of the comprehensive EDC profile at various concentrations. Standard EDC solutions were prepared in triplicate at 5, 1, 0.5, 0.1 and 0.05 g mL-1. Two separate calibration analyses were performed, each containing all of the EDCs in the study and the surrogate chrysene-d12 (added to each EDC solution at a concentration of 0.5 g mL-1). Calibration 1 was performed to determine the detection limit of the EDCs without any derivatiz ation reagent added. Calibration 1 analys is was achieved by reconstituting the EDC residue with isooctane Calibration 2 was performed to measure the detection limit on two levels : the derivatization-require d compounds and the GC-ready compounds in the presence of derivatization reagent. Ca libration 2 was achieved by reconstituting the EDC mixture with deriva tization reagent, followed by performing the derivatization reaction (microwa ve Â– 900 W for 1 minute). Post-r eaction, the solution was blown down with nitrogen and reconst ituted with isooctane. Blank solu tions in methanol (with and without chrysene-d12) were evaluated with and without deri vatizing reagent. The blank solutions were reconstituted with isooctane prior to GC analysis and were run in triplicate. The semiquantitative calibration was perf ormed by approximating the effective concentration ranges (peaks observed with signal to noise ratio (S /N) > 10) of the EDCs analyzed with the comprehensive profiling method. The detection lim it levels were selected based on finding an appropriate concentration to detect as many EDCs as possible. The method was not designed as a
142 single component assay, thus some EDCs may have much lower detection limits when run individually. Here, we focused on obtaining the best detection lim its for a comprehensive list of EDCs (both polar and non-pol ar) in a single analysis. Pilot-Study for Water Analysis Surface water was collected from various locations at Lake Apopka (Florida) and was stored at Â–20 C. Lake Apopka was selected be cause of its high level of contamination and turbidity. The EDC mixture was sp iked in the Lake Apopka water samples at a concentration of 0.1 g mL-1. Due to the high turbidity of the water samples, solid-phase extraction was necessary. The water sample was loaded (3 mL ) onto a Strata C-18E cartridge (Phenomenex, Torrance, CA), which had been preconditioned with ethyl acetate (2 mL), acetonitrile (2 mL), and deionized water (2 mL). Post sample load ing, the cartridge was dried for 15 minutes. The EDCs were eluted with ethyl acetate (1 mL). The resulting extract was blown down at room temperature with nitrogen using a PrepSep vacuum manifold (Fisher Scientific). BSTFA with 1% TMCS was added to the residue (200 L) and reacted in the microwave for 1 min at 900 W. The derivatized sample was then blown down, r econstituted in isooctane, and subsequently injected into the GC. The analysis of unspiked Lake Apopka wa ter was performed using the liquid-liquid cartridge extraction (LLCE) met hod previously described in Ch apter 2. After the sample was loaded onto the LLCE cartridge, 2 mL of methyl tertbutyl ether was added to each cartridge for elution of the EDCs. No filtration step was required. The sample was then evaporated to dryness and reconstituted prior to GC/MS analysis.
143 Results and Discussion Characterization of Potential EDC Compounds Non-polar EDCs, including some pesticides, phthalates and polycyclic aromatic hydrocarbons, pose no analytical difficulty usin g GC/MS due to their volatility and thermal stability; therefore, the key for the developmen t of a comprehensive EDC profile was to expand the range of compounds amenable to GC/MS by incorporating more polar EDCs, including alkylphenols, bisphenol A, steroids, and non-polar EDC by-products, by employing silyl derivatization. The compre hensive method permits a profile of both non-polar and polar EDCs in a single analysis. The success of the comprehensive method was ev aluated on two levels: (1) the derivatization reaction not aff ecting the GC-ready compounds, a nd (2) the derivatization method providing effective derivatizati on of the polar EDCs. Although BSTFA with 1% TMCS was used in this analysis, it was also found that other silylating reagents, such as MSTFA, were also efficient for derivatiz ation of the comprehe nsive EDC mixture. Comprehensive EDC Profile Figure 5-1 shows three chromatograms reflecti ng variation in the EDC mixture (each at a concentration of 5 g mL-1): (a) no derivatiz ation reaction, (b) after thermal derivatization (70 C Â– 30 min), and (c) after micr owave derivatization (900 W for 1 min). Chromatogram (a) demonstrates the successful se paration of all the GC -ready (non-polar) EDCs (peaks 1-5, 7, 9-14, 16, 18-22, 24-26, and 30) and many of the polar EDCs underivatized (peaks 6, 8, 15, 17, 23, 2729). The underivatized polar EDCs, which typically require derivatization for characterization by GC/MS, are present due to the high concentration used in this analysis. Chromatograms (b) and (c) illustrate the separation achieved of the EDC profile after the derivatization reaction was employed. As shown in both chromatograms (b) and (c), the separation of the 21 GC-ready EDCs was not affected by the derivatization re action. Furthermore, the polar EDCs (peaks 6, 8,
144 15, 17, 23, 27, 28, and 29 in chromatogram (a)) were de rivatized and separated effectively and in most cases with an increase in sensitivity (pea ks A-C, E-I, and K-N). The comprehensive method with derivatization extended the polarity range of compounds separated by also characterizing derivatized HPTE and p,pÂ’-DDA (peaks D and J), which were not detected in Figure 5-1. The method also characterized derivatized coumestrol (tr > 28 min) and resorcinol (tr < 12 min) but are not shown in the figure. The chromat ograms from both the thermal and microwave derivatization reactions were comparable, demonstrating a si gnificant reduction in analysis time with no loss of derivatized EDCs by employing microwave heating. Non-Derivatized EDCs Figure 5-2 shows the RRF values for th e GC-ready (non-polar) EDCs (each at a concentration of 5 g mL-1). Figure 5-2 shows little differen ce between the RRF values obtained when derivatization was employed (either by bloc k or microwave heating) and those obtained with no derivatization. It was an ticipated that dicofol, which has a hydroxyl group, would need derivatization to make it thr ough the GC, but it was found predom inantly in the underivatized form. Figure 5-2 also shows a reduced RRF value for diethyl phthalate when derivatizing reagent is added. The inset in Figure 5-2 shows that when derivatization is applie d, the ester group of the diethyl phthalate reacts with the derivatizati on reagent to form an alternate derivatization product, mono-tms-diethyl phthalate. Alternate deriva tization products, or artifacts, are the result of the formation of partial derivatives due to ex cessive reaction conditions or side reactions. The comprehensive method attempted to minimize arti facts such as mono-tms-diethyl phthalate. Derivatized Steroid EDCs Figure 5-3 shows the RRF values for the ster oid EDCs (each at a concentration of 5 g mL-1), which typically require deri vatization for GC/MS analysis. The RRF values for the target
145 derivatized steroids (d i-TMS-DES, mono-TMS-E1, di-TMS-E2, mono-TMS-EE2) were approximately ten times larger than the RRF values for all of the underivatized steroids examined (M+.), which illustrates the value of derivatization in providing greater sensitivity. The RRF value for di-TMS-EE2 (the complete derivatization product) was lower than the RRF value for the underivatized species. U nder the current derivatization c onditions examined in this study, the mono-derivative of EE2 (mono-TMS-EE2) was favored and had the highest RRF values. The RRF value for progesterone, which doesnÂ’t requir e derivatization to make it through the GC, is much higher than the mono-TMS-progesterone, wh ich indicates that the derivatization method does not promote the formation of this artifact. Di ethylstilbestrol (DES) had two isomeric peaks for each derivative (monoand di-) formed; they were summed to represent total DES products. Di-TMS-DES was found to be the pre dominant derivatization product of DES. Other Hydroxylated EDCs Figure 5-4 shows the RRF values for the non-st eroidal EDCs which required derivatization (each at a concentration of 5 g mL-1), including the alkylphenols, bisphenol A and polar byproducts. The two polar by-produc ts studied in this analys is were HPTE and p,pÂ’-DDA, byproducts of p,pÂ’-methoxychlor and p,pÂ’-DDT, resp ectively. Derivatization was necessary for these polar EDCs, as the underivati zed species was either not detected or much lower than the derivatized species under the GC/MS conditions in this analysis. 2,4-dichlorophenol was not detected using the current method, and thus was not included in the Figure 5-4. Microwave Derivatization The RRF values for the derivatized EDCs were typically the lowest using the microwave heating at 500 W for one minute. However, microwave heating at 900 W for one minute was better or comparable to the thermal heating me thod at 70 C for 30 minutes for all the EDCs
146 derivatized (as shown in Figures 5-3 and 5-4). No distinct di sadvantage was apparent for the derivatization of th e EDC mixture with microwave heating. Semi-Quantitative Calibration Analysis A series of semi-quantitative experiments were performed to estimate the potential detection limits of the polar and non-polar ED Cs using the comprehensive profiling method. The approximate detection limits (in ppb, shown in Ta ble 5-3) were estimated by observing peaks with S/N greater than 10. With the addition of derivatization reagent, some of the GC-ready EDCs experienced poorer detection lim its (dicofol, dieldrin, heptachlor, -endosulfan, vinclozolin, trifluralin, heptachlor epoxide, benzophenone and di ethyl phthalate). The decrease in approximate detection limits for these compound s was likely due to an increase in background noise and/or a decrease in signal due to unw anted partial derivatization of derivatization susceptible functional groups. The application of derivatiza tion reagent does sacrifice the detection limit of some GCready EDCs; however, the tradeoff comes with a gain in the number of co mponents analyzed in a single analysis. The focus was toward a single method for screening EDCs (both polar and nonpolar) using the comprehensive list. This method ma y be the most effective for analyzing smaller subsets, for example, the analysis of DDT and its polar degradation produ cts (such as DDA) in a single analysis. Previous analyses for this subset (and for similar mixtures of polar and non-polar EDC subsets) have required separa te analyses for both polar types.116, 136 The rest of the GC-ready EDCs were not aff ected by the derivatization reagent and were detected well below 0.05 g mL-1 (0.05 g mL-1 was the lowest concentration level examined in this study). The detection limits for all of the derivatized polar EDCs were much lower than those for the underivatized species. The polar ED Cs di-tms-coumestrol, mono-tms-DDA, di-tms-
147 resorcinol, and mono-tms-ethinyles tradiol were the only derivatized species which did not yield estimated detection limits below 0.05 g mL-1. Spiked Water Analysis Figure 5-5 displays a chromatogr am of the EDC mixture spiked into a water sample from Lake Apopka. Although measurable levels of some native EDCs were detect ed in the water, the focus of this pilot-study was to demonstrate and evaluate the effectiveness of the comprehensive method for the characterization of EDCs exhibiti ng a wide range of polarities. Even with the high turbidity, the use of solid-phase extraction al lowed the characterization of most of the EDCs in this study at a concentration of 0.1 g mL-1. As shown in Figure 5-5, many of the GC-ready (peaks 1-5, 7, 9-14, 18, 21, 24, 25, 30) and derivatiza tion-required (peaks A-F, H, J-N) EDCs were successfully detected us ing the comprehensive method. Although intended as a screening method, the percent recovery of th e spiked Lake Apopka water (0.1 g mL-1) was performed by using a semi-quantitative calibration plot (0.05 Â– 5000 g mL-1). The average percent recovery was above 90% for 13 out of the 25 EDCs in th e spiked water. Since the comprehensive list covers a variety of EDC species (with differe nt polarities), the remaining 12 EDCs had recoveries between 25 and 90%. The EDCs used for the recovery study are shown in Figure 5-5. The underivatized species of the polar EDCs were not found at this concentration. Lake Apopka Water In 1980, Lake Apopka experienced a large spill of dicofol and DDT.37, 62, 72, 73 To date, Lake Apopka has also been noted to contai n several other poten tial EDCs, including trans nonachlor,37 DDE,37, 52 dieldrin,52, 70 aldrin,52 DDD,52 toxaphene,52 and several others.62, 75 The analysis of unspiked Lake Apopka water (FWL 1874) using the LLCE me thod from Chapter 2 allowed the detection of eight indigenous EDCs: p,pÂ’-DDT, diel drin, aldrin, endrin, endrin
148 ketone, heptachlor, lindane, and DEET. A chroma togram showing the detected EDCs (except DEET) is shown in Figure 5-6. This analysis not only suggests further work is needed to improve the characterization of Lake Apopka water but also indicates a need for improved clean up methods for water contamination. In addition, se veral studies report elevated serum and egg concentrations of EDCs, like DDE, in alligators living in Lake Apopka.37, 70, 72, 292 Thus, future analyses will also include an investigation of EDCs found in th e serum of alligators living in Lake Apopka. Conclusion The presence of multiple EDC types in biological and environmental samples has pushed the need for analytical methods that expand the po larity range of compounds able to be analyzed in a single analysis. Here it was demonstrated th at the superior profili ng capabilities of GC/MS with a standard silyl derivatiza tion reaction for the development of a comprehensive EDC profile is capable of analyzing not only compounds trad itionally amenable to GC, but also those compounds that require derivatizat ion. Microwave derivatization pr ovided comparable results to the thermal derivatizat ion method with a sign ificant decrease in analysis time. The comprehensive EDC method was evaluated semi-qua ntitatively at severa l concentrations to explore the effective concentration ranges fo r each EDC. Finally, the comprehensive EDC profile was effective in detec ting several polar and non-polar ED Cs in a spiked water sample. Future analyses will examine and compare nativ e EDC species detected from both Lake Apopka and reference lakes.
149 Table 5-1. Characteristic ions for the underi vatized endocrine disrupting compounds (EDCs) Name MW Function Underivatized Ions (m/z) tr Coumestrol 268.2 Phytoestrogen ND p,p'-DDA 281.1 DDD Metabolite ND HPTE 317.5 Methoxychlor Degradation ND 2,4Dichlorophenol 163.0 Pesticide Intermediate 63 162 7.29 Resorcinol 110.1 Plasticizer 82 110 8.46 Diethyl Phthalate 222.2 Plasticizer 149 177Â† 12.05 Benzophenone 182.2 Plastic 77 182 12.73 Trifluralin 335.3 Pesticide 264 306Â† 13.01 Atrazine 215.6 Herbicide 200 215 14.18 Lindane 290.8 Insecticide 181 219Â† 14.24 4-Octylphenol 206.3 Surfactant 107 206 14.40 Anthracene 178.2 PAH 152 178 15.02 4-Nonylphenol 220.3 Surfactant 107 220 15.80 Vinclozolin 286.1 Fungicide 178 286 16.02 Alachlor 269.7 Herbicide 160 188Â† 16.14 Heptachlor 373.3 Insecticide 272 371Â† 16.37 Dibutyl Phthalate 278.3 Plasticizer 149 278 16.88 Dicofol 370.4 Pesticide 139 251Â† 17.61 Heptachlor Epoxide 389.3 Heptachlor Metabolite 353 355Â† 18.34 Triclosan 289.5 Antibacterial Agent 218 289 18.96 -Endosulfan 406.9 Insecticide 195 241Â† 19.33 Bisphenol A 228.3 Plastics 213 228 19.77 p,p'-DDE 318.0 DDT Degradation 246 318 19.87 Dieldrin 380.9 Insecticide 79 263Â† 20.05 Endrin 380.9 Insecticide 281 317Â† 20.60 p,p'-DDD 320.0 DDT Degradation 165 235Â† 20.98 Kepone 490.6 Insecticide 237 272Â† 21.57 Diethylstilbestrol 268.3 Synthetic Estrogen 145 268 21.78 p,p'-DDT 354.4 Pesticide 165 235Â† 21.97 Chrysene-d12 240.3 Surrogate (IS) 240 241 23.36 p,p'-Methoxychlor 345.6 Insecticide 227 228Â† 23.43 Estrone 270.3 Natural Estrogen 185 270 25.36 17 -Estradiol 272.3 Natural Estrogen 213 272 25.57 Ethinylestradiol 296.4 Synthetic Estrogen 213 296 26.37 Progesterone 314.4 Natural Progestogen 124 314 27.80 tr = average (n=3) retention time in minutes. ND = not detected. Column 1 = base peak ions. Column 2 = the molecular ion (except if not presentÂ†).
150 Table 5-2. Characteristic ions for the deri vatized endocrine disr upting compounds (EDCs) Name Mono-TMS Ions ( m/z ) tr Di-TMS Ions ( m/z ) tr p,p'-DDA 200 337 19.98 ND Dicofol 73 323 23.20 NA Bisphenol A ND 357 372 20.17 Coumestrol ND 207 412 30.29 4-Nonylphenol 179 292 16.41 NA 4-Octylphenol 179 278 15.07 NA 2,4Dichlorophenol 219 235 9.42 NA HPTE 343 390 24.58 ND Resorcinol 167 182 9.13 239 254 9.50 Triclosan 200 362 19.17 NA 17 -Estradiol 244 344 ND 285 416 25.73 Estrone 257 342 25.34 399 414 ND Ethinylestradiol 285 368 26.38 425 440 27.09 Diethylstilbestrol 311 340 20.85/21.84217 412 20.65/21.88 Progesterone 371 386 27.48 ND Chrysene-d12 NA NA tr = average (n=3) retention time in minutes. ND = not detected. NA = not applicable. Column 1 = base peak ions. Column 2 = the molecular ion.
151 Table 5-3. Approximate detection li mits using comprehensive EDC method No Reagent With Reagent No Reagent With Reagent Non-polar EDCs (ppb) (ppb) R2 Value Polar EDCs (ppb) (ppb) R2 Value p,p'-DDE < 50 < 50 0.9998 p,p'-DDA ND 100Â† p,p'-DDT < 50 < 50 0.9981 Bisphenol A 100-500 < 50Â‡ 0.9984 p,p'-DDD < 50 < 50 0.9986 Resorcinol ND 100Â‡ Dicofol 50 100-500 4-Nonylphenol 100 < 50Â† 0.9992 p,p'-Methoxychlor 500 500 4-Octylphenol 100 < 50Â† 0.9997 Dieldrin 50 500 HPTE ND < 50Â† 0.9948 Endrin 500 500 Triclosan 500 < 50Â† 0.9991 Heptachlor 100 100-500 17 -Estradiol 1000-5000 < 50Â‡ 0.9983 Heptachlor Epoxide < 50 50-100 Estrone 500 < 50Â† 0.9996 Lindane < 50 < 50 0.9953 Ethinylestradiol 500-1000 100Â† Dibutyl Phthalate < 50 < 50 0.9996 Diethylstilbestrol 5000 < 50Â‡ 0.9962 -Endosulfan 50-100 500 Coumestrol ND 500Â‡ Anthracene < 50 50 0.9951 Vinclozolin < 50 50-100 Trifluralin < 50 50-100 Kepone 500-1000 1000 Atrazine < 50 < 50 0.9977 Benzophenone < 50 100-500 Alachlor < 50 < 50 0.9982 Diethyl Phthalate < 50 50-100 Progesterone 100-500 100-500 Â† Mono-TMS species, Â‡ Di-TMS species. The concen trations investigated were 50, 100, 500, 1000, and 5000 ppb. R2 values are presented for the analytes which were dete cted (after derivatization) over all five concentrations. Concentrations below 50 ppb were not investigated in this study. SPE wa s not included in calibration study.
