<%BANNER%>

Optimization of Sensitivity of Electrospray Ionization Mass Spectrometry for Metabolite Analysis

Permanent Link: http://ufdc.ufl.edu/UFE0024133/00001

Material Information

Title: Optimization of Sensitivity of Electrospray Ionization Mass Spectrometry for Metabolite Analysis
Physical Description: 1 online resource (158 p.)
Language: english
Creator: Mautjana, Nare
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: dopamine, electrochemistry, electrospray, ionization, one, oxidation, purines, spectrometry, thiols
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Improvements in the sensitivity of on-line Electrochemistry Electrospray Ionization Fourier Transformation Ion Cyclotron Resonance Mass Spectrometry (EC/ESI FT-ICR MS) for uric acid, cysteine, homocysteine, and other low oxidation potential (lower than 1 V vs SHE) metabolites is presented. Inclusion of a variable resistor to the EC cell circuit to regulate applied voltage allowed more detailed elucidation of the mechanisms of electrochemical reactions occuring during ESI MS analysis of the stated metabolites. Various factors which can affect the detection sensitivity of these metabolites in positive ion mode ESI MS and EC/ESI MS are discussed. Changes in the intensity profiles of the various species as a function of applied EC cell potential, ESI flow rate, and analyte concentration provide information for signal optimization. Additional improvements in sensitivity (over ten-fold intensity increases) were observed when the standard cylindrical MS capillary inlet was replaced with a cone-shaped inlet. This new design and the associated dynamics leading to increased sensitivity are discussed. The new modification of ESI inlet is particularly valuable for the analysis of small metabolites, which tend to be spatially distributed in the electrospray interface and are radially segregated. Some of the metabolites used in this work have been reported as antioxidants, and results in this dissertation support antioxidant activity as indicated by the proposed radical mechanisms. In addition to revealing antioxidant activity of selected analytes, step-wise one-electron, one-proton oxidation reactions are observed for dopamine and uric acid with the aid of electrochemistry coupled on-line with ESI MS. These results suggest that positive ion mode ESI MS offers a new radiation-free technique for studying radical pathways. On-line EC/ESI MS data also show that dopamine and purines follow similar oxidation pathways leading to the development of a generic model for their behavior during ESI MS.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Nare Mautjana.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Brajter-Toth, Anna F.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024133:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024133/00001

Material Information

Title: Optimization of Sensitivity of Electrospray Ionization Mass Spectrometry for Metabolite Analysis
Physical Description: 1 online resource (158 p.)
Language: english
Creator: Mautjana, Nare
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: dopamine, electrochemistry, electrospray, ionization, one, oxidation, purines, spectrometry, thiols
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Improvements in the sensitivity of on-line Electrochemistry Electrospray Ionization Fourier Transformation Ion Cyclotron Resonance Mass Spectrometry (EC/ESI FT-ICR MS) for uric acid, cysteine, homocysteine, and other low oxidation potential (lower than 1 V vs SHE) metabolites is presented. Inclusion of a variable resistor to the EC cell circuit to regulate applied voltage allowed more detailed elucidation of the mechanisms of electrochemical reactions occuring during ESI MS analysis of the stated metabolites. Various factors which can affect the detection sensitivity of these metabolites in positive ion mode ESI MS and EC/ESI MS are discussed. Changes in the intensity profiles of the various species as a function of applied EC cell potential, ESI flow rate, and analyte concentration provide information for signal optimization. Additional improvements in sensitivity (over ten-fold intensity increases) were observed when the standard cylindrical MS capillary inlet was replaced with a cone-shaped inlet. This new design and the associated dynamics leading to increased sensitivity are discussed. The new modification of ESI inlet is particularly valuable for the analysis of small metabolites, which tend to be spatially distributed in the electrospray interface and are radially segregated. Some of the metabolites used in this work have been reported as antioxidants, and results in this dissertation support antioxidant activity as indicated by the proposed radical mechanisms. In addition to revealing antioxidant activity of selected analytes, step-wise one-electron, one-proton oxidation reactions are observed for dopamine and uric acid with the aid of electrochemistry coupled on-line with ESI MS. These results suggest that positive ion mode ESI MS offers a new radiation-free technique for studying radical pathways. On-line EC/ESI MS data also show that dopamine and purines follow similar oxidation pathways leading to the development of a generic model for their behavior during ESI MS.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Nare Mautjana.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Brajter-Toth, Anna F.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024133:00001


This item has the following downloads:


Full Text

PAGE 1

1 OPTIMIZATION OF SENSITIVITY OF ELECTROSPRAY IONIZATION MASS SPECTROMETRY FOR M ETABOLITE ANALYSIS By NARE ALPHEUS MAUTJANA 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

PAGE 2

2 2009 Nare Alpheus Mautjana

PAGE 3

3 To my Grandmother, Selaki MaChuene Seemole Mautjana

PAGE 4

4 ACKNOWLEDGMENTS I am very grateful to my mother Melida Ma utjana for motivating me since my beginning of school at gaMahoai village th rough college in Vanderbijl Park and Cape Town South Africa, up until graduate school here in Gainesville, Florida USA. I acknowledge with much gratitude the (parent-like and financial) s upport I received from Dr. Hunt Davis and his wife Jeanne Davis anytime I needed it during the course of my PhD studies. I thank Dr. Anna Brajter-Toth for her teachi ng, her guidance, and for being my guardian in many respects. What seemed like a long haul in the beginning turned out to be a brief period of professional and personal grow th. I also thank members of the Toth group (Mehj, Andrews, and Abraham) for interesting chemis try discussions and friendship. I thank Dr. John Eyler for giving me tota l access to the FT-ICR mass spectrometer and for being available to assist me with experiment s and to discuss my research. I also acknowledge the invaluable exchange of ideas with members of the Eyler group, particul arly Cesar Contreras. Sincere thanks go to Dr. Sergiu Palii for giving me the initial FT-ICR MS training and Dr. John Toth for giving me on-the-job mass spec training. I also wish to express my appreciation to Drs. Vaneica Young, Weihong Ta n, John R. Eyler, and Haniph Latchman for serving on my PhD supervisory committee. I th ank Dr. Williams for her invaluable comments and for proof readi ng this dissertation. Special thanks go to my wife, Kgadi, for her ever-present support, her patience, friendship and her kind love. I appreciate immensely her spending time with our son Lesego, while I was working in the lab. To Lesego, I am happy to say NOW I am done with my school work. Everyone has been a blessing to me and I thank God!

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................8LIST OF FIGURES .........................................................................................................................9LIST OF SCHEMES......................................................................................................................13ABSTRACT ...................................................................................................................... .............15 CHAP TER 1 INTRODUCTION .................................................................................................................. 17Electrospray Ionization Mass Spectrometry (ESI MS) .......................................................... 17Electrochemical Nature of Electrospray Ionization ................................................................18On-Line Electrochemistry Mass Spectrometry (EC/MS) ....................................................... 22On-Line Electrochemical Cell Designs ........................................................................... 24Control of EC Cell Potential in EC/ESI MS ................................................................... 26The EC/ESI MS of Dopamine .........................................................................................27The EC/ESI MS of Uric Acid ..........................................................................................28The EC/ESI MS of Thiols ...............................................................................................30Study Overview ......................................................................................................................312 EXPERIMENTAL .................................................................................................................. 39Methods and Instrumentation .................................................................................................39Construction of the EC/ESI MS System .................................................................................39Cone-Shaped MS Capillary Inlet ............................................................................................40Fourier Transform Ion Cyclotron Res onance (FT-ICR) Mass Spectrometry ......................... 40Operational Safety ............................................................................................................ ......40Cyclic Voltammetry ................................................................................................................41Fundamentals of Methods Used .............................................................................................41ESI MS ....................................................................................................................................41EC/ESI MS .............................................................................................................................44FT-ICR Mass Spectrometry ....................................................................................................45Hydrogen/Deuterium (H/D) Exchange Methods .................................................................... 46Tandem Mass Spectrometry (MS/MS or MSn) ...................................................................... 47Cyclic Voltammetry ................................................................................................................48Experimental Conditions ........................................................................................................49Solution Preparation .......................................................................................................... .....49Cyclic Voltammetry ................................................................................................................50

PAGE 6

6 FT-ICR MS Data Analysis ..................................................................................................... 50Photo-Induced Dissociation and MS/MS ...............................................................................51H/D Exchange Experiment .....................................................................................................513 ONE-ELECTRON OXIDATION OF DOPAM INE IN ESI AND E C/ESI MS .................... 58Introduction .................................................................................................................. ...........58Results and Discussion ........................................................................................................ ...60The ESI MS of Dopamine (DA) ......................................................................................60Cone-Shaped vs Cylindrical Inlet .................................................................................... 61The H/D Exchange of DA ...............................................................................................62The MS/MS of DA Dimer ............................................................................................... 63The EC/ESI MS of DA ....................................................................................................63Effect of Flow Rate .........................................................................................................65Cyclic Voltammetry of DA ............................................................................................. 65The ESI MS of DA in the Presence of Cysteine (CySH) ................................................ 66The EC/ESI MS of DA in the Presence of CySH ........................................................... 67Conclusions .............................................................................................................................674 ONE-ELECTRON OXIDATION AND DETECTION SENSITIVITY OF URIC ACID IN ESI AND EC/ESI MS ....................................................................................................... 78Introduction .................................................................................................................. ...........78Results and Discussion ........................................................................................................ ...79Ionization of Uric Acid in Electrospray (ES) .........................................................................79The EC/ESI MS of Uric Acid ..........................................................................................85The ESI MS of Uric Acid in Urine ..................................................................................86Conclusions .............................................................................................................................875 SENSITIVITY OF POSITIVE MODE ESI AND EC/ESI MS TO THE ANALYSIS OF THIOL METABOLITES ........................................................................................................95Introduction .................................................................................................................. ...........95Results and Discussion ........................................................................................................ .100The ESI MS of glutathione (GSH), cyst eine (CySH) and homocysteine (hCySH) ...... 100Effect of GSH Concentration on ESI MS ...................................................................... 102The ESI MS of GSH, CySH and hCySH Mixture .........................................................104The EC/ESI MS of GSH ................................................................................................105The ESI MS of GSH and hCySH in the Presence of Dopamine (DA) .......................... 106Effect of GSH Concentrati on on Thiol/DA Mass Spectra ............................................ 107The ESI MS of GSH in the Presence of Uric Acid ....................................................... 108The ESI MS of GSH, CySH and hCyS H Mixture in Presence of DA .......................... 109The EC/ESI MS of Thiols in the Presence of DA .........................................................110Summary of Thiol Mixture Analysis .............................................................................111Evidence of Catalysis of CySH Oxidation by Metal Ions ............................................. 111Proposed Mechanism of Catalysis of CySH Oxidation by Iron (II) ............................. 112Conclusions ...........................................................................................................................113

PAGE 7

7 6 OXIDATION OF PURINES DURING ESI MS AND EC/ESI MS .................................... 126Introduction .................................................................................................................. .........126Results and Discussion ........................................................................................................ .129The ESI MS and EC/ESI MS of Guanine (Gua) ...........................................................129The ESI MS and EC/ESI MS of Adenine (Ad) .............................................................132The ESI MS and EC/ESI MS of Hypoxanthine (hXan) ................................................ 134The ESI MS and EC/ESI MS of Xanthine (Xan) ..........................................................135Conclusions ...........................................................................................................................1377 CONCLUSIONS .................................................................................................................. 145LIST OF REFERENCES .............................................................................................................148BIOGRAPHICAL SKETCH .......................................................................................................158

PAGE 8

8 LIST OF TABLES Table page 4-1. Theoretical and average measured m/z va lu es of identified io ns and their isotopic abundances (n = 15). .......................................................................................................... 885-1. Average intensities (n = 3) of thio l derived ions in the presence of DA. ........................ 1145-2. Average intensities (n = 3) of cysteine disulfide dimer (m/z 241) indicating metal ion catalysis of cysteine oxidation. ........................................................................................ 114

PAGE 9

9 LIST OF FIGURES Figure page 1-1. Electrospray Ionization Interface for LC/MS [Adapted from Fenn et al., 1989] .............. 341-2. Electrospray ionization process [Adapted from Cech and Enke, 2001] ............................341-3. Current vs voltage curve for a current-limited device ....................................................... 351-4. Equivalent electric circuit re presentation of the ESI process ............................................351-5. Thermospray interface for LC/MS ..................................................................................... 361-6. Particle beam interface for LC/MS ....................................................................................361-7. On-line electrochemical cell for EC /TSI MS [Adapted from Hambitzer and Heitbaum, 1986] ............................................................................................................... .371-8. On-line electrochemical cell for EC/PBI MS and EC/TSI MS [Adapted from Regino and Brajter-Toth, 1997] .....................................................................................................371-9. On-line electrochemical cell for EC/ESI MS (ss, stainless steel capillary; Pd, palladium electrode) [Adapted from Zhang et al., 2002] ................................................... 381-10. Distribution diagram of different dopamine species as a function of pH [Adapted from Sanchez-Rivera et al., 2003] ..................................................................................... 382-1. The EC/ESI MS system with the el ectrochemical cell intergrated into the electrospray capillary. A) Schematic with dimensions; B) Expansion of on-line EC cell. ......................................................................................................................... ............522-2. Electrical circuit diagram of the on-line EC/ ESI MS system ...........................................532-3. Standard linear track variable resistor. ...............................................................................532-4. Projected voltage profile s along the ES capillary with ~ 0.5V increments of applied EC cell voltage (Vapp). The low EC cell voltage is floated at the high voltage (HV) of the electrospray. The solid line repr esents the reported voltage profile of a standard ES emitter [Pozniak and Cole, 2007]. ................................................................. 542-5. The ICR cell showing the el ectronic circuit through which rf electric field is applied to excite, trap and detect ions. ............................................................................................ 542-6. Fourier transform IR absorption spectrum of dopamine. The band at 10.65 m was laser targeted for the IRMPD MS/MS experiment. ........................................................... 552-7. Fourier transform IR absorption sp ectrum of uric acid. The band at 10.16 m was laser targeted for the IRMPD MS/MS experiment. ........................................................... 56

PAGE 10

10 2-8. Typical cyclic voltammogram. .......................................................................................... 573-1. Positive ion ESI MS of DA (2.5 mM) with (A) cylindrical inlet and (B) cone-shaped inlet; flow rate 30 L/h; 50/1/49 (vol %) H2O/HAc/MeOH, pH~4.2; HV ~3 kV. ............ 693-2. The ESI mass spectrum of dopamine (0.5 mM) in 50/49/1 vol%, D2O/methanol/acetic acid (A). Peaks repr esenting the number of exchangeable protons for the ions [DA-NH3]+, [DA]+ and [2DA-H]+ are shown in (B), (C) and (D), respectively. See structures in Scheme 3-1. .......................................................................703-3. Changes in ion intensities in ESI mass spectra of DA due to change in concentration. Dots indicate off-scale intensity. Flow rate 60 L/h; cone-shaped capillary inlet; 50/1/49 (vol %) H2O/HAc/MeOH, pH~4.2; HV ~3 kV. ...................................................713-4. The ESI MS of 2.5 mM dopamine (A); ESI MS after ejection of m/z < 307 and m/z > 307 ions i.e isolation of dopamine dimer [2DA-H]+ (m/z 307) ion (B); MS/MS of [2DA-H]+ following CO2 laser irradiation (>0.5s) Notice product ion peaks at m/z 154, most likely [DA]+ and at m/z 174, unassigned (C). ................................................... 723-5. The EC/ESI MS of DA (0.25 mM). Conditions as in Figure 3-3; moving average in black. ........................................................................................................................ ..........733-6. The ESI MS of DA (2.5 mM) as a function of flow rate: (A) cylinder capillary inlet; (B) conical capillary inlet; (C) conical cap illary inlet in EC/ESI MS (1.5 V). Other conditions as in Figure 3-3. ................................................................................................743-7. Cyclic voltammetry of DA (400 M) at stainless steel el ectrode in (A) phosphate buffer (31 mM), pH~7.4; (B) 50/1/49 vol%, water/acetic acid/methanol, pH~4.2; (C) 99/1 vol%, water/acetic acid, pH~4.0. Disk radius 50.8 m; scan rate 50 mVs-1. ........... 753-8. The ESI MS of cysteine (CySH) and DA with CySH: (A) CySH (0.5 mM); (B) DA (2.5 mM), CySH (0.5 mM). Flow rate 45 L/h. Other conditions as in Figure 3-3. ..........................................................................................................................763-9. The EC/ESI MS of DA with Cy SH. Conditions as in Figure 3-3. .................................... 774-1. Positive ion mass spectra of uric acid. Cone-shaped capillary inlet; 40/60 vol%, water/ methanol, 0.001M ammonium acetate, pH~6.3* ; flow rate 40L/h; HV 3kV. .....894-2. The ESI mass spectrum of ur ic acid (50 M) in 40/60 vol%, D2O/MeOH, 1mM NH4Ac (A). Peaks representing the number of exchangeable protons for the ions [H2U+H]+, [H2U+K]+ and [2H2U+H]+ are shown in (B), (C ) and (D), respectively. See structures in Scheme 4-1. ............................................................................................904-3. The ESI MS of 50 M uric acid (A); ESI MS after ejection of m/z < 337 and m/z > 337 ions i.e isolation of uric acid dimer [2H2U+H]+ (m/z 337) ion (B); MS/MS of [2H2U+H]+ (m/z 337) following CO2 laser irradiation (>0. 5s) Notice the product

PAGE 11

11 peak at m/z 169, likely due to [H2U+H]+ ion and smaller unassigned product peaks at m/z values <300 (C). ..........................................................................................................914-4. The ln (Intensity) vs ln (concentrat ion, mol/L) plots for the ions [K(Allnt)+K]+ (m/z 235) (A) and [K(Allnt)+Ac+2K]+ (m/z 333) (B). ..............................................................924-5. Intensity of uric acid (50 M) ions in EC/ESI MS as a function of on-line EC cell voltage. Cone-shaped inlet; 40/60 vol%, H2O/MeOH, 10-3 M NH4Ac, pH 6.3; Flow rate 40 L/h; HV 3 kV. ...................................................................................................... 934-6. Positive ion mode ESI MS mass spectra of human urine: A) 1000 fold diluted; B) 1000 fold diluted and spiked with 20 M uric acid. Same conditions as in Figure 4-1. ... 945-1. Positive mode ESI MS of (A) GS H (0.05 mM), (B) CySH (0.05 mM) and (C) hCySH (0.5 mM) in 40/60 vol%, H2O/MeOH containing 1 mM NH4Ac, pH~6.3; Flow rate 50 L/h; HV 3 kV. ...........................................................................................1155-2. The H/D exchange ESI MS of GSH (0.05 mM) in 40/60 vol%, D2O/MeOH, 1 mM NH4Ac, pH 6.3; Flow rate 50 L/h; HV 3 kV See structures in Scheme 5-3. ............... 1165-3. Lowest concentration of GSH (2 x 10-3 mM or 2 M) detected with positive mode ESI MS; Same conditions as in Figure 5-1. ..................................................................... 1175-4. Positive mode ESI MS of GSH (0.5 mM); Same conditions as in Figure 5-1. ............... 1175-5. The ESI MS of mixed thiols, GSH (0.05 mM), CysH (0.05 mM) and hCySH (0.5 mM). Same conditions as in Figure 5-1 ........................................................................... 1185-7. Positive mode ESI MS of (A) GSH (0.5 mM) and (B) hCysH (0.5 mM), each in the presence of DA (2.5 mM); 50/49/1 vol%, H2O/MeOH/HAc, pH~4.2; flow rate 50L/h; HV 3 kV. ............................................................................................................ 1195-8. Positive mode ESI MS of GSH (various concentrations) in the presence of DA (2.5 mM); Same conditions as in Figure 5-7. .................................................................. 1205-9. Positive mode ESI MS of GSH (0.01 mM or 10 M) in the presence of uric acid (0.06 mM or 60 M); Same conditions as in Figure 5-1. ................................................ 1215-10. Positive mode ESI MS of a thiol mi xture, GSH (0.05 mM), CySH (0.05 mM) and hCySH (0.5 mM), in the presence of DA ( 2.5 mM). Same conditions as in Figure 57........................................................................................................................................1225-11. Effect of applied EC cell voltage on oxidation products of (A) GSH and (B) hCySH, each in presence of DA (2.5 mM); Same conditions as in Figure 5-7. ............................ 1235-12. Cyclic voltmmograms ( = 50 mV/s) of cysteine (0.5 mM) alone (A) and with Fe2+ (1 mM) (B); cysteine (2.0 mM alone) (C) and with Fe2+ (1 mM) (D) on stainless steel electrode (r=50 m); Ref = SCE. Blank = 50/49/1 vol%, H2O/MeOH/HAc. .................124

PAGE 12

12 5-13. Positive mode ESI MS of CySH (0.5 mM) with Fe2+ (100 M). ....................................1256-1. The ESI MS of guanine (50 M) in 50/49/1 vol%, H2O/MeOH/HAc, pH~4.2; HV 3 kV; Flow rate 30 L/h. ....................................................................................................1406-2. The EC/ESI MS of guanine (50 M) in 50/49/1 vol%, H2O/MeOH/HAc, pH~4.2; HV 3 kV; Flow rate 30 L/h; EC cell voltage 1.5V. ................................................... 1406-3. The EC/ESI MS of guanine (50 M ). Other conditions as in Figure 6.2. ....................... 1416-4. The ESI MS of guanine (50 M) in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, pH~6.3; HV 3 kV; Flow rate 50 L/h. ............................................................................1416-5. The ESI MS of adenine (50 M) in 50/49/1 vol%, H2O/MeOH/HAc, pH 4.2; HV 3 kV; Flow rate 30 L/h; EC cell voltage 1.5V. ................................................................. 1426-6. The ESI MS of adenine (50 M) in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, pH 6.3. Other conditions as in Figure 6-4. .................................................................................... 1426-7. The ESI MS of hypoxanthine (50 M) in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, pH 6.3. Other conditions as in Figure 6-4. .......................................................................1436-8. The EC/ESI MS of hypoxanthine (50 M). Other conditions as in Figure 6-4. .............. 1436-9. The ESI MS of xanthine (50 M) in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, Other conditions as in Figure 6.4. ..............................................................................................1446-10. The EC/ESI MS of xanthine (50 M). Other conditions as in Figure 6-4. ...................... 144

PAGE 13

13 LIST OF SCHEMES Schem e page 1-1. Three proton dissociations of DA [Sanchez-Rivera et al., 2003]. .....................................271-2. Tautomeric structures of uric acid ..................................................................................... 281-3. Proposed one-electron, one-proton oxidati on of uric acid [Buettner and Jerkiewicz, 1996]. ........................................................................................................................ .........291-4. Metabolic pathway of purines in humans (ATP, adenosine triphosphate; GTP, guanosine triphosphate; AMP, adenosine monophosphate; IMP, inosine monophosphate; XMP, xanthosine monophosphate; GMP, guanosine monophosphate). ............................................................................................................... .301-5. Biosynthesis of homocyst eine, cysteine and glutathione [adapted from Himmelfarb et al., 2002] ................................................................................................................. .......323-1. Dopamine oxidation in positive mode ESI MS. Hydrogens that form H-bonds are not exchangeable with deuterium in the presence of D2O (see mass spectra in Figure 32). ........................................................................................................................... ............613-2. Proposed cleavage mechanism of the DA dimer in the infrared multiphoton dissociation (IRMPD) experiment. .................................................................................... 643-3. Formation of DAQ-CySH adduct in positive ion mode ESI MS. ...................................... 674-1. Proposed reactions during ESI of uric acid (H2U). Urate (HU-) is present in the pH 6.3 carrier solution of H2U (pKa1=5.4). Hydrogens that form H-bonds and those between C=O groups are not exchangeable with deuterium in the presence of D2O (see mass spectra in Figure 4-2). Proton adducts of uric acid [H2U+H]+ (m/z 169) and uric acid dimer (m/z 337) are detected in this carrier solu tion in ESI MS. The H+ is generated during positive ion mode electrospray. ..........................................................81 4-2. Proposed cleavage mechanism of the uric acid dimer in the infrared multiphoton dissociation (IRMPD) experiment. .................................................................................... 834-3. Oxidation pathway of uric acid to 5-hydroxyhydantoinparabanic acid adduct [OHhyd+Parab], alloxan monohydrate [Allxnhyd] and allantoin [Allant] [Volk et al., 1992; Volk et al., 1999]. Protonate d hydroxyhydantoinparabanic acid adduct (m/z 231) is detected in ESI MS. Allant oin and alloxan monohydrate are detected as K+ adducts [K(Allnt)+K]+ (m/z 235) and [K(Allxnhyd)+K]+ (m/z 237), respectively. .... 844-4. Proposed dimer formation pathway (A) during ESI of uric acid (H2U). HUis present in pH 6.3 ca rrier solution of H2U (pKa1=5.4). Protonated uric acid [H2U+H]+ (m/z 169) and the protonated dimer (m/z 337) are detected in ESI MS. The H+ is generated during positive ion m ode electrospray. (B) H-atom transfer leading to the

PAGE 14

14 uric acid dimer and the neutral radical, which further gives diimine and final oxidation products. .............................................................................................................865-1. Structures and pKa values of thiol metabolites [Nek rassova et al., 2003; Budavari et al., 1989]. ...........................................................................................................................965-2. Glutathione disulfide dimer detected as [GSSG+H]+ (m/z 613) in ESI MS ....................1005-3. Proposed oxidation of GSH during positive mode ESI MS. Hydrogens that form Hbonds are not exchangeable with deuterium in the presence of D2O (see the mass spectrum in Figure 5-2). ...................................................................................................1015-4. Fragmentation of GSH during ESI MS [adapted from Rubino et al., 2006]. .................. 1045-5. The GSSG fragmentations observed duri ng ESI MS [Adapted from Rubino et al., 2006]. ........................................................................................................................ .......1055-6. Formation of [DAQ+GSH] adduct in positive ion mode ESI MS. .................................. 1086-1. Structure of the nucleotide, adenosine triphosphste (ATP). ............................................1276-2. Structures of purine bases, pKa values [Budavari, 1989; Rogstad et al., 2003] and one-electron oxidation potentials (E1 vs SHE) [Jovanovic and Simic, 1986] ................. 1286-3. The H-bonded guanine tetramer with a metal ion center (M+ = sodium ion, Na+ or potassium ion, K+) ........................................................................................................... 1306-4. Proposed mechanism of oxidati on of guanine in 40/60 vol%, H2O/MeOH, 10-3 M NH4Ac, pH 6.3, during positive mode ESI MS. .............................................................. 1316-5. Proposed mechanism of oxidation of adenine during positive mode ESI MS. ............... 1336-6. Proposed mechanism of oxidation of hypoxanthine during positive mode ESI MS. ......1356-7. The H-bonded xanthine tetramer with a metal ion center (M+ = sodium ion, Na+).........1366-8. Proposed oxidation mechanism of xanthi ne to xanthine radi cals [Adapted from Kathiwala et al., 2008]. The final product (2 Xan-H) is detected as a [(2Xan-H).+Na]+ (m/z 326). .................................................................................................................... .....1376-9. Proposed general path of oxidation of purines during positive mode ESI MS (m = purine metabolite). ...........................................................................................................138

PAGE 15

15 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 OPTIMIZATION OF SENSITIVITY OF ELECTROSPRAY IONIZATION MASS SPECTROMETRY FOR M ETABOLITE ANALYSIS By Nare Alpheus Mautjana May 2009 Chair: Anna Brajter-Toth Major: Chemistry Improvements in the sensitivity of on-line Electrochemistry Electrospray Ionization Fourier Transformation Ion Cy clotron Resonance Mass Spectrometry (EC/ESI FT-ICR MS) for uric acid, cysteine, homocysteine, and other lo w oxidation potential (low er than 1 V vs SHE) metabolites is presented. Inclusion of a variable re sistor to the EC cell ci rcuit to regulate applied voltage allowed more detailed elucidation of the mechanisms of electrochemical reactions occuring during ESI MS analysis of the stated me tabolites. Various factors which can affect the detection sensitivity of these metabolites in positive ion mode ESI MS and EC/ESI MS are discussed. Changes in the intensity profiles of th e various species as a function of applied EC cell potential, ESI flow rate and analyte concentration provide information for signal optimization. Additional improvements in sens itivity (over ten-fold intensity increases) were observed when the standard cylindrical MS capillary inlet was replaced with a cone-shaped inlet. This new design and the associated dynamics leading to increased sensitivity are discussed. The new modification of ESI inlet is particularly valuab le for the analysis of small metabolites, which tend to be spatially distributed in the electrospray inte rface and are radially segregated. Some of

PAGE 16

16 the metabolites used in this work have been reported as antioxidants, and results in this dissertation support antioxidant activity as indicated by the proposed radical mechanisms. In addition to revealing antioxidant activity of selected analytes, step-wise one-electron, one-proton oxidation reactions are observed fo r dopamine and uric acid with the aid of electrochemistry coupled on-line with ESI MS. Th ese results suggest that positive ion mode ESI MS offers a new radiation-free technique for studying radical pathways On-line EC/ESI MS data also show that dopamine and purines follow similar oxidation pathways leading to the development of a generic model fo r their behavior during ESI MS.

PAGE 17

17 CHAPTER 1 INTRODUCTION Electrospray Ionization Mass Spectrometry (ESI MS) Mass Spectrom etry (MS) was introduced some 100 years ago by Thompson [1913], recipient of the 1906 Nobel Prize in physics for th e discovery of the elec tron in 1898. The initial applications of mass spectrometry were carried out by F.W. Aston, a former member of Thompsons research group. Astons work led to the discovery of naturally occurring isotopes and resulted in the awarding of the Nobel Pr ize in chemistry to Aston in 1922. Today, mass spectrometry is used to obtain molecular weight a nd structural information for a diverse range of compounds including peptides, proteins, pharm aceuticals, natural and synthetic products, metabolites, and more. A key step in acquiring a mass spectrum is i onization of the analyte molecules [Vestal, 2001]. For non-volatile biological samples the most prominent ionization methods are matrixassisted laser desorption/ioniza tion (MALDI) [Karas and Hille nkamp, 1988] and electrospray ionization (ESI) [Whitehouse et al., 1985; Fenn et al., 1990]. Both MALDI and ESI allow MS detection of large proteins intact, which was previously nearly impossible to achieve. The convenient coupling to separation techniques and resultant analysis efficiency, particularly in proteome analyses, has given ESI MS a lead ing edge over MALDI MS, since this latter technique requires off-line identification of the analytes [Shen et al., 2005]. The electrospray phenomenon, whereby a sample solution is dispersed into small charged droplets by electrostatic field, was reported as early as 1917 by Zenely [1917]. However, its use to produce gas-phase ions was not demonstrated until about 50 years later by Dole et al. [1968] and Mack et al. [1970]. Fenn et al. [1989] were the first to couple elect rospray ionization with

PAGE 18

18 mass spectrometry (Figure 1-1). For his part in developing electro spray ionization mass spectrometry (ESI MS), Fenn received the N obel Prize in chemistry in 2002 [Fenn, 2003]. Electrochemical Nature of Electrospray Ionization Electrospray ionization is accomplished by appl ying high voltage ( 2-5 kV) to a solution flowing at a slow rate (L/h) through a narrow metal capillary which faces a counter electrode. Electrospray can be generated by applying either a positive or negative vo ltage to the solution, producing either positively charged or negatively ch arged droplets. Girault and coworkers [Rohner et al., 2004] have described three stages of the electrospray process. At first, before any voltage is applie d, the liquid surface in th e capillary has a curved front surface due to surface tension and the hydr ostatic pressure. With application of a positive potential to the capillary, the liquid/air inte rface becomes polarized and the emerging liquid forms a Taylor cone (Figure 1-2) as the electric field drives the cations at the liquid surface away from the capillary. The cone-shaped formation by liqui ds in an electric fi eld was first studied by G. I. Taylor [1964]; hence the de signation Taylor cone. At highe r potentials th e electric field strength overcomes the surface tension and the c one is destabilized, giving way to a liquid jet which breaks off into tiny segments that become charged droplets [Mutoh et al., 1979]. As the solvent evaporates, the droplets shrink, and the charge density at the droplet surface increases. Eventually, each droplet explodes due to Coulombi c repulsion, leading to smaller droplets and ultimately to gas phase ions [Iribane and Thomson, 1976]. The removal of positively charged droplets from the emerging liquid causes charge imbalance in the analyte solution, resulting in electrochemical oxidation; i.e., transfer of electrons from the solution components into the wall of the electrospray needle, in order to maintain ch arge balance in solution. The electrosprayed ions flow continuously under the influence of the electric field across the interface and give rise to electrospray current, iES. The high voltage applied to the

PAGE 19

19 electrospray ion source has only a minimal eff ect on the electrospray current, because the electrical conductivity of air (~3.0 x 10-20 mS/cm or 3 fS/m) [Aplin, 2005] is very small. In fact, once established, the electrospray current is una ffected by further increase in applied, high ESI voltage. Although previously debated [Van Berkel et al., 2000; Van Berk el, Kertesz, 2007], the observed continuous flow of current suggests that the electrospray current is due to Faradaic processes in solution [Jackson and Enke, 1999]. Compounds with the lowest oxidation potential are oxidized first followed by those with highe r oxidation potentials including water (in the carrier solution) and these maintain current at a constant level [Jack son and Enke, 1999]. Figure 1-3 shows a typical current versus applied voltage curve by an elec trospray ion source similar to that of a current-limited device. The actual voltage drop asso ciated with electrochemical processes [Vec = ( i R)ec] at the tip region of the ES capilla ry has been measured by Pozniak and Cole [2004; 2007; Li et al., 2003] as 2.5V vs SHE, which covers the range of most electrochemical oxidation reactions. The migration of charged droplets away from the electrospray capillary (a working electrode) under high applie d ESI voltage leading to reduction of ions at the MS inlet (a counter electrode) makes the electrospr ay interface a special kind of an electrochemical cell and a voltage drop [Vapp = ( i R)gap] is associated with this process [Blades et al., 1991; Van Berkel et al., 1995a; Van Berkel et al., 1995b ; Van Berkel, Kertesz, 2007]. Migration of charged droplets across the ES-interface is analogous to diffusion in a conventional electrochemical cell. The rate of charge neutralization i.e. reduction which occurs at the MS inlet surface (Vinl) upon contact of radially distributed ions with th e inlet surface determines the observed ESI current. Note that this does not include ions that enter the mass spectrometer since they are neutralized at the detector (Vdet). Ions neutralized at the inle t and at the detector represent a total of all preformed ions,

PAGE 20

20 including those formed as adducts and ions produced by oxidation at the tip of the ES capillary, provided no ion loss occurs along the MS transfer optics. Based on this assumption, a complete equivalent electrical circuit repr esentation can be constructed (Fi gure 1-4) and the total voltage in the ES interface (VES) can be given by the following relatio nship [adapted from Jackson and Enke, 1999]: VES = ( i R)ec +( i R)gap + Vinl + Vdet + V (1-1) Direct control of the voltage responsible for electrochemical oxidations at the ES capillary (Vec) is difficult. However, various experiments have been carried out to minimize undesirable electrochemical reactions which occur in the el ectrospray ion source without compromising the controlled-current properties. One way is the us e of fused silica capillary (liner) inside the stainless steel capillary (sleeve), which is connected to the high voltage [Kertesz and Van Berkel, 2001]. Most commonly the end of the silica capillary liner is flush with the stainless steel sleeve or it is slightly protruding to restrict electrochemical processes to the stainless steel rim. When the fused silica capillary was pu lled back by a few millimeters to expose the solution to more stainless steel, electrospray current sustaining re actions such as oxidation of water have been found to increase, indicated by the appearance of higher charge stat es of proteins [Konermann et al., 2001]. Kertesz and Van Berkel controlled und esired electrochemical reactions by using a copper electrospray capillary. This guaranteed that the electrode interface potential (Vec) was held constant around the potential for equilibrium of the corrosion reaction of copper [Kertesz and Van Berkel, 2001]. Using this adaptation, the oxidation of N-phenyl-1,4-phenylenediamine to N-phenyl-1,4-phenylenediimine was inhibited becau se it occurs at a more positive potential than oxidation of copper [Kertesz and Van Berkel, 2001].

PAGE 21

21 Yet another way to control the electrochemical processes at the ES capillary tip is the addition of redox buffering agents such as iodide (I-; Eo (I-/I2) = -0.53V vs SHE) in the form of CsI [Van Berkel et al., 1997] or KI [Konerm ann et al., 2001] to the analyte solution. Redox buffers are electrolytes that are more easily oxidi zed than most analytes. Thus redox buffers play a sacrificial role, thereby preventing analyte oxidation. In other applications, a potentiostat was inco rporated into the ES capillary circuit [Van Berkel et al., 2004; Van Berkel et al., 2005] to control the voltage at the capillary/solution interface in order to study elec trochemical processes occurring during ESI [Van Berkel et al., 2004; Van Berkel et al., 2005]. The earliest reports of incorporati ng a potentiostat into the ESI source are those of Cole and coworkers [Xu et al., 1996; Lu et al., 1997], who used this kind of ESI cell to study oxidation of pol yaromatic hydrocarbons and to determine products of diphenyl sulfide oxidation. Kertesz and Van Berkel [2006] used a porous flow-through working electrode and a quasi-reference electrode, a battery plus a chain of resistors, to design a two-electrode cell [Kertesz and Van Berkel, 2006]. This on-line electr ochemical cell was floated at the high voltage of the electrospray capillary and was used to follow the oxidation processes of methylene blue and reserpine at different current magnitudes [Kertesz and Van Berkel, 2006]. Nyholms group developed a chip for on-line electrochemistry ES I MS (EC/ESI MS) by incorporating an array of gold microcoil electrodes into a po lydimethylsiloxane (PDMS) substrate, which also formed the wall of a microchannel. A graphite tip intergated into the PDMS block served as an ESI emitter [Liljegren et al., 2005]. Control over on-line electroch emical reactions in the chip was gained by: 1) varying the number of turns of the gold coil s to adjust the electroactive surface area, 2) including appropriate insulation, and 3) connecting the chip vo ltage (floated on the high ESI voltage) to a potentiostat. The capability of the microcoil electrode to control electrochemical

PAGE 22

22 rections prior to ESI MS was demonstrated th rough changes in signal intensity of various oxidation products of dopamine. A different approach is used in the presen t study. An electrochemical cell is coupled online with ESI MS to facilitate instead of suppressing, electroc hemical reactions in order to enhance ionization efficiency, particularly fo r neutral and negatively charged analytes. The system (described in detail in Chapter 2) includes a variable resistor to control voltage applied to the electrochemical cell. The orif ice of the MS capillary inlet has been widened into a coneshape to further augment the improvements in sensitivity due to the applied EC cell voltage. Advantages derived from potential control of this EC/ESI MS syst em are demonstrated in this work. On-Line Electrochemistry Mass Sp ectrometry (EC/MS) Electrochemistry plays an important ro le in quantitative determinations, and electroanalytical studie s of electroactive metabolites and drugs in biological systems are necessary for better understandi ng of disease and drug developmen t. Electrochemical processes are often studied in aprotic solven ts, which are ideal for fundamental investigations, as well as in aqueous solutions, which allow evaluation of the behavior of diffe rent reaction products generated in biological systems. Furthermore, electrochemistry is relatively easy to control via external voltage. For these reasons, a number of electrochemical methods, including biosensors, have been developed and are used extensively in bioanalytical research. The bioanalytical utility of electrochemistry is further a dvanced when this technique is coupled to mass spectrometry, a universal detection system, in on-line electr ochemistry mass spectrometry (EC/MS) [Lohman and Karst, 2008]. As described herein and else where [Van Berkel, 2004; Diehl and Karst, 2002], EC/MS is a rapidly de veloping technique.

PAGE 23

23 As described above, on-line electrochemistry has been incorporated into ESI MS to control electrochemical processes associated with electr ospray ionization. But on-line electrochemistry mass spectrometry (EC/MS) has also been developed as an analytical tec hnique in its own right [Volk et al., 1992]. Development of EC/MS st arted in the 1970s, when semi-permeable membranes were used as interfaces between th e electrochemical cell and the mass spectrometer [Bruckenstein and Gadde, 1971; Brockman and A nderson, 1984]. At that time, the utility of EC/MS was limited only to those reduction-oxidati on reactions that produce gaseous and volatile products which could pass through the membranes and enter the mass spectrometer. This was a severe limitation, because most electrochemical processes produce non-volatile products which remain in solution. One of the major challenges, perhaps the greatest, in the development of EC/MS was dealing with the liquid solvent cont aining the analyte prio r to performing mass spectrometry which is a low-pressure gas phase technique. The same challenge confronted the developers of liquid chromatography-mass spect rometry (LC/MS), who designed several types of interfaces, including thermospray, which was i nvented in 1983 by Blackley and Vestal [1983], particle beams, and electrospray. All three of thes e interfaces have also been successfully applied to EC/MS. Figure 1-5 is a diagram of a typical thermosp ray interface. Resistiv e heating of a metal capillary leads to rapid volat ilization of the analyte solution flowing through it. Analyte molecules are subsequently ionized by electron impact and directed towards the mass spectrometer by electrostatic repulsion from th e repeller electrode while neutral solvent molecules are pumped away. A typical particle beam interface is shown in Figure 1-6. A flow of helium concentric to the LC outlet generates an aerosol/spray by nebulization. Upon exiting the aerosol/spray

PAGE 24

24 chamber, the lighter solvent molecules and helium drift outward from the concentric center and are pumped away, leaving heavier analyte molecule beam which enters the ionization chamber. This is the point in particle beam ionization process referred to as the momentum separation stage. Either electron impact (E I) or chemical ionization (CI) methods can be used to generate charged analyte species which are directed to the mass spectrometer. The development of the particle beam interface, formerly called a m onodisperse aerosol genera tion interface (MAGIC), was described by Willoughby and Browner [1984; 1988] and has also been reviewed by Creaser and Stygal [1993], and Capiello and Bruner [1993]. On-Line Electrochemical Cell Designs Volk et al. [ 1992] reviewed on-line EC/MS in strumentation available prior to 1992, when electrospray ionization was yet developing, for monitoring reactants, shor t-lived intermediates and products of electrochemical reactions of bi ologically active molecules as a function of electrode potential. Thermospray Interface: On-line electrochemistry-thermospray ionization mass spectrometry (EC/TSI MS) was first designed in 1986 by Hambitzer an d Heitbaum [1986] who investigated the oxidatio n of dialkylanalines. Their EC/TSI MS system included an HPLC pump which was used to pump the carrier solution from the reservoir and to force the analyte solution out of the three-electrode thin-l ayer electrochemical cell through the working electrode placed at the cell outlet (Figure 1-7). The analyte solution was then introduced into the thermospray ion source. Following Hambitzer and Heitbaums successful demonstration of EC/TSI MS, Yost, Brajter-Toth and Volk [1986; 1990] reported the mechanism of oxida tion of purines and thiopurines studied using EC/T SI MS/MS. This group used a commercially available threeelectrode flow-through cell with porous reticula ted vitreous carbon working electrode and