152 (a) (c) (b)1 24 1 1 23 4,5 22 27 3 25,26 7,A 7 3 30 6 4,5 9 18 7,A 16 17 2 2 2 21 8 29 12 4,5 9 30 20 3 14 13 15 28 30 19 9 10 11 8 N M 6 18 16 10 16 10 18 14 12 C 13 14 12 11,B 11,B 13 L C N 22 22 G L G M 21 21 19 20,F D 20,F E E 19 D J J 25,26 K 25,26 K H,I 24 24 H,I TIC 7.49E6 TIC 5.37E6 TIC 7.59E6 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Time (min) 0 20 40 60 80 100 20 40 60 80 100Relative Abundance 20 40 60 80 100 X X(a) (c) (b)1 24 1 1 23 4,5 22 27 3 25,26 7,A 7 3 30 6 4,5 9 18 7,A 16 17 2 2 2 21 8 29 12 4,5 9 30 20 3 14 13 15 28 30 19 9 10 11 8 N M 6 18 16 10 16 10 18 14 12 C 13 14 12 11,B 11,B 13 L C N 22 22 G L G M 21 21 19 20,F D 20,F E E 19 D J J 25,26 K 25,26 K H,I 24 24 H,I TIC 7.49E6 TIC 5.37E6 TIC 7.59E6 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Time (min) 0 20 40 60 80 100 20 40 60 80 100Relative Abundance 20 40 60 80 100 X X 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Time (min) 0 20 40 60 80 100 20 40 60 80 100Relative Abundance 20 40 60 80 100 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Time (min) 0 20 40 60 80 100 20 40 60 80 100Relative Abundance 20 40 60 80 100 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Time (min) 0 20 40 60 80 100 20 40 60 80 100Relative Abundance 20 40 60 80 100 X X Figure 5-1. Chromatograms of co mprehensive EDC profile with (a ) no derivatizatio n, (b) derivatization usin g block heating of 7 0 C for 30 minutes, and (c) derivatization using microwave heating 900 watts for 1 minute. 1 = diethyl phthalate, 2 =
153 benzophenone, 3 = trifluralin, 4 = atr azine, 5 = lindane, 6 = 4-octylphenol, 7 = anthracene, 8 = 4-nonylphenol, 9 = vinclozolin, 10 = alachlor, 11 = heptachlor, 12 = dibutyl phth alate, 13 = dicofol, 14 = heptac hlor epoxide, 15 = triclosan, 16 = -endosulfan, 17 = bisphenol A, 18 = p,pÂ’-DDE, 19 = di eldrin, 20 = endrin, 21 = p,pÂ’-DDD, 22 = kepone, 23 = diethylstilbestrol (DES), 24 = p,pÂ’-DDT, 25 = chrysene-d12, 26 = p,pÂ’-methoxychlor, 27 = estrone, 28 = 17 -estradiol, 29 = ethinylestradiol, 30 = progest erone. A = mono-tms-4-octylp henol, B = mono-tms-4-nonylphenol, C = mono-tms-triclosan, D = mono-tms-p,pÂ’-DDA, E = di-tms-bisphenol A, F = di-t ms-DES-1, G = mono-tms-DES-1, H = di-tms-DES-2, I = mono-tms-DES-2, J = mono-tms-HPTE, K = mono-tms-estrone, L = di-tms-17 -estradiol, M = mono-tms-ethinylestradiol, N = di-tms-ethinylestradiol. Mono-tms-resorcinol (tr < 12 min) and di-tms-coumestrol (tr > 28 min) not shown in figure. Peak labeled X = phthalate contaminant.
154 0.0 0.5 1.0 1.5 2.0 2.5 3.0Benzophenone Trifluralin Atrazine Lindane Anthracene Vinclozolin Alachlor Heptachlor Di-Butyl Phthalate Hepachlor Epoxide a-Endosulfan DDE Dieldrin Endrin DDD Kepone DDT Diethyl Phthalate Dicofol Methoxychlor Relative Response Factor No Derivatization Reagent BSTFA Block 70C 30min BSTFA Micowave 500W 1min BSTFA Microwave 900W 1min 0.000 0.500 1.000 1.500 2.000Mono-TMS-Diethyl Phthalate 0.0 0.5 1.0 1.5 2.0 2.5 3.0Benzophenone Trifluralin Atrazine Lindane Anthracene Vinclozolin Alachlor Heptachlor Di-Butyl Phthalate Hepachlor Epoxide a-Endosulfan DDE Dieldrin Endrin DDD Kepone DDT Diethyl Phthalate Dicofol Methoxychlor Relative Response Factor No Derivatization Reagent BSTFA Block 70C 30min BSTFA Micowave 500W 1min BSTFA Microwave 900W 1min 0.000 0.500 1.000 1.500 2.000Mono-TMS-Diethyl Phthalate Figure 5-2. Relative response fact ors (RRFs) for the underivatized (GC-ready) EDCs in the comprehensive profile with and witho ut the presence of derivatizing reagen t (each at a concentration of 5 g mL-1). The inset displays an unwanted derivatization product of the ester group of diethyl pht halate. Error bars represent the sta ndard deviation of the mean (n=3).
155 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Total Mono-TMSDES Total Di-TMSDES Mono-TMSEstrone Di-TMS-17BEstradiol Mono-TMSEthinylestradiol Di-TMSEthinylestradiol Mono-TMSProgesteroneRelative Response Factor No Derivatization Block Heating Mono/Di-TMS Microwave Heating 1 Mono/Di-TMS Microwave Heating 2 Mono/Di-TMSM+. 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Total Mono-TMSDES Total Di-TMSDES Mono-TMSEstrone Di-TMS-17BEstradiol Mono-TMSEthinylestradiol Di-TMSEthinylestradiol Mono-TMSProgesteroneRelative Response Factor No Derivatization Block Heating Mono/Di-TMS Microwave Heating 1 Mono/Di-TMS Microwave Heating 2 Mono/Di-TMSM+. 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Total Mono-TMSDES Total Di-TMSDES Mono-TMSEstrone Di-TMS-17BEstradiol Mono-TMSEthinylestradiol Di-TMSEthinylestradiol Mono-TMSProgesteroneRelative Response Factor No Derivatization Block Heating Mono/Di-TMS Microwave Heating 1 Mono/Di-TMS Microwave Heating 2 Mono/Di-TMSM+. Figure 5-3. Relative response factors (RRFs) for the derivatized steroids with and wi thout derivatization ( each at a concentra tion of 5 g mL-1). No derivatization = detected underi vatized species, Derivatization = bloc k and microwave heating Â– detected mono/di-tms species. Thermal derivatiz ation was performed at 70 C for 30 min, microwave 1 and microwave 2 derivatization methods were pe rformed at 500W and 900W for 1 minute, respec tively. Error bars re present the standard deviation of the mean (n=3).
156 0.0 0.5 1.0 1.5 2.0 2.5 Mono-TMS-4Octylphenol Mono-TMS-4Nonylphenol Mono-TMSTriclosan Di-TMSResorcinol Mono-TMSDDA Mono-TMSHPTE Di-TMSCoumestrol Di-TMSBisphenol ARelative Response Factor No Derivatization Block Heating Mono/Di-TMS Microwave Heating 1 Mono/Di-TMS Microwave Heating 2 Mono/Di-TMSM+. 0.0 0.5 1.0 1.5 2.0 2.5 Mono-TMS-4Octylphenol Mono-TMS-4Nonylphenol Mono-TMSTriclosan Di-TMSResorcinol Mono-TMSDDA Mono-TMSHPTE Di-TMSCoumestrol Di-TMSBisphenol ARelative Response Factor No Derivatization Block Heating Mono/Di-TMS Microwave Heating 1 Mono/Di-TMS Microwave Heating 2 Mono/Di-TMSM+. Figure 5-4. Relative response fact ors (RRFs) for the other deriva tized polar EDCs with and without derivatization (each at a concentration of 5 g mL-1). No derivatization = detected underivatized species, Deriva tization = block and microwave heating Â– detected mono/di-tms species. Thermal derivatiz ation was performed at 70 C for 30 min, microwave 1 and microwave 2 derivatization methods were performed at 500W and 900W for 1 minute, respec tively. Error bars represent the standard deviation of the mean (n=3).
157 25 14 30 18 11,B 10 24 13 21 12 9 7,A 4 2 5 3 1 M J N L K H D F E C RT: 11.85 -30.20 12 14 16 18 20 22 24 26 28 30 Time (min) 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 25 14 30 18 11,B 10 24 13 21 12 9 7,A 4 2 5 3 1 M J N L K H D F E C 25 14 30 18 11,B 10 24 13 21 12 9 7,A 4 2 5 3 1 M J N L K H D F E C RT: 11.85 -30.20 12 14 16 18 20 22 24 26 28 30 Time (min) 0 10 20 30 40 50 60 70 80 90 100Relative Abundance RT: 11.85 -30.20 12 14 16 18 20 22 24 26 28 30 Time (min) RT: 11.85 -30.20 12 14 16 18 20 22 24 26 28 30 Time (min) 0 10 20 30 40 50 60 70 80 90 100Relative Abundance Figure 5-5. Chromat ogram of 100 ppb (0.1 g mL-1) EDC mixture spiked in extracted Lake Apopka water. The EDCs detected were: 1 = diethyl phthalate, 2 = benzophenone, 3 = trifluralin, 4 = atrazine 5 = lindane, 6 = 4-octylphenol, 7 = anthracene, 9 = vinclozolin, 10 = alachlor, 11 = heptachlor, 12 = dibutyl phthalate, 13 = dicofol, 14 = heptachlor epoxide, 18 = p,pÂ’-DDE, 21 = p,pÂ’-DDD, 24 = p,pÂ’-DDT, 25 = chrysene-d12, 30 = progesterone. A = mono-tms-4-octylphenol, B = mono-tms-4nonylphenol, C = mono-tms-triclosan, D = m ono-tms-p,pÂ’-DDA, E = di-tms-bisphenol A, F = di-tms-DES-1, H = di-tmsDES-2, J = mono-tms-HPTE, K = mono-tms-estrone, L = di-tms-17 -estradiol, M = mono-tms-et hinylestradiol, N = ditms-ethinylestradiol. Mono-tms-resorcinol (tr < 12 min) is not shown in figure.
158 100 Endrin ketone Time (min) 14 15 16 17 18 19 20 21 22 23 24 0 10 20 30 40 50 60 70 80 90 Relative Abundance Heptachlor Lindane Endrin Dieldrin DDT Aldrin 100 Endrin ketone Time (min) 14 15 16 14 15 16 17 18 19 20 21 22 23 24 0 10 20 30 40 50 60 70 80 90 Relative Abundance Heptachlor Lindane Endrin Dieldrin DDT Aldrin Figure 5-6. Chromatogram of several pot ential EDCs found in Lake Apopka water (FWL1874). The DDT detected was p,pÂ’-DDT.
159 CHAPTER 6 INVESTIGATION OF ESTROGENS IN ALLIGATOR PLASMA BY GAS CHROMATOGRAPHY/MASS SPECTROMETRY (GC/MS) COUPLED WITH SOLIDPHASE MICROEXTRACTION ON-FIBER DERIVATIZATION (SPME-OFD) Introduction Hormones produced by the endocrine system pl ay important roles re gulating physiological functions, such as growth, development, and reproduction. Given the important roles that steroids play during embryonic development, puberty, reproductive cycl icity, immune system upkeep and stress response,1, 2, 336 there is a fundamental need to develop methods to increase the understanding of these steroid pathways. Steroids, both natural and synthetic, have been used to enhance athletic performance,2, 83-85 improve livestock quality,159-165 and control reproductive processes.86, 91 Furthermore, the endogenous profile of ster oids has also served as a biomarker for endocrine disorders, from either natural occurrences6, 154, 187, 188, 190, 195, 199-203 or exposure to anthropogenic contaminants.70 With the ever-expanding knowl edge base of known and unknown steroid pathways and responses, new methods for the characterization of steroids are becoming increasingly imperative. To date, no method exists for the characterization of all relevant steroids in a single analysis; however, se veral methods have been applie d for the analysis of related subsets of steroids, including immunoassay (I A) based techniques, liquid chromatography/mass spectrometry (LC/MS), and gas chroma tography/mass spectrometry (GC/MS). Steroid Analysis by IA-Based Techniques IA based techniques, such as radio(RIA), enzyme(EIA), and enzyme-linkedimmunoassay (ELISA), have been used frequently to detect specific st eroids in biological samples. These techniques are based on the ability to measure the binding of specific antibodies (proteins) to specific antigens (steroids).181 Thus, IA-based methods have high sensitivity (detection at concentrations as low as pg mL-1).174 However, steroids with similar chemical
160 strcutures have shown the ability to interfere with the assays and pr oduce false positives. Along with cross reactivity, IA-based techniques also suffer from the limited availability of antibodies and the fact for the investigation of multiple ster oids, separate assays have to performed for each one.6, 182-186 Steroid Analysis by Liquid Chromatography/Tandem Mass Spectrometry (LC/MS/MS) LC/MS has been the most commonly used techni que for the detection of steroids and has been used to detect steroids at levels as low as ng mL-1 in biological samples (with MS/MS).98 The wide application of LC/MS for steroid analysis arises from its ability to avoid extra tedious sample preparation strategi es, such as derivatization.85, 150, 166-169, 171 Practical LC/MS analysis typically has a lower separative power in compar ison to capillary GC, as determined by a lower peak capacity.172 LC/MS analysis also suffers from an in ability to separate the entire range of polarities in one LC separation, which is possi ble by GC/MS after derivatization. In addition, since separation by LC/MS has often been restri cted to characterizing small sub-classes of chemically related steroids due to their intera nd intra-class chemical differences, compatibility issues often arise between the steroids, colu mn, and mobile phase setup, consequentially requiring prohibitively long methods (>30 min.) to profile across cl asses. Furthermore, LC is typically coupled to either electrospray (ES I) or atmospheric-pressure chemical (APCI) ionization mechanisms, with each ionization met hod working more efficiently at opposite ends of the polarity spectrum.6, 185, 193, 194 ESI typically is best-suited for polar steroids (such as 17 estradiol) while APCI is suited for the analys is of low to mid-polar steroids (such as progesterone). Steroid Analysis by Gas Chromat ography/Mass Spectrometry (GC/MS) GC/MS has been the most effective technique fo r the comprehensive analysis of steroids in biological samples due to its high chromatographic resolving power and extensive spectral
161 libraries.172, 188, 245 However, most steroids require chem ical derivatization to improve their volatility and thermal stabil ity. Derivatization, although cons idered time-consuming and tedious,119 can be implemented to expand the range of compounds amenable in a single analysis. In addition to optimized deriva tization procedures, the GC/MS analysis of steroids from biological samples typically requires sample extraction. Analysis of Steroid by Solid-Phase Microex traction with On-Fiber Derivatization (SPMEOFD) The successful analysis of steroids by GC/MS is strongly dictated by the effectiveness of the sample preparation procedures employed prio r to injection. An important procedure often required for the analysis of st eroids in plasma is extrac tion. The most common extraction technique is solid-phase extracti on (SPE), which involves the reten tion of an analyte on a sorbent packing inside an extraction cartr idge, followed by an elution step.260 Although extraction by SPE is often considered the ideal strategy, it ca n suffer from cartridge plugging and sample loss during required offline derivatization steps. An interesting alternative to SPE is solid-phase microextraction (SPME). SPME, employing a poly acrylate fiber, can routinely extract and preconcentrate steroids on the fiber.104, 222, 241, 242, 270 These methods typically employ on-fiber derivatization (OFD), which aids in increasing the GC amenabil ity of the extracted compounds. SPME-OFD is also noted for its simplicity and it s ability to form single derivatized species.241, 242, 270 In this study, changes in the estrogen cascade of E2-injected alligators were examined using GC/MS coupled with SPME-OFD. Innova tive techniques designed to improve the extraction of steroids from pl asma by SPME-OFD were also examined. The developed SPMEOFD method was capable of charac terizing the endogenous estrogens, ( and ) E2 and estrone (E1), along with several other ster oids. A comparison of the con centrations obtained using the
162 developed SPME-OFD method was compared to values obtained using an RIA method. The presence of the endogenous steroids was also confirmed by LC/MS/MS. Experimental Section Steroid Standards Standards of the steroids: testoste rone (T), androstenedione (AE), 5-dihydrotestosterone (DHT), 17-estradiol (E2), estrone (E1), progesterone (P), and pr egnenolone (PREG) were obtained from Sigma-Aldrich (St. Louis, MO). The internal standard, estradiol-17 -acetate (E2Ac) was also acquired from Sigma. The solvents used were all HPLC-grade solvents (water and methyl tert -butyl ether) and were obtain ed from Fischer Scientific (Fair Lawn, NJ). The steroid solutions were prepared at various concentrations from a stock solution made at a concentration of 100 mg L-1. The derivatizing reagent was N -methylN -trimethylsilyl-trifluoroacetamide (MSTFA) (Pierce, Rockford, IL). Liquid Chromatography/Tandem Mass Spectrometry (LC/MS/MS ) The LC/MS/MS analysis was performed on a ThermoFinnigan LCQ quadrupole ion trap mass spectrometer. The column employed was a L una C18 packed-particulate column (with the dimensions of 250 mm x 4.6 mm i.d. and a 5 m particle size) The sample was loaded onto the column by a Paradigm MS4 Magic autosampler. The autosampler was set to inject 20L of sample. The APCI and tandem mass spectrome tric (MS/MS) parameters were followed according to previous work.337 The MS/MS analysis of the es trogens was performed using the [M H2O+H]+ precursor ion, along with another charac teristic ion obtained from examining standard solutions. Solid-Phase Microextraction with On-Fiber Derivatization (SPME-OFD) Prior to the extraction by SPME, a liquid-liqui d extraction step was implemented to clean up the sample. The plasma was added to a vial followed by 2 mL of MTBE. The plasma/MTBE