PAGE 25

25 palladium reference and counter el ectrodes. A similar electrochemical cell was used by Getek et al. [1989] to investigate on-line electroche mical reactions producing the acetaminophenglutathione conjugate. Similar fl ow cells for on-line EC MS are commercially available from companies such as ESA Inc. and Antec Leyden, Netherlands. Particle Beam Interface: The first demonstration of online electrochemistry particle beam ionization mass spectrometry (EC/PBI MS ) was reported by Regino and Brajter-Toth [1997] who developed a homemade thin-layer, flow-through electrochemi cal cell that could withstand the high back pressure s encountered in PBI MS (Figure 1-8). Their electrochemical cell could also be used with TSI MS and had removable electrodes for surface cleaning and/or modification. The conversion efficiency of the cell, as well as the sensitivity and reproducibility of the EC/PBI MS were found to be affect ed by solvent composition, including aqueous-toorganic solvent ratio and supporti ng electrolyte. A significant e nhancement of the signal was observed for triphenylamine when tetrabutylammonium perchlorate was used as an electrolyte [Zhang and Brajter-Toth, 2000]. Electrospray Interface: While well established and commercialized (e.g. by Vestec), thermospray and particle beam techniques ar e not as popular as el ectrospray for EC/MS interfacing today. The advantages of on-line electrochemistry electrospray ionization mass spectrometry (EC/ESI MS) were re alized in 1995 by Bond et al. [1995] who used a simple flow through tubular two-electrode EC cell on-line with ESI MS to study metal diethylcarbamate complexes. That same year, Zhou and Van Berkel [1995] described three different types of online EC cells, which include a tubul ar two-electrode cell which is floated at the ESI high voltage. Working electrodes included the ES capillary emitt er, as well as the ESI-decoupled thin-layer and porous working electrode type s, both of which require a poten tiostat. As described above,

PAGE 26

26 Xu et al. [1996] and Lu et al. [1997] designed an EC/ESI MS system w ith three-electrode EC cell incorporated into the ES em itter, and they demonstrated its performance using polyaromatic hydrocarbons, diphenyl sulfide and nitrobenzene. Zhang, Brajter-Toth and co-workers designed a two-configurations-in-one on-li ne EC cell using a polyetheret herketone (PEEK) 4-way channel (Figure 1-9) [Zhang et al., 2002]. Two palladium electrodes were mounted along one axis of the PEEK cross, with stainless stee l (ss) capillaries along the other axis. The cell operated as a thinlayer configuration when Pd-elect rodes were used and as a tubul ar two-electrode configuration when the stainless steel capillaries were us ed. Using the latter ce ll configuration with triphenylamine as analyte, this group demonstrat ed the enhancement in sensitivity of ESI MS by on-line electrochemical reactions [Zhang et al., 2 002]. Karsts review of applications reported prior to 2004 [Karst, 2004], descri bes clearly the diversity and s uperior information content of EC/ESI MS in metabolic research. Control of EC Cell Potential in EC/ESI MS It is d emonstrated in this work that a variable resistor can be incorporated into the EC cell circuit portion of the EC/ESI MS system to adjust voltages app lied to the EC cell from a 9 V battery. Applied EC cell voltage enhances analyte ionization efficiency in EC/ESI MS by providing electrochemical reactions in addition to those occurr ing at the ES capillary tip. Coupling of the electrochemical cell to ESI MS expands the cove rage of this technique to include formerly neutral and negatively charge d analytes as discusse d above. Furthermore, electrochemistry coupled on-line to ESI MS diversifies the range of ESI MS applications to studies of electrochemical react ions of many analytes of pharm aceutical and biological interest. The need to acquire insight into biological oxida tion reactions and to discover different pathways related to normal growth, aging and disease, wh ich are the hall-marks of metabolomics, calls for

PAGE 27

27 techniques such as EC/ESI MS, which can mimic enzyme-catalyzed oxidation reactions [Permentier et al., 2008]. The EC/ESI MS of Dopamine The performance of the EC/ESI MS designs has been evaluated using dopam ine as the model analyte, both in previous reports [Deng and Van Berkel, 1999; Lilj egren et al., 2005] and in this research. Dopamine (3,4-dihydroxyphenethylamine, DA) ha s three dissociable protons (pKa1, pKa2 and pKa3 of 8.9, 10.6 and 12.1, respectively) asso ciated with the deprotonation reactions (Scheme 1-1). A distribution diagram of the different DA species (Figure 1-10) shows that DA exists as the H3DA+ ion (referred to as [DA]+ in this thesis) in acidic to neutral solutions [Sanchez-Rivera et al., 2003]. Scheme 1-1. Three proton dissociations of DA [Sanchez-Rivera et al., 2003]. The oxidation potential of DA (E0 (DA) = -0.12V vs SHE) [Blank et al., 1976] is lower than E0 of most metabolites. Being a preformed ion and having a low Eo value, which indicate relatively easy ESI MS detec tion and oxidation, have made DA a suitable model analyte for testing new EC/ESI MS systems [Liljegren et al ., 2005; Deng, Van Berkel, 1999; Mautjana et al., 2008a]. Oxidation of DA in EC/MS systems produces 1e-, 1H+ and 2e-, 2H+ oxidation products. Endogenous DA functions as a neurotransmitter in the nervous system. Oxidation of DA in vivo by loss of 2eand 2H+ to produce dopamine quinone (DAQ) has been associated with the development of Parkinsons disease [Spina et al., 1989]. It has been proposed that DAQ can

PAGE 28

28 modify essential proteins through electrophilic addition at the sulfhydryl group of cysteine residues, leading to dopaminergic neuron deaths and a neurodegenerative condition [Whitehead et al., 2001; LaVoie and Hastings, 1999]. The electrophilic addition of quinones to thiols was first proposed by Dryhurst and co-workers [S hen et al., 1996; Shen and Dryhurst, 1996a,b], based on cyclic voltammetry in conjunction with off-line nuclear magnetic resonance (NMR) spectrocopy for structure elucida tion. In the present researc h, the use of EC/ESI MS to investigate DA oxidation and the addition reaction of DAQ to thiols provides direct evidence for this reaction via exact mass determination of electrochemically generated products, and observation of finer mechanistic details. The EC/ESI MS of Uric Acid Another im portant test compound for EC/ESI MS is uric acid which exists in two tautomeric forms, the ketoand enol-forms (Scheme 1-2) [Simic and Javanovic, 1989]. The ketoform is predominant at very low pH values, wh ile the enol-form becomes more predominant as the pH approaches pKa1 (= 5.4) [Allen et al., 2004]. HN N H O O N H H N O 3 2 4 5 8 9 6 7 1HN N H O O N H N OH 3 2 4 5 8 9 6 7 Keto Enol Scheme 1-2. Tautomeric structures of uric acid It has been proposed that upon one-electron ox idation, uric acid lose s a proton from the hydroxyl group, which is more acidic than the sec ondary amine group, to form a neutral radical

PAGE 29

29 [Simic and Javanovic, 1989] (Scheme 1-3). The lone electron in uric acid radical is therefore centered on the oxygen bonded to C8. HN N H O O N H N O 3 2 4 5 8 9 6 7HN N H O O N H N OH 3 2 4 5 8 9 6 7 1,-H -1e 1HU H2U Eo = 0.59V vs SHE Scheme 1-3. Proposed one-electr on, one-proton oxidation of uric ac id [Buettner and Jerkiewicz, 1996]. One-electron oxidation pathways have also been suggested by Ames et al. [1981] in their proposal that uric acid is an antioxidant. The experiments carried out by Ames et al. [1981] included purging a uric acid so lution with singlet oxygen prepared by irradiation of oxygen gas with a tungsten/halogen lamp. Th eir HPLC results showed that singlet oxygen oxidized urate, and they observed the same behavior for the re action of urate with hydr oxyl radicals produced from gamma ( )-irradiated water. Uric acid radicals were detected by Maples and Mason [1988] using electron spin resonance spectroscopy as well as by Simic and Javanovic [1989] who used pulse radiolysis to elucidate the antioxidant m echanism. The present work demonstrates that positive ion mode ESI MS is a radiation-free alte rnative to these techniques for investigating pathway mechanisms of metabolites with low E0 which undergo 1e-, 1H+ oxidation during ESI. While difficulties in detection of uric acid by st andard ESI MS often leads to its omission in purine assays [La Marca et al., 2006 ; Ito et al., 2000], the EC/ESI MS system developed in this

PAGE 30

30 work produces uric acid mass spectra with in tense peaks at m/z 169 and 337, due to proton adducts of uric acid and uric acid dime r formed through electrochemical reactions. In man and other primates, uric acid is the final product of the breakdown of energycarrying molecules, adenosine triphosphate (ATP) and guanosine triphosphate (GTP) [Becerra and Lazcano, 1998] (Scheme 1-4). Wh ile there are salvage pathways for other purines (adenine, hypoxanthine, and guanine), excess uric acid is removed from the body by excretion through the kidneys and intestines. IMP XMP AMP GMP Hypoxanthine Xanthine Uric Acid Inosine Guanosine Adenosine Adenine Guanine Xanthosine GTP ATP Scheme 1-4. Metabolic pathway of purines in humans (ATP, adenosine triphosphate; GTP, guanosine triphosphate; AMP, adenosine monophosphate; IMP, inosine monophosphate; XMP, xanthosine monophosphate; GMP, guanosine monophosphate). The EC/ESI MS of Thiols The low m olecular weight thiol compounds, gl utathione (GSH), cy steine (CySH) and homocysteine (hCySH) are involved in many impo rtant physiological and pathological processes [Rahman et al., 2005]. GSH and CySH are found at millimolar concentrations in different types

PAGE 31

31 of cells and in blood plasma. Physiological f unctions of endogenous thiols include protecting cells against reactive oxygen species (ROS) a nd reactive electrophiles [Forman and Dickinson, 2003]. Decrease in concentrations of GSH and Cy SH from normal (1 10 mM) correlate with numerous disease conditions including Parkin sons and Alzheimers diseases [Shen and Dryhurst, 2001]. Abnormally high concentrations of hCySH, which is an intermediate in the metabolism of methionine to CySH [Nekrass ova et al., 2003; Himmelfarb et al., 2002], a precursor for GSH, have been found to correlat e with atherosclerosis and venous thrombosis [Demuth et al., 2002; Van den Brandhof et al. 2001] However, the mechanism of vascular injury is still unknown and could possibly involve GSH and CySH, given the common biosynthetic pathway of these three thiols (Scheme 1-5) [Himmelfarb et al ., 2002]. Simultaneous analysis of thiols in biological samples is therefore importa nt for clinical applications, as well as for understanding their ro les in physiology. Using ESI MS and EC/ESI MS to detect thiols allows direct and si multaneous detection of all three thiols, which is challenging when using electrochemical methods, given similar oxidation potentials, or using liquid chromat ography with UV or fluorescence detectors which require derivatization because thiols do not pos sess any chromopores in their sructures. Study Overview This thes is investigates the sensitivity asp ects of positive ion mode electrospray ionization mass spectrometry (ESI MS), including on-line electrochemistry(EC/ ESI MS), focusing on uric acid, thiols and other purine metabolites, which ar e clinically important metabolites but are not reliably detected by positive mode ESI MS. Th e electrochemical processes inherent to the electrospray ionization operation are described in relation to thei r effects on the detection of easily oxidized analytes. The application of el ectrochemistry coupled on-line with ESI MS

PAGE 32

32 Scheme 1-5. Biosynthesis of homocysteine, cysteine and glutathione [adapted from Himmelfarb et al., 2002] (EC/ESI MS) to the analysis of the abovementi oned analytes and the resulting enhancement in sensitivity is demonstrated.This chapter has pr ovided background information about ESI MS and the attempts towards gaining control of the electrochemical processes associated with electrospray ionization. Efforts to gain control over these processes culminated into on-line electrochemistry, which has become a technique in its own right. Chapter 2 explains the experimental methods and instrumentation. Detailed schematics are given and fundametals of the techniques used ar e described to help the reader appreciate the science behind experiments conducted in this rese arch. Chapter 3 describes the performance of EC/ESI MS at different flow rates, concentra tions and applied EC cell voltages, using dopamine as the test compound. Oxidation pro cesses that are all too evident are explained in terms of ion formation pathways.

PAGE 33

33 Oxidation and sensitivity of uric acid in ESI MS is the subject of Chapter 4. In this chapter factors involved in the detection of uric ac id are discussed and co rresponding changes in observed mass spectra are explained. Results fo r EC/ESI MS of three thiol metabolites, glutathione, cysteine and ho moxysteine, discussed in Chapter 5, show that ESI mass spectrometry is more selective than any of the other commonly used methods. Chapter 6 discusses ESI MS of other purines (guanine, adenine, hypoxanthin e and xanthine) and mechanisms of oxidation of these purine metabolites during ESI MS are proposed. Chapter 7 summarizes conclusions drawn from the research re sults of the entire thesis and proposes ideas for future work.

PAGE 34

34 Figure 1-1. Electrospray Ionization Interface fo r LC/MS [Adapted from Fenn et al., 1989] electrons Figure 1-2. Electrospray ionization proce ss [Adapted from Cech and Enke, 2001]

PAGE 35

35 Figure 1-3. Current vs voltage cu rve for a current-limited device A VinlSolution Flow Vapp+ InletDetector Vdet ( i R)ec ( i R)gap e Figure 1-4. Equivalent el ectric circuit representa tion of the ESI process Normal ESI operating range E (volts) i (A) 1 2 3 corona discharge

PAGE 36

36 Pump Solvent Ion Beam To the Mass Spectrometer Repeller Electrode Electron Filament Heated Capillary AnalyteSolution Heated Spray Chamber Figure 1-5. Thermospray interface for LC/MS Figure 1-6. Particle beam interface for LC/MS

PAGE 37

37 Figure 1-7. On-line electrochemical cell for EC/TSI MS [Adapted from Hambitzer and Heitbaum, 1986] Figure 1-8. On-line electrochemical cell for EC/PBI MS and EC/TSI MS [Adapted from Regino and Brajter-Toth, 1997] WORKING ELECTR ODE REFERENCE ELECTRODE AUXILIARY ELECTRODE Counter Electrode Reference Electrode From HPLC Pump To thermospray Chamber Inlet Working Electrode

PAGE 38

38 Figure 1-9. On-line electrochemical cell for EC/ESI MS (ss, st ainless steel capillary; Pd, palladium electrode) [Adapted from Zhang et al., 2002] Figure 1-10. Distribution diagram of different dopamine species as a function of pH [Adapted from Sanchez-Rivera et al., 2003] ss Pd Pd ss ES needle To the mass spectrometer Flow dc supply Or [DA]+

PAGE 39

39 CHAPTER 2 EXPERIMENTAL Methods and Instrumentation Construction of the EC/ESI MS System An on-line electrochem istry/electrospray io nization Fourier transform ion cyclotron resonance mass spectrometry (EC/ESI FT-ICR MS) sy stem was developed in this work, with the EC cell integrated into the ion source (Figure 21). A simple series, two-electrode EC cell circuit was incorporated into the ES emitter by dividing th e stainless steel ES capill ary (80 m i.d.) into two (4 and 5 cm long) sections (ss1 and ss2). Thes e were joined with plastic tubing (100 m i.d.) approximately 0.5 mm in length. A 9V battery wa s connected via a variable resistor (50 k max.) to the two stainless steel ES capillaries to supply low, controlled, varying dc voltages. This produced an electrochemical cell with ss1 as the anode (working electrode) and ss2 as the cathode (counter electrode) a nd total cell volume (left end of ss2 to the ESI gap) of approximately 0.46 L. The EC cell was fixed onto an adjustable xyz -micropositioner (World Precision Instruments, Sarasota, FL, USA). The electri cal circuit of the entire system is shown in Figure 2-2. There is no upstream grounding point (i.e. between the counter electrode and the syringe). With the long plastic tubing (~140 mm) between the syringe and ss2, and without upstream ground, electrochemical oxidations cannot occur at the counter electrode but are limited to the working electrode [V an Berkel and Kertesz, 2007]. The resistors R1 (50 K max) and R2 (10 K max) are commercially available, standard linear track variable resistors wi th manually movable (via a spindl e) third contact (Figure 2-3). Connected as shown in Figure 2-2, resistors R1 and R2 form a two-turn rheostat for rough and fine adjustment, respectively, of the voltage applied to the on-line electrochemical cell. The voltage drop across the electrochemical cell (between counter and working electrodes) is

PAGE 40

40 represented as ( i R)ec online, and that due to solution at the tip of the electrospray capillary is ( i R)ec, while ( i R)gap represents the voltage drop due to the air gap. Charge neutralization at the inlet and detector are represented by Vinl and Vdet, respectively. Jackson and E nke [1999] found that the amount of current flowing from the electrospray capillary back to the syringe during positive ion mode ESI MS is very small and negligible in theoretical equations for modeling the electrical behavior of ESI MS. Cone-Shaped MS Capillary Inlet A cone-shaped MS capillary inlet (Figure 21A) which is sim ilar to the flared inlet introduced by Wu et al. [2006], was formed by r eaming out about 25 mm depth of the orifice of the standard cylindrical inlet. This produced an orifice diameter of 6.12 mm, which tapers to 1.58 mm with a cone angle of 50o. As described belo w, the results obtained with the cone-shaped inlet were compared to those of a standard cyli ndrical metal capillary MS inlet with 1.58 mm i.d. Both metal capillary inlets are made of brass. Fourier Transform Ion Cyclotron Resonance (FT-ICR) Mass Spectrometry An APEX 4.7-T FT-ICR m ass spectrometer (Bruker Daltonics, Billerica, MA, USA) was used with an operating ESI ion source region pre ssure of ~5.3 x 10-6 mbar. The other parameters included the ion transfer region and the ICR cell pressure ~3.4 x 10-10 mbar, needle voltage ~3 kV, and capillary inlet temperature of 120oC. A syringe pump (Cole Parmer 74900, Vernon Hills, IL, USA) provided constant flow rate of ca rrier solution continuously into the electrospray. Mass spectra for comparison of the cylindrical and cone-shaped inlets were acquired using identical settings for transfer of the ions through the optics in the external source and for the ICR detector. Operational Safety Safety protocols for high voltage and superconducting m agnets were observed.

PAGE 41

41 Cyclic Voltammetry A Bio-Analytical System s (BAS-100) electroc hemical analyzer (Bioanalytical Systems, Inc., West Lafayette, IN, USA) was used for slow scan (50 mV/s) cyclic voltammetry. A homemade stainless steel disk electrode (r = 50.8 m) was used as the working electrode, with a standard calomel electrod e (SCE) as reference. Electrode Fabrication: Stainle ss steel microelectrodes were fabricated according to the method reported by Bravo et al. [1998]. The stainle ss steel wire was inserted into a micropipette tip and sealed using an epoxy mixture of Sh ell Epon 828 resin (Miller-Stephenson Chemicals, Danbury, CT, USA) and m-phenylenediamine hard ener (Miller-Stephenson Chemicals, Danbury, CT, USA). The mixture was heated at 70 oC in a water-bath until the epoxy was liquid and transparent. The liquid epoxy was transferred into the micropipette tip with embedded stainless steel wire, and the electrode was le ft to dry for 72 hrs and then cured in an oven at 150 oC for 1h. After curing, the electro de tip was sanded using an Ecomet I polishing wheel (Buehler, Evaston. IL, USA) with 600-grit silicon carbide paper, followed by an Alpha-A felt cloth (Mark V Labortory, East Granby, CT, USA) for smooth-polishing. Fundamentals of Methods Used ESI MS The flow of ions generating the ES current ( iES) depends on the potential difference ( E ) between the ES capillary and the MS in let, the carrier so lution conductivity ( ), surface tension ( ), flow rate ( ), permittivity ( ); and on permittivity of the vacuum ( o) according to Equation 2-1 [Rohner et al., 2004]. iES = [(4 / )3(9 )2o 5]( E )3/7( )4/7 (2-1)

PAGE 42

42 When droplet sizes and charges are considered, the effect of the potential difference is negligibly small compared to solution conductivity, and E can be neglected leading to Equation 2-2 [de la Mora, 1992] iES = ( r)( r)1/2 (2-2) where ( r) is a function of the dielectric constant ( ( r) 18 for water and methanol whose dielectric constant, r 30). Because the resistance of th e gap between the ES capillary and the MS inlet is very large [Apli n, 2005] and the observed current ( iES) is constant, it is apparent that the observed current ( iES) is mostly due to the Faradaic processes. Part of the electrospray current ( iES) is due to preformed ions which are simply repelled at the ES capillary tip and migrate to the MS inlet. This fraction is negligible because of the weak influence of the electric field. Therefore the Faradaic current ( iF) at the ES capillary tip, which is sustained by oxidation of compounds with lowest oxidation potentials (Eo) first, and progressively by those with higher Eo values, can be assumed approximately equal to the electrospray current ( iES): iF = iES (2-3) The diffusion controlled Faradaic cu rrent of reactants is given by iF= F jnjcjmj (2-4) where F is the Faraday constant (9.6485 x 104 C/mol), j the number of different electrolysis reactions that occur, nj the number of electrons involved in the producti on of one mole of electrolysis product in reaction j cj the concentration of the reduced species and mj the mass transfer (diffusion) coefficient of the reduced species in reaction j [Van Berkel and Kertesz, 2007; Rohner et al., 2004]. The equivalence of Faradaic and observed electrospray currents stated in Equation 2-3 was verified by Blades et al. [1991] who measured iES for electrochemically generated Zn2+ and found it to be equal to that from a calibration of Zn2+