163 mixture was vortexed for 45 seconds. The sample wa s then put into a freezer (20 C). After approximately one hour, the top layer was remove d and added to another vial. The LLE extract was then evaporated with UHP-nitrogen and r econstituted with analytical grade water. The sample was then sonicated for 1 minute and ex tracted using a SPME fiber coated with an 85 m polyacrylate (PA) phase (Supelc o, Bellefonte, PA). Directimmersion-SPME was performed by exposing the PA fiber into the plasma extract fo r 30 minutes at room temperature. Extraction was facilitated by the use of a magnetic stir bar. Po st-extraction, the fiber wa s head-space derivatized in a vial containing 50 L of MSTFA heated in a water bath at 55 C. The PA fiber was exposed to the derivatization reag ent vapor for 30 minutes. The fiber wa s then injected into the GC, and the analytes were desorbed from the fiber in the injection port for five minutes. The internal standard employed was 17-E2-Ac (at a concentration of 100 ng L-1) and was derivatized on the fiber. The standard calibration curve was performed (by SPME-OFD, as previously described) with steroid standards spiked into water at concentrati ons of 0.2, 0.7, 2, 20, 100, 200, and 875 ng mL-1. Gas Chromatography/Mass Spectrometry (GC/MS) GC/MS analysis of plasma extract was pe rformed using a ThermoFinnigan Trace GC 2000 gas chromatograph/quadrupole ion trap ma ss spectrometer, equipped with an AS3000 Autosampler (San Jose, CA) and Xcalibur 1.4 da ta acquisition software. An SLB-5ms capillary column (Supelco, Bellefonte, PA) w ith dimensions of 30 m x 0.25 mm x 0.25 m (film thickness), was utilized for th e separation of components in the Trace GC. The ion source and transfer line temperatures were set to 200 C and 300 C, respec tively. Splitless injection was employed at a temperature of 280 C and a split flow of 50 mL min-1. The carrier gas, ultra-highpurity (UHP) helium (99.99%), was us ed at a flow rate of 1 mL min-1. For the analysis of
164 derivatized extracts by GC/MS, a temperature program was employed at an initial temperature of 120 C (2 minutes). The temperature ramp then increased to 2 50 C at 15 C min-1. After a temperature of 250 C was reached, the ra mp increased to 300 C at 5 C min-1 and was held for 5 minutes. Analysis of E2-Injected Alligator Plasma The eleven alligators used in the study were injected with E2 at a concentration of 2270 g/kg every 2 days. The injections were administer ed for 10 days. The samp les analyzed in this study were from the 10th day (last day of plasma collection) More information on the alligators (sex, origin, and label) is shown in Table 6-1. The concentrations calc ulated by the RIA method were performed by graduate students in the Gu illette research group in the Department of Zoology. Results and Discussion Overall SPME-OFD GC/MS Procedure Due to the high viscosity of th e alligator plasma samples, SPE methods could not be used to extract the estrogens from the plasma because of severe cartridge clogging. Futhermore, the high abundance of proteins created difficulties fo r traditional SPME methods, as proteins tend to interrupt the fiber phase. Thus, a method to ci rcumvent the high abundance of proteins was created. The implementation of a LLE method (using MTBE) prior to SPME enabled the extraction of the free estrogens without protein disruption. SPME-OFD GC/MS Calibration The calculated concentrations of 17 -E2 (with RSD) for all eleven alligators using both the SPME-OFD (with GC/MS) and the RIA method are shown in Table 6-1. The calibration curves for both E1 and 17 -E2 are shown in Figures 6-1 and 6-2. In all cases, the aver age concentrations calculated using the GC/MS method was below the RIA method (except for animal 3.4). The
165 difference in calculated E2 concentrations between both methods may be due to several factors. First, the SPME-OFD extraction method was initia lly designed to extract in the low ng mL-1 concentration range. Upon the in itial analysis of pooled E2 plasma, it was discovered that the calculated concentrations of the E2 plasma were outside of the calibration curve (a result also experienced by the Guillette group when performi ng the RIA). At higher concentrations, it was determined that the fiber and analyte reach a saturation point in which the extraction of the highly concentrated analyte is no longer linear. To correct for the high concentration levels, the LLE plasma extract was diluted by a factor of te n. Unfortunately, the internal standard (17 -E2Ac) was added prior to LLE but after the sample dilution. Thus, some error could be a result of the dilution of the intern al standard since its linearity wa s not examined (although the internal standard had peak area ratios with an RSD of approximately 10%). Differences Between RIA and GC/MS (with SPME-OFD) Other factors which may have contributed to the difference in concentration of E2 between the RIA and GC/MS methods could involve the ine fficiency of extraction associated with the LLE method (MTBE). In previous work, the ex traction by MTBE had hi gh relative extraction efficiencies for the estrogens, but had a high propagation of error. Furthermore, crucial differences in sample treatment between the tw o methods (RIA, GC/MS) could account for some of the differences in concentration levels. A protein precipitation method was investigated, employing acetontrile (3:1 to plasma); however, the resultant extract was no t clean enough for SPME analysis. A benefit of employing GC/MS with SPME-OFD is the collection of additional information. GC/MS analysis also provided the ability to calculate concentrations of 17 -E2 and E1. In addition, the analysis also allowed for the detection of 17 -E2, cholesterol, vitamin E, and
166 several fatty acids. The chromatogram in Figure 63 shows all of the ster oids (and IS) separated by the SPME-OFD GC/MS method. Analysis of the plasma sample by RIA only provided information on the E2 concentrations (additional analyses are required for information on other analytes). Confirmation with LC/MS/MS Analysis Since most steroids are low to mid-polar, APCI was the best-suited ionization strategy for the analysis of steroids over a wider range of polarities. The analysis of E2-injected alligator plasma by LC/MS/MS was capable of detecting both and -E2 and E1. The LC/MS/MS method also detected estriol (E3) in several of the samples. A chromatogram and mass spectrum of E2 detected in E2-injected alligator plasma is shown in Figure 6-4. 17 -E2 The method developed with SPME-OFD potentia lly identified the e ndogenous estrogen, 17 -E. Although there have been a fe w reports which describe 17 -E2 and its medicinal applications,338, 339 the role of 17 -E2 in the body is currently unknown.338, 340 During the examination of the E2-injected alligators, two chromatogr aphic peaks were observed for the characteristic ions of di-TMS-E2 (416 and 285). The additional peak was not observed during the analysis of normal alligator and human plasma. The potential -E2 has a nearly identical mass spectrum (as shown in Figure 6-4) but with a di stinctly different retention time (di-TMS-17 -E2 25.09 minutes and di-TMS-17 -E2 25.60 minutes). The 17-E2 chromatographic peak is in low abundance in comparison to the 17-E2 peak, as 17-E2/17 -E2 is typically about 1.5% (shown in Table 6-2). Estradiol is relatively easy to derivatize (at both hydroxyl positions, and since estradiol typically only exists as an and hydroxyl at the 17-posit ion, it is reasonable to
167 assume that the other peak is 17-E2. Further analysis using 17-E2 standards need to be performed for complete verification. Conclusion In this study, the developed SPME-OFD me thod demonstrated the capability of characterizing endogenous steroids in alligator plasma. Alt hough extra sample preparation procedures were required (addi tional LLE), it was able to detect several estrogens in a single analysis. Moreover, with the addition of the sample dilution step (post-LLE), multiple SPMEOFD analyses of the same sample could be performed. The investigation of changes in the es trogen cascade, in response to an E2-injection, not only provided information on the clearance of E2, but it also provided information about how other steroids can be affected. Th e extremely high level of estrogens in the al ligators indicates a slow clearance of the st eroids upon stimulation. The next stage of the analysis will involve characterizing the pl asma extracts for the eleven alligators over the other nine days of the study. In addition, future work will involve an examination of sample stability. Plasma samples from the same aliquot will be analyzed by both labs and the resulting E2 concentrations will be compared. Fu ture work will also involve the improvement of the sample preparation steps used prior to SPME.
168 Table 6-1. Estimated concentration levels (in ppb) for the E2-injected alligators using both RIA and GC/MS (with SPME-OFD) methods. Animal Lake Year Collected Sex E2 RIA E2 GC/MS E1 GC/MS 1.1 Apopka 2002 Female 637 540 65 5.0 0.2 1.2 Apopka 2002 Male 643 461 28 4.5 1.2 1.3 Woodruff 2002 Male 801* 604 66 6.3 0.9 2.1 Apopka 2002 Female 960* 469 42 4.8 0.8 2.2 Apopka 2003 Female 1084* 448 27 2.8 0.4 2.3 Apopka 2003 Male 626 438 18 5.9 1.3 2.4 Woodruff 2003 Male 764 549 16 6.5 1.1 3.1 Apopka 2002 Female 880* 455 41 7.1 4.0 3.2 Apopka 2001 Male 612 563 84 7.8 1.6 3.3 Woodruff 2002 Male 641 594 77 5.2 0.4 3.4 Unknown Unknown Female 630 630 82 5.6 1.5 The values with the indicate that the con centration was outside of the standard curv e used to determine the concentration. Th e concentrations are reported the relativ e standard deviation (RSD, in ppb).
169 Table 6-2. Estimated percentage of -E2 in comparison to the -E2 in the E2-injected alligators usi ng GC/MS coupled with SPMEOFD Animal Lake Year Collected Sex E2/ E2 1.1 Apopka 2002 Female 1.4% 1.2 Apopka 2002 Male 1.2% 1.3 Woodruff 2002 Male 1.7% 2.1 Apopka 2002 Female 2.1% 2.2 Apopka 2003 Female 1.5% 2.3 Apopka 2003 Male 1.2% 2.4 Woodruff 2003 Male 1.9% 3.1 Apopka 2002 Female 0.7% 3.2 Apopka 2001 Male 1.9% 3.3 Woodruff 2002 Male 1.9% 3.4 Unknown Unknown Female 1.7%
170 Peak area ratio = 0.0074(conc) 0.0024 R2 = 0.9999 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 050100150200250Concentration (ppb)Peak Area Ratioppb RSD 0.245% 0.710% 234% 2019% 2002% Peak area ratio = 0.0074(conc) 0.0024 R2 = 0.9999 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 050100150200250Concentration (ppb)Peak Area Ratioppb RSD 0.245% 0.710% 234% 2019% 2002% Figure 6-1. Calibration curve for E1 using the SPME-OFD method with GC/MS. The corresponding RSD values are also present for each concentration level.
171 Peak Area Ratio = 0.0061(conc) 0.0325 R2 = 0.9987 0 1 2 3 4 5 6 01002003004005006007008009001000 Concentration (ppb)Peak Area Ratioppb RSD 223% 2010% 1003% 2007% 8755% Peak Area Ratio = 0.0061(conc) 0.0325 R2 = 0.9987 0 1 2 3 4 5 6 01002003004005006007008009001000 Concentration (ppb)Peak Area Ratioppb RSD 223% 2010% 1003% 2007% 8755% Figure 6-2. Calibration curve for 17 -E2 using the SPME-OFD method with GC/MS. Th e corresponding RSD values are also present for each concentration level.
172 RT: 11.93 34.65 12 14 16 18 20 22 24 26 28 30 32 34 Time (min) 0 10 20 30 40 50 60 70 80 90 100 Relative Abundance x10 x10 Estrone 17-Estradiol 17-Estradiol-acetate Vitamin E Cholesterol 17-Estradiol RT: 11.93 34.65 12 14 16 18 20 22 24 26 28 30 32 34 Time (min) 0 10 20 30 40 50 60 70 80 90 100Relative Abundance x10 x10 Estrone 17-Estradiol 17-Estradiol-acetate Vitamin E Cholesterol 17-Estradiol Figure 6-3. Chromatogram of e ndogenous steroids (and vitamin E) in the SPME-OFD extract of E2-injected alligator plasma.
173 RT: 0.00 48.00 SM: 7G 0 5 10 15 20 25 30 35 40 45 Time (min) 0 10 20 30 40 50 60 70 80 90 100 Relative Abundance 35.78 36.26 29.80 34.49 27.4140.03 NL: 5.08E5 m/z= 132.50-133.50+ 158.50-159.50 F: + c APCI t Full ms2 firstname.lastname@example.org [ 70.00-400.00] MS E2inducedPlasmanewmethod E2inducedPlasma-newmethod # 3627-3700 RT: 35.46-36.26 AV: 24 NL: 1.96E5 F: + c APCI t Full ms2 email@example.com [ 70.00-400.00] 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 159.49 133.36 255.80 255.07 173.12 146.06 199.29 174.16 213.97 237.89 108.79 256.54 132.06 95.37 80.38 301.27 272.78 352.21 344.01 311.67 17 -Estradiol RT: 0.00 48.00 SM: 7G 0 5 10 15 20 25 30 35 40 45 Time (min) 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 35.78 36.26 29.80 34.49 27.4140.03 NL: 5.08E5 m/z= 132.50-133.50+ 158.50-159.50 F: + c APCI t Full ms2 firstname.lastname@example.org [ 70.00-400.00] MS E2inducedPlasmanewmethod E2inducedPlasma-newmethod # 3627-3700 RT: 35.46-36.26 AV: 24 NL: 1.96E5 F: + c APCI t Full ms2 email@example.com [ 70.00-400.00] 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 159.49 133.36 255.80 255.07 173.12 146.06 199.29 174.16 213.97 237.89 108.79 256.54 132.06 95.37 80.38 301.27 272.78 352.21 344.01 311.67 17 -Estradiol Figure 6-4. LC/MS/MS chromat ogram and mass spectrum of E2 detected in E2-injected alligator plasma.
174 RT: 24.93 25.87 25.0 25.1 25.2 25.3 25.4 25.5 25.6 25.7 25.8 Time (min) 0 20 40 60 80 100 Relative Abundance 25.57 25.09 25.7225.81 25.79 24.95 25.45 25.0125.3125.47 25.18 25.16 25.2425.3725.40 25.26 NL: 2.06E5 m/z= 284.50-285.50+ 415.50-416.50 MS spme-alligator-ofdnospikespme-alligator-ofd-nospike # 1804-1815 RT: 25.53-25.63 AV: 12 NL: 4.09E4 T: + c Full ms [50.00-650.00] 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 m/z 0 20 40 60 80 100Relative Abundance 285.21 416.16 326.21 73.15 286.22 232.25 284.24 417.17 218.26 244.25 205.25 325.22 327.20 269.26 129.24 75.16 418.21 115.23 189.25 179.25 159.29 401.22 105.24 67.19 388.17 344.19 374.22 spme-alligator-ofd-nospike # 1761-1765 RT: 25.07-25.11 AV: 5 NL: 1.78E3 T: + c Full ms [50.00-650.00] 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 m/z 0 20 40 60 80 100Relative Abundance 285.18 416.18 73.14 207.12 326.21 286.21 284.19 417.19 232.24 75.15 218.23 325.20 244.25 327.19 205.22 269.23 129.18147.22 91.20 105.22 418.21 177.21 163.20 55.15 355.08 401.15 369.15 388.19 17-estradiol 17-estradiol 17-estradiol 17-estradiolO O Si C H3CH3CH3Si CH3C H3CH3 RT: 24.93 25.87 25.0 25.1 25.2 25.3 25.4 25.5 25.6 25.7 25.8 Time (min) 0 20 40 60 80 100Relative Abundance 25.57 25.09 25.7225.81 25.79 24.95 25.45 25.0125.3125.47 25.18 25.16 25.2425.3725.40 25.26 NL: 2.06E5 m/z= 284.50-285.50+ 415.50-416.50 MS spme-alligator-ofdnospikespme-alligator-ofd-nospike # 1804-1815 RT: 25.53-25.63 AV: 12 NL: 4.09E4 T: + c Full ms [50.00-650.00] 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 m/z 0 20 40 60 80 100Relative Abundance 285.21 416.16 326.21 73.15 286.22 232.25 284.24 417.17 218.26 244.25 205.25 325.22 327.20 269.26 129.24 75.16 418.21 115.23 189.25 179.25 159.29 401.22 105.24 67.19 388.17 344.19 374.22 spme-alligator-ofd-nospike # 1761-1765 RT: 25.07-25.11 AV: 5 NL: 1.78E3 T: + c Full ms [50.00-650.00] 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 m/z 0 20 40 60 80 100Relative Abundance 285.18 416.18 73.14 207.12 326.21 286.21 284.19 417.19 232.24 75.15 218.23 325.20 244.25 327.19 205.22 269.23 129.18147.22 91.20 105.22 418.21 177.21 163.20 55.15 355.08 401.15 369.15 388.19 17-estradiol 17-estradiol 17-estradiol 17-estradiolO O Si C H3CH3CH3Si CH3C H3CH3 Figure 6-5. Chromatogram and ma ss spectra for the two di-TMS-E2 peaks ( and -E2). Both and -E2 were di-TMS-derivatized.