PAGE 43

43 standards. However, it is difficult to determin e the exact magnitude of the potential of the capillary electrode, because it is not fixed during the ESI MS expe riment, but it adjusts according to the number of interactive variables to maina tin the required current [Van Berkel and Zhou, 1995]. The concentration of excess charge in the ES droplets [Q] (mol/L), which leads to coulombic explosion and release of ions into the gas phase, can be calculated from Faradays first law of electrolysis [Van Berkel and Kertesz, 2007, R ohner et al., 2004]: [Q] = iF/(F ) (2-5) Therefore, the largest coulombic explosions occu r in experiments where the lowest flow rate ( ) of carrier solution is used. In addition to the parameters already menti oned, surface activity, proton affinity (PA) and the electrochemical properties of analytes, can also influence the number of ions produced in positive ion mode ESI and the detection sensitivit y. Surface activity refers to the tendency of the analyte to migrate from the droplet center to the surface, where it can easily enter the gas phase. Relatively non-polar, hydrophobic metabolites show a greater degree of surface activity than polar, hydrophilic compounds. Polar metabolites generally acquire charge by forming H+-, Na+-, K+-, Li+-, or NH4 +-adducts in solution and they ente r the gas phase as adduct ions. Proton affinity (PA) refers to the favorable enthalpy change (negative H) for protonation of a given metabolite in the gas phase. In the ga s phase, molecules with high PA, which were not already protonated in solution, can abstract protons from soluti on-protonated molecules with low PA. Analytes of interest can be detected reliably if they are dissolved in solvents with lower PA values [Cech and Enke, 2001]. Analytes that are neutral or even negatively charged but have

PAGE 44

44 relatively low oxidation potentials (Eo) can be oxidized in positiv e ion mode ESI MS and form species that give a respon se [Mautjana et al., 2008b]. MS capillary inlet design: For an MS inlet capillary with a small orifice, uniform ion intensities for metabolites with different PAs an d surface activities are obtained using very low flow rates ( L/h) and narrow ES capillaries to promote formation of small droplets. Efficient ion desolvation is achieved by heating th e MS inlet capillary to about 120 oC for efficient evaporation of the carrier solu tion, which often includes water. However, large losses in ion transmission can occur at the MS capilla ry inlet because of the small area (internal diameter) of the orifice of the inlet relative to the size of the charged droplet/ion plume produced by electrospray. Internal diameters of standard MS capillary inlets are kept small to preserve low pressure (vacuum ) conditions. But there is no observable effect on the pressure (no additional vacuum pumping is requi red) if the inlet capillary orifice is enlarged into a cone (Figure 2-1A) [Mautjana et al., 2008 a,b] or has a flared orifice [Wu et al., 2006]. Under conditions where positive ions are electrost atically repelled from the ES capillary tip rather than attracted to the capi llary inlet surface (since the inlet is held at a lower, but still positive, potential), a wide orifice allows collecti on of radially distributed ions [Nemes et al., 2007] and increases ESI MS sensitivity several fold. In contrast to other ion transmission enhancing approaches, such as focusing the ES dropl et/ion plume with static dc fields at the inlet [Zhou et al., 2006; Schneider et al., 2001] or adding electrod ynamic funnel at the skimmer position [Kelly et al., 2007; Kim et al., 2000], the c one-shaped capillary inlet is easy to make and use. EC/ESI MS As a result of the applied EC cell voltage, other electro chemical reactions occur, in addition to those at the tip of the electrospray capillary (Figure 2-8). Cole s group has shown that

PAGE 45

45 the electrode-solution interface po tential is highest at the tip of the electrospray capillary [Pozniak and Cole, 2007; Pozniak and Cole, 2004]. Their measurements indicate that voltage decreases exponentially from highest ( 2.5 V vs SHE) at the ES capillary tip ( 0.25mm from the ES capillary tip opening) and approaches zero in the upstream direction, so that much of the inner surface of the ES capillary is not elec trochemically active during ESI. Application of voltage to the other end of the cap illary (left end of ss1 in Figur e 2-1) increases the active inner surface area of the ES capillary (s ee Figure 2-4), as indicated by th e increase in ES current as applied EC cell voltage is increased [Mautjana et al., 2008a]. It has also been pointed out in a recent review [Van Berkel and Kertesz, 2007] th at the magnitude of pot ential reached at all points along the ES capillary is dir ectly related to the ES current. FT-ICR Mass Spectrometry Pioneered by Marshall et al. [1974] a Fourier transfor m ion cyclotr on resonance (FT-ICR) mass spectrometer uses combined magnetic and elec tric fields to determine the mass-to-charge (m/z) ratio of an ion. Ions in an FT-ICR mass spectrometer possess very low kinetic energy and, unlike the case in the magnetic sector mass spectrometer, they do not pass through the magnetic field but are actually trapped in the magnetic fiel d. When entering a region of constant magnetic field strength in a direction that is perpendicular to the field, the i on path is bent into a circle, a motion referred to as cyclotron motio n and mathematically described by c = qB/m (2-6) where c is the angular frequency, q the charge, B the magnetic field and m is the mass of the ion. Ion cyclotron motion is spatially confined and thus made observable by applying an electric field at frequency (rf) that is the same as (or resonating with) the ion cyclotron frequency. Application of rf excitation field cau ses the ions to spiral outward to larger orbits, and all ions of a given mass move coherently (as a clump). Di stinct ion oscillation (or image) currents are

PAGE 46

46 induced at (and therefore detected by) the plates of the cell (Figure 2-5) generating the ICR (time domain) signal whose digital form is converted by Fourier transformation into a frequencydomain spectrum and further (using the magnetic field strength value) to a mass spectrum [Grosshans and Marshall, 1991; Marshall et al ., 1998; Baykut and Eyler, 1986]. The high precision of digital frequency measurement, in c onjunction with large magnetic field strengths (> 7 Tesla), results in very high mass resolution (1 0000+), a very desirable and important property of FT-ICR MS. Furthermore, because the image cu rrent is induced as long as the ions are moving coherently, the frequencies can be m easured hundreds of times, thereby improving the signal-to-noise (S/N) ratio. Hydrogen/Deuterium (H/D ) Exch ange Methods Hydrogen/deuterium exchange methods are used for distinguishing covalently bonded species from non-covalently bonded structures and for elucidation of the formation of complexes in solution. Solution phase H/D exchange is us ed for structures with hydrogen-bond donor and acceptor groups. The donor group usually contains oxygen (O), nitrogen (N), sulfur (S) or a strong electron-withdrawing group to which the hydrogen (denoted H in H/D exchange terms) is attached. If an active H forms an H-bond either w ith another atom or group in the same molecule (i.e intra-molecular H-bonding) or with an atom or goup of another molecule forming a complex, that H is protected and will not be replaced by deuterium [Liu et al., 2008; Jiang et al., 2007]. Generally, active (or acidic, or labile) hydroge ns that are bonded to heteroatoms (not carbon) in organic molecules are replaced by deuterium when exposed to high deuterium concentrations (e.g. in D2O). The reaction can be written as R-XH + D2O R-XD +DOH (2-7)

PAGE 47

47 where R represents a partial molecular structur e, X represents oxygen, nitrogen or sulfur, H is hydrogen and D, deuterium. When n active hydrogens (n x 1Da) of a given molecule are exchanged with n deuterium atoms ( n x 2Da), the molecular mass increases by n. The increase in mass due to H/D exchange corresponds to the number of exchageable hydrogens, because the neutral molecule Mn H (where nH represents the number of exchangeable hydrogens) becomes Mn D after the exchange. In ESI MS, residual Mn H and Mn D form proton adducts (given the presence of acid electroly tes, e.g. acetic acid, or H+-producing solvent oxidation during ESI). The relative intensities of [Mn H+H]+ and n[MD+H]+ ions reflect molar fractions of molecules with exchanged hydrogens [Liu et al., 2008; Jiang et al., 2007]. Tandem Mass Spectrometry (MS/MS or MSn) Tandem mass spectrometry (MS/MS or MSn) involves at least two stages of mass spectrometry. The ion of interest is isolated in th e first MS. It is then activated to the point of dissociation either by collisions with neutral atoms (collisionactivated dissociation CAD or collision-induced dissociation CID) or by irra diation with a laser (e .g. 10.6 m photons from a CO2 laser; infrared multiple photon dissociation IRMPD). Activation results in secondary (or daughter) ions which are analyzed by the second MS stage to identify the primary (parent) ion. The generation of secondary ions is represented by mp + md + + mn (2-8) where mp + is the primary (or parent) ion, md + the secondary (or daughter) ion, and mn is a neutral fragment. IRMPD has the advantage of being vacuum-friendly which allows immediate FT-ICR detection of the secondary ions Other activation techniques in clude sustained off-resonance irradiation (SORI) where selected ions are irradiated off-reso nance (i.e. at a different, nonresonating frequency) causing them to collide as they spiral away from and back into orbit, and

PAGE 48

48 black-body infrared dissociation (BIRD) in which ions are confined in a hot (40 215oC) ICR cell for 10 1000 s, during which time they ab sorb black-body photons, are excited, and undergo unimolecular dissociation [Price et al., 1996; Glish and Vachet, 2003; Marshall, 1998]. Cyclic Voltammetry In cyclic voltamm etry, the current is monitore d as potential is sca nned linearly in the forward and reverse directions repetitively, thus following a triangular waveform in time, typically with sub-second scan cycles. A potential (E) scan towards more positive values results in oxidation of the analyte which displays an anodic (oxidation) peak current ( ipa) at Epa. Reverse potential scan, towards less positive values, resu lts in a cathodic (reduction) peak current ( ipc) at Epc, provided the electrochemical pr ocess is reversible. A typical cyclic voltammogram is shown in Figure 2-8. For reversible reactions ipa = ipc and the potentials corres ponding to the peaks are related by Epa Epc = 2.22RT/nF (2-9) where n is the number of electrons in each (either oxidation or redu ction) half-reaction. For a reversible reaction at 25 oC, the peak current is given by ip = (2.69 x 108)n3/2AcD1/21/2 (2-10) whereas for electrochemically irre versible (i.e very slow) reacti ons peak current is given by ip = (2.99 x 108)n(na)1/2AcD1/21/2 (2-11) where A is electrode area (m2), c is concentration (mol/L), D and are the diffusion coefficient (m2/s) and the transfer coefficient of the analyte (unitless), respectively, is the scan rate (V/s), and na is the number of electrons in the rate determining step [Bard and Faulkner, 2001]. As Equations 2-10 and 2-11 show, peak current is proportional to 1/2.

PAGE 49

49 Cyclic voltammetry is used both for quantit ative determination of analytes and for studying the kinetics of reactions occurring at the electrode surface. The use of microelectrodes, which have very small active surface area and very small iR drops, is advantageous for analysis of samples with poor conductivity, including no naqueous solutions. Furthermore, the low capacitance associated with the very small su rface areas allows very small charging currents relative to the Faradaic curre nts of many oxidation reactions resulting in significant improvement in limits of detecti on [Harris, 2007; Willard et al., 1988]. Experimental Conditions Solution Preparation All solutions were prepared and stored in glass volum etric flasks. When stored in plasticware, solutions produced mass spect ra with numerous unidentifiable peaks. Reagents: Dopamine hydrochloride (98%), sodium dihydrogen phosphate (99%), uric acid (H2U) (99%), glutathione (99%), L-cysteine hydr ochloride (98%), DL-homocysteine (95%), and disodium hydrogen phosphate (99%) were purchas ed from Aldrich (St. Louis, MO, USA); ammonium acetate (NH4Ac), glacial acetic acid (HAc) (99.9%), KOH (88%)and methanol (HPLC grade) were from Fisher (Pittsburgh, PA, USA); deuterium oxide (99.9%) was from Cambridge Isotope Laboratories (Andover, MA). All chemicals were used as received. Dopamine: The carrier solution was 50/1/49vol%, water/acetic aci d/methanol with a pH of 4.2 (by a pH-meter). Dopamine solutions were prepared in the carrier solution unless specified otherwise. Specific conductivity was measured with a conductivity meter (Wissenschaftlich Technise Werkstatten, Weilheim, Germany) as 103 and 312 S cm-1 for the carrier solution and 2.5 mM dopamine solutions, respectively. Uric acid (and other purines): The carrier solution was 40/60 vol%, water/methanol containing 10-3 M NH4Ac, pH ~6.3 or containing 0.10 M KOH and 0.04 M HAc, pH ~12.7. Uric

PAGE 50

50 acid was dissolved in the water fraction (40% of final volume) of the carrier solution with moderate heating followed by dissolution of NH4Ac. Methanol was added and the solution was allowed to cool to room temperature and then was made up to volume with methanol. Any uric acid solution that appeared turbid was discarde d. Specific conductivities of the pH 6.3 and the pH 12.7 carrier solutions were measured using a conductivity meter (Wissenschaftlich Technise Werkstatten, Weilheim, Germany) as 0.132 mS/cm and 3.30 mS/cm, respectively. Thiols: Analyte solutions were made in 40/60 vol%,, water/methanol containing 0.001 M ammonium acetate, pH ~6.3 or 50/49/1 vol %,, water/methanol/acetic acid, pH~4.2. Cyclic Voltammetry A BAS-100 electrochem ical analyzer was used for analysis of both dopamine and cysteine (either without or with 100 M Fe2+). A 50 mV/s scan rate was used for all anlyses, and the typical potential scan window was -0.2 V to 1.3 V vs SCE. Duplicate cyclic voltammograms were recorded for each smple. For analysis of dopamine, a total of four 400 M DA solutions were prepared in: 31 mM sodium dihydrogen phosphate/disodium hydrogen phosphate buffer, pH~7.4; in the carrier solution (50/49/1 vol% water/methanol/acetic acid, pH~4.2); and 99/1 vol%, water/acetic acid, pH~4.0. For cysteine analysis 0.5 mM cysteine solution was prepared in 50/49/1 vol%,, water/methanol/acetic acid, pH~4.2. FT-ICR MS Data Analysis The Predator (version 1.2) [Blakney et al., 2005] and Modular ICR Da ta Acquisition and Analysis System (MIDAS) software [Senko et al., 1996], written at the National High Magnetic Field Laboratory (Florida State a nd University of Florida) were used to obtain frequency-domain (ion peak height) mass spectra, following magnitude mode Fourier Transformation of the timedomain (transient signal) spectra. The data from 100 scans were signal averaged. In order to confirm peak assignments, the IsoPro 3.0 program [ http://members.aol.com/msmssoft/ ] was used

PAGE 51

51 for comparison of theoretical rela tive isotope peak in tensities and measured isotope abundances. Averages of at least duplicate measurements are reported. Photo-Induced Dissociation and MS/MS A 7W grating-tuned CO2 laser (Access Laser Company, Ma rysville, WA, USA) with its output set to either 10.65 and 10.16 m (corresponding to dopamine and uric acid IR absorption bands, Figures 2-6 and 2-7), was used for infra-red multiple photon dissociation (IRMPD). H/D Exchange Experiment A 0.5 m M DA solution was pr epared in 50/1/49 vol%, D2O/acetic acid/methanol and was allowed to stand overnight for pr oton/deuterium exchange. To veri fy the structure of the uric acid dimer (MW 336.06), a proton/deuterium exchange experiment was carried out using a solution of uric acid in 40/60 vol%,, D2O/MeOH, 10-3 M NH4Ac. The reaction was allowed to proceed overnight (18 hrs) to ensure completion of H/D exchange.

PAGE 52

52 Figure 2-1. The EC/ESI MS system with the electrochemical cell intergrated into the electrospray capillary. A) Schematic with dimensions; B) Expansion of on-line EC cell. A B

PAGE 53

53 A Vinl( i R)ec onlineSolution Flow A Vapp+ Inlet Detector Vdet9V Battery ( i R)ec-+ ( i R)gap R2R1 e Figure 2-2. Electrical circ uit diagram of the on-lin e EC/ ESI MS system Figure 2-3. Standard linear track variable resistor.

PAGE 54

54 Figure 2-4. Projected voltage prof iles along the ES capillary with ~0.5V increments of applied EC cell voltage (Vapp). The low EC cell voltage is fl oated at the high voltage (HV) of the electrospray. The solid line represents th e reported voltage profile of a standard ES emitter [Pozniak and Cole, 2007]. Figure 2-5. The ICR cell showing the electronic circuit through which rf electric field is applied to excite, trap and detect ions.

PAGE 55

55 Figure 2-6. Fourier transform IR absorption spectrum of dopamine. The band at 10.65 m was laser targeted for the IRMPD MS/MS experiment. 10.65 m

PAGE 56

56 Figure 2-7. Fourier transform IR absorption spectrum of uric acid. The band at 10.16 m was laser targeted for the IRMPD MS/MS experiment. 10.16 m

PAGE 57

57 Figure 2-8. Typical cyclic voltammogram.

PAGE 58

58 CHAPTER 3 ONE-ELECTRON OXIDATION OF DOPAMINE IN ESI AND EC/ESI MS Introduction Electrospray ionization m ass spectrometry (ESI MS) is an important tool in proteomics [Aebersold and Mann, 2003; Bogdanov and Smith, 2005] and, because of its high throughput characteristics and relative ease of coupling to capillary liquid chromatography and microfluidic devices, it is attractive for the an alysis of complex mixtures in metabolomics [Shen et al., 2005]. Electrochemical processes encountered during ESI, reviewed recen tly by Van Berkel and Kertesz [2007], have been exploited in proteomi cs research [Roussel et al., 2004; Maleknia et al., 1999] and are of interest in meta bolomics [Gamache et al., 2004]. In ESI MS, high voltage (HV) is applied to a metal capillary through which the analyte solution is pumped. In positive mode ESI MS, charge separation of ions in solution results when positively charged droplets form [Blades et al., 1 991; Xu et al., 1994]. As discussed in Chapter 1, electrochemical (oxidation) reactions maintain charge balance during ESI and the ESI interface operates as an electrochemical cell. Electrochemical reactions which occur during ESI are unique and can be exploited in analysis [Karst, 2004]. Water, which is commonly present in aqueous/organic carrier solutions, is oxidized during positive mode ESI generating protons (H+) [Moini et al., 1999; Konermann et al., 2001]. As mentioned in Chapter 1, oxidation bu ffers can be added to the carrier solution in order to control electrochemi cal reactions of the analyte during ESI. Lately though, new instrumental designs are being considered [Kerte sz and Van Berkel, 2006; Liljegren et al., 2005]. In on-line electrochemistry ESI MS (EC/ESI MS) experiments described in this dissertation, low voltage is appl ied to the electrospray needle in addition to the HV of ES (Chapter 2, Figure 2-1). The ES capillary, connected through low volume plastic tubing to a

PAGE 59

59 second capillary section into which the sample is injected, is the working electrode and the second capillary section is the counter/reference electrode [Zhou and Van Berkel, 1995; Zhang et al., 2002]. Thus the low voltage cell is floated on the HV of the ES. When low voltage is applied, ionization efficiency increases as does the electrospray current and sensitivity improves [Zhang et al., 2002; Kertesz, Van Berkel, 2005]. Radicals of dopamine (2,4-dihydroxyphe nethylamine; DA), a catecholamine neurotransmitter, have been proposed to functio n as biological antioxidants against reactive oxygen species [Anderson and Harris, 2002]. In a ddition, the formation of Cys radicals by 1e-, 1H+ oxidation processes has been proposed from el ectrochemical studies of cysteine (CySH), a known biological antioxidant [Spataru et al., 2001; Zhou et al., 2007]. Sensitive and selective measurements of these molecules as markers of disease, including by LC MS methods [Zhou et al., 2007; Ogasawara et al., 2007], ar e of interest; and developmen t of sensitive LC MS methods can be facilitated through in formation provided by ESI MS. The formation of one-electron (1e-) oxidation products of DA and CySH, as well as DA quinone-CySH adduct, via a radical pathway duri ng positive mode ESI are described in this chapter. ESI MS thus provi des direct insight into 1eoxidation processes of biological interest. The DA quinone-CySH adduct has been identified as an important metabolite in studies of oxidative stress in the brain [Zhang and Dryhurst 1994; LaVoie and Hastings, 1999; Spencer et al., 1998]. The ESI MS system used in this work, with a cone-shaped capilla ry inlet (Figure 2-1) allowed efficient collection of the electrospray plume of ions. The use of cone-shaped inlet results in significant enhancement in sensitivity compared to the standard cylindrical capillary inlet, which only collects the axially centered porti on of the ES plume. Highe r sensitivity aids the investigations of 1eprocesses inherent to ESI.

PAGE 60

60 Results and Discussion The ESI MS of Dopamine (DA) Figure 3-1 shows ESI mass spectra of DA using a cylind rical inlet (A) and a cone-shaped inlet (B). The three major peaks in either spectrum were assi gned to deaminated dopamine [DANH3]+ m/z 137 (peak 1), the DA cation [DA]+ m/z 154 (peak 2) and the DA dimer [2DA-H]+ m/z 307 (peak 3). A small peak (peak 4) at m/z 152 was assigned to DA quinone [DAQ]+. Scheme 3-1 shows the proposed ion formation pathways, whic h are discussed in more detail later in this chapter. Since DA is a cation in the carrier solution (pKa = 8.9) [Sanchez-Rivera et al., 2003; Linert et al., 1996; Hawley et al., 1967] it forms gas phase [DA]+ ions simply through solvent evaporation. The [DA-NH3]+ ions likely form in the heated MS capillary inlet due to gas phase expansion within a small space. The low formal oxidation potential (Eo= 0.12V vs SCE) [Tse et al., 1976] favors DA oxidation during ESI [Van Be rkel et al., 2007]. However, the low pH reached during ESI [Moini et al., 1999] can inhibit 2e-, 2H+ oxidation of DA, which may explain low intensity of the DA quinone [DAQ]+ signal. Studies show that the initial electrolytic 1eoxidation occurs with a concomitant loss of H+ regardless of pH [Constentin et al., 2007]. Therefore, oxidation of DA by 1e-, 1H+ processes can produce radicals and then, rapidly, the DA dimer [2DA-H]+ by nucleophilic reactions common to cat echolamines [Hawley et al., 1967; Tse et al. 1976] (Scheme 3-1). Further 1eoxidation of the DA dimer (Sch eme 3-1) during ESI, to a doubly charged dimer [2DA-H]2+ (m/z 153.5), is indicated by EC/ESI MS results discussed below. In EC/ESI MS, as shown below, 2e-, 2H+ oxidation of DA to DA quinone (DAQ) is made more efficient than in ESI MS, and DA dimer intensity decreases as a result.