175 CHAPTER 7 CONCLUSION AND FUTURE WORK Conclusions The underlying theme of this dissertation wa s to develop innovative methods for the characterization of endocrine di srupting compounds (EDCs) and st eroids in environmental and biological samples, respectively. The developed methods were based upon the combination of optimized chromatographic methods with optimized sample preparation strategies. The developed method employed gas chromatogra phy/mass spectrometry due to its superior chromatographic resolution and extensive spectral libraries. Novel extraction and derivatization strategies were explored to improve and exte nd the qualitative and quantitative information obtained in comparison to previous strategies. Several SPE cartridges were explored in thei r ability to extract analytes over a wide polarity range in a single extraction. Extrac tion with Oasis HLB cartridges had the best reproducibility; however, the LLCE cartridges ty pically had higher relative extraction efficiencies for the spiked steroids and EDCs in water. Furthermore, extraction using the LLCE method was simple and possessed the ability to extr act a diverse mixture of analytes in a single analysis. The application of SPME and SPME-OFD for the analysis of steroids and EDCs exhibited great potential for the simultaneous extrac tion of both non-polar and polar compounds. The addition of OFD to SPME improved the amenab ility of the polar steroids/EDCs for GC/MS analysis and allowed the detection of these compounds as low as 0.2 ng mL-1 in spiked plasma. Optimization of derivatization strategies fo r the analysis of steroids by GC/MS was investigated by varying the experimental para meters (reagent, reaction time and reaction temperature) to determine the optimal c onditions for steroids on an individual and
176 comprehensive level. Three methods of derivati zation enhancement were also investigated: the use of sonication, the use of a microwave heatin g, and the addition of solvents to the reaction mixture. Microwave reactions (using a domestic microwave oven) were capable of increasing the derivatization yield (for all the silyl reagen ts) in one minute (compared to the thermal method performed for 30 minutes). The addition of so lvent during the reaction also enhanced the deriviatization reaction, but only for BSTFA/TMCS and BSA. MSTFA reactions were not enhanced by the addition of solvent. Further examinations of the capabilities of microwave-accelerated derivatization (MAD) were performed using a laboratory synthesi s microwave system. MAD using a synthesis microwave system was evaluated and compared to traditional thermal derivatization methods in terms of yield, reproducibility, and overall analys is time. Parameters affecting MAD, including reaction temperature, time, and pow er, were systematically optimized for several silyl reagents (BSTFA with TMCS, MSTFA, and BSA) and othe r derivatization proced ures (MOX reagent and MTBSTFA). MSTFA was found to derivatize best w ith the microwave, as demonstrated by the enhanced relative response factors (RRFs). BSTFA with TMCS, on the other hand, did not couple as well, but RRF values improved signifi cantly upon addition of polar solvents. The rapid (1 min) derivatization reactions associated with MAD had comparab le RRFs for all reagents with those obtained with thermal heating (>30 mi n). Finally, the MAD methodology improved the derivatization of steroids in spiked plasma. This study highlig hts the best methods for analyzing a comprehensive variety of steroids, and also pr ovides ideal strategies fo r the MAD of steroids on an individual and class level. The expanding list of EDCs has increased th e need for the development of improved monitoring methods to evaluate their presence in the environment. Furthermore, the diverse
177 physiochemical properties of EDCs impose i nherent analytical lim itations and thus new comprehensive methods that can simultaneously analyze numerous EDCs in one chromatographic analysis would be a signifi cant improvement over current EDC detection methods. Gas chromatography/mass spectrometr y (GC/MS) offers promising profiling capabilities; however, many polar EDCs require derivatization for adequate detection. Hence, a novel method for the comprehensive profiling of EDCs that employs a silyl derivatization strategy to expand the polarity ra nge of compounds able to be se parated and detected in a single chromatographic analysis was developed. The comprehensive method successfully separated twenty-one GC-ready and twelve derivatization-required EDCs in one chromatographic analysis. Thermal and microwave derivatiz ation methods were effectiv e for the comprehensive EDC mixture, although the microwave derivatization ofte n proved more effective in less analysis time. A pilot-study of surface water from Lake Apopka (Florida) demonstrated the potential of the comprehensive EDC profiling method. The characterization of water examined from Lake Apopka illustrated that not only were several EDCs present at relatively high concentrations, but several of the EDCs found are not currently used in agricultural practices. In particular, DDT was found at a very high concentration, as its detection was possible without any extraction. The developed SPME-OFD method clearly dem onstrated the capability of monitoring endogenous steroids in alligator plasma. The monitoring of endoge nous steroids helps provide a better understanding of the levels by which normal steroidogenesis can be altered. The detection of 17 -E2, 17 -E2, and E1 was demonstrated in the E2-injected alligator plasma. The concentration levels for 17-E2 obtained by GC/MS were typically lower than the values obtained using radioimmunoassay, like due to differences in sample treatment (between the RIA and GC/MS methods) and the inherent precis ion associated with the SPME-OFD method.
178 Future Work For the analysis of alligator plasma, further research is needed in the development of analytical methods capable of de tecting steroids at lower concen trations. Several strategies to achieve this include the investigation of better ex traction strategies prior to the derivatization and GC/MS analysis. In particular, for the improve ment of the SPME OFD method, investigations into the application of different pH levels, salt concentrations, and extraction temperatures should be investigated as alternative means to reach equilibrium more efficiently (between the fiber and the analyte). In addi tion, several OFD parameters should be optimized, including the head-space volume and derivatiz ation temperature. Beyond th e optimization of SPME-OFD parameters, an effort to scale down both th e overall analysis time and extra preparation procedures should be performed. The investigation of E2-injected alligators provided an ex ample of how alligators respond to external stimulation. An inve stigation of steroid profiles of alligators from different lakes (contaminated vs. non contaminated lakes) should be performed to gain a better understanding of how EDCs can alter normal steroidogenesis for allig ators in wildlife. In addition, an examination of alligators of different ages, sex and size would be helpful in the understanding in the differences by which EDCs can affect the endocrine system within a species.
179 LIST OF REFERENCES 1. Hadley, M. E., Endocrinology 4th ed.; Prentice Hall: U pper Saddle River, NJ, 1996. 2. Solomon, E. P.; Berg, L. R.; Martin, D. W., Biology 5th ed.; Harcourt Brace College Publishers: Fort Worth, TX, 1999. 3. Tyler, C. R.; Jobling, S.; Sumpter, J. P., Endocrine Disruption in Wildlife: A Critical Review of the Evidence. Crit. Rev. Toxicol. 1998, 28 (4), 319 361. 4. Mueller, S. O.; Korach, K. S., Mechanisms of Estrogen Receptor-M ediated Agonistic and Antagonistic Effects. In The Handbook of Environmental Chemistry Metzler, M., Ed. Springer-Verlag: Berlin, Ge rmany, 2001; Vol. 3-L. 5. Mueller, S. O., Xenoestrogens: mechan isms of action and detection methods. Anal. Bioanal. Chem. 2004, 378 (3), 582-587. 6. Holst, J. P.; Soldin, O. P.; Guo, T.; Sold in, S. J., Steroid Hormones: relevance and measurement in the clinical laboratory. Clin. Lab. Med. 2004, 24 105-118. 7. Guillette Jr., L. J.; Crain, D. A., Environmental Endocrine Disruptors: An Evolutionary Perspective Taylor & Francis: New York, NY, 2000. 8. Colborn, T.; Saal, F. S. v.; Soto, A. M., De velopmental Effects of Endocrine-Disrupting Chemicals in Wildlife and Humans. Environ. Health Perspect. 1993, 101 (5), 378-384. 9. Cooper, R. L.; Kavlock, R. J., Endocrine disruptors and reproductive development: a weight-of-evidence overview. J. Endocrinol. 1997, 152 159-166. 10. Kavlock, R. J.; Daston, G. P.; DeRosa, C.; Fe nner-Crisp, P.; Gray, L. E.; Kaattari, S.; Lucier, G.; Luster, M.; Mac, M. J.; Maczka, C.; Ron, M.; Moore, J.; Rolland, R.; Scott, G.; Sheehan, D. M.; Sinks, T.; Tilson, H. A., Research Needs for the Risk Assessment of Health and Environmental Effects of Endocri ne Disruptors: A Repor t of the U.S. EPASponsored Workshop. Environ. Health Perspect. 1996, 104 715-740. 11. Colborn, T.; Dumanoski, D.; Myers, J. P., Our Stolen Future Dutton: New York, NY, 1996; p 306. 12. Fossi, C. M.; Marsili, L., Effects of endocrine disrutpors in aquatic mammals. Pure Appl. Chem. 2003, 75 (11-12), 2235-2247. 13. LeBlanc, G. A.; Bain, L. J., Chronic Toxic ity of Environmental Contaminants: Sentinels and Biomarkers. Environ. Health Perspect. 1997, 105 65-80. 14. Vos, J. G.; Dybing, E.; Greim, H. A.; Ladefoged, O.; Lambreras, C.; Tarazona, J. V.; Brandt, I.; Vethaak, A. D., Health Eff ects of Endocrine-Disrupting Chemicals on Wildlife, with Special Referen ce to the European Situation. Crit. Rev. Toxicol. 2000, 30 (1), 71 133.
180 15. Guillette Jr., L., Endocrine Disrupting Envi ronmental Contaminants and Developmental Abnormalities in Embryos. Hum. Ecol. Risk Assess. 1995, 1 (2), 25-36. 16. U.S. Department of Health and Human Serv ices, Toxicological Profile for DDT, DDE, and DDD. Service, P. H., Ed. Agency for T oxic Substances and Disease Registry: 2002. 17. Stapleton, D. H., A Lost Chapte r in the Early History of DDT. Tech. Cult. 2005, 46 (3), 513-540. 18. Snedeker, S. M., Pesticides and breast cance r risk: a review of DDT, DDE, and dieldrin. Environ. Health Perspect. 2001, 109 (Suppl 1), 35-47. 19. Hunter, D. J.; Hankinson, S. E.; Laden, F.; Colditz, G. A.; Manson, J. E.; Willett, W. C.; Speizer, F. E.; Wolff, M. S., Plasma Orga nochlorine Levels and the Risk of Breast Cancer. N. Engl. J. Med. 1997, 337 (18), 1253-1258. 20. Daston, G. P.; Gooch, J. W.; Breslin, W. J.; S huey, D. L.; Nikiforov, A. I.; Fico, T. A.; Gorsuch, J. W., Environmental Estrogens a nd Reproductive Health : Discussion of the Human and Environmental Data. Reprod. Toxicol. 1997, 11 (4), 465-481. 21. Sharpe, R. M.; Skakkebaek, N. E., Are oest rogens involved in falling sperm counts and disorders of the male repr oductive tract? (Hypothesis). The Lancet 1993, v341 (n8857), p1392(4). 22. Wilcox, A. J.; Baird, D. D.; Weinberg, C. R.; Hornsby, P. P.; Herbst, A. L., Fertility in Men Exposed Prenatally to Diethylstilbestrol. N. Engl. J. Med. 1995, 332 (21), 14111416. 23. Giusti, R. M.; Iwamoto, K., Diethylstilbestrol revisited: A review of the long-term health effects. Ann. Intern. Med. 1995, 122 (10), 778-788. 24. Berkson, L., Hormone Deception McGraw-Hill: New York, NY, 2000. 25. Suh, R. S.; Wang, X.; Ohl, D. A.; vom Saal, F. S.; Smith, G. D., Influence of Prenatal Diethylstilbestrol (DES) and dietary phytoestrogens on sper m function and reproductive organ development in male mice. Fertil. Steril. 2003, 80 (Supplement 3), 235-235. 26. McLachlan, J. A.; Newbold, R. R.; Shah, H. C.; Hogan, M. D.; Dixon, R. L., Reduced fertility in female mice exposed transpl acentally to diethylstilbestrol (DES). Fertil. Steril. 1982, 38 (3), 3640371. 27. McLachlan, J. A., Prenatal exposure to diethy lstilbestrol in mice: toxicological studies. J. Toxicol. Environ. Health 1977, 2 (3), 527-537. 28. Carson, R., Silent Spring Houghton Mifflin Company: New York, NY, 1962.
181 29. Ankley, G. T.; Johnson, R. D.; Detenbeck, N. E.; Bradbury, S. P., Development of a research strategy for assessing the ecol ogical risk of endocrine disruptors. Rev. Toxicol. 1997, 1 71-106. 30. Myers, J. P.; Guillette, L. J.; Palanza, P.; Parmigiani, S.; Swan, S. H.; Saal, F. S. v. In The emerging science of endocrine disruption International Seminar on Nuclear War and Planetary Emergencies, Erice, Italy, Erice, It aly, 2003; pp 1-13. 31. Boas, M.; Feldt-Rasmussen, U.; Skakkebaek, N. E.; Main, K. M., Environmental chemicals and thyroid function. Eur. J. Endocrinol. 2006, 154 (5), 599-611. 32. Sonnenschein, C.; Soto, A. M., An Update d Review of Environmental Estrogen and Androgen Mimics and Antogonists. J. Steroid Biochem. Mol. Biol. 1998, 65 (1-6), 143150. 33. Korach, K. S., Surprising places of estrogenic activity. Endocrinology 1993, 132 (6), 2277-2278. 34. Feldman, D. M., Editorial: Estrogens from Plastic--Are We Being Exposed? Endocrinology 1997, 138 (5), 1777-1779. 35. Krishnan, A. V.; Stathis, P.; Permuth, S. F.; Tokes, L.; Feldman, D., Bisphenol-A: an estrogenic substance is released from polycarbonate flasks during autoclaving. Endocrinology 1993, 132 (6), 2279-2286. 36. Crews, D.; Willingham, E.; Skipper, J. K., Endocrine Disruptors: Present Issues, Future Directions. Q. Rev. Biol. 2000, 75 (3), 244-260. 37. Guillette Jr., L. J.; Crain, D. A.; Gunderson, M. P.; Kools, S. A. E.; Milnes, M. R.; Orlando, E. F.; Rooney, A. A.; Woodward, A. R., Alligators and Endocrine Disrupting Contaminants : A Current Perspective. Amer. Zool. 2000, 40 438-452. 38. Kelce, W. R.; Stone, C. R.; Laws, S. C.; Gr ay, L. E.; Kemppainen, J. A.; Wilson, E. M., Persistent DDT metabolite p,p'Â–DDE is a potent androgen receptor antagonist. Nature 1995, 375 (6532), 581-585. 39. Kelce, W. R.; Wilson, E. M., Antiandroge nic Effects of Environmental Endocrine Disruptors. In The Handbook of Environmental Chemistry Metzler, M., Ed. SpringerVerlag: Berlin, Germany, 2001; Vol. 3-L. 40. Barcelo, D.; Kettrup, A., Endocrine disruptors. Anal. Bioanal. Chem. 2004, 378 (3), 547548. 41. Thornton, J. W.; McCally, M.; Houlihan, J ., Biomonitoring of Industrial Pollutants: Health and Policy Implications of the Chemical Body Burden. Public Health Rep. 2002, 117
182 42. Boyd, G. R.; Palmeri, J. M.; Zhang, S.; Grim m, D. A., Pharmaceuticals and personal care products (PPCPs) and endocrine disrupting chemicals (EDCs) in stormwater canals and Bayou St. John in New Orleans, Louisiana, USA. Sci. Total Environ. 2004, 333 (1-3), 137-148. 43. Bound, J. P.; Voulvoulis, N., Household Dispos al of Pharmaceuticals as a Pathway for Aquatic Contamination in the United Kingdom. Environ. Health Perspect. 2005, 113 (12), 1705-1711. 44. Acerini, C. L.; Hughes, I. A., Endocrine di srupting chemicals: a new and emerging public health problem? Arch. Dis. Child. 2006, 91 (8), 633-641. 45. Vanderford, B. J.; Pearson, R. A.; Rexing, D. J.; Snyder, S. A., Analysis of Endocrine Disruptors, Pharmaceuticals, and Persona l Care Products in Water Using Liquid Chromatography/Tandem Mass Spectrometry. Anal. Chem. 2003, 75 (22), 6265-6274. 46. Rhind, S. M., Are Endocrine Disrupting Com pounds a Threat to Farm Animal Health, Welfare and Productivity? Reprod. Domest. Anim. 2005, 40 282-290. 47. Metzler, M.; Pfeiffer, E., Chemistry of Natural and Anthropogenic Endocrine Active Compounds. In The Handbook of Envir onmental Chemistry Metzler, M., Ed. SpringerVerlag: Berlin, Germany, 2001; Vol. 3-L. 48. Danzo, B. J., Environmental Xenobiotics Ma y Disrupt Normal Endocrine Function by Interfering with the Binding of Physiological Ligands to Steroid Receptors and Binding Proteins. Environ. Health Perspect. 1997, 105 (3), 294-301. 49. Labadie, P.; Budzinski, H., Alteration of Steroid Hormone Balance in Juvenile Turbot ( Psetta maxima ) Exposed to Nonylphenol, Bisphenol A, Tetrabromodiphenyl Ether 47, Diallylphthalate, Oil, and O il Spiked with Alkylphenols. Arch. Environ. Contam. Toxicol. 2006, 50 (4), 552-561. 50. Folmar, L. C.; Denslow, N. D.; Rao, V.; C how, M.; Crain, D. A.; Enblom, J.; Marcino, J.; Guillette, L. J., Jr., Vitellogenin Induction and Reduced Serum Testosterone Concentrations in Feral Male Carp (Cyprinus carpio ) Captured near a Major Metropolitan Sewage Treatment Plant. Environ. Health Perspect. 1996, 104 (10), 10961101. 51. Toft, G.; Guillette, J. L. J., Decreased sper m count and sexual behavior in mosquitofish exposed to water from a pesticide-contaminated lake. Ecotoxicol. Environ. Saf. 2005, 60 (1), 15-20. 52. Toft, G.; Edwards, T. M.; Baatrup, E.; Guillette Jr, L. J., Disturbed Sexual Characteristics in Male Mosquitofish ( Gambusia holbrooki ) from a Lake Contaminated with Endocrine Disruptors. Environ. Health Perspect. 2003, 111 (5), 695.