PAGE 61

61 Scheme 3-1. Dopamine oxidation in positive mo de ESI MS. Hydrogens that form H-bonds are not exchangeable with deuterium in the presence of D2O (see mass spectra in Figure 3-2). Cone-Shaped vs Cylindrical Inlet A significant im provement in detection sens itivity for dopamine (> 100% increase in intensity of the base peak) is observed when the cone-shaped inlet is used (Figure 3-1B) compared to a standard cylindric al inlet (Figure 3-1A). A sma ll peak at m/z 152 (Figure 3-1B peak 4), which corresponds to the dopamine quinone [DAQ]+ (Scheme 3-1), is observed with the cone-shaped inlet ESI MS owing to its great sensitivity.

PAGE 62

62 In the normalized version of spectra shown in Figure 3-1, the relative intensity of the dopamine dimer [2DA-H]+ (m/z 307) is higher in the cylindri cal inlet ESI MS, most likely due to loss of radially disributed lighter ions [Nem es et al., 2007]. The smaller ions of unoxidized dopamine are spatially distributed upon coulombic explosion of electrosp ray droplets [Rohner et al., 2004, Watson, 1997] and are not colle cted in the cylindrical inlet. When radially distributed ions re collected using the cone -shaped MS inlet, the sensitivity and the relative intensity of [DA]+ ions increases (Figure 3-1), which allows de tection of low concentrations of dopamine. Considering that the loss of the NH3 group from [DA]+ likely occurs during the expansion and desolvation of the electrospray droplets in the MS inlet capillary, it is not surpri sing that higher intensity of the [DA-NH3]+ ion (peak 1) relative to intact [DA]+ ion (peak 2) is observed in the more confined cylindrical inlet (F igure 3-1A). The DA dimer, a [2M-H]+ species (not to be confused with a proton-bound dimer [2M + H]+) can only be explained by a covalently bound arrangement such, as that shown in Scheme 3-1, given that DA (pKa =8.9) exists as a cation at pH 4.2 [Sanchez-Rivera et al., 2003; Linert et al., 1996]. Furt hermore, dopamine hydrochloride was used in this work, which dissociates into positive dopamine ions and negative chloride ions in solution. It is unlikely theref ore, that at pH 4.2 a dopamine di mer formed by two electrically neutral dopamine molecules exists. The H/D Exchange of DA The H/D exchange experim ents show that the DA dimer has five exchangeable protons (Figure 3-2D), in agreement with the proposed c ovalent structure of the dimer shown in Scheme 3-1. In addition, H/D exchange shows that [DA]+ has four exchangeable protons (Figure 3-2C), consistent with a closed ring structure of [DA]+ in the gas phase. Scheme 3-1 shows the closed ring structure in equilibrium with the open ring for ions of DA, which is common among catecholamines [Blank et al., 1976]. High sensitivity is achieved with the cone-shaped inlet ESI

PAGE 63

63 MS. The lowest measured concentration is 0.02 mM (20 M), where only the [DA]+ (m/z 154) ion is observed (Figure 3-3). The intensities of DA ions change with experimental conditions and with DA concentration (Figure 3-3). The MS/MS of DA Dimer That the O-C and intramolecular hydrogen bonds link the dopam ine semi-quinone and dopamine (see structure of [2DA-H]+ in Scheme 3-1) is supported by results of the H/D exchange experiment much more explicitly than MS/MS experiment. The MS/MS spectrum of the DA dimer, obtained after activation and subs equent dissociation of DA dimer by CO2 laser irradiation (for > 0.5s) showed both the [DA]+ (m/z 154) and the DA dimer [2DA-H]+ (m/z 307) peaks (Figure 3-4). Considering that the tw o DA molecules are bound by one covalent bond and a hydrogen bond, this result is expected since the new covalent bond is more li kely to cleave than any of the covalent bonds in the DA dimer stru cture. The persistent presence of the DA dimer under prolonged irradiation indicates the presence of a (difficult to cleave) covalent bond holding together the dimer structure, presumably the O-C covalent bond linking the two DA structures. The proposed cleavage mechanism is shown in Scheme 3-2. If the H-bound DA dimer were present, it would be cleaved readily and the DA dimer peak (m/z 307) would disappear completely upon laser irradiation. The EC/ESI MS of DA In EC/ESI MS, the intensity of DA ions is de te rmined by the low voltage applied in the online EC cell. As shown in Figure 3-5, the intensities of [DA]+ and [DA-NH3]+ ions increase with the applied voltage. Electrospray current is higher in EC/ESI MS than in ESI MS, an indication of improved ionization efficiency and conditions that are conduciv e to the release of preformed [DA]+ ions from solution to the gas phase and co rresponding increase in fragmentation of [DA]+ to [DA-NH3]+. In EC/ESI MS the DA quinone signal increases and the DA dimer signal

PAGE 64

64 decreases as applied EC cell voltage is incresed. Thus 2e-, 2H+ oxidation of DA is promoted in EC/ESI MS. Scheme 3-2. Proposed cleavage mechanism of the DA dimer in the infrared multiphoton dissociation (IRMPD) experiment. During EC/ESI MS of 2.5 mM DA, the electrospray current increases to ~400 nA from ~200 nA in ESI MS. Assuming a uniform current di stribution in a capillary of a circular cross section, the limiting current should be ~26 A, based on ilim = 1.61zFC(DA/r)2/3 1/3 (3-1) [Bard and Faulkner, 2001] where ilim is the mass transport-limited current during laminar flow, z= 1 (number of electrons), F = 96485 C mol-1 (Faraday constant), l= 5.0x10-2 m (length of the capillary tube), C= 2.5 mol m-3 (concentration), D= 6.0x10-10 m2s-1 [Gerhardt and Adams, 1982] (diffusion coefficient), = 8.3x10-12 m3s-1 (30 L/h) (flow rate), r = 4.0x10-5 m (cross sectional capillary radius) and A = 2 rl = 1.3 x 10-5 m2 (area inside the capillary). However, the

PAGE 65

65 observed electrospray current in EC/ESI MS of DA is lower than the theoretical ilim calculated from Equation 3-1. The presencxe of large signals of [DA]+ and [DA-NH3]+ ions verify that in both ESI and EC/ESI MS, oxidation efficiency of DA is less than 100% (Figure 3-1), which reflects non-uniform distribution of current in the electrospray capillary needle. Effect of Flow Rate Sensitiv ity of ESI MS depends on flow rate [Tang et al., 2001]. In ESI MS with the cylindrical capillary inlet, the ion signal (and th us sensitivity) does not depend strongly on flow rate as shown by Figure 3-6A. W ith the cone-shaped inlet (Figur e 3-6B), the ion signal is a stronger function of flow rate because of efficien t collection of spray droplets by the large orifice of the MS inlet. In EC/ESI MS, the maximum si gnal is reached at lowe r flow rates (Figure 36C), because higher current increas es ion signal at all flow rates. With a cylindrical inlet, good sensitivity has been reported at flow rates 200 nL/min (or 12 L/h) [Vanhoutte et al., 1998]. These results show that the cone-shaped inle t ESI MS system developed in this work accommodates much higher optimum flow rates, which is practical for ESI MS coupling to micro-column HPLC and other micro-fluidic separations. Cyclic Voltammetry of DA A cyclic voltamm ogram of DA in pH 7.4 phos phate buffer at a stainless steel disk electrode (r = 50.8 m) shows one broad shoulder at ~400 mV vs SCE before the increase in the background current at positive potentials (Figure 37). This wave is not observed for DA in the carrier solution (50/1/49 vol% H2O/HAc/MeOH, pH 4.2) (Figure 3-7B) or in 99/1 vol% H2O/HAc, pH 4.0 (Figures 3-7C). DA electrooxi dation is thus slow in the 50/1/49 vol% H2O/HAc/MeOH, pH 4.2 carrier so lution at a stainless steel el ectrode, and it likely occurs together with oxidation of water.

PAGE 66

66 The ESI MS of DA in the Presence of Cysteine (CySH) The ESI m ass spectrum of CySH (Figure 3-8A; di scussed in detail in Chapter 5) shows the base peak at m/z 241, assigned to the proton adduct of the cysteine dimer [CySSCy+H]+ (Scheme 3-3), and the proton adduct of CySH [CySH+H]+ (m/z 122). Peak 6 of the proton adduct of monomeric CySH [CySH+H]+ (m/z 122) is much smaller compared to DA ions because CySH is neutral in the carrier soluti on. ESI MS sensitivity to CySH is low under the present conditions, which are opt imized for DA, and the backgr ound signal is relatively high. The ESI mass spectrum of DA with CySH reflects the individual mass spectra of DA and CySH (Figure 3-8B). In addition, an m/z 273 ion is detected and assigned as the DAQ/CySH adduct [DAQ+CySH]+. The intensities of this adduct and of the DA dimer increase at higher CySH concentration from 0.1 to 2.0 mM at a constant DA concentration (2.5 mM ). Thus, during ESI, oxidation of DA to DAQ is facilitated by the pres ence of CyS. radicals, probably because of the difference in Eo values of DA (Eo = -0.12 V vs SHE) [Tse et al., 1976] and CySH (0.92 V vs SHE) [Buettner and Jerkiewicz, 1996], and is followed by nucleophilic addition of CySH to DAQ, which produces the DAQ/CySH adduct. As shown in Figure 3-8A a disulfide dimer CySSCy forms during ESI of CySH Ionization of CySH is faci litated by higher electrospray current and by the presence of DA, which chemically reduces CyS. radicals. In this case, DA acts as an antioxidant. Ion formation pathways are summarized in Scheme 3-3. Nucleophilic addition of CySH to DAQ is significant in Alzheimers and Parkinsons disease [Shen and Dryhurst, 1996a; Shen and Dryhurst, 1996b; Shen et al., 1996;]. In these ESI MS experiments (where CySH is present), high intensities of the DA oxidation products, DA dimer and DAQ, relative to the CySH oxidation product and the disulfide dimer of cysteine, may be due in part to faster oxidation of DA+ to DA+ (k ~ 7.2x109 M-1s-1) [Anderson and Harris, 2002] relative to oxidation of CySH to CyS. (k ~ 1.2x 103 M-1s-1) [Uchiyama and Sekioka, 2005].

PAGE 67

67 The EC/ESI MS of DA in the Presence of CySH High electro spray current in EC/ESI MS improves ES efficiency of DA and CySH ionization. There is less formation of the DA and CySH oxidation products, DA and CySH dimers, and DAQ-CySH adduct (Figure 3-9). At higher applied voltages ( 1.0V), oxidation of DA by 2e-, 2H+ is facilitated, and the intensity of the DAQ/CySH adduct increases while the intensity of the DA dimer decreases. Scheme 3-3. Formation of DAQ-CySH adduct in positive ion mode ESI MS. Conclusions In positive mode ESI of DA and CySH in aqueous/m ethanol carrier solution, DA and CySH dimer signals are detecte d. Thus during ESI, DA and CySH radicals appear to form and

PAGE 68

68 rapidly dimerize. Cone-shaped capillary inlet ES I MS is more sensitive than the standard cylindrical capillary inlet ESI MS. Formation of DA quinone by a 2e-, 2H+ process is not efficient, possibly because of the low pH during ESI. Real-time chemical identification of products of 1eand 2eionization processes is possible in ESI and EC/ESI MS. High flow rates that can be used with cone-shaped inlet ESI MS can be practical for dir ect coupling of ES to microcolumn separations. New insights into oxidation and antioxidant ac tivity can be obtained by ESI and EC/ESI MS. The limiting step in the overall 2e-, 2H+ oxidation of DA, which is positively charged, may be the first 1e-, 1H+ loss. In vivo 2e-, 2H+ oxidation of DA in the presen ce of CySH was shown to form DAQ-CySH adducts [LaVoie and Hastings, 1999] which may lead to the depletion of the cysteine pool. ESI MS results suggest th at this reaction can be relatively slow.

PAGE 69

69 062006 DA#13m/z 600 500 400 300 200 100 Abundance 100 90 80 70 60 50 40 30 20 10 m/z 137 [DA-NH3]+m/z 154 [DA]+m/z 307 [2DA-H]+ 071907 DAflow#13bm/z 60 0 500 400 300 200 100 Abundance 200 150 100 50 Figure 3-1. Positive ion ESI MS of DA (2.5 mM) with (A) cylindrical inlet and (B) cone-shaped inlet; flow rate 30 L/h; 50/1/49 (vol %) H2O/HAc/MeOH, pH~4.2; HV ~3 kV. 4 4 Intensity m/z 160 155 150 1 1 [DA-NH3]+ 2 [DA]+ 3 [2DA-H]+ [ [ D D A A Q Q ] ]+ +2 [DA]+ 4 4 ~ 1 1 Intensity [DA-NH3]+ + 2 [ DA ] + 3 [ 2DA-H ] + A B

PAGE 70

70 010809 dDA#1 (m/z 137)m/z 600 500 400 300 200 100 Rel. Intensity 100 90 80 70 60 50 40 30 20 10 010809 dDA#1 (m/z 137)m/z 140 135 Rel. Intensity 80 70 60 50 40 30 20 10 010809 dDA#1 (m/z 154)m/z 160 155 150 Rel. Intensity 100 90 80 70 60 50 40 30 20 10 010809 dDA#1 (m/z 307)m/z 315 310 305 Rel. Intensity 55 50 45 40 35 30 25 20 15 10 5 Figure 3-2. The ESI mass spectrum of dopamine (0.5 mM) in 50/49/1 vol%, D2O/methanol/acetic acid (A). Peaks representing the number of exchangeable pr otons for the ions [DA-NH3]+, [DA]+ and [2DA-H]+ are shown in (B), (C) and (D), respectively. See structures in Scheme 3-1. m/z 154 m/z 137 m/z 307 A m/z 137.07 d1-[DA-NH3]+ m/z 137.08 d 2 -[DA-NH3]+ m/z 137.07 [DA-NH3]+ B m/z 154.09 [DA]+ m/z 156.10 d 2 [ DA ] + m/z 157.11 d 3 [ DA ] + m/z 158.12 d4-[DA]+ m/z 155.09 d 1 [ DA ] + C m/z 307.16 [2DA-H]+ m/z 308.17 d1-[2DA-H]+ m/z 311.19 d4[ 2DA-H ] +m/z 310.17 d3-[2DA-H]+ m/z 312.19 d5-[2DA-H]+ m/z 309.18 d2-[2DA-H]+ D

PAGE 71

71 Figure 3-3. Changes in ion intensit ies in ESI mass spectra of DA due to change in concentration. Dots indicate off-scale intensity. Flow rate 60 L/h; cone-shaped capillary inlet; 50/1/49 (vol %) H2O/HAc/MeOH, pH~4.2; HV ~3 kV.

PAGE 72

72 m/z 500 400 300 200 100 Rel. Intensity 100 90 80 70 60 50 40 30 20 10 m/z 500 400 300 200 100 Rel. Intensity 100 90 80 70 60 50 40 30 20 10 m/z 500 400 300 200 100 Rel. Intensity 100 90 80 70 60 50 40 30 20 10 Figure 3-4. The ESI MS of 2.5 mM dopamine (A); ESI MS after ejection of m/z < 307 and m/z > 307 ions i.e isolation of dopamine dimer [2DA-H]+ (m/z 307) ion (B); MS/MS of [2DA-H]+ following CO2 laser irradiation (>0.5s) Notice product ion peaks at m/z 154, most likely [DA]+ and at m/z 174, unassigned (C). m/z 307 [2DA-H]+ B Intensity (%) m/z 307 [2DA-H]+ m/z 154 [DA]+ m/z 307 [DAN H3]+ A Intensity (%) m/z 154 [DA]+ m/z 307 [2DA-H]+ m/z 174 unassigned C Intensity (%)

PAGE 73

73 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 0.00.51.01.52.02.53.03.54.04.55.05.56.06.57.07.58.0EC Cell Voltage (V)Relative Intensity (%) m/z 137 [DA NH3]+ m/z 152 [DAQ]+ m/z 153.5 [2DA H]2+ m/z 154 [DA]+ m/z 307 [2DA H]+ m/z 307 [2DA H]+ Figure 3-5. The EC/ESI MS of DA (0.25 mM). Cond itions as in Figure 3-3; moving average in black.

PAGE 74

74 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 20.025.030.035.040.045.050.055.060.065.070.0Flow Rate (uL/h)Intensity m/z 137 [DA-NH3]+ m/z 154 [DA]+ m/z 307 [2DA-H]+ 0 100 200 300 400 500 600 700 800 0 20 40 60 80 100 120 140Flow Rate (uL/h)Intensity m/z 137 [DA-NH3]+ m/z 154 [DA]+ m/z 307 [2DA-H]+ 0 100 200 300 400 500 600 700 020406080100120140Flow Rate (uL/h)Intensity m/z 137 [DA-NH3]+ m/z 154 [DA]+ m/z 307 [2DA-H]+ Figure 3-6. The ESI MS of DA (2.5 mM) as a func tion of flow rate: (A) cylinder capillary inlet; (B) conical capillary inlet; (C) conical cap illary inlet in EC/ESI MS (1.5 V). Other conditions as in Figure 3-3. C A B

PAGE 75

7510008006004002000-200 -40 -20 0 AM2A31 10008006004002000-200 -60 -40 -20 0 20 AM2A31 10008006004002000-200 -100 -80 -60 -40 -20 0 20 AM2A31 Figure 3-7. Cyclic voltammetry of DA (400 M) at stainless steel el ectrode in (A) phosphate buffer (31 mM), pH~7.4; (B) 50/1/49 vol%, water/acetic acid/methanol, pH~4.2; (C) 99/1 vol%, water/acetic acid, pH~4.0. Disk radius 50.8 m; scan rate 50 mVs-1. E ( mV ) vs SCE i ( nA ) A E ( mV ) vs SCE B i ( nA ) E ( mV ) vs SCE i ( nA ) C Blank Blank Blank

PAGE 76

76 072907 Cys#3bm/z 60 0 500 400 300 200 100 Abundance 40 35 30 25 20 15 10 5 110607 DACys#8am/z 60 0 500 400 300 200 100 Abundance 100 90 80 70 60 50 40 30 20 10 Figure 3-8. The ESI MS of cysteine (CySH) and DA with CySH: (A) CySH (0.5 mM); (B) DA (2.5 mM), CySH (0.5 mM). Flow rate 45 L/h. Other conditions as in Figure 3-3. Intensity (%) 1 B 3 [2DA-H]+ [CySSCy+H]+ 7 [DA-NH3]+ 2 [DA]+ 8 [DAQ+Cys]+ [CysH+H]+ 6 Intensity (%) m/z 122 [CysH+H]+ A m/z 241 [CySSCy+H]+7 6

PAGE 77

77 Figure 3-9. The EC/ESI MS of DA with CySH. Conditions as in Figure 3-3. EC Cell Voltage (V) m/z 122 [CysH+H]+ m/z 137 [DA-NH3]+ m/z 154 [DA]+ m/z 241 [CySSCy + H] + m/z 273 [DAQ+ C y sH ] + m/z 307 [2DA-H]+ (%)

PAGE 78

78 CHAPTER 4 ONE-ELECTRON OXIDATION AND DETECTION SENSITIVITY OF URIC ACID IN ESI AND E C/ESI MS Introduction Purine m etabolites are important as markers of physiological disorders, and their analysis is of interest in clinical di agnostics and metabolomics [Simmonds et al., 1997; La Marca et al., 2006; Gamache et al., 2004]. Uric acid, which is the final product of nucleotide catabolism in humans [La Marca et al., 2006], has been identified as a marker of cardiovascular disease [Waring et al., 2000; Alderman and Kala, 2004]. Ur ic acid has also been reported to have protective functions during oxidative stress, when high concentrati ons of superoxide (O2.-) drive the reaction of O2.w ith nitric oxide (NO.) to peroxynitr ite (ONOO-) [Skinne r et al., 1998; Hooper et al., 2000]. The consumption of NO ., which regulates bl ood vessel dilation, is undesirable. However, uric acid appears to s cavenge the peroxynitrite and liberate NO. [Skinner et al., 1998]. Thus uric acid may minimize tissue damage via the release of NO. [Hooper et al., 2000]. Purine metabolites can be analyzed by high performance liquid chromatography (HPLC) [La Marca et al., 2006; Gamache et al., 2004, Hartman et al., 2006; Ito, et al., 2000] and capillary electrophoresis (CE) [Benavente, et al., 2006] w ith ESI MS detection. However, uric acid is difficult to detect by ESI MS, in spite of its relatively high concentration in biological fluids [La Marca et al., 2006]. Uric acid wa s identified as a potent biologi cal antioxidant from HPLC and UV analysis of 1eoxidation products in singlet oxygen a nd hydroxyl radical reactions [Ames et al., 1981]. In the previous work 1eoxidation and radical formation was confirmed by ESR and by transient UV measurements [Maples and Mason, 1988; Simic, Javanovic, 1989]. To enhance the sensitivity of uric acid dete ction in ESI MS, on-line electroc hemistry ESI MS (EC/ESI MS), described in Chapter 2, was used in this work (see Figure 2-1).