183 53. Berg, C.; Halldin, K.; Fridolfsson, A.-K.; Br andt, I.; Brunstrm, B., The avian egg as a test system for endocrine disr upters: effects of diethylstilbestrol a nd ethynylestradiol on sex organ development. Sci. Total Environ. 1999, 233 (1-3), 57-66. 54. Bull, J. J.; Gutzke, W. H. N.; Crews, D., Sex reversal by estradiol in three reptilian orders. Gen. Comp. Endocrinol. 1988, 70 (3), 425-428. 55. Hayes, T. B.; Collins, A.; Lee, M.; Mendoza, M.; Noriega, N.; Stuart, A. A.; Vonk, A., Hermaphroditic, demasculinized frogs after ex posure to the herbicide atrazine at low ecologically relevant doses. Proc. Natl. Acad. Sci. U. S. A. 2002, 99 (8), 5476-5480. 56. Bergeron, J. M.; Crews, D.; McLachlan, J. A ., PCBs as Environmental Estrogens: Turtle Sex Determination as a Biomarker of Environmental Contamination. Environ. Health Perspect. 1994, 102 (9), 780-781. 57. Willingham, E.; Crews, D., Sex Reversal Effects of Environmentally Relevant Xenobiotic Concentrations on the Red-Eared Sl ider Turtle, a Species with TemperatureDependent Sex Determination. Gen. Comp. Endocrinol. 1999, 113 (3), 429-435. 58. Bogh, I. B.; Christensen, P.; Dantzer, V.; Groot M.; Thofner, I. C. N.; Rasmussen, R. K.; Schmidt, M.; Greve, T., Endocrine disr upting compounds: effect of octylphenol on reproduction over three generations. Theriogenology 2001, 55 (1), 131-150. 59. Birnbaum, L. S.; Fenton, S. E., Cancer and Developmental Exposure to Endocrine Disruptors. Environ. Health Perspect. 2003, 111 (4), 389. 60. Holdway, D. A.; Hefferman, J.; Smith, A., Multigeneration assessment of nonylphenol and endosulfan using a model Australian freshwater fish, Melanotaenia fluviatilis Environ. Toxicol. 2008, 23 (2), 253-262. 61. Milnes, M. R.; Woodward, A. R.; Rooney, A. A.; Guillette, L. J., Plasma steroid concentrations in relation to size and age in juvenile alligators fr om two Florida lakes. Comp. Biochem. Phys. A 2002, 131 (4), 923-930. 62. Semenza, J. C.; Tolbert, P. E.; Rubin, C. H.; Guillette Jr, L. J.; Jackson, R. J., Reproductive Toxins and Alligator Abnormalities at Lake Apopka, Florida. Environ. Health Perspect. 1997, 105 (10), 1030-1032. 63. Guillette Jr., L. J.; Crain, D. A.; Rooney, A. A.; Pickford, D. B., Organization versus Activation: The Role of Endocrine-Disrup ting Contaminants (EDCs) during Embryonic Development in Wildlife. Environ. Health Perspect. 1995, 103 157-164. 64. Geyer, H. J.; Rimkus, G. G.; Scheunert, I. ; Kaune, A.; Schramm, K.-W.; Kettrup, A.; Zeeman, M.; Muir, D. C. G.; Hansen, L. G.; Mackay, D., Bioaccumulation and Occurence of Endocrine-Disrupting Compounds (EDCs), Persistent Organic Pollutants (POPs), and Other Organic Compounds in Fi sh and Other Organisms including Humans. In The Handbook of Environmental Chemistry Beek, B., Ed. Spring-Verlag: Berlin, Germany, 2000; Vol. 2-J.
184 65. 65. Milnes, M. R.; Roberts, R. N.; Guill ette, L. J., Jr., Effects of Incubation Temperature and Estrogen Exposure on Aromatas e Activity in the Brain and Gonads of Embryonic Alligators. Environ. Health Perspect. 2002, 110 393-396. 66. Crain, D. A.; Louis J. Guillette; Rooney, A. A.; Pickford, D. B., Alterations in Steroidogenesis in Alligators ( Alligator Mississippiensis ) Exposed Naturally and Experimentally to Environmental Contaminants. Environ. Health Perspect. 1997, 105 528-533. 67. Pickford, D. B.; Guillette, L. J.; Crain, D. A.; Rooney, A. A.; Woodward, A. R., Plasma Dihydrotestosterone Concentrations and Phallus Size in Juvenile American Alligators ( A. mississippiensis ) from Contaminated and Reference Populations. J. Herpetol. 2000, 34 (2), 233-239. 68. Guillette Jr., L. J.; Woodward, A. R.; Crai n, D. A.; Pickford, D. B.; Rooney, A. A.; Percival, H. F., Plasma Steroid Concentra tions and Male Phallus Size in Juvenile Alligators from Seven Florida Lakes. Gen. Comp. Endocrinol. 1999, 116 (3), 356-372. 69. Guillette Jr., L. J.; Crain, D. A.; Rooney, A. A.; Woodward, A. R., Effect of Acute Stress on Plasma Concentrations of Sex and Stress Hormones in Juvenile Alligators Living in Control and Contaminated Lakes. J. Herpetol. 1997, 31 (3), 347-353. 70. Guillette Jr., L. J.; Brock, J. W.; Rooney, A. A.; Woodward, A. R., Serum Concentrations of Various Environmental Contaminants and Their Relationship to Sex Steroid Concentrations and Phallus Size in Juvenile American Alligators. Arch. Environ. Contam. Toxicol. 1999, 36 (4), 447-455. 71. Gunderson, M. P.; LeBlanc, G. A.; Guillette Jr L. J., Alterations in Sexually Dimophic Biotransformation of Test osterone in Juvenile American Alligators ( Alligator mississippiensis ) from Contaminated Lakes. Environ. Health Perspect. 2001, 109 (12), 1257. 72. Vonier, P. M.; Crain, D. A.; McLachlan, J. A.; Guillette, L. J., Jr.; Arnold, S. F., Interaction of Environmental Chemicals w ith the Estrogen and Progesterone Receptors from the Oviduct of the American Alligator. Environ. Health Perspect. 1996, 104 (12), 1318-1322. 73. Guillette Jr., L. J.; Pickford, D. B.; Crain, D. A.; Rooney, A. A.; Percival, H. F., Reduction in Penis Size and Plasma Testoster one Concentrations in Juvenile Alligators Living in a Contaminated Environment. Gen. Comp. Endocrinol. 1996, 101 32-42. 74. Guillette Jr., L. J.; Gross, T. S.; Masson, G. R.; Matter, J. M.; Percival, H. F.; Woodward, A. R., Developmental Abnormalities of the Gonad and Abnormal Sex Hormone Concentrations in Juvenile Alligators from Contaminated and Control Lakes in Florida. Environ. Health Perspect. 1994, 102 (2), 680-688. 75. Gunderson, M. P.; Bermudez, D. S.; Bryan, T. A.; Degala, S.; Edwards, T. M.; Kools, S. A. E.; Milnes, M. R.; Woodward, A. R.; Guillett e, L. J., Variation in sex steroids and
185 phallus size in juvenile American alligators ( Alligator mississippiensis ) collected from 3 sites within the Kissimmee-Everglad es drainage in Florida (USA). Chemosphere 2004, 56 (4), 335-345. 76. Woodward, A. R.; Percival, H. F.; Jennings, M. L.; Moore, C. T., Low clutch viability of American alligators in Lake Apopka. Florida Scientist 1993, 56 (1), 52-62. 77. Lind, P. M.; Milnes, M. R.; Lundberg, R.; Bermud ez, D. S.; Orberg, J.; Guillette Jr, L. J., Abnormal Bone Composition in Female Juven ile American Alligators from a PesticidePolluted Lake (Lake Apopka, Florida). Environ. Health Perspect. 2004, 112 (3), 359-362. 78. Rooney, A. A.; Bermudez, D. S.; Louis J. Guil lette, J., Altered histology of the thymus and spleen in contaminant-exposed juvenile American alligators. J. Morphol. 2003, 256 (3), 349-359. 79. Sonnenschein, C.; Soto, A. M.; Fernandez, M. F.; Olea, N.; Olea-S errano, M. F.; RuizLopez, M. D., Development of a marker of estrogenic exposure in human serum. Clin. Chem. 1995, 41 (12), 1888-1895. 80. Weber, R. F. A.; Pierik, F. H.; Dohle, G. R.; Burdorf, A., Environmental influences on male reproduction. BJU International 2002, 89 (2), 143-148. 81. Keith, L. H., Environmental endocrine disruptors. Pure Appl. Chem. 1998, 70 (12), 23192326. 82. Barton, H. A.; Andersen, M. E., Endocrine Active Compounds: From Biology to Dose Response Assessment. Crit. Rev. Toxicol. 1998, 28 (4), 363-423. 83. Haupt, H. A.; Rovere, G. D., Anabolic steroids: A review of the literature. Am. J. Sports Med. 1984, 12 (6), 469-484. 84. Kutscher, E. C.; Lund, B. C.; Perry, P. J., An abolic Steroids: A Re view for Clinicians. Sports Med. 2002, 32 (5), 285-296. 85. Yu, N. H.; Ho, E. N. M.; Leung, D. K. K.; Wan, T. S. M., Screening of anabolic steroids in horse urine by liquid chromatography-tandem mass spectrometry. J. Pharm. Biomed. Anal. 2005, 37 (5), 1031-1038. 86. Katzung, B. G., Basic and Clinical Pharmacology 6th ed.; Appleton and Lange: Norwalk, CT, 1995. 87. Fenton, S. E., Endocrine-Disrupting Co mpounds and Mammary Gland Development: Early Exposure and Later Life Consequences. Endocrinology 2006, 147 (6), s18-24. 88. Guillette, E. A.; Conard, C.; Lares, F.; Aguilar, M. G.; McLachlan, J.; Guillette Jr, L. J., Altered Breast Development in Young Girls from an Agricultural Environment. Environ. Health Perspect. 2006, 114 (3), 471-475.
186 89. Olea, N.; Pulgar, R.; Perez, P.; Olea-Serrano, F. ; Rivas, A.; Novillo-Fer trell, A.; Pedraza, V.; Soto, A. M.; Sonnenschein, C., Estroge nicity of Resin-Based Composites and Sealants Used in Dentistry. Environ. Health Perspect. 1996, 104 (3), 298-305. 90. Sathyanarayana, S.; Karr, C. J.; Lozano, P.; Brown, E.; Calafat, A. M.; Liu, F.; Swan, S. H., Baby Care Products:Possible Sources of Infant Phthalate Exposure. Pediatrics 2008, 121 (2), e260-268. 91. Sharpe, R. M.; Irvine, D. S., How strong is the evidence of a link between environmental chemicals and adverse effect s on human reproductive health? Brit. Med. J. 2004, 328 447-451. 92. Myers, P.; Hessler, W., Does the dose make the poison? Environmental Health News 2007 1-6. 93. Breithaupt, H., A cause without a disease. EMBO Rep. 2004, 5 (1), 16-18. 94. vom Saal, F. S.; Welshons, W. V., NIH Pa nel Confirms that Endocrine Disrupting Chemicals Cause Effects at very Low Doses. Risk Policy Report 2000, 7 47-50. 95. Safe, S., Endocrine disruptors and human health: is there a problem. Toxicology 2004, 205 (1-2), 3-10. 96. Rapp, D. A., Our Toxic World: A Wake Up Call Environmental Medical Research Foundation: Buffalo, NY, 2004. 97. Soto, A. M.; Sonnenschein, C.; Chung, K. L.; Fe rnandez, M. F.; Olea, N.; Serrano, F. O., The E-SCREEN Assay as a Tool to Iden tify Estrogens: An Update on Estrogenic Environmental Pollutants. Environ. Health Perspect. 1995, 103 113-122. 98. Campbell, C. G.; Borglin, S. E.; Green, F. B. ; Grayson, A.; Wozei, E.; Stringfellow, W. T., Biologically directed environmental m onitoring, fate, and transport of estrogenic endocrine disrupting compounds in water: A review. Chemosphere 2006, 65 (8), 1265-1280. 99. Petrovic, M.; Eljarrat, E.; Lopez de Alda, M. J.; Ba rcelo, D., Endocrine disrupting compounds and other emerrging contaminants in the environment : A survey on new monitoring strategies and occurrence data. Anal. Bioanal. Chem. 2004, 378 549-562. 100. Petrovic, M.; Barcelo, D., Determination of Anionic and Nonionic Surfactants, Their Degradation Products, and Endocrine-Dis rupting Compounds in Sewage Sludge by Liquid Chromatography/Mass Spectrometry. Anal. Chem. 2000, 72 (19), 4560-4567. 101. Brossa, L.; Marc, R. M.; Borrull, F.; Pocuru ll, E., Determination of endocrine-disrupting compounds in water samples by on-line solid -phase extraction-programmed-temperature vaporisation-gas chromatography-mass spectrometry. J. Chromatogr., A 2003, 998 (1-2), 41-50.
187 102. Chusaksri, S.; Sutthivaiyakit, S.; Sutthiva iyakit, P., Confirmato ry determination of organochlorine pesticides in surface waters using LC/APCI/tandem mass spectrometryÂ‹Â„. Anal. Bioanal. Chem. 2006, 384 (5), 1236-1245. 103. Fernandez, M. P.; Ikonomou, M. G.; Buchana n, I., An assessment of estrogenic organic contaminants in Canadian wastewaters. Sci. Total Environ. 2007, 373 250-269. 104. Carpinteiro, J.; Quintana, J. B.; Rodrguez, I.; Carro, A. M.; Lorenzo, R. A.; Cela, R., Applicability of solid-phase microextract ion followed by on-fiber silylation for the determination of estrogens in water sa mples by gas chromatography-tandem mass spectrometry. J. Chromatogr., A 2004, 1056 (1-2), 179-185. 105. Koh, Y. K. K.; Chiu, T. Y.; Boobis, A.; Cart mell, E.; Lester, J. N.; Scrimshaw, M. D., Determination of steroid estrogens in wastewater by high performance liquid chromatography-tandem mass spectrometry. J. Chromatogr., A 2007, 1173 (1-2), 81-87. 106. Jeannot, R.; Sabik, H.; Sauvard, E.; Dagnac, T.; Dohrendorf, K., Determination of endocrine-disrupting compounds in envir onmental samples using gas and liquid chromatography with mass spectrometry. J. Chromatogr., A 2002, 974 (1-2), 143-159. 107. Alda, M.; Barcel, D., Review of analytical methods for the determination of estrogens and progestogens in waste waters. Fresenius. J. Anal. Chem. 2001, 371 (4), 437-447. 108. Braun, P.; Moeder, M.; Schrader, S.; Popp, P.; Kuschk, P.; Engewald, W., Trace analysis of technical nonylphenol bisphenol A and 17 -ethinylestradiol in wastewater using solid-phase microextraction and ga s chromatography-mass spectrometry. J. Chromatogr., A 2003, 988 (1), 41-51. 109. Xiao, X.-Y.; McCalley, D. V.; McEvoy, J., An alysis of estrogens in river water and effluents using solid-phase extraction and gas chromatography-negative chemical ionisation mass spectrometry of th e pentafluorobenzoyl derivatives. J. Chromatogr., A 2001, 923 (1-2), 195-204. 110. Lee, H.-B.; Peart, T. E.; Svoboda, M. L., De termination of endocri ne-disrupting phenols, acidic pharmaceuticals, and personal-care produ cts in sewage by solid-phase extraction and gas chromatography-mass spectrometry. J. Chromatogr., A 2005, 1094 (1-2), 122129. 111. Cspedes, R.; Lacorte, S.; Ralda, D.; Ginebr eda, A.; Barcel, D.; Pia, B., Distribution of endocrine disruptors in the Llobregat River basin (Catalonia, NE Spain). Chemosphere 2005, 61 (11), 1710-1719. 112. Rodriguez-Mozaz, S.; Lopez de Alda, M. J.; Barcelo, D., Monitoring of estrogens, pesticides, and bisphenol A in natural waters and drinking water treatment plans by solidphase extraction liquid chromatography-mass spectrometry. J. Chromatogr., A 2004, 1045 85-92.
188 113. Heisterkamp, I.; Gandrass, J. ; Ruck, W., Bioassay-directed chemical analysis utilizing LC/MS: a tool for identifying estr ogenic compounds in water samples? Anal. Bioanal. Chem. 2004, 378 (3), 709-715. 114. Gatidou, G.; Thomaidis, N. S.; Stasinakis A. S.; Lekkas, T. D., Simultaneous determination of the endocrine di srupting compounds nonylphenol, nonylphenol ethoxylates, triclosan and bi sphenol A in wastewater and sewage sludge by gas chromatography-mass spectrometry. J. Chromatogr., A 2007, 1138 (1-2), 32-41. 115. Esperanza, M.; Suidan, M. T.; Nishimura, F. ; Wang, Z. M.; Sorial, G. A.; Zaffiro, A.; McCauley, P.; Brenner, R.; Sayles, G., De termination of Sex Hormones and Nonylphenol Ethoxylates in the Aqueous Matrixes of Two Pilot-Scale Municipal Wastewater Treatment Plants. Environ. Sci. Technol. 2004, 38 (11), 3028-3035. 116. Trenholm, R. A.; Vanderford, B. J.; Holady, J. C.; Rexing, D. J.; Snyder, S. A., Broad range analysis of endocrine disruptors and pharmaceuticals using gas chromatography and liquid chromatography tandem mass spectrometry. Chemosphere 2006, 65 (11), 1990-1998. 117. Pojana, G.; Gomiero, A.; Jonkers, N.; Marcom ini, A., Natural and Synthetic endocrine disrupting compounds (EDCs) in water, sediment, and biota of a coastal lagoon. Environ. Int. 2007, 33 929-936. 118. Ballesteros, O.; Zafra, A.; Navaln, A.; Vlc hez, J. L., Sensitive gas chromatographicmass spectrometric method for the determina tion of phthalate esters, alkylphenols, bisphenol A and their chlorinated de rivatives in wastewater samples. J. Chromatogr., A 2006, 1121 (2), 154-162. 119. Hernando, M. D.; Mezcua, M.; Gmez, M. J.; Malato, O.; Agera, A.; Fernndez-Alba, A. R., Comparative study of analytical methods involving gas chromatography-mass spectrometry after derivatiza tion and gas chromatography-t andem mass spectrometry for the determination of selected endocrine disrupting compounds in wastewaters. J. Chromatogr., A 2004, 1047 (1), 129-135. 120. Ternes, T. A.; Andersen, H.; Gilberg, D.; B onerz, M., Determination of Estrogens in Sludge and Sediments by Liqui d Extraction and GC/MS/MS. Anal. Chem. 2002, 74 (14), 3498-3504. 121. Peng, X.; Wang, Z.; Yang, C.; Chen, F.; Mai, B., Simultaneous determination of endocrine-disrupting phenols and steroid estr ogens in sediment by gas chromatographymass spectrometry. J. Chromatogr., A 2006, 1116 (1-2), 51-56. 122. Tan, B. L. L.; Hawker, D. W.; Mller, J. F.; Leusch, F. D. L.; Tremblay, L. A.; Chapman, H. F., Comprehensive study of endocrine di srupting compounds using grab and passive sampling at selected wastewater treatment pl ants in South East Queensland, Australia. Environ. Int. 2007, 33 (5), 654-669.
189 123. Kuch, H. M.; Ballschmiter, K., Determinati on of endogenous and exogenous estrogens in effluents from sewage treatment plants at the ng/L-level. Fresenius. J. Anal. Chem. 2000, 366 (4), 392-395. 124. Liu, R.; Zhou, J. L.; Wilding, A., Simultaneous determination of endocrine disrutping phenolic compounds and steroids in water by solid-phase extraction-gas chromatographymass spectrometry. J. Chromatogr., A 2004, 1022 179-189. 125. Zheng, W.; Yates, S. R.; Bradford, S. A., An alysis of steroid hormones in typical dairy waste disposal system. Environ. Sci. Technol. 2008, 42 530-535. 126. Kolodziej, E. P.; Sedlak, D. L., Rangeland Grazing as a Source of Steroid Hormones to Surface Waters. Environ. Sci. Technol. 2007, 41 (10), 3514-3520. 127. Richardson, S. D., Water Analysis: Emer ging Contaminants and Current Issues. Anal. Chem. 2007, 79 (12), 4295-4324. 128. Amalric, L.; Henry, B. t.; Berrehouc, A., Determination of chlordecone in soils by GC/MS. Int. J. Environ. Anal. Chem. 2006, 86 (1/2), 15-24. 129. Rudel, R. A.; Camann, D. E.; Spengler, J. D.; Korn, L. R.; Brody, J. G., Phthalates, Alkylphenols, Pesticides, Polybrominated Diphenyl Ethers, and Other EndocrineDisrupting Compounds in Indoor Air and Dust. Environ. Sci. Technol. 2003, 37 (20), 4543-4553. 130. Soleas, G. J.; Yan, J.; Hom, K.; Goldberg, D. M., Multiresidue analysis of seventeen pesticides in wine by gas chromat ography with mass-selective detection. J. Chromatogr., A 2000, 882 (1-2), 205-212. 131. Flamini, R.; Panighel, A., Mass Spectrometry in Grape and Wine Chemistry. Part II: The Consumer Protection. Mass Spectrom. Rev. 2006, 25 741-774. 132. Fernandez, M.; Pico, Y.; Manes, J., Co mparison of Gas and Liquid Chromatography Coupled to Mass Spectrometry for the Residue Analysis of Pesticides in Oranges. Chromatographia 2001, 54 302-308. 133. Carabias-Martnez, R.; Rodrguez-Gonzalo, E.; Revilla-Ruiz, P., Determination of endocrine-disrupting compounds in cereals by pressurized liquid extraction and liquid chromatography-mass spectrometry: Study of background contamination. J. Chromatogr., A 2006, 1137 (2), 207-215. 134. Guenther, K.; Heinke, V.; Thiele, B.; Kleist E.; Prast, H.; Raecker, T., Endocrine Disrupting Nonylphenols Are Ubiquitous in Food. Environ. Sci. Technol. 2002, 36 (8), 1676-1680. 135. Drewes, J. E.; Hemming, J.; Ladenburger, S. J.; Schauer, J.; Sonzogni, W., An Assessment of Endocrine Disrupting Activit y Changes during Wastewater Treatment
190 through the use of Bioassays and Chemical Measurements. Water Environ. Res. 2005, 77 (1), 12-23. 136. Wan, Y.; Hu, J.; Liu, J.; An, W.; Tao, S.; Jia, Z., Fate of DDT-related compounds in Bohai Bay and its adjacent Haihe Basin, North China. Mar. Pollut. Bull. 2005, 50 (4), 439-445. 137. Brossa, L.; Marc, R. M.; Borrull, F.; Pocu rull, E., Application of on-line solid-phase extraction-gas chromatography-mass spectrome try to the determination of endocrine disruptors in water samples. J. Chromatogr., A 2002, 963 (1-2), 287-294. 138. Nakamura, S.; Daishima, S., Simultaneous dete rmination of 64 pesticides in river water by stir bar sorptive extraction and ther mal desorption-gas chromatography-mass spectrometry. Anal. Bioanal. Chem. 2005, 382 (1), 99-107. 139. Petrovic, M.; Eljarrat, E.; Lop ez de Alda, M. J.; Barcelo, D ., Recent advances in the mass spectrometric analysis related to en docrine disrupting compounds in aquatic environmental samples. J. Chromatogr., A 2002, 974 (1-2), 23-51. 140. Gomez, M. J.; Mezcua, M.; Martinez, M. J.; Fernandez-Alba, A. R.; Aguera, A., A new method for monitoring oestrogens, N -octyl phenol, and bisphenol A in wastewater treatment plants by solid-phase extr action-gas chromatography/tandem mass spectrometry. Int. J. Environ. Anal. Chem. 2006, 86 (1/2), 3-13. 141. Brossa, L.; Pocurull, E.; Borrull, F.; Marce, R. M., Solid-Phase Extraction/High Performance Liquid Chromatography-Electro spray Mass Spectrometry to Determine Endocrine Disruptors in Water Samples. Chromatographia 2004, 59 419-423. 142. Di Corcia, A.; Crescenzi, C.; Guerriero, E.; Samperi, R., Ultratrace Determination of Atrazine and Its Six Major Degradation Produc ts in Water by Solid-Phase Extraction and Liquid Chromatography-Electro spray/Mass Spectrometry. Environ. Sci. Technol. 1997, 31 (6), 1658-1663. 143. Lopez-Roldan, P.; de Alda, M. J. L.; Barcel o, D., Simultaneous determination of selected endocrine disrupters (pesticides, phenols and pht halates) in water by in-field solid-phase extraction (SPE) using the prototype PR OFEXS followed by on-line SPE (PROSPEKT) and analysis by liquid. Anal. Bioanal. Chem. 2004, 378 (3), 599-609. 144. Haber, E.; Muoz-Guerra, J. A.; Soriano, C.; Ca rreras, D.; Rodriguez, C.; Rodriguez, F. A., Automated sample preparation and gas chromatographic-mass spectrometric analysis of urinary androgenic anabolic steroids. J. Chromatogr. B 2001, 755 (1-2), 17-26. 145. MacNeil, J. S., GC/MS Drug Testers Face Olympian Challange. Today's Chemist at Work 2001, 10 (6), 28-34. 146. Thevis, M.; Schanzer, W., Mass Spectrometry in Sports and Drug Testing: Structure Characterization and Analytical Assays. Mass Spectrom. Rev. 2007, 26 79-107.
191 147. Pozo, O. J.; Van Eenoo, P.; Deventer, K.; Delbeke, F. T., Development and validation of a qualitative screening method for the detecti on of exogenous anabolic steroids in urine by liquid chromatography-t andem mass spectrometry. Anal. Bioanal. Chem. 2007, 389 (4), 1209-1224. 148. Council of the European Communities, Offici al Journal of the European Communities, Luxemborg, 1996; Vol. Council Directive EC/96/22. 149. Wu, F. C. W., Endocrine asp ects of anabolic steroids. Clin. Chem. 1997, 43 (7), 12891292. 150. Deventer, K.; Eenoo, P. V.; Delbeke, F. T., Screening for anabolic steroids in doping analysis by liquid chromatography/el ectrospray ion trap mass spectrometry. Biomed. Chromatogr. 2006, 20 (5), 429-433. 151. Rossi, S.-A.; Johnson, J. V.; Yost, R. A., Optimization of short-column gas chromatography/electron ionization mass spect rometry conditions for the determination of underivatized anabolic steroids. Biol. Mass Spectrom. 1992, 21 (9), 420-430. 152. Marques, M. A. S.; Pereira, H. M. G.; A quino Neto, F. R. d., Improvements in steroid screening in doping control with special em phasis to GC-MS analytical conditions and method validation. J. Braz. Chem. Soc. 2006, 17 382-392. 153. Gambelunghe, C.; Sommavilla, M.; Ferranti, C. ; Rossi, R.; Aroni, K.; Manes, N.; Bacci, M., Analysis of anabolic st eroids in hair by GC/MS/MS. Biomed. Chromatogr. 2007, 21 (4), 369-375. 154. Wolthers, B. G.; Kraan, G. P. B., Clinical applications of gas chromatography and gas chromatography-mass spectrometry of steroids. J. Chromatogr., A 1999, 843 (1-2), 247274. 155. Aguilera, R.; Becchi, M.; Casabianca, H.; Hatton, C. K.; Catlin, D. H.; Starcevic, B.; Harrison G. Pope, J., Improved method of detection of testosterone abuse by gas chromatography/combustion/isotope ratio mass sp ectrometry analysis of urinary steroids. J. Mass Spectrom. 1996, 31 (2), 169-176. 156. Houghton, E.; Teals, P.; Dumasia, M. C., Improved Capillary Gas-chromatographic Mass Spectromtric Method for the Determination of Anabolic Steroid and Corticosteroid Metabolites in Horse Urine Using On-colum n Injection with High-boiling Solvents. Analyst (Cambridge, U. K.) 1984, 109 273-275. 157. Bonnaire, Y.; Dehennin, L.; Plou, P.; Toutai n, P. L., Testosterone Administration to Mares: Criteria for Detecti on of Testosterone Abuse by Analysis of Metabolites in Plasma and Urine. J. Anal. Toxicol. 1995, 19 175-181. 158. Anielski, P.; Thieme, D.; Schlupp, A.; Grosse, J. ; Ellendorff, F.; Mueller, R., Detection of testosterone, nandrolone and precursors in horse hair. Anal. Bioanal. Chem. 2005, 383 (6), 903-908.
192 159. Impens, S.; Van Loco, J.; Degroodt, J. M.; De Brabander, H., A downscaled multiresidue strategy for detection of anabo lic steroids in bovi ne urine using gas chromatography tandem mass spectrometry (GC-MS3). Anal. Chim. Acta 2007, 586 (12), 43-48. 160. Dubois, M.; Taillieu, X.; Colemonts, Y.; Lans ival, B.; De Graeve, J.; Delahaut, P., GCMS determination of anabolic steroids after multi-immunoaffinity purification. Analyst (Cambridge, U. K.) 1998, 123 2611-2616. 161. Hartmann, S.; Steinhart, H., Simultaneous determ ination of anabolic and catabolic steroid hormones in meat by gas chromatography-mass spectrometry. J. Chromatogr. B 1997, 704 (1-2), 105-117. 162. Seo, J.; Kim, H.-Y.; Chung, B. C.; Hong, J., Simultaneous determination of anabolic steroids and synthetic hormones in meat by freezing-lipid f iltration, solid-phase extraction and gas chromatography-mass spectrometry. J. Chromatogr., A 2005, 1067 (12), 303-309. 163. Kootstra, P. R.; Zoontjes, P. W.; van Tricht, E. F.; Sterk, S. S., Multi -residue screening of a minimum package of anabolic steroids in urine with GC-MS. Anal. Chim. Acta 2007, 586 (1-2), 82-92. 164. Vyncht, G. V.; Gaspar, P.; Depauw, E.; Ma ghuin-Rogister, G., Mu lti-residue screening and confirmatory analysis of anabolic ster oids in urine by gas chromatography coupled with tandem mass spectrometry. J. Chromatogr., A 1994, 683 67-74. 165. Rambaud, L.; Monteau, F.; Deceuninck, Y.; Bichon, E.; Andre, F.; Le Bizec, B., Development and validation of a multi-residue method for the detection of a wide range of hormonal anabolic compounds in hair using gas chromatography-tandem mass spectrometry. Anal. Chim. Acta 2007, 586 (1-2), 93-104. 166. Mazzarino, M.; Botr, F., A fast liquid chro matographic/mass spectrometric screening method for the simultaneous detection of synthe tic glucocorticoids, some stimulants, antioestrogen drugs and synthe tic anabolic steroids. Rapid Commun. Mass Spectrom. 2006, 20 (22), 3465-3476. 167. Thevis, M.; Bommerich, U.; Opfermann, G. ; Schnzer, W., Characterization of chemically modified steroids for doping control purposes by elec trospray ionization tandem mass spectrometry. J. Mass Spectrom. 2005, 40 (4), 494-502. 168. Ho, E. N. M.; Leung, D. K. K.; Wan, T. S. M.; Yu, N. H., Comprehensive screening of anabolic steroids, corticosteroids, and acidic drugs in horse urine by solid-phase extraction and liquid chroma tography-mass spectrometry J. Chromatogr., A 2006, 1120 38-53. 169. van Ginkel, L. A.; Jansen, E. H. J. M.; St ephany, R. W.; Zoontjes, P. W.; Schwillens, P. L. W. J.; van Rossum, H. J.; Visser, T ., Liquid chromatographic purification and detection of anabolic compounds. J. Chromatogr., A 1992, 624 (1-2), 389-401.
193 170. Muiz-Valencia, R.; Gonzalo-Lumbreras, R. ; Santos-Montes, A.; Izquierdo-Hornillos, R., Liquid chromatographic method development for anabolic androgenic steroids using a monolithic column: Application to animal feed samples. Anal. Chim. Acta 2008, 611 (1), 103-112. 171. Poucke, C. V.; Velde, M. V. D.; Pe teghem, C. V., Combination of liquid chromatography/tandem mass spectrometry a nd gas chromatography/mass spectrometry for the detection of 21 anabolic steroid residues in bovine urine. J. Mass Spectrom. 2005, 40 (6), 731-738. 172. Dettmer, K.; Aronov, P. A.; Hammock, B. D., Mass spectrometry-based metabolomics. Mass Spectrom. Rev. 2007, 26 (1), 51-78. 173. Sawaya, W. N.; Lone, K. P.; Husain, A.; Dashti, B.; Al-Zenki, S., Screening for estrogenic steroids in sh eep and chicken by the application of enzyme-linked immunosorbent assay and a comparison w ith analysis by gas chromatography-mass spectrometry. Food Chem. 1998, 63 (4), 563-569. 174. Isobe, N.; Toshihiko, N., Direct enzyme immunoassay of pr ogesterone in bovine plasma. Anim. Sci. J. 2003, 74 369-373. 175. Fortune, J. E.; Tsang, P. C., Production of androgen and estradiol-17 by Xenopus ovaries treated with gonadotropins in vitro. Gen. Comp. Endocrinol. 1981, 43 (2), 234242. 176. Draisci, R.; Volpe, G.; Compagnone, D.; Puri ficato, I.; delli Quadri, F.; Palleschi, G., Development of an electrochemical ELISA for the screening of 17 -estradiol and application to bovine serum. Analyst (Cambridge, U. K.) 2000, 125 1419-1423. 177. Monfort, S. L.; Wasser, S. K.; Mashburn, K. L.; Burke, M.; Brewer, B. A.; Creel, S. R., Steroid metabolism and validation of noninvasi ve endocrine monitoring in the African wild dog ( Lycaon pictus ). Zoo Biol. 1997, 16 (6), 533-548. 178. Stanczyk, F. Z.; Cho, M. M.; Endres, D. B.; Morrison, J. L.; Patel, S.; Paulson, R. J., Limitations of direct estradiol and testosterone immunoassay kits. Steroids 2003, 68 (14), 1173-1178. 179. Janzen, N.; Peter, M.; Sander, S.; Steuerwal d, U.; Terhardt, M.; Holtkamp, U.; Sander, J., Newborn Screening for Congenital Adrenal Hype rplasia: Additional Steroid Profile using Liquid Chromatography-Tandem Mass Spectrometry. J. Clin. Endocrinol. Metab. 2007, 92 (7), 2581-2589. 180. Munro, C. J.; Stabenfeldt, G. H.; Cragun, J. R. ; Addiego, L. A.; Overstreet, J. W.; Lasley, B. L., Relationship of serum estradiol and pr ogesterone concentrati ons to the excretion profiles of their major urinary metabolites as measured by enzyme immunoassay and radioimmunoassay. Clin. Chem. 1991, 37 (6), 838-844.
194 181. Pratt, J. J., Steroid immunoa ssay in clinical chemistry. Clin. Chem. 1978, 24 (11), 18691890. 182. Nozaki, O., Steroid analysis for medical diagnosis. J. Chromatogr., A 2001, 935 (1-2), 267-278. 183. Selby, C., Interference in immunoassay. Ann. Clin. Biochem. 1999, 36 704-721. 184. Ichimura, K.; Yamanaka, H.; Chiba, K.; Shinozuka, T.; Shiki, Y.; Saito, K.; Kusano, S.; Ameniya, S.; Oyama, K.; Nozaki, Y.; Kat o, K., Simultaneous quantitative measurement of fourteen adrenal steroids by capi llary column gas chromatography---mass spectrometry, and its clinical application. J. Chromatogr. B 1986, 374 5-16. 185. Guo, T.; Chan, M.; Soldin, S. J., Steroid Profiles using Liquid Chromatography-Tandem Mass Spectometry with Atmospheric Pressure Photoionization Source. Arch. Path. Lab. Med. 2004, 128 469-475. 186. Taieb, J.; Benattar, C.; Birr, A. S.; Lindenbaum A., Limitations of Steroid Determination by Direct Immunoassay. Clin. Chem. 2002, 48 (3), 583-585. 187. Wudy, S. A.; Hartmann, M. F., Gas Chroma tography-Mass Spectrometry Profiling of Steroids in Times of Molecular Biology. Horm. Metab. Res. 2004, (6), 415-422. 188. Wudy, S. A.; Homoki, J.; Teller, W. M., Current Practice of Gas Chromatography-Mass Spectrometry Marcel Dekker: New Yor k, NY, 2001; Vol. 86, p 528. 189. Horning, E. C.; Horning, M. G., Metabolic Pr ofiles: Gas-Phase Methods for Analysis of Metabolites. Clin. Chem. 1971, 17 (8), 802-809. 190. Adatia, R.; Cooke, M., An Improved Method for Steroid Profile Analys is Using Capillary Gas Chromatography. Chromatographia 1988, 25 (7), 598-602. 191. Bekker, L. C.; Bester, M. J.; Hurter, P., Re ference steroid profiles in 0 to 3 day old neonates. J. Microcolumn Sep. 2001, 13 (6), 250-254. 192. Trocha, P.; D'Amato, N. A., Method fo r screening urinary steroids by gas chromatography. Clin. Chem. 1978, 24 (2), 193-199. 193. Halket, J. M.; Waterman, D.; Przyborowska, A. M.; Patel, R. K. P.; Fraser, P. D.; Bramley, P. M., Chemical derivatization and mass spectral libraries in metabolic profiling by GC/MS and LC/MS/MS. J. Exp. Bot. 2005, 56 (410), 219-243. 194. Thurman, E. M.; Ferrer, I.; Barcelo, D., Choosing between Atmospheric Pressure Chemical Ionization and Electrospray Ionizati on Interfaces for the HPLC/MS Analysis of Pesticides. Anal. Chem. 2001, 73 (22), 5441-5449. 195. Honour, J. W., Urinary steroid profile analysis. Clin. Chim. Acta 2001, 313 (1-2), 45-50.