PAGE 79

79 As has been shown in Chapter 3 [Mautjana et al., 2008a], ESI and EC/ESI MS can provide unique insights into 1ereactions relevant to biological an tioxidant properties. Furthermore, it has been demonstrated that oxi dation of analytes during electro spray ionization can enhance the sensitivity in ESI MS measurements, because the higher electrospray current increases ionization efficiency [Mautjana et al., 2008a; Zhang et al ., 2002]. In EC/ESI MS, th e sensitivity can be further improved because the electrospray current in creases when current flows in the on-line EC cell. A conical MS inlet capillary [Wu et al., 2006], used for study of one-electron oxidation reactions of dopamine [Mautjana et al., 2008a], can be used in the measurements of uric acid. Low m/z ions of biological metabolites, such as uric acid, are radia lly distributed in the electrospray and are not collect ed by the standard cylindrical inlet ESI MS, resulting in low detection sensitivity. The enhancement of sensitivity is expected as a result of improved collection of ions by the large-orifice conical MS inlet capillary [Wu et al., 2006] compared to the standard cylindrical capillary inlet. Positive ion mode ESI MS is used in this wo rk because of its high sensitivity to proton adducts of analytes. The results of this work suggest that during positive mode ESI MS, uric acid, which is negatively charged in the carrier solution, is oxidized by 1eprocesses to neutral radicals, which form dimers. The signals of the protonated monomeric and dimeric species contribute to the total uric acid signal. Furt hermore, the low solubility of uric acid (~0.009g/100mL or 5.35 x 10-4 M at 25oC in water) [Mentasti et al., 1983] which can limit sensitivity in ESI MS, was addressed here thro ugh control of the carrier solutions composition. Results and Discussion Ionization of Uric Acid in Electrospray (ES) Figure 4-1 s hows the positive ion mass spectra of uric acid in a 40/60 vol%, H2O/MeOH carrier solution with 1 mM NH4Ac, pH~6.3 (A and B) and with 0.10 M KOH, 0.04 M HAc, pH

PAGE 80

80 ~12.7 (C). The base peak at m/z 169 (Figure 4-1A) was assigned to the proton adduct of uric acid [H2U+H]+ and the peak at m/z 337 [Skinner et al., 1998; Binyamin et al., 2001] was assigned to the proton adduct of uric acid dimer [2H2U+H]+. The dimer can form during ESI by 1eoxidation of urate followed by rapid dimerizatio n of the radicals that are formed during oxidation. The proposed ion formation pathway is shown in Scheme 4-1. A small peak at m/z 231 (Figure 4-1A) was identified as the protona ted adduct of 5-hydroxyhydantoin with parabanic acid [OHhyd+Parab+H]+ (Scheme 4-2). Thus, 2e-, 2H+ oxidation of uric acid (Scheme 4-2) [Volk et al., 1992; Volk et al., 1999], can occur during ESI, but with low efficiency, as indicated by the low intensity of the [OHhyd+Parab+H]+ adduct. A consecutive 1eoxidation, with radical format ion, has been reported in the electrochemical and chemical oxidation of uric acid and other purines [Subramanian, et al., 1987; Griffiths, 1952]. In agreement with the 1eoxidation and radical dimerization during ESI, formation of covalent dimers has been reported in ESI MS of dopamine (Eo = -0.12 V vs SHE), caffeic acid (Eo = 0.20 V vs SHE) and cysteine (Eo = 0.92 V vs SHE) [Mautjana et al., 2008a; Arakawa et al., 2004]. H/D exchange experiments show that three hydrogents of uric acid (H2U) are exchangeable (Figure 4-2B) in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac. The N1-proton between C2=O and C6=O groups is the least acidic of all pr otons in the uric acid molecule and might be exchangeable if a completely deuterated solvent 40/60 vol%, D2O/CD3OD, 1 mM ND4(CD3CDOO). The uric acid dimer (2H2U) has five exchangeable protons (Figure 4-2D), in agreement with the structure proposed in Scheme 4-1. See molecular structures in Scheme 4-1 and their H/D exchange mass spectra in Fi gure 4-2. A hydrogen-bonded dimer (unlikely for purines due to the ener getically unfavored 90o arrangement of the bonded molecules) would have

PAGE 81

81HN ON H N O H N O HN ON H N O H N O NH O N H N H N O MW167 [HU-] MW167 [HU] -e-HN ON H N O H N O O H [H2U]NH O N H N H H N O HN ON H N O H N O O H [H2U] [HU] MW336 [2H2U] 1 3 8 6 4 2 5 7 9 HN ON H N OH H N O 1 3 8 6 4 2 5 7 9 MW168 [H2U] pKa1=5.4 -H+ Scheme 4-1. Proposed reactions during ESI of uric acid (H2U). Urate (HU-) is present in the pH 6.3 carrier solution of H2U (pKa1=5.4). Hydrogens that form H-bonds and those between C=O groups are not exchangeable with deuterium in the presence of D2O (see mass spectra in Figure 4-2). Proton adducts of uric acid [H2U+H]+ (m/z 169) and uric acid dimer (m/z 337) are detected in this carrier solution in ESI MS. The H+ is generated during positive i on mode electrospray. fewer exchangeable protons. Stable H-bonding between purines requires four molecules (Chapter 6). At higher uric acid concentrations, Na+ and K+ adducts of uric acid species were detected in addition to H+ adducts (Figure 4-1B). The formation of adducts of various analytes

PAGE 82

82 with Na+ and K+ is commonly observed because of Na+ and K+ ions from glass which have been estimated to reach concentrations of ~10-5 M [Kebarle, 1997]. The Na+ and K+ adducts become more apparent at higer pH values where the proton concentration is relatively small (~10-6 M at pH 6.3). Similar to the dopamine dimer discussed in Chapter 3, MS/MS of the uric acid dimer ion [2H2U+H]+ (m/z 337) following its di ssociation under infrared CO2 laser irradiaton (~ 0.5s) in an infrared multiple photon dissociation (IRMPD) experiment, shows an ion peak at m/z 169 without complete disappearance of the uric acid dimer peak at m/z 337 (Figure 4-3). This result could be an indication of the re lative strength of the covalent bond linking the two uric acid molecules as opposed to H-bonds. The proposed cleavage mechanism of uric acid dimer is shown in Scheme 4-2. The 40/60 vol%, water/methanol, 10-3 M NH4Ac carrier solution was chosen because it provides the maximum solubility of H2U with feasible electrospraying conditions. In this carrier solution, the ectrospray current is 10 0 200 nA and the solubility of H2U is 50 150 M. When the water fraction is lower (<40%) uric acid precipitates; a larger water fraction ( 50%) causes arcing. To avoid the low sensitivity for uric acid in the 50/50 vol%, water/methanol carrier used for negative ion mode HPLC ESI MS [La Marca, et al., 2006; Ito, et al., 2000; Shi, et al., 2003], 0.1 M KOH was used to dissolve uric acid [Bi nyamin, et al., 2001], and 0.04 M HAc was added to stabilize the current in 40/ 60 vol%, water/methanol. The hi gh specific conductivity of the 40/60 vol%, water/methanol, 0.1 M KOH, 0.04 M HAc, carrier solution of ~3.30 mS/cm, compared to ~0.132 mS/cm for 40/60 vol%, water/methanol, 10-3 M NH4Ac, and ~0.103 mS/cm

PAGE 83

83 Scheme 4-2. Proposed cleavage mechanism of th e uric acid dimer in the infrared multiphoton dissociation (IRMPD) experiment. for 50/49/1 vol%, water/methanol/acetic acid, can contribute to the hi gh electrospray current that was observed. In the carrier so lution with KOH of pH ~12.7, 2e-, 2H+ oxidation of uric acid (Scheme 4-3) generates allantoin (MW 158; pKa ~8.96), which was detected as [K(Allnt)+K]+ (Figure 4-1C). A K+ adduct of unoxidized urate (U2-) (pKa1 5.4; pKa2 9.8) [Simic and Javanovic, 1989] was detected with a standard cylindrical capillary inlet ESI MS, where the electrospray current is lower (not shown). This adduct was not detected under the experimental conditions of Figure 4-1C, however. Allantoin a nd parabanic acid observed in th is work have also been

PAGE 84

84 Scheme 4-3. Oxidation pathway of uric acid to 5-hydroxyhydantoinpa rabanic acid adduct [OHhyd+Parab], alloxan monohydrate [Allxnhyd], and allantoin [Allant] [Volk et al., 1992; Volk et al., 1999]. Protonated hydr oxyhydantoinparabanic acid adduct (m/z 231) is detected in ESI MS. Allantoin and alloxan monohydrate are detected as K+ adducts [K(Allnt)+K]+ (m/z 235) and [K(Allxnhyd)+K]+ (m/z 237), respectively. observed during radiolytic oxidation of uric acid at pH 3.4 and 7.4 using HPLC with UV detection,where significantly more parabanic acid than allantoin was produced at pH 3.4 [Hicks et al., 1993]. Radiolysis produces high concentration of OH radicals at pH 7.4 which can attack allantoin to give parabanic acid, but Hicks et al. observed little ch ange in allantoin concentration at the higher pH. It is possible th at the formation of allantoin at the high pH (~12.7) seen in this work is accompanied by the formation of undetect ably low amounts of parabanic acid, and that

PAGE 85

85 formation of parabanic acid at pH~6.3 used in this work is accompanied by formation of undetectably low amounts of allantoin, in which cas e the two studies produce consistent results. The ions identified in ESI MS of uric acid are summarized in Table 4-1. Assignments are based on the relative intensity of M+2 ions of the K-41 isotope and are in good agreement with theoretical isotope abun dances. The [K(Allxnhyd)+K]+ (m/z 237) signal overlaps with the M+2 isotope signal of K-41 of [K(Allnt)+K]+ (m/z 235) and the intensity of the M+2 isotopic cluster [K(Allnt)+Ac+2K]+ (m/z 333) is inflated by the overlapping signal of K+-alloxan-acetate cluster [K(Allxnhyd)+Ac+2K]+ (m/z 335) (Table 4-1). Alloxan monohydrate (MW 160) is a product of the 2e-, 2H+ oxidation of uric acid (Scheme 4-2) and was also detected as [K(Allxnhyd)+K]+. A linear relationship with an av erage correlation (R2) > 0.7 was obs erved between ln (ion intensity) and ln(uric acid concentration) for [K(Allnt)+K]+ (m/z 235) as well as [K(Allnt)+Ac+2K]+ (m/z 333) (Figure 4-4). The EC/ESI MS of Uric Acid In EC/ESI MS, the ion intensities of the proton adducts of uric acid [H2U+H]+ (m/z 169), uric acid dimer [2H2U+H]+ (m/z 337) and [OHhyd+Parab+H]+ (m/z 231) adduct are a function of the voltage applied to the on-line low voltage EC cell (Figure 4-5). Thus ion distribution is determined by the efficiency of electrooxidation. At higher applied voltages the dimer signal decreases while the 2e-, 2H+ oxidation of uric acid increasingly generates parabanic acid (Figure 4-5). Since purine dimers can be oxidized at hi gh positive potentials [Subramanian et al., 1987], oxidation of the dimer may contribute to the in crease in the parabanic acid signal at higher applied voltages. The intensities of Na+ and K+ adducts of uric acid species follow those of H+ adducts (not shown).

PAGE 86

86 A general stepwise 1e-, 1H+ oxidation pathway [Hicks et al., 1993] can account for the ionization of urate, forming neutral radicals [HU.], which rapidly dimerize to the [HU.+H2U+H]+ radical and then to the [2H2U+H]+ (m/z 337) dimer. Scheme 4-4A summarizes this pathway. The pathway may involve H-atom transfer (Scheme 4-4B) [Siegbahn et al., 1997]. Scheme 4-4. Proposed dimer formation pa thway (A) during ESI of uric acid (H2U). HUis present in pH 6.3 ca rrier solution of H2U (pKa1=5.4). Protonated uric acid [H2U+H]+ (m/z 169) and the protonated dimer (m/z 337) are detected in ESI MS. The H+ is generated during positive ion m ode electrospray. (B) H-atom transfer leading to the uric acid dimer and the neutral radical, which further gives diimine and final oxidation products. The ESI MS of Uric Acid in Urine In 1000-fold diluted norm al urine (pH 5.3 8.0) the concentration of uric acid is 2 M [Yue-Dong, 1998; Kupeli, et al., 2005] Direct ESI MS analysis of a more concentrated urine sample leads to irreproducible mass spectra. In 1 000-fold diluted urine, the concentration of uric acid is lower than that of the solutions of sta ndards that were tested (~20 M). Nevertheless a K+ adduct of Na+ urate [NaHU+K]+ (m/z 229) was detected in 40/60 vol%, water/methanol, 1 mM

PAGE 87

87 NH4Ac, pH ~6.3 carrier solution (Figure 4-6A). In 1000-fold d iluted urine spiked with 20 M uric acid, protonated uric acid [H2U+H]+ (m/z 169) was detected together with [NaHU+K]+ (Figure 4-6B). Na+ and K+ found in urine may form the urate complexes (NaHU, pKsp ~ 4.6 at 25oC) [Wang and Konigsberger, 1998] and [NaHU+K]+ that were detected. Conclusions In the pos itive ion conical capillary inlet ESI MS of uric acid, ion signals of the protonated uric acid dimer and two-electron oxidation pr oducts of uric acid, pa rabanic acid, hydantoin, allantoin and alloxan monohydrate we re detected. The results thus indicate that negatively charged urate ions are oxidized during positive ion mode ESI to neutral uric acid radicals, which form the uric acid dimer. The dimer and the uno xidized uric acid are detected, which improves the sensitivity in ESI MS of uric acid. H/D exchange experiments indicate a covalent structure of the uric acid dimer. ESI of ur ic acid may involve stepwise 1e-, 1H+ oxidation and dimerization pathway similar to those for dopamine [Mautjana et al., 2008a] and caffeic acid [Arakawa et al., 2004]. Thus ESI MS can provide in sight into the 1e--oxidation r eactions of biological interest, especially to individuals st udying the antioxidant properties of easily oxidized biological analytes, such as uric acid. The results additiona lly show that concentration and composition of the carrier solution determine ion distribution and ESI MS signal in tensity of uric acid ions. In EC/ESI MS with an on-line EC cell voltage floated on the HV of the ES, the 2e-, 2H+ oxidation of uric acid is more efficient than in ESI MS, and the intensity of the 1e oxidation product, the protonated dimer, decreases. The high sensitivit y of the conical capillary inlet ESI MS allows measurements of uric acid. In 1000-fold dilu ted normal human urine 2 M uric acid was detected as a Na+ urate complex with K+.

PAGE 88

88 Table 4-1. Theoretical and average measured m/z values of identified ions and their is otopic abundances (n = 15). Theoretical Measured m/z Isotopic abundance(%) m/z Isotopic abundance(%) Ion Identity M M+1 (M+1)/M (M+2)/M M M+1 (M+1)/M (M+2)/M a[H2U+H]+ 169.04 170.04 7.20 169.04 170.04 5.44 a[H2U+Na]+ 191.02 192.02 7.19 191.02 192.03 5.39 a[2H2U+H]+ 337.06 338.07 14.39 337.07 338.08 11.73 a[2H2U+Na]+ 359.05 360.05 14.46 359.05 360.05 10.28 a[OHhyd+parab+H]+ 231.04 232.04 8.46 1.51 231.08 232.08 7.09 1.19 a[OHhyd+parab+Na]+ 253.02 254.02 8.45 1.51 253.06 251.06 6.55 n/d b[K2U+K]+ 282.90 283.90 7.13 22.47 282.91 283.91 5.22 22.35 b[K(Allnt)+K]+ 234.96 235.96 6.08 15.19 234.92 235.92 3.40 18.28c b[K(Allxnhyd)+K]+ 236.93 237.93 5.37 15.44 236.94 237.94 d d b[K(Allnt)+Ac+2K]+ 332.94 333.94 8.44 22.96 332.90 333.90 7.61 33.45c b[KAllxnhyd+Ac+K]+ 334.91 335.91 7.68 23.24 334.92 335.92 d d aCarrier solution: 40/60 vol%,, H2O/CH3OH containing 0.001 M NH4CH3COO, pH~6.3. bCarrier solution: 40/60 vol%,, H2O/CH3OH containing 0.10 M KOH and 0.044 M CH3COOH, pH~12.7. cUncharacteristically high abundances due to spectral overlap of M+2 peaks with those of ions of equivalent m/z values, [K(Allxn hyd)+K]+ (m/z 236.93) and [K(Allxnhyd)+Ac +K]+ (m/z 334.91), respectively. dValues were omitted because these peaks were obs erved with poor S/N ratio in some mass spectra.

PAGE 89

89 041807 UA#4m/z 60 0 550 500 450 400 350 300 250 200 150 100 Abundance 400 350 300 250 200 150 100 50 041808 UA#1m/z 60 0 500 400 300 200 100 Intensity 110 100 90 80 70 60 50 40 30 20 10 Figure 4-1. Positive ion mass spectra of uric aci d. Cone-shaped capillary inlet; 40/60 vol%, water/ methanol, 0.001M ammonium acetate, pH~6.3* ; flow rate 40L/h; HV 3kV. m/z 169 [H2U+H]+ m/z 191 [H2U+Na]+ m/z 231 [OHhyd+Parab+H]+m/z 337 [2H2U+H]+ m/z 359 [2H2U+Na]+ m/z 253 [OHhyd+Par ab+Na ] + Intensity B m/z 333 [K(Allnt)+Ac+2K]+ m/z 235 [K(Allnt)+K]+ C 041807 UA#2m/z 60 0 550 500 450 400 350 300 250 200 150 100 Abundance 200 150 100 50 Intensity m/z 169 [ H 2 U+H ] +m/z 231 [OHhyd+ Parab+H]+ m/z 337 [ 2H2U+H ] +A m/z 207 [H2U+K]+ m/z 375 [2H2U+K]+

PAGE 90

90 010809 dUA#1am/z 550 500 450 400 350 300 250 200 150 100 Rel. Intensity 100 90 80 70 60 50 40 30 20 10 010809 dUA#1a (m/z 169)m/z 175 170 165 Rel. Intensity 90 80 70 60 50 40 30 20 10 010809 dUA#1a (m/z 207)m/z 215 210 205 Rel. Intensity 100 90 80 70 60 50 40 30 20 10 010809 dUA#1a (m/z 337)m/z 345 340 335 330 Rel. Intensity 80 70 60 50 40 30 20 10 Figure 4-2. The ESI mass spectrum of uric acid (50 M) in 40/60 vol%, D2O/MeOH, 1mM NH4Ac (A). Peaks representing the number of exchangeable protons for the ions [H2U+H]+, [H2U+K]+ and [2H2U+H]+ are shown in (B), (C) and (D), respectively. See structures in Scheme 4-1. m/z 169 [H2U+H]+ m/z 207 [H2U+K]+ m/z 375 [2H2U+K]+ m/z 337 [2H2U+H]+m/z 191 [H2U+Na]+ m/z 169 [H2U+H]+ m/z 170 d1[H2U+H]+ m/z 171 d2[H2U+H]+ m/z 172 d3[H2U+H]+ m/z 338 d1[2H2U+H]+ m/z 339 d2[2H2U+H]+ m/z 340 d3[2H2U+H]+ m/z 341 d4[2H2U+H]+ m/z 338 d5[2H2U+H]+ m/z 337 [2H2U+H]+ m/z 207 [H2U+K]+ m/z 208 d1[H2U+K]+ m/z 209 d2[H2U+K]+ m/z 210 d3[H2U+K]+ m/z 211 d2[H2U+41K]+ A B C D Intensity (%) Intensity (%) Intensity (%) Intensity (%)

PAGE 91

91 m/z 500 400 300 200 100 Rel. Intensity 100 90 80 70 60 50 40 30 20 10 m/z 500 400 300 200 100 Rel. Intensity 100 90 80 70 60 50 40 30 20 10 m/z 500 400 300 200 100 Rel. Intensity 100 90 80 70 60 50 40 30 20 10 Figure 4-3. The ESI MS of 50 M uric acid (A); ESI MS after ejection of m/z < 337 and m/z > 337 ions i.e isolation of uric acid dimer [2H2U+H]+ (m/z 337) ion (B); MS/MS of [2H2U+H]+ (m/z 337) following CO2 laser irradiation (>0. 5s) Notice the product peak at m/z 169, likely due to [H2U+H]+ ion and smaller unassigned product peaks at m/z values <300 (C). m/z 337 [2H2U+H]+ m/z 169 [H2U+H]+ m/z 337 [2H2U+H]+ A m/z 337 [2H2U+H]+ m/z 169 [H2U+H]+ B C Intensity (%) Intensity (%) Intensity (%)

PAGE 92

92 R = 0.7858-1.00 4.00 9.00 14.00 19.00 24.00 29.00 -80-60-40-200Ln (m/z 235 Intensity)Ln (Uric acid concentration, mol/L)A R = 0.91770.00 10.00 20.00 30.00 40.00 50.00 60.00 -70-60-50-40-30-20-100Ln (m/z 333 Intensity)Ln (uric acid concentration, mol/L)B Figure 4-4. The ln (Intensity) vs ln (concentr ation, mol/L) plots for the ions [K(Allnt)+K]+ (m/z 235) (A) and [K(Allnt)+Ac+2K]+ (m/z 333) (B).