195 196. Deng, C.; Ji, J.; Zhang, L.; Zhang, X., Diagnos is of congenital ad renal hyperplasia by rapid determination of 17a-hydroxyprogest erone in dried blood spots by gas chromatography/mass spectrometry following microwave-assisted silylation. Rapid Commun. Mass Spectrom. 2005, 19 2974-2978. 197. Ueshiba, H.; Segawa, M.; Hayashi, T.; Miyach i, Y.; Irie, M., Seru m profiles of steroid hormones in patients with Cushing's syndr ome determined by a new HPLC/RIA method. Clin. Chem. 1991, 37 (8), 1329-1333. 198. Xiao, X.-y.; McCalley, D., Quantitative analysis of estrogens in human urine using gas chromatography/negative chemical ionisation mass spectrometry. Rapid Commun. Mass Spectrom. 2000, 14 (21), 1991-2001. 199. Legrand, C.; Dousset, B.; Tronel, H.; Bellevi lle, F.; Nabet, P., Measurement of plasma testosterone by gas chromatographynegative-ion mass spectrometry using pentafluoropropionic derivatives. J. Chromatogr. B 1995, 663 (2), 187-192. 200. Schmidt, N. A.; Borburgh, H. J.; Penders, T. J.; Weykamp, C. W., Steroid profiling--an update. Clin. Chem. 1985, 31 (4), 637-639. 201. Vicente, F. B.; Smith, F. A.; Sierra, R.; Wa ng, S., Measurement of serum testosterone using high-performance liquid chroma tography/tandem mass spectrometry. Clin. Chem. Lab. Med. 2006, 44 (1), 70-75. 202. Nelson, R. E.; Grebe, S. K.; O'Kane, D. J. ; Singh, R. J., Liquid Chromatography-Tandem Mass Spectrometry Assay for Simultaneous Me asurement of Estradiol and Estrone in Human Plasma. Clin. Chem. 2004, 50 (2), 373-384. 203. Shimada, K.; Mitamura, K.; Higashi, T ., Gas chromatography and high-performance liquid chromatography of natural steroids. J. Chromatogr., A 2001, 935 (1-2), 141-172. 204. Homma, K.; Hasegawa, T.; Masumoto, M.; Takeshita, E.; Watanabe, K.; Chiba, H.; Kurosawa, T.; Takahashi, T.; Matsuo, N., Re ference Values for Urinary Steroids in Japanese Newborn Infants: Gas Chromat ography/Mass Spectrometry in Selected Ion Monitoring. Endocr. J. 2003, 50 (6), 783-792. 205. Nassar, A. E. F.; Varshney, N.; Getek, T.; Cheng, L., Quantitative Analysis of Hydrocortisone in Human Urine Using a High-Performance Liquid ChromatographicTandem Mass Spectrometric-Atmospheric-Pr essure Chemical Ionization Method. J. Chromatogr. Sci. 2001, 39 59-64. 206. Horning, E. C.; Brooks, C. J. W.; Johnson, L.; Gardiner, W. L., Separation, Identification, and Estimation of Human Steroid Hormones and Their Metabolites: Applications to Adrenocortical Steroids. Sep. Sci. Technol. 1966, 1 (5), 555 573. 207. Hmlinen, E.; Fotsis, T.; Adlercreutz, H., A gas chromatographic method for the determination of neutral steroid profiles in urine, including studi es on the effect of
196 oxytetracycline administration on these profiles in men. Clin. Chim. Acta 1991, 199 (2), 205-220. 208. Rule, G.; Henion, J., High-throughput sample preparation and analysis using 96-well membrane solid-phase extraction and liqui d chromatography-tandem mass spectrometry for the determination of steroids in human urine. J. Am. Soc. Mass Spectrom. 1999, 10 (12), 1322-1327. 209. Daz-Cruz, M. S.; Alda, M. J. L. d.; Lpez, R.; Barcel, D ., Determination of estrogens and progestogens by mass spectrometric tec hniques (GC/MS, LC/MS and LC/MS/MS). J. Mass Spectrom. 2003, 38 (9), 917-923. 210. Almeida, C.; Serodio, P.; Florencio, M. H.; Nogue ira, J. M. F., New strategies to screen for endocrine-disrupting chemicals in the Po rtuguese marine environment utilizing large volume injection-capillary gas chromatography-mass spectrometry combined with retention time locking lib raries (LVI-GC-MS-RTL). Anal. Bioanal. Chem. 2007, 387 2569-2583. 211. Lopez de Alda, M. J.; Barcelo, D., Determin ation of steroid sex hormones and related synthetic compounds considered as endoc rine disrupters in water by liquid chromatography-diode array detection-mass spectrometry. J. Chromatogr., A 2000, 892 (1-2), 391-406. 212. Vulliet, E.; Baugros, J.-B.; Flament-Waton, M. -M.; Grenier-Loustalot, M.-F., Analytical methods for the determination of selected steroid sex hormones and corticosteriods in wastewater. Anal. Bioanal. Chem. 2007, 387 (6), 2143-2151. 213. Carabias-Martnez, R.; Rodrguez-Gonzalo, E.; Revilla-Ruiz, P., Determination of weakly acidic endocrine-disrupting compounds by liquid chromatography-mass spectrometry with post-column base addition. J. Chromatogr., A 2004, 1056 (1-2), 131138. 214. Stehmann, A.; Meesters, R. J. W.; Schroder, H. F., Mass spectrometric analytical methods for the determination of en docrine disrupting chemicals (EDCs). Water Sci. Technol. 2004, 50 (5), 165-171. 215. Ding, W.-H.; Chiang, C.-C., Derivatization proc edures for the detection of estrogenic chemicals by gas chromatography/mass spectrometry. Rapid Commun. Mass Spectrom. 2003, 17 (1), 56-63. 216. Gibson, R.; Becerril-Bravo, E.; Silva-Castro, V. ; Jimenez, B., Determination of acidic pharmaceuticals and potential endocrine disrup ting compounds in wastewaters and spring waters by selective elution and analysis by gas chromatography-mass spectrometry. J. Chromatogr., A 2007, 1169 31-39. 217. Famiglini, G.; Palma, P.; Siviero, A.; Rezai M. A.; Cappiello, A., Determination of Endocrine Disrupting Compounds in Marine Water by Nanoliquid
197 Chromatography/Direct-Electr on Ionization Mass Spectrometry. Anal. Chem. 2005, 77 (23), 7654-7661. 218. Weigel, S.; Kallenborn, R.; Hhnerfuss, H., Simultaneous solid-pha se extraction of acidic, neutral and basic pharmaceuticals from aqueous samples at ambient (neutral) pH and their determination by gas chromatography-mass spectrometry. J. Chromatogr., A 2004, 1023 (2), 183-195. 219. Xue, N.; Xu, X., Composition, Distributi on, and Characterization of Suspected Endocrine-Disrupting Pesticides in Beijing GuanTing Reservoir (GTR). Arch. Environ. Contam. Toxicol. 2006, 50 (4), 463-473. 220. Zhang, Z. L.; Hibberd, A.; Zhou, J. L., Optimisat ion of derivatisation for the analysis of estrogenic compounds in water by solid-pha se extraction gas chromatography-mass spectrometry. Anal. Chim. Acta 2006, 577 (1), 52-61. 221. Labadie, P.; Budzinski, H. l. n., Devel opment of an analy tical procedure for determination of selected estrogens and progestagens in water samples. Anal. Bioanal. Chem. 2005, 381 (6), 1199-1205. 222. Yang, L.; Luan, T.; Lan, C., Solid-phase mi croextraction with on-fiber silylation for simultaneous determinations of endocrine disrupting chemicals a nd steroid hormones by gas chromatography-mass spectrometry. J. Chromatogr., A 2006, 1104 (1-2), 23-32. 223. Almeida, C.; Nogueira, J. M. F., Determin ation of steroid sex hormones in water and urine matrices by stir bar sorptive extracti on and liquid chromatography with diode array detection. J. Pharm. Biomed. Anal. 2006, 41 (4), 1303-1311. 224. Xu, X.; Veenstra, T. D.; Fox, S. D.; Roman, J. M.; Issaq, H. J.; Falk, R.; Saavedra, J. E.; Keefer, L. K.; Ziegler, R. G., Measuri ng Fifteen Endogenous Estrogens Simultaneously in Human Urine by High-Performance Liqui d Chromatography-Mass Spectrometry. Anal. Chem. 2005, 77 (20), 6646-6654. 225. Shou, W. Z.; Jiang, X.; Naidong, W., Development and valida tion of a high-sensitivity liquid chromatography/tandem mass spectrome try (LC/MS/MS) method with chemical derivatization for the determination of ethinyl estradiol in human plasma. Biomed. Chromatogr. 2004, 18 (7), 414-421. 226. Hill, M.; Havlikova, H.; Vrbikova, J.; Kancheva R.; Kancheva, L.; Pouzar, V.; Cerny, I.; Starka, L., The identification and simu ltaneous quantificati on of 7-hydroxylated metabolites of pregnenolone, dehydroepiandrosterone, 3 17 -androstenediol, and testosterone in human serum using gas chromatography-mass spectrometry. J. Steroid Biochem. Mol. Biol. 2005, 96 187-200. 227. Chatman, K.; Hollenbeck, T.; Hagey, L.; Vallee, M.; Purdy, R.; Weiss, F.; Siuzdak, G., Nanoelectrospray Mass Spectrometry and Pr ecursor Ion Monitoring for Quantitative Steroid Analysis and Attomole Sensitivity. Anal. Chem. 1999, 71 (13), 2358-2363.
198 228. Kancheva, L.; Hill, M.; Vcelkov, H.; Vrb kov, J.; Peliknov, T.; Strka, L., The identification and simultaneous quantification of neuroactiv e androstane steroids and their polar conjugates in the serum of adult men, using gas chromatography-mass spectrometry. Steroids 2007, 72 (11-12), 792-801. 229. Budzinski, H.; Devier, M. H.; Labadie, P.; To gola, A., Analysis of hormonal steroids in fish plasma and bile by coupling solid-phase extraction to GC/MS. Anal. Bioanal. Chem. 2006, 386 (5), 1429-1439. 230. Gonzalo-Lumbreras, R.; Pimentel-Trapero, D.; Izqierdo-Hornillos, R., Solvent and solidphase extraction of natural and synthetic anabolic steroids in human urine. J. Chromatogr. B. 2001, 754 (2), 419-425. 231. Heikkinen, R.; Fotsis, T.; Adlercreutz, H., Re versed-phase C18 cartridge for extraction of estrogens from urine and plasma. Clin. Chem. 1981, 27 (7), 1186-1189. 232. Samuels, T. P.; Nedderman, A.; Seymour, A.; Houghton, E., Study of the metabolism of testosterone, nandrolone, a nd estradiol in cattle. Analyst (Cambridge, U. K.) 1998, 123 2401-2404. 233. Amendola, L.; Garribba, F.; Botre, F., De termination of endogenous and synthetic glucocorticoids in human urine by gas chromatography-mass spectrometry following microwave-assisted derivatization. Anal. Chim. Acta 2003, 489 (2), 233-243. 234. Clifton, V. L.; Bisits, A.; Zarzycki, P. K ., Characterization of human fetal cord blood steroid profiles in relation to fetal sex and mode of deliver y using temperature-dependent inclusion chromatography and princi pal component analysis (PCA). J. Chromatogr. B 2007, 855 (2), 249-254. 235. Newman, A. E. M.; Chin, E. H.; Schmidt, K. L.; Bond, L.; Wynne-Edwards, K. E.; Soma, K. K., Analysis of steroids in songbi rd plasma and brain by coupling solid phase extraction to radioimmunoassay. Gen. Comp. Endocrinol. 2008, 155 (3), 503-510. 236. Biancotto, G.; Angeletti, R.; Traldi, P.; Silvestri, M.; Sacc on, M.; Guidugli, F., Determination of 17 -estradiol in bovine plasma: development of a highly sensitive technique by ion trap gas chromatogra phy-tandem mass spectrometry using negative chemical ionization. J. Mass Spectrom. 2002, 37 (12), 1266-1271. 237. Dumasia, M. C., In vivo biotransformation of 17 -methyltestosterone in the horse revisited: identifica tion of 17-hydroxymethyl metabolites in equine urine by capillary gas chromatography/mass spectrometry. Rapid Commun. Mass Spectrom. 2003, 17 (4), 320329. 238. Akre, C.; Fedeniuk, R.; MacNeil, J. D., Valid ation of a simple, sensitive method for the determination of -estradiol in bovine urine usi ng gas-chromatography negative-ion chemical ionization mass spectrometry. Analyst (Cambridge, U. K.) 2004, 129 145-149.
199 239. Gibson, R.; Tyler, C. R.; Hill, E. M., Anal ytical methodology for the identification of estrogenic contaminants in fish bile. J. Chromatogr., A 2005, 1066 (1-2), 33-40. 240. Hjelmborg, P. S.; Ghisari, M.; Bonefeld-Jor gensen, E. C., SPE-HPLC purification of endocrine-disrupting compounds from human serum for assessment of xenoestrogenic activity. Anal. Bioanal. Chem. 2006, 385 (5), 875-887. 241. Snow, N. H., Solid-phase micro-extracti on of drugs from biological matrices. J. Chromatogr., A 2000, 885 (1-2), 445-455. 242. Okeyo, P. D.; Snow, N. H., Analysis of Es trogens and Anabolic Steroids by SPME with On-Fiber Derivatization and GC/MS. J. Microcolumn Sep. 1998, 10 (7), 551-556. 243. Bartle, K. D.; Myers, P., History of gas chromatography. Trends Anal. Chem. 2002, 21 (9-10), 547-557. 244. Ditz, R., Preparative Chromatography Wiley-VCH: Weinheim, Germany, 2005. 245. Poole, C. F., The Essence of Chromatography 1st ed.; Elsevier: New York, NY, 2003; p 925. 246. Watson, J. T., Introduction to Mass Spectrometry 3rd ed.; Lippincott-Raven: New York, NY, 1997; p 496. 247. Lavangnini, I.; Magno, F.; Seraglia, R.; Traldi, P., Quantitative Applications of Mass Spectrometry John Wiley and Sons Inc.: Hoboken, NJ, 2006; p 138. 248. Niessen, W. M. A., Liquid Chromatography-Mass Spectrometry 3rd ed.; CRC Press: Boca Raton, FL, 2006; p 608. 249. Lough, W. J.; Wainer, I. W., High performance liquid chromatography: Fundamental Principles and Practice CRC Press: Boca Raton, FL, 1995; p 276. 250. Abian, J., The coupling of gas and liqui d chromatography with mass spectrometry. J. Mass Spectrom. 1999, 34 (3), 157-168. 251. Tanaka, N.; Kobayashi, H.; Ishizuka, N.; Mi nakuchi, H.; Nakanishi, K.; Hosoya, K.; Ikegami, T., Monolithic silica columns for hi gh-efficiency chromatographic separations. J. Chromatogr., A 2002, 965 (1-2), 35-49. 252. Garcia, D. M.; S, K. H.; Stansbury, W. F ., Optimization of the Atmospheric Pressure Chemical Ionization Liquid Chromatography Mass Spectrometry Interface. J. Am. Soc. Mass Spectrom. 1996, 7 (1), 59-65. 253. March, R. E., An Introduction to Qu adrupole Ion Trap Mass Spectrometry. J. Mass Spectrom. 1997, 32 (4), 351-369.
200 254. Johnson, J. V.; Yost, R. A.; Kelley, P. E.; Br adford, D. C., Tandem-in-space and tandemin-time mass spectrometry: triple quadrupoles and quadrupole ion traps. Anal. Chem. 1990, 62 (20), 2162-2172. 255. Hennion, M.-C., Solid-phase extraction: met hod development, sorbents, and coupling with liquid chromatography. J. Chromatogr., A 1999, 856 (1-2), 3-54. 256. Theodoridis, G.; Koster, E. H. M.; de Jong, G. J., Solid-phase microextraction for the analysis of biological samples. J. Chromatogr. B. 2000, 745 (1), 49-82. 257. Santos, F. J.; Galceran, M. T., The applica tion of gas chromatography to environmental analysis. Trends Anal. Chem. 2002, 21 (9-10), 672-685. 258. Pichon, V., Solid-phase extraction for multiresid ue analysis of organic contaminants in water. J. Chromatogr., A 2000, 885 (1-2), 195-215. 259. Huck, C. W.; Bonn, G. K., Recent developments in polymer-based sorbents for solidphase extraction. J. Chromatogr., A 2000, 885 (1-2), 51-72. 260. Fritz, J. S., Analytical Solid-Phase Extraction Wiley-VCH: New York, NY, 1999; p 209. 261. Arthur, C. L.; Pawliszyn, J., Solid phase mi croextraction with th ermal desorption using fused silica optical fibers. Anal. Chem. 1990, 62 (19), 2145-2148. 262. Pawliszyn, J., Solid Phase Microextraction: Theory and Practice Wiley-VCH: New York, NY, 1997; p 247. 263. Pan, L.; Pawliszyn, J., Derivatization/Solid-P hase Microextraction: New Approach to Polar Analytes. Anal. Chem. 1997, 69 (2), 196-205. 264. Dietz, C.; Sanz, J.; Cmara, C., Recent de velopments in solid-phase microextraction coatings and related techniques. J. Chromatogr., A 2006, 1103 (2), 183-192. 265. Lord, H.; Pawliszyn, J., Evolution of solid-phase microextraction technology. J. Chromatogr., A 2000, 885 (1-2), 153-193. 266. Beltran, J.; Lpez, F. J.; Hernndez, F., Solid -phase microextraction in pesticide residue analysis. J. Chromatogr., A 2000, 885 (1-2), 389-404. 267. Queiroz, M. E. C.; Lancas, F. M., Practical tips on preparing plasma samples for drug analysis using SPME. LC-GC Eur. 2005, 18 (3), 145-154. 268. Ulrich, S., Solid-phase microextr action in biomedical analysis. J. Chromatogr., A 2000, 902 (1), 167-194. 269. Eisert, R.; Levsen, K., Solid-phase microext raction coupled to gas chromatography: A new method for the analysis of organics in water. J. Chromatogr., A 1996, 733 (1-2), 143-157.