PAGE 93

93 0.0 200.0 400.0 600.0 0.01.02.03.04.05.06.07.08.0IntensityEC Cell Voltage (V) m/z 231 [(OH)hyd+Parab+H]+ m/z 337 [2H2U+H]+ m/z 169 [H2U+H]+ Figure 4-5. Intensity of uric acid (50 M) ions in EC/ESI MS as a function of on-line EC cell voltage. Cone-shaped inlet; 40/60 vol%, H2O/MeOH, 10-3 M NH4Ac, pH 6.3; Flow rate 40 L/h; HV 3 kV. 2 2 + + +

PAGE 94

94 Figure 4-6. Positive ion mode ESI MS mass spectra of human urine: A) 1000 fold diluted; B) 1000 fold diluted and spiked with 20 M uric acid. Same conditions as in Figure 4-1. 062807 UrineC#1am/z 240 230 220 210 200 190 180 170 160 Abundance 35 30 25 20 15 10 5 zoom 062807 UrineA#1bm/z 240 230 220 210 200 190 180 170 160 Abundance 40 35 30 25 20 15 10 5 m/z 229 [NaHU+K]+ A Intensit y m/z 169 [H2U+H]+ m/z 229 [NaHU+K]+ B Intensit y

PAGE 95

95 CHAPTER 5 SENSITIVITY OF POSITIVE MODE ESI AND EC/ESI MS TO THE ANALYSIS OF THIOL METABOLITES Introduction Disease biom arker and biological pathway discovery has led to the analysis of thiol metabolites, glutathione (GSH), cysteine (CySH) and homocysteine (hCySH) which are involved in many important physiological processes [Rah man et al., 2005]. GSH and CysH are found at millimolar concentrations in different types of cells and in blood plasma. They function as endogenous antioxidants, which protect cells ag ainst reactive oxygen spec ies (ROS) and reactive electrophiles, and which serve to restore vital proteins to their redu ced form [Forman and Dickinson, 2003]. Changes in concentrations of GSH and CysH from normal (1 10 mM) correlate with numerous disease conditions. For example, depleti on of GSH has been reported to occur in type II diabetes [Sami ec et al., 1998] and HIV [Staal et al., 1992], while elevated levels of CysH have been reported in cases of Parkinsons and Alzheimers diseases [Shen and Dryhurst, 2001]. Elevated concentrations of hCysH, which is an intermediate in the metabolism of methionine to CysH [Nekrassova et al., 2003 ; Himmelfarb et al., 2002], a precursor for GSH, have been found to correlate w ith cardiovascular diseases such as atherosclerosis and venous thrombosis [Demuth et al., 2002; Van den Brandhof et al., 2001]. Despite th is correlation of high hCySH concentrations and cardiov ascular disease, the role of hCySH in the mechanism of vascular injury is unknown and could involve GSH and CysH, which share a common biosynthetic pathway with hCySH [Himmelfarb et al., 2002]. Simultaneous analysis of thiols in biological samples is important fo r clinical applications, as well as for understand ing their roles in physiology and pathology. Since they require enzymatic catalysis to undergo oxidation reactions with H2O2 within the cells [Gilbert, 1995; Meis ter, 1983] thereby pr oducing disulfides, it is evident that thiol metabolites are stable in vivo in their reduced form, which is consistent

PAGE 96

96 with their relatively high oxidation potentials (Eo 0.92 V vs SHE). In view of this, the GSH/GSSG concentration ratio ha s been proposed as an efficient measure of oxidative stress [Curello et al., 1987; Schafer and Buettner, 2001; Kemp et al., 2008]. Similarities in chemical structures, react ivity and properties between homocysteine, cysteine and glutathione (see Sc heme 5-1 below) make their simultaneous analysis difficult by most analytical methods. Scheme 5-1. Structures and pKa values of thiol metabolites [Nek rassova et al., 2003; Budavari et al., 1989]. High performance liquid chromatography (HPLC) is often used for thiol separations [Garcia et al., 2008; Giustrarini et al., 2008; Nolin et al., 200 7; Bald, 2004; Yoshida, 1996; Winters et al., 1995] with electroc hemical detection, or with UV a nd fluorescence detection after derivatization because thiols are poor chromophores. Various derivatizing agents and offand on-line derivatization methods are being evaluate d in an ongoing effort to optimize derivative yields for quantitative spectropho tometric detection of thiols. However, derivatization and LC tend to be time consuming. Comptons group repor ted a method which involved tagging of thiols

PAGE 97

97 with electrochemically generated quinoids to faci litate their quantitative determination by cyclic voltammetry [White et al., 2001a; White et al., 2001b]. However, the tem poral resolution based on varying diffusion coefficients, rates of oxidation reactions (k(CySH/CyS ) = 1.2 x 103 M-1 s-1) [Uchiyama and Sekioka, 2005] and thiol addition to the quinoid, which is associated with electrochemical (EC) methods and which could result in short analys is time, is not realized in thiol analysis. It has been observed that larg e overpotentials are requi red for the oxidation of thiols, often resulting in poorly defined and ove rlapping thiol or thiol adduct oxidation/reduction peaks [White et al., 2002; Zhou et al., 2007]. Furthermore, electrode materials, mostly metals, may catalyze thiol oxidation and may produce varying oxidation peak potentials for each thiol compound [White et al., 2002, Sahlin et al., 2002]. Thus measured peak potentials are se ldom close to oxidation potentials calculated from thermodynamic data, such as 0.92 V vs SHE for CySH and for GSH reported by Buettner and Jerkiewicz [Buettner and Jerkiewicz, 1996]. Mc Carley and co-workers [Pacsial-Ong et al., 2006] used catechol-analogues (Flurone blac k, 1-methyl-1,2,3,4-tetrhydr o-6,7-isoquionolinediol hydrobromide and 3,5-di-tert-butyl catechol) to form monothiol derivatives with GSH, CysH and hCysH, respectively. The monothiols were found to be oxidized further at di fferent potentials to form bisthiols, allowing discrimination between GSH, CysH and hCysH. LC/EC methods with mercury electrodes allow accumulation of thiols at the surface at positive potential. The thiols are released upon reduction, ther eby providing high sensitivity, but these methods are not practical because of toxicity of mercury. Complex mediators and deriva tives of hexacyanometallates and metallophthalocyanines, which catalytically oxidize thiols and are subseque ntly reoxidized at the electrode surface, have been used in electrochemical thiol detection [L iu et al., 2004], and thei r development is ongoing.

PAGE 98

98 Electrospray ionization mass spectrometry (ESI MS) coupled with separation techniques is attractive for thiol analysis. Yet few reports of thiol analysis by ESI MS are found [Rubino et al., 2004; Bouligand et al., 2006; Gucek et al., 2002] The advantage of HPLC/ESI MS in thiol analysis is selective, mass-based detection. However, long analysis time may be a limitation. With direct injection ESI MS, thiols can be re solved based on exact mass, and analysis time can be significantly reduced. However, standard calibration and quantitation with direct injection ESI MS is more challenging than with LC/MS, which allows better ion counti ng statistics when the mass spectrometer is focused on a single m/z value corresponding to the eluting compound [Watson, 1997]. Direct injection MS is limited to comprehensive data collection where the detection time is shared by all ions, thus producing ion count variations and S/N ratio that is less than that achieved with selected ion monitoring. But more significant are the processes undergone by analyte molecules during analysis. Understanding the processes govern ing ionization of analyte mo lecules during electrospray and ion transmission could be useful for improving quantitatitation accuracy. Gucek et al. [2002] attempted the determination of GSH in extracts of needles of a spruce plant by positive mode ESI MS using standard addition method with limited success, at which they proposed to use an isotopically (15N) labeled GSH as an internal standard. Gu cek et al. attributed the difficulties in determing GSH content to the different GSH sp ecies formed even under controlled (dark, low temperature storage) conditions. However, Gucek et al. observed no change in LC/MS peak area for samples stored at 4oC and considerably little amounts of decomposition products when GSH solutions were left standing in the light at 25oC, when they examined the stability of GSH in aqueous solutions [Gucek et al., 2002]. The GS H species observed by Gucek et al. could be

PAGE 99

99 attributed to the oxidation pr oducts of GSH formed during ESI, which the present work demonstrates. Electrochemical processes which are inherent to electrospray ionization [Van Berkel and Kertesz, 2007; Van Berkel et al., 2000] can cause oxidation of thiols to disulfides, and poor ion collection efficiency can affect detection sensitivity. In Chapters 3 and 4 we demonstrated that metabolites with low oxidation potentials such as dopamine (Eo = -0.12 V vs SHE) and uric acid (Eo = 0.59 V vs SHE) can undergo oxidation during positive mode ESI MS. We also demonstrated that electrospray ions can be e fficiently collected using a cone-shaped MS inlet [Mautjana et al., 2008a,b]. Similarly, thiols could undergo oxidation during positive mode ESI MS even though they have relatively higher Eo values ( 0.92V vs SHE for CySH or GSH) [Buettner and Jerkiewicz, 1996] and could form disulfide dimers. Oxidation and fragmentation, found in the present work to be the two major pro cesses leading to the formation of different ions of thiols during positive mode ESI MS, were elucidated. Th is was achieved through on-line electrochemistry ESI MS which can enhance ion formation and increase sensitivity. Sensitivity of reduced thiols was also increased by adding dopamine, which may undergo 1e-, 1H+ electrochemical oxidation and may be further oxidized chemically to DAQ by thiol radicals as proposed [Mautjana et al., 2008a]. In turn, thiol radicals can be reduced to thiols and form a thiol/DAQ adduct through nucleophilic add ition. This chapter pr esents the ESI MS spectra of GSH, CySH and hCySH obtained und er high sensitivity conditions brought about by the use of cone-shaped metal capillary inlet a nd on-line electrochemistry. The formation of different ions due to electrochemical processes of the ES i on source is discussed, and the application of ESI MS for thiol mixture analysis is described.

PAGE 100

100 Results and Discussion The ESI MS of glutathione (GSH), cysteine (CySH) and ho mocyste ine (hCySH) ESI mass spectra of GSH, CySH and hCySH in 40/60 vol% H2O/MeOH, 1 mM NH4c, pH 6.3 with which relatively high detection sensitivity for all three thiols wa s obtained, are shown in Figure 5-1. GSH (Figure 5-1A) produces i ons of m/z 308, 330 and 346, which have been assigned to proton, sodium and potassium adducts of GSH, respectively. Other ion peaks at m/z 613 and 615 (Figure 5-1A insert ) are assigned to proton addu cts of the covalently bound (radical-radical) dimer [GSSG] and the hydrogen bound dimer [2GSH]. The structure of GSSG is shown in Scheme 5-2. Scheme 5-2. Glutathione disulf ide dimer detected as [GSSG+H]+ (m/z 613) in ESI MS The peak at m/z 154 (Figure 5-1A) is assigned to the proton adduct of an oxygen-centered glutathione cation radical [GSH.+H]2+, which is proposed to form by one-electron oxidation of the zwitterion form of GSH (Scheme 5-3). Altho ugh it is formed with a +1 charge and m/z of 307 (Scheme 5-3), the oxygen-centered glutathione radical was detected as a doubly-charged proton adduct [GSH.+H]2+ (m/z 154). The formation of the oxygen-centered glutathione cation radical (MW 307) is supported by H/D exchange data which indicate that this GSH radical (m/z 154) has five exchangeable hydrogens (Figure 5-2) implying an intramolecular H-bonding of the GSSG MW = 612

PAGE 101

101 sulfhydryl hydrogen (Scheme 5-3). In addition to H/D exchange, the structure of the oxygen centered radical [GSH.+H]2+ (m/z 154) is supported by its relative isotope distribution. O HN O O S H N NH3OO O H O HN O O H N NH3OO H -e-O N H HO O S H N NH2HOO O O N H HO O S H N NH2HOO O -(e-,H+) GS MW306 GSH MW307 O H N HO O S N H NH2OH O O O N H HO O S H N NH2OH O O GSSG MW612 GSH MW307 GSH MW307 GS H H O S H Scheme 5-3. Proposed oxidation of GSH during positive mode ESI MS. Hydrogens that form Hbonds are not exchangeable with deuterium in the presence of D2O (see the mass spectrum in Figure 5-2). The detection of a disulfide dimer, assigned as [GSSG+H]+ (m/z 613), suggests oxidation of the thiol group in a n on-zwitterionic form of the thiol by a loss of one electron and one proton,

PAGE 102

102 despite intramolecular H-bonding of the sulfhydryl hydrogen. The intermed iate thiol radicals (GS.) couple rapidly to form the dimer (Scheme 5-3). ESI MS of cysteine (CySH) produces peaks at m/z 122 and 241 which have been assigned to the proton adduct [CySH+H]+ and the disulfide dimer [CySSCy+H]+ as in Figure 5-1B. In the 40/60 vol% H2O/MeOH, 1 mM NH4Ac, pH 6.3 carrier solution, th e small ion peak at m/z 243, assigned to the proton adduct of a hydrogen bound dimer [2CySH+H]+, which was not detected when 50/49/1 vol% H2O/MeOH/HAc was used (Chapt er 3), is detected. This is attributed to a different tuning of the instrument parameters to optimize thiol detection, whereas in Chapter 3, parameters were tuned for DA dete ction. A different mobile phase may also cause this increase in sensitivity to [2CySH+H]+ m/z 243 ion. This assignment is supported by the observed isotopic distribution. Homocysteine (hCySH ; Figure 5-1C), produces ion pe aks that have been assigned to the proton adduct [hCySH+H]+ (m/z 136) and the proton a dduct of its disulfide dimer [hCSSCh+H]+ (m/z 269). All solutions were fr eshly prepared and kept at ~4oC until their injection, and all oxidati on products observed are elec trochemically generated. The ESI MS detection sensitivity for GSH is high relative to that for CySH and hCySH, possibly because in a protein-like folded stat e GSH is more hydrophobic and forms gas phase ions more easily than CySH and hCySH. That Cy SH is detected with hi gher sensitivity than hCySH could be due to the ability of CySH to form an intramolecular H-bond between the sylfhydryl hydrogen and the carbonyl oxygen rend ering CySH less polar (or more hydrophobic) than hCySH. Intramolecular H-bonding is difficult within th e hCySH structure. Effect of GSH Concentration on ESI MS W ith dilution of 50 M GSH solution, a ll the proton adducts including [GSH+H]+ (m/z 308) and [GSSG+H]+ (m/z 613) decrease in intensity. Simi lar to previous observations made during uric acid analysis in 40/60 vol%, H2O/MeOH carrier solution containing 10-3 M NH4Ac,

PAGE 103

103 pH ~6.3 (Chapter 4) [Mautjana et al ., 2008b], sodium and potassium adducts [GSH+Na]+/[GSH+K]+ (m/z 330/ m/z 346) and [GSSG+Na]+ (m/z 635) have higher intensities than the proton adducts in large (up to 20 fold) dilutions with the same carrier solution; and [GSH+Na]+ (m/z 330) becomes the base peak (Figure 5-3). This is so because at pH ~6.3 the H+ concentration (~10-6 M) is less than that of Na+ and K+ estimated to be 10-5 M [Kebarle, 1997] and the [thiol][proton] ion product is simply lo w. The other prominent peak observed at m/z 372 is assigned to the ammonium adduct of the glutamate-less glutathione disulfide [(GSSG2Pyr)+NH4]+. The lowest GSH concentration detected with 40/60 vol% H2O/MeOH, 10-3 M NH4Ac, pH 6.3 carrier solution, is 2 M which is ten times lowe r than the lowest detected concentration of dopamine (20 M) with 50/49/1 vol% H2O/MeOH/HAc, pH 4.2 carrier solution (Chapter 3). The difference in sensitivity could be attributed to rela tive hydrophilicity of DA (which exists as a cation in solu tion) or relative hyr ophobicity of GSH; and relative volatility of MeOH promotes the transfer of GSH to the ga s phase. GSH, CySH and hCySH were generally detected with better sensitivity in 40/60 vol% H2O/MeOH, 10-3 M NH4Ac, pH 6.3 than in the 50/49/1 vol% H2O/MeOH/HAc, pH 4.2 carrier solution. Being a tripeptide, GSH can be folded, lik e proteins, thus becoming hydrophobic [Forman and Dickinson, 2003] given that the [GSH+H]+ ion signal remains strong during ESI MS (Figure 5-4), and that relatively lo w signals of GSH fragment i ons are observed. The little GSH fragmentation that is observed, co uld occur during transition to the gas phase in the inlet, in the same manner as described previ ously for dopamine [Mautjana et al., 2008a]. GSH fragmentation seen at high concentrations ( 0.5 mM) (Figure 5-4) occurs th rough the loss of the amino acid glutamate as pyroglutamic acid (Pyr; MW = 129) as well as loss of gl ycine (Gly; MW = 75)

PAGE 104

104 (Scheme 5-4) to produce cysteinyl glycine a nd cysteinyl glutamate which are observed as [(GSH-Pyr)+H]+ (m/z 179) and [(GSH-Gly)+H]+ (m/z 233), respectively (Figure 5-4). O N H OH O HS H N H2N HOO O O N H OH O HS NH2HN HOO O GSH MW307 GSH-Pyr MW178 Pyr MW129 + Scheme 5-4. Fragmentation of GSH during ES I MS [adapted from Rubino et al., 2006]. In contrast, glutathione disulfid e appears to fragment relatively easily through the loss of glycine and glutamate leading to fragment ions of various m/z values given its relatively large size. The loss of (2Pyr) from GSSG (Sch eme 5-5) produces [(GSSG-2Pyr)+H]+ (m/z 355) at high concentration (Figure 5-4) or [(GSSG-2Pyr)+NH4]+ (m/z 372) at low con centration (Figure 5-3). The ESI MS of GSH, CySH and hCySH Mixture A m ass spectrum of a mixture of thiols is shown in Figure 5-5. Ions observed in Figure 5-5 are those of the disulf ide dimers, [CySSCy+H]+ (m/z 241) and [hCySSCyh+H]+ (m/z 269), proton adducts of glutathione [GSH+H]+ (m/z 308), and interest ingly, a mixed disulfide [hCySSCy+H]+ (m/z 255) formed by hCySH and CySH, which may suggest a kinetically facile reaction between hCyS. and CyS. radicals. The formation of a mixed disulfide [hCySSCy+H]+ (m/z 255) was significantly inhibited by the addi tion of dopamine (Table 5-1). However, more work is necessary to investigate ways to exploit or prevent its fo rmation for analytical purposes.

PAGE 105

105 Scheme 5-5. The GSSG fragment ations observed during ESI MS [Adapted from Rubino et al., 2006]. The EC/ESI MS of GSH Changes in signal intensities of GSH-derived ions which occur with increase in applied low EC cell voltage in on-line electrochemistry ESI MS (EC/ESI MS) can be seen in Figure 5-6. The ion of m/z 308 [GSH+H]+ remains the base peak over the tested EC cell voltage range. The formation of oxygen-cen tered radical [GSH.+H]2+ (m/z 154) appears to compete with generation of the dimers, [GSSG+H]+ and [2GSH+H]+, when the low EC cell voltage is applied. The increase in relative intensity of m/z 154 ion at higher applied EC ce ll voltage (Figure 5-6) suggests that the formation of [GSH.+H]2+ is facilitated during on-line EC/ESI MS, which may be so because [GSH.+H]2+ ion is produced electrochemically and its ionization by proton adduct formation is promoted by the increase in applied EC cell voltage. Relative ion signal intensities

PAGE 106

106 of [GSSG+H]+ (m/z 613) and [2GSH+H]+ (m/z 615) increase to a maximum, likely due to increased ionization efficiency, a nd then they decrease gradually as the applied EC cell voltage is increased. The latter, a hydrogen bound dimer not an electrochemical pr oduct decreases much more sharply compared to [GSSG+H]+ (m/z 613), presumably because GSH is oxidizeable. GSSG appears to undergo oxidation as more peaks at higher m/z values were observed at EC cell voltages >3.5 V (data not shown), indicating GSH/GSSG decomposition accompanied possibly by high mass adducts, trimers, tetramers and so on. The hydrogen/deuterium (H/D) exchange experi ment indicates that the m/z 154 ion has five exchangeable hydrogens, which implie s intramolecular H-bonding (of the sulfhydryl hydrogen) as proposed in Scheme 5-3 whereas GSH has six exchangeable hydrogens (Figure 5-2). The ESI MS of GSH and hCySH in the Presence of Dopamine (DA) It has been proposed that one-electron oxidation of thiols and of DA, in the presence of each other, which occurs during pos itive mode ES I MS, leads to chemical oxidation of dopamine by the thiol radical to dopamine quinone ( DAQ) [Mautjana et al., 2008a], given the low oxidation potential of DA (Eo = -0.12V vs SHE) re lative to the oxidation pot ential of thiols (Eo (GSH or CySH) = 0.92V vs SHE) [Buettner and Jerkiewicz, 1996]. It has been proposed in Chapter 3 that oxidation of DA by CySH re sults in a small increase in the intensity of CySH signal as [CySH+H]+ (m/z 122) in addition to forming a CySH/DAQ adduct. Given the possibility of improving thiol sensitivity by ad dition of DA, a study of ESI MS and EC/ESI MS of GSH and hCySH in the presence of DA was conducted. Fu rthermore, the thiol/DAQ reaction has been proposed as part of the toxicity mechanism cau sing Parkinsons disease [Shen et al., 1996, Shen and Dryhurst, 1996a,b; Whitehead et al., 2001; LaVoie and Hastings 1999] and EC/ESI MS is used here to determine possible roles of GSH and hCySH in the mechanism.