201 270. Okeyo, P.; Rentz, S. M.; Snow, N. H., Anal ysis of Steroids fr om Human Serum by SPME with Headspace Derivatization and GC/MS. J. High Resolut. Chromatogr. 1997, 20 171173. 271. Pawliszyn, J.; Pedersen-Bjergaard, S., Analy tical Microextraction: Current Status and Future Trends. J. Chromatogr. Sci. 2006, 44 291-307. 272. Zhang, Z.; Pawliszyn, J., Headsp ace solid-phase microextraction. Anal. Chem. 1993, 65 (14), 1843-1852. 273. Knapp, D. R., Handbook of Analytical De rivatization Reactions Wiley: New York, NY, 1979; p 449-514. 274. Nicholson, J. D., Derivative Formation in th e Quantitative Gas-chromatographic Analysis of Pharmaceuticals. Part II. The Analyst 1978, 103 (1224), 193-222. 275. Segura, J.; Ventura, R.; Jurado, C., Derivati zation procedures for gas chromatographicmass spectrometric determination of xenobiotic s in biological samp les, with special attention to drugs of abuse and doping agents. J. Chromatogr. B 1998, 713 (1), 61-90. 276. Blau, K.; Halket, J., Handbook of Derivatives for Chromatography 2nd ed.; Wiley: New York, NY, 1993; p 52-100. 277. Zhang, K.; Zuo, Y., Pitfalls and solution for simultaneous determina tion of estrone and 17 -ethinylestradiol by gas chromatography-ma ss spectrometry after derivatization with N,O -bis(trimethylsilyl)trifluoroacetamide. Anal. Chim. Acta 2005, 554 (1-2), 190-196. 278. Stanford, B. D.; Weinberg, H. S., Isotope d ilution for quantitation of steroid estrogens and nonylphenols by gas chromatography with tandem mass spectrometry in septic, soil, and groundwater matrices. J. Chromatogr., A 2007, 1176 26-36. 279. Song, J.; Wadhwa, L.; Bejjani, B. A.; O'Brie n, W. E., Determination of 3-keto-4-ene steroids and their hydroxylated metabolites catalyzed by r ecombinant human cytochrome P450 1B1 enzyme using gas chromatography-mass spectrometry with trimethylsilyl derivatization. J. Chromatogr. B 2003, 791 (1-2), 127-135. 280. Little, J. L., Artifacts in trimethylsilyl deri vatization reactions and ways to avoid them. J. Chromatogr., A 1999, 844 (1-2), 1-22. 281. Zuo, Y.; Lin, Y., "Solvent effects on the silylation-gas ch romatography-mass spectrometric determination of natural a nd synthetic estrogeni c steroid hormones" Comment on "Formation of chlorinated estrones via hypochlorous disinfection of wastewater effluent contai ning estrone" by Hideyuki Nakamura, Ryoko Kuruto-Niwa, Mitsuo Uchida and Yoshiyasu Terao [Chemosphere 66 (2007) 1441-1448]. Chemosphere 2007, 69 (7), 1175-1176. 282. Shareef, A.; Angove, M. J.; Wells, J. D., Optimization of silylation using N -methylN (trimethylsilyl)-trifluoroacetamide, N,O -bis-(trimethylsilyl)-trifluoroacetamide and N-
202 (tert-butyldimethylsilyl)-N-methyltrifluor oacetamide for the determination of the estrogens estrone and 17 -ethinylestradiol by gas chro matography-mass spectrometry. J. Chromatogr., A 2006, 1108 (1), 121-128. 283. Gleispach, H., The use of different silylating agents for structure an alysis of steroids. J. Chromatogr. 1974, 91 407-412. 284. Kelly, C., Analysis of steroids in envi ronmental water samples using solid-phase extraction and ion trap gas chromatogra phy-mass spectrometry and gas chromatographytandem mass spectrometry. J. Chromatogr., A 2000, 872 309-314. 285. Mol, H. G. J.; Sunarto, S.; Steijger, O. M., Determination of endocrine disruptors in water after derivatization with N-methyl-N-(tert.butyldimethyltrifluoroacetamide) using gas chromatography with mass spectrometric detection. J. Chromatogr., A 2000, 879 (1), 97112. 286. Chambaz, E. M.; Horning, E. C., Conversion of steroids to trimethylsilyl derivatives for gas phase analytical studies: Re actions of silylating reagents. Anal. Biochem. 1969, 30 (1), 7-24. 287. Zuo, Y.; Zhang, K.; Lin, Y., Microwave-accel erated derivatization for the simultaneous gas chromatographic-mass spectrometric analys is of natural and s ynthetic estrogenic steroids. J. Chromatogr., A 2007, 1148 (2), 211-218. 288. Agatha, G.; Kauf, E., GC Analysis of St eroids, Fatty Acids, Organic Acids and Catecholamine Metabolites with Microwave A ccelerated Derivatization for the Diagnosis of Metabolic Disorders. Clin. Lab. 1999, 45 (7+8), 387-397. 289. Deng, C.; Yin, X.; Zhang, L.; Zhang, X., Development of microwave-assisted derivatization followed by gas ch romatography/mass spectrometry for fast determination of amino acids in neonatal blood samples. Rapid Commun. Mass Spectrom. 2005, 19 (16), 2227-2234. 290. Guillette Jr., L.; Gunderson, M. P., Alterati ons in development of reproductive and endocrine systems of wildlife populations exposed to endocrine-disrupting contaminants. Reprod. 2001, 122 (6), 857-864. 291. Breton, R.; Boxall, A., Pharmaceuticals and Pe rsonal Care Products in the Environment: Regulatory Drivers and Research Needs. QSAR Comb. Sci. 2003, 22 (3), 399-409. 292. Heinz, G. H.; Percival, H. F.; Jennings, M. L., Contaminants in American alligator eggs from lakes Apopka, Griffin, and Okeechobee, Florida. Environ. Monit. Assess. 1991, 16 277-285. 293. Deventer, K.; Van Eenoo, P.; Mikulckov, P.; Van Thuyne, W.; Delbeke, F. T., Quantitative analysis of androst-4-ene-3,6,17-trione and metabolites in human urine after the administration of a food supplement by liquid chromatography/ion trap-mass spectrometry. J. Chromatogr. B 2005, 828 (1-2), 21-26.
203 294. Berrueta, L. A.; Gallo, B.; Vicente, F., A Review of Solid Phase Extraction: Basic Principles and New Developments. Chromatographia 1995, 40 (7/8), 474. 295. Hajkova, K.; Pulkrabova, J.; Schurek, J.; Ha jslova, J.; Poustka, J.; Napravnikova, M.; Kocourek, V., Novel approaches to the analysis of steroid estrogens in river sediments. Anal. Bioanal. Chem. 2007, 387 (4), 1351-1363. 296. Zhang, S.; Zhang, Q.; Darisaw, S.; Ehie, O. ; Wang, G., Simultaneous quantification of polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), and pharmaceuticals and personal care products (PPC Ps) in Mississippi ri ver water, in New Orleans, Louisiana, USA. Chemosphere 2007, 66 (6), 1057-1069. 297. Boussahel, R.; Bouland, S.; Moussaoui, K. M. ; Baudu, M.; Montiel, A., Determination of chlorinated pesticides in water by SPME/GC. Water Res. 2002, 36 (7), 1909-1911. 298. Magdic, S.; Pawliszyn, J. B., Analysis of organochlorine pesticides using solid-phase microextraction. J. Chromatogr., A 1996, 723 (1), 111-122. 299. Volante, M.; Pontello, M.; Valoti, L.; Cattane o, M.; Bianchi, M.; Colzani, L., Application of Solid Phase Micro-Extraction (SPME) to the analysis of pesticide residues in vegetables. Pest Manage. Sci. 2000, 56 618-636. 300. Ouyang, G.; Pawliszyn, J., SPME in environmental analysis. Anal. Bioanal. Chem. 2006, 386 (4), 1059-1073. 301. Boyd-Boland, A. A.; Magdic, S.; Pawliszyn, J. B., Simultaneous Determination of 60 Pesticides in Water Using Solid-phase Microextraction and Gas Chromatography-Mass Spectrometry. Analyst 1996, 121 929-938. 302. Valor, I.; Prez, M.; Cortada, C.; Apra iz, D.; Molt, J. C.; Font, G., SPME of 52 pesticides and polychlorinated biphenyls: Extraction efficien cies of the SPME coatings poly(dimethylsiloxane), polyacr ylate, poly(dimethylsiloxane )-divinylbenzene, Carboxen-poly(dimethylsiloxane), and Carbowax-divinylbenzene. J. Sep. Sci. 2001, 24 (1), 39-48. 303. Guidotti, M.; Vitali, M., Determination of Urinary Pentachlorophenol by SPME and GC/MS. J. High Resolut. Chromatogr. 1998, 21 (2), 137-139. 304. Quintana, J. B.; Carpinteiro, J.; Rodriguez, I.; Lorenzo, R. A.; Carro, A. M.; Cela, R., Determination of natural and synthetic estr ogens in water by gas chromatography with mass spectrometric detection. J. Chromatogr., A 2004, 1024 (1-2), 177-185. 305. Fiehn, O., Extending the breadth of metabo lite profiling by gas chromatography coupled to mass spectrometry. Trends Anal. Chem. 2008, 27 (3), 261-269. 306. Ma, Y.-C.; Kim, H.-Y., Determination of Steroids by Liquid Chromatography/Mass Spectrometry. J. Am. Soc. Mass Spectrom. 1997, 8 (9), 1010-1020.
204 307. Isobe, T.; Shiraishi, H.; Yasuda, M.; Shinoda A.; Suzuki, H.; Morita, M., Determination of estrogens and their conjuga tes in water using solid-phase extraction followed by liquid chromatography-tandem mass spectrometry. J. Chromatogr., A 2003, 984 195-202. 308. Lagana, A.; Fago, G.; Marino, A.; Santarel li, D., Liquid Chromatography Tandem Mass Spectrometry Applied to the Analysis of Natu ral and Synthetic Steroids in Environmental Waters. Anal. Lett. 2001, 34 (6), 913 926. 309. Ingrand, V.; Herry, G.; Beausse, J.; de Roubi n, M.-R., Analysis of steroid hormones in effluents of wastewater treatment plants by liquid chromatography-tandem mass spectrometry. J. Chromatogr., A 2003, 1020 99-104. 310. Horning, M. G.; Moss, A. M.; Horning, E. C ., Formation and gas-liquid chromatographic behavior of isometric steroid ketone methoxime derivatives. Anal. Biochem. 1968, 22 (2), 284-294. 311. Thenot, J. P.; Horning, E. C., MO-TMS Deri vatives of Human Urinary Steroids for GC and GC-MS Studies. Anal. Lett. 1972, 5 (1), 21-33. 312. Koh, T.-S., Microwave drying of biological tissues for trace element determination. Anal. Chem. 1980, 52 (12), 1978-1979. 313. Wang, X.; Chen, H.; Luo, K.; Shao, J.; Yang, H., The Influence of Microwave Drying on Biomass Pyrolysis. Energ. Fuel. 2008, 22 (1), 67-74. 314. Jin, Q.; Liang, F.; Zhang, H.; Zhao, L.; Huan, Y.; Daqian, S., Application of microwave techniques in analytical chemistry. Trends Anal. Chem. 1999, 18 (7), 479-484. 315. Kingston, H. M. S., Standardization of Sample Preparation for Trace Element Determination Through Microwave-Enhanced Chemistry. At. Spectrosc. 1998, 19 (2), 27-65. 316. Lamble, K. J.; Hill, S. J., Microwave digest ion procedures for environmental matrices: A critical review. The Analyst 1998, 123 103R-133R. 317. Agazzi, A.; Pirola, C., Fundamentals, methods and future trends of environmental microwave sample preparation. Microchem. J. 2000, 67 (1-3), 337-341. 318. Rice, S. L.; Mitra, S., Microwave-assisted solvent extraction of solid matrices and subsequent detection of pharmaceuticals and personal care products (PPCPs) using gas chromatography-mass spectrometry. Anal. Chim. Acta 2007, 589 (1), 125-132. 319. Ramil Criado, M.; Pombo da Torre, S.; R odrguez Pereiro, I.; Cela Torrijos, R., Optimization of a microwave-assisted de rivatization-extractio n procedure for the determination of chlorophenols in ash samples. J. Chromatogr., A 2004, 1024 (1-2), 155163.
205 320. Caddick, S., Microwave assisted organic reactions. Tetrahedron 1995, 51 (38), 1040310432. 321. Langa, F.; Cruz, P. d. l.; Hoz, A. d. l. ; Diaz-Ortiz, A.; Diez-Barra, E., Microwave irradiation: more than just a method for accelerating reactions. Contemp. Org. Synth. 1997, 4 (5), 373-386. 322. Perreux, L.; Loupy, A., A tentative rationalizat ion of microwave effects in organic synthesis according to th e reaction medium, and mech anistic considerations. Tetrahedron 2001, 57 (45), 9199-9223. 323. Tierney, J. P.; Lidstrom, P., Microwave Assisted Organic Synthesis Blackwell Publishing: Boca Raton, FL, 2005. 324. Lidstrom, P.; Tierney, J.; Wathey, B.; We stman, J., Microwave assisted organic synthesis--a review. Tetrahedron 2001, 57 (45), 9225-9283. 325. Momenbeik, F.; Khorasani, J. H., Separation and determination of sugars by reversedphase high-performance liquid chromatogra phy after pre-column microwave-assisted derivatization. Anal. Bioanal. Chem. 2006, 384 (3), 844-850. 326. Chavez, G.; Bravo, B.; Pina, N.; Arias, M.; Vivas, E.; Ysambertt, F.; Marquez, N.; Caceres, A., Determination of aliphatic al cohols after on-line microwave-assisted derivatization by liquid chromat ography-photodiode array detection. Talanta 2004, 64 (5), 1323-1328. 327. Strassnig, S.; Gfrerer, M.; Lankmayr, E. P ., Microwave-assisted derivatization of 2,5hexanedione in urine: evalua tion using GC-MS and GC-ECD. J. Chromatogr. B 2004, 813 (1-2), 151-158. 328. Chu, T.-Y.; Chang, C.-H.; Liao, Y.-C.; Chen, Y.-C., Microwave-accelerated derivatization processe s for the determination of pheno lic acids by gas chromatographymass spectrometry. Talanta 2001, 54 (6), 1163-1171. 329. Hoz, A. d. l.; Diaz-Ortiz, A.; Moreno, A., Micr owaves in organic synthesis. Thermal and non-thermal microwave effects. Chem. Soc. Rev. 2005, 34 (2), 164-178. 330. Kappe, C. O.; Stadler, A., Microwave in Organic and Medicinal Chemistry Wiley-VCH: Weinheim, Germany,: 2005; Vol. 25. 331. Baghurst, D. R.; Mingos, D. M. P., Superhea ting Effects Associat ed with Microwave Dielectric Heating. J. Chem. Soc., Chem. Commun. 1992, (9), 674-677. 332. Gedye, R. N.; Wei, J. B., Rate enhancemen t of organic reactions by microwaves at atmospheric pressure. Can. J. Chem. 1998, 76 (5), 525-532. 333. Weast, R. C., CRC Handbook of Chemistry and Physics 67th ed.; CRC Press: Boca Raton, FL, 1987-88.
206 334. Milnes, M. R.; Bermudez, D. S.; Bryan, T. A.; Gunderson, M. P.; Guillette, L. J., Jr., Altered Neonatal Developmen t and Endocrine Function in Alligator mississippiensis Associated with a Contaminated Environment. Biol. Reprod. 2005, 73 (5), 1004-1010. 335. Oosterkamp, A. J.; Hock, B.; Seifert, M.; Irth, H., Novel monitoring strategies for xenoestrogens. Trends Anal. Chem. 1997, 16 (10), 544-553. 336. Hill, J. W.; Baum, S. J.; Scott-Ellis, R. J., Chemistry and Life: An Introduction to General, Organic, and Biological Chemistry 6th ed.; Prentice Hall: Upper Saddle River, NJ, 2000. 337. Colosi, D. M.; Bowden, J. A.; Mora-Montero, D.; Garrett, T. J.; Yost, R. A., Investigation of High-Performance Li quid Chromatography/Mass Spectrometry (HPLC/MS) Using a Monolithic Column to Separate a Diverse Suite of Steroids. J. Chromatogr. Sci. in press 338. Hobe, G.; Schn, R.; Goncharov, N.; Katsiya, G.; Koryakin, M.; Gesson-Cholat, I.; Oettel, M.; Zimmermann, H., Some new aspects of 17 -estradiol metabolism in man. Steroids 2002, 67 (11), 883-893. 339. Toran-Allerand, C. D.; Tinnikov, A. A.; Singh, R. J.; Nethrapalli, I. S., 17 -Estradiol: A Brain-Active Estrogen? Endocrinology 2005, 146 (9), 3843-3850. 340. Moos, W. H.; Dykens, J. A.; Howell, N., 17 -estradiol: a less-feminizing estrogen. Drug Dev. Res. 2008, 69 (4), 177-184.
207 BIOGRAPHICAL SKETCH John A. Bowden was born in 1980, in Madison, Wisconsin. John and his family moved to Venice, Florida, in 1996. He graduated from Venice High School in 1999. John started at the University of Florida in 1999 and graduated cum laude in May 2003 with a B. S. in chemistry. John was accepted into graduate sc hool at the University of Fl orida in 2003. During his time as an analytical chemistry graduate student, John was advised by Dr. Richard A. Yost. John served as a teaching assistant in the chemis try department at the University of Florida and received multiple departmental and univers ity level teaching awards. John was also a member of the inaugural Howard Hughes Medi cal Institute G.A.T.O. R mentoring program, where he mentored two undergraduate students fo r more than a year. During his research, John collaborated with Dr. Louis J. Guillette (Zoology ) to incorporate and develop mass spectrometric techniques capable of solving bi ologically relevant questions. J ohn is a member of the American Society for Mass Spectrometry. John was awarded the Ph. D. in May 2009.