PAGE 107

107 ESI mass spectra of GSH (0.4 mM) and hCyS H (0.5 mM) with DA (2.5mM) are shown in Figure 5-7. A 50/49/1 vol% H2O/MeOH/HAc, pH 4.2 carrier solution was chosen because it was found to give relatively high se nsitivity of DA and thiol/DA a dduct ions. ESI mass spectra of CySH/DA in the same carrier solution were discussed in Chapter 3 [Mautjana et al., 2008a]. For both GSH/DA and hCySH/DA solutions, ions obse rved include previously reported DA derived ions namely, [DA-NH3]+ (m/z 137), [DA]+ (m/z 154) and [2DA-H]+ (m/z 307). In the presence of DA, the GSH derived ions, [GSH+H]+ (m/z 308), [GSH+Na]+ (m/z 330), [(GSSG-Gly)+H]+ (m/z 538) and [GSSG+H]+ (m/z 613) are observed with (appa rent) improved intensity (Figure 57A) compared to the absence of DA (Figure 5-1A) in agreement with the effect of DA on CySH mass spectra reported in Chapter 3. Homocyst eine (hCySH) derived ions, i.e [hCySH+H]+ and [hCSSCh+H]+, are suppressed in the presence of DA (Figure 5-7B) but the low sensitivity may also be due, in part, to different mobile phase than that used in Figure 5-1C. The adducts of GSH with DAQ [GSH+DAQ]+ (m/z 459) (Figure 5-7A) and hCySH with DAQ [hCySH+DAQ]+ (m/z 287) (Figure 5-7B) are detected as observed for cysteine/DAQ. The radical formation pathway reported previously [Mautjana et al., 2008a] may be responsible for the adduct formation. Effect of GSH Concentration on Thiol/DA Mass Spectra The adduct [DAQ+ GSH]+ (m/z 459) increases as the GS H concentration increases from 0.1 to 0.5mM (Figure 5-8) while DA concentration is kept constant at 2.5 mM. A plot of ion intensity vs GSH concentration (Figure 5-8), wher e DA concentration (2.5 mM) is held constant, shows that the adduct signal intensity increases w ith an increase in GSH concentration. There is an increase in intensitie s of DAQ adduct, [DAQ+GSH]+ (m/z 459), and [GSH+H]+ (m/z 308), as well as the H-bonded GSH dimer, [2GSH+H]+ (m/z 615) (Scheme 5-6). There is similar increase in [2DA-H]+ and [GSSG+H]+ dimers (Figure 5-8).

PAGE 108

108 Scheme 5-6. Formation of [DAQ+GSH] adduct in positive ion mode ESI MS. The ESI MS of GSH in the Presence of Uric Acid A possible form ation of an adduct between thiols and diimine, a product of 2e-, 2H+ oxidation of uric acid, proposed by Dutt et al. [2 003] based on changes in UV absorption spectra with addition of thiol, was investigated. An ESI mass spectrum of a solution containing GSH (10 M) and H2U (60 M) in 40/60 vol%, H2O/MeOH, 10-3 M ammonium acetate, pH~6.3, is shown in Figure 5-9. Solutions with increasing uric acid/GSH concentration ratio, where uric acid (H2U) concentration (60 M) was kept constant while GSH was increased from 10 through 60 M, were analyzed. None of the observed signals in the mass spectrum could be ascribed to GSH and H2U adduct or their oxidation products Observed in Figure 5-9 are H2U derived ions, [H2U+H]+ (m/z 169) and [2H2U+H]+ (m/z 337) as well as GSH derived ions that have already been discussed. It is worth noting that under these conditions, the dimer [GSSG+H]+ (m/z 613) increases and becomes the base peak. Whether this in crease is due to the presence of uric acid is yet unclear. The apparent inability of GSH to fo rm an adduct with UA could be due to the fact that the second oxidation of H2U radicals by GS radicals would produce a diimine which is unstable [Volk et al., 1992; Volk et al.; 1999].

PAGE 109

109 The ESI MS of GSH, CySH and hCySH Mixture in Pres ence of DA Analysis of all three thiols mixed in solution, in presen ce of DA was also performed (Figure 5-10). While both CySH [CySH+H]+ (m/z 122) and hCySH [hCySH+H]+ (m/z 136) ions are observed in the mass spectra of individual CySH (Figure 5-1B) and hCySH (Figure 5-1C), these monomeric ions are not detected in the ma ss spectrum of the thiol mi xture in the presence of 2.5 mM DA (Figure 5-10), and the signal inte nsity of hCySH disulfide dimer [hCySSCyh+H]+ (m/z 269) ion is very low. The monomer ic ions of CySH and hCySH, [CySH+H]+ (m/z 122) and hCySH [hCySH+H]+ (m/z 136), are not detected in the mass spectrum of the thiol mixture in the absence of DA (Figure 5-5) either. The absence of CySH+H]+ (m/z 122) and hCySH [hCySH+H]+ (m/z 136) ions in the thiol mixture without DA, may be an indication of fac ile cross-coupling of radi cals given the presence of a mixed disulfide dimer [hCySSCy+H]+ (m/z 255). The absence of CySH+H]+ (m/z 122) and hCySH [hCySH+H]+ (m/z 136) ions could also be due to the relatively poor sensitivity of ESI MS to the monomeric [CySH+H]+ (m/z 122) and [hCySH+H]+ (m/z 136) ions compared to [GSH+H]+ (m/z 308) whose signal is observed. In the presence of DA, only a small peak of the mixed disulfide dimer [hCySSCy+H]+ (m/z 255) is observed, which may indicate suppression of this ion, and the monomeric CySH+H]+ (m/z 122) and hCySH [hCySH+H]+ (m/z 136) ions by DA ions, or it could indicate competitive thio l/DA adduct formation where the thiols are completely consumed considering that DA is present in excess. The intensity of the disulfide dimer [hCySSCyh+H]+ (m/z 269) decreases (from 37 to 5 counts; Table 5-1) in the hCySH/DA mixture because of formation of [DAQ+hCySH]+ (m/z 287) adduct which competes with thiol dimerization.

PAGE 110

110 The EC/ESI MS of Thiols in the Presence of DA Sim ilar to CySH/DAQ adduct, the adducts of [GSH+DAQ]+ (m/z 459) (Figure 5-7A) and [hCySH+DAQ]+ (m/z 287) (Figure 5-7B) increase in intensity with applied EC cell voltage (Figures 5-11A&B). The DAQ/thiol adducts incr ease with applied EC cell voltage while the respective signal intensities of disulfide dimers, [GSSG+H]+ (m/z 613; Figure 5-11A) and [hCySSCyh+H]+ (m/z 269; Figure 5-11B), a ppear to be unchanging. The CySH disulfide dimer, with relatively hi gh intensity (Chapter 3) than observed here for GSSG and hCySSCyh, but lower than the DAQ/ CySH adduct intensity, appear to increase along with the DAQ/CySH adduct as EC cell volta ge is increased. It is possible that the formation of CySSCy is kinetically competitiv e with DAQ/CySH formation as discussed in Chapter 3, whereas the formation of GSSG and hCySSCyh is slow. In either case, the radical pathway mechanism proposed previously for the DAQ/CySH adduct formation [Mautjana et al., 2008a] is supported by the experime ntal results. Brie fly, both GSH (or hCySH) and DA undergo 1e-, 1H+ oxidation during ESI to form their respec tive radicals which, given the difference in oxidation potentials (Eo (GSH or hCySH) = 0.92 V vs SHE and Eo (DA) = -0.12 V vs SHE), undergo chemical oxidation reaction where the GSHor hCySH-radical gets reduced and DA-radical oxidized. The resulting products, GSH (or hC ySH) and DAQ, undergo a nucleophilic, 1,4Michael addition r eaction to give the adduct [DAQ+GSH]+ (Scheme 5-6) or [hCySH+DAQ]. The additional adduct formed by two hCySH molecules and DAQ [DAQ+2hCySH]+ (m/z 420) is observed in the mass spectrum of hCyS H/DA solution (Figure 5-7B) and its intensity increases with applied EC cell voltage too. The same 50/49/1 vol% H2O/MeOH/HAc, pH 4.2, carrier solution was used for all th iol/DA mass spectra compared above.

PAGE 111

111 Summary of Thiol Mixture Analysis The presence of DA in 50/49/1 vol% H2O/MeOH/HAc, pH 4.2 carri er solution used in these experiments, and application of on-line EC cell voltage affect the different thiol ions and adducts in different ways. In EC/ESI MS described below the [DAQ+hCySH]+ adduct ion intensity increases from 70 to 111 counts when the EC cell is turned on and is the highest hCySH related signal produced (Table 5-1) The highest signal for CySH is that of the CySSCy disulfide dimer (m/z 241) (with EC cell ON) whereas for GSH it is that of GSH proton adduct [GSH+H]+ (m/z 308) (with EC cell OFF). The [DAQ+GSH]+ (m/z 459) has the least signal intensity even with the EC cell ON, which could be due to th e slow kinetics of the nucleophilic addition of GSH to DAQ relative to CySH and hCySH. Close inspection of data shown in Table 5-1 reveals that the best experimental condi tions for thiol analysis, with which all thiol species can be detected, are when DA is present in the carri er solution and about 1.5 V EC cell voltage is applied. However, at 2.5 mM concentration in the mixture, DA appears to suppress monomeric thiol ions. Additional studies could be conducted in the future to determine the DA concentration that can allow maximum thiol detection sensitivity. Evidence of Catalysis of CySH Oxidation by Metal Ions Off-line cyclic voltammetry of CysH (0.5 m M) in the presence of Fe2+ (1 mM) shows increase in anodic current for CySH at ~1.1 V vs SCE (Figure 5-12), which indicates that CySH oxidation is likely catalyzed, and this is further su pported by ESI MS data (Table 5-2). The mechanism of catalysis of CySH oxidation by the Fe2+ ion likely involves the Fe2+/CySH complex [Tanaka et al., 1955a; Tanaka et al., 1955b; Wang and Stanbury, 2008]. The ESI MS spectra of CySH (0.5 mM) show an increase in intensity of the CySH disulfide dimer (m/z 241) when Fe2+ (100 M) is added to CySH (50 M ) in 50/49/1 vol% H2O/MeOH/HAc, pH 4.2 carrier solution (Table 5-2). Signal intensitie s of [CySH+H]+ (m/z 122) and [CySSCy+H]+ (m/z

PAGE 112

112 241) in the presence of Fe2+ or Cu2+ are much higher compared to those obtained without these ions (compare intensities in Tabl e 5-1 to those in Table 5-2). The base peak in the mass spectra of Fe2+/CySH solution appears at m/z 538 (Figure 5-13), and has been assigned to the redox active [Fe(CySH)4]+ [Rose et al.; 1998; Cotton et al., 1987] formed in solution. Addition of Cu2+ (100 M) has even greater ca talytic effect on CySH given the high intensity of [CySSCy+H]+ (m/z 241) relative to that obtained in presence of Fe2+ (Table 5-2). The intensity of cysteine disulfid e dimer (m/z 241) in the presence of Cu2+ (100 M), increases with EC cell ON (Table 5-2. Proposed Mechanism of Catalysis of CySH Oxidation by Iron (II) Much research on catalytic oxida tion of thiols such as cysteine, by iron and copper ions is being done [Wang and Stanbury, 2008; Tyapoc hkin and Kozliak, 2005] to understand the m echanisms involved, which might be benefi cial for analysis of thiol compounds, and undertansing metal catalyzed oxidation reactions of endogenous thiols including proteins (with sulfhydryl groups) which may occur in vivo, leading to disease. In the case of iron, however, iron-sulfur centers in some proteins such as rubredoxin (found in bacterium C. Pasturianum ) have been well characterized [Rose et al.; 1998; Cott on et al., 1987]. Rubredoxins are relatively low-molecular weig ht proteins (~6000Da) which contain one iron atom surrounded by a distorted tetrahedron of cysteinyl sulf ur atoms [Cotton et al., 1987]. Iron is normally in the Fe(III) ox idation state but can be reduced to the Fe(II) oxidation state (Eo = 0.77 V vs SHE) with only a slight increase in the FeS distances and no change in the tetrahedral coordination [Rose et al.; 1998; Cotton et al., 1987]. Ir on in both oxidation states has been shown through the Mossbauer spectroscopy (w hich detects magnetic resonance shifts of gamma-irradiated nuclei) to be in the unpaired, high spin state [Ph illips et al., 1970; Rose et al., 1998]. The high spin state with unpa ired electrons allows rubredoxins to participate in electron

PAGE 113

113 transfer reactions. Therefore, it is possi ble that the CySH ligands of the Fe(CySH)4 complex can be exchanged with free cysteine in solution, with th e released cysteine in oxidized form (i.e. as thiol radicals), which couple to form the cysteine disulfide dimer. This mechanism is seemingly supported by the increase in the disulfide dimer [CySSCy+H]+ (m/z 241) as applied EC cell voltage is increased (Table 5-2). Conclusions Reduced thiols, GSH, CySH and hCySH and their oxidized f orms GSSG, CySSCy and hCSSCh were detected by positive mode ESI MS Electrospray induces oxidation of these analytes leading to formation of sulfur-centered th iol radicals which dimerize rapidly to form the disulfide dimers. On-line electrochemistry improve s the detection sensitiv ity of ESI MS to the thiols. Relatively low concentration of ~2 M GSH was detected. Relatively stable oxygencentered radicals are also forme d. The stability of oxygen-centere d GSH radicals likely results from the intramolecular hydrogen bonding, and the resultant electron delo calization along the OC-O-H-O-C-O chain (Sheme 5-3). The formation of GSSG observed from the pH ~6.3 solution (of GSH, pKa(SH) = 8.7) suggests that electrolytic oxidation of thiol proceeds regardless of pH seeing that it is relatively low with respect to the pKa. In the presence of DA, GSH (and hCySH) together with DA undergo 1e-, 1H+ oxidation. The respective radicals can reac t to form the dimers, [2DA-H]+ (m/z 307) and [GSSG+H]+ (m/z 612), [CySSCy+H]+ (m/z 241) or [hCSSCh+H]+ (m/z 269), and can also undergo chemical redox reaction (given Eo (GSH) = 0.92 V vs SHE and Eo (DA) = -0.12 V vs SHE) which leads to formation of GSH (CySH or hCySH) and DAQ and subse quent nucleophilic addition of GSH (CySH or hCySH) to the electron deficient DAQ to form dopamine adducts as already described. It is possible that GSH and uric acid radicals undergo redox reaction as well but unlike DAQ, the diimine formed from oxidation of uric acid ra dicals is unstable. Thus no adduct is formed.

PAGE 114

114 Table 5-1. Average intensities (n = 3) of th iol derived ions in the presence of DA. m/z Value Assignment Intensity (without dopamine) Intensity (with dopamine) EC Cell OFF EC Cell ON (1.5V) EC Cell OFF EC Cell ON (1.5V) 122 [CySH +H]+ Not detected Not detectedNot detected Not detected 136 [hCySH +H]+ Not detected Not detectedNot detected 6 137 [DA-NH3]+ n/a n/a 2 4 154 [GSH.] 2+ ; *incl. [DA]+ 5 5 151* 214* 241 [CySSCy+H]+ 41 98 51 56 255 [hCSSCy+H]+ 131 196 29 25 269 [hCSSCh+H]+ 37 47 6 5 273 [DAQ+CySH]+ n/a n/a 158 244 275 [GSH.+CySH]2+; *incl.[DA+ CySH]+ 3 3 5* 8* 287 [DAQ+hCySH]+ n/a n/a 70 111 307 [2DA-H]+ n/a n/a 4 4 308 [GSH+H]+ 135 115 37 32 459 [DAQ+GSH]+ n/a n/a 6 8 Table 5-2. Average intensities (n = 3) of cystei ne disulfide dimer (m/z 241) indicating metal ion catalysis of cysteine oxidation. Intensity (without metal ions) Intensity (with Fe(II)) Intensity (with Cu(II)) m/z Value Assignment EC Cell OFF EC Cell ON (1.5V) EC Cell OFF EC Cell ON (1.5V) EC Cell OFF EC Cell ON (1.5V) 122 [CySH+H]+ 8.7 64.1 83.4 104.7 10.7 26.3 241 [CySSCy+H]+ 49.7 213.9 88.7 280.0 231.4 290.0 296.7 [CySFeSCy]+ 2.6 12.1 16.5 20.4 n/a n/a 538 [Fe+4CySH]+ 2.8 6.2 222.4 394.2 14.9 8.1 304 [CySSCy+Cu]+ n/a n/a n/a n/a 9.7 10.9

PAGE 115

115 050708 GSH#1am/z 700 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 081208 Cys#1am/z 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 071108 HCys#4m/z 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 5-1. Positive mode ESI MS of (A) GSH (0.05 mM), (B) CySH (0.05 mM) and (C) hCySH (0.5 mM) in 40/60 vol%, H2O/MeOH containing 1 mM NH4Ac, pH~6.3; Flow rate 50 L/h; HV 3 kV. Cm/z 136 [hCysH+H]+ m/z 269 [hCySSCyh+H]+ m/z 122 [CysH+H]+ Bm/z 241 [CySSCy+H]+ m/z 243 [2CysH+H]+ m/z 154 [GSH.+H]2+ m/z 308 [GSH+H]+ Am/z 330 [GSH+Na]+ m/z 346 [GSH+K]+ m/z 613 [GSSG+H]+m/z 615 [2GSH+H]+ 050708 GSH#1am/z 617 616 615 614 613 612 Intensity (%) 25 20 15 10 5 m/z 613 m/z 615

PAGE 116

116 Figure 5-2. The H/D exchange ESI MS of GSH (0.05 mM) in 40/60 vol%, D2O/MeOH, 1 mM NH4Ac, pH 6.3; Flow rate 50 L/h; HV 3 kV. See structures in Scheme 5-3.

PAGE 117

117 070508 GSH#L2am/z 700 600 500 400 300 200 100 Intensity 9 8 7 6 5 4 3 2 1 Figure 5-3. Lowest concen tration of GSH (2 x 10-3 mM or 2 M) detected with positive mode ESI MS; Same conditions as in Figure 5-1. 111307 GSH#1m/z 700 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 5-4. Positive mode ESI MS of GSH (0.5 mM); Same conditions as in Figure 5-1. 111307 GSH#1m/z 617 616 615 614 613 612 611 610 609 Intensity (%) 0.1 m/z 308 [GSH+H]+ m/z 330 [GSH+Na]+ m/z 346 [GSH+K]+ m/z 372 [(GSSG-2Pyr)+NH4]+ m/z 613 [GSSG+H]+ m/z 635 [GSSG+Na]+ m/z 154 [GSH.+H]2+ m/z 308 [GSH+H]+ m/z 355 [(GSSG2Pyr)+H]+ m/z 613 [GSSG+H]+ m/z 179 [(GSH-Pyr)+H]+ m/z 233 [(GSHGly)+H]+

PAGE 118

118 Figure 5-5. The ESI MS of mixed thiols, GS H (0.05 mM), CysH (0.05 mM) and hCySH (0.5 mM). Same conditions as in Figure 5-1 Figure 5-6. Effect of on-line EC cell voltage on ESI MS of GSH (0.05 mM); Same conditions as in Figure 5-1. [GSH.+H]2+ [GSH+H]+[GSSG+H]+[2GSH+H]+

PAGE 119

119 032108 DAGSH#4bm/z 700 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 071108 DAHCys#1am/z 600 550 500 450 400 350 300 250 200 150 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 5-7. Positive mode ESI MS of (A) GSH (0.5 mM) and (B) hCysH (0.5 mM), each in the presence of DA (2.5 mM); 50/49/1 vol%, H2O/MeOH/HAc, pH~4.2; flow rate 50L/h; HV 3 kV. 032108 DAGSH#4bm/z 311 310 309 308 307 306 305 304 Intensity (%) 100 90 80 70 60 50 40 30 20 10 307 30 8 A m/z 154 [GSH.+H]2+ incl. [DA]+ m/z 137 [DA-NH3]+ m/z 307 [2DA-H]+ m/z 308 [GSH+H]+ m/z 330 [GSH+Na]+ m/z 459 [GSH+DAQ]+ m/z 538 [(GSSGGly)+H]+ m/z 613 [GSSG+H]+ m/z 154 [DA]+ B m/z 137 [DA-NH3]+ m/z 269 [hCySSCy+H]+ m/z 287 [hCysH+DAQ]+ m/z 307 [2DA-H]+ m/z 420 [DAQ+2hCysH]+

PAGE 120

120 Figure 5-8. Positive mode ESI MS of GSH (various concentrations) in the presence of DA (2.5 mM); Same conditions as in Figure 5-7. [DA]+/[GSH.+H]+[2DA-H]+[GSH+H]+[GSH+DAQ]+[GSSG+H]+[2GSH+H]+

PAGE 121

121 040308 UAGSH3#1am/z 700 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 5-9. Positive mode ESI MS of GSH (0.01 mM or 10 M) in the presence of uric acid (0.06 mM or 60 M); Same conditions as in Figure 5-1. m/z 179 [(GSHPyr)+H]+ m/z 154 [ GSH.+H ] + m/z 169 [H2U+H]+ m/z 308 [GSH+H]+ m/z 330 [GSH+Na]+ m/z 346 [GSH+K]+ m/z 337 [2H2U+H]+m/z 484 [(GSSGPyr)+H]+m/z 538 [(GSSGGly)+H]+ m/z 613 [GSSG+H]+ m/z 635/637 [GSSG+Na]+ /[2GSH+Na]+ m/z 651/653 [GSSG+K]+ /[2GSH+K]+

PAGE 122

122 072808 DAMIX#10m/z 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 5-10. Positive mode ESI MS of a thio l mixture, GSH (0.05 mM), CySH (0.05 mM) and hCySH (0.5 mM), in the presence of DA ( 2.5 mM). Same conditions as in Figure 5-7. m/z 154 [DA]+ m/z 137 [DA-NH3]+ m/z 269 [hCySSCy+H]+ m/z 287 [hCysH+DAQ]+ m/z 308 [2DA-H]+ m/z 241 [CySSCy+H]+ m/z 273 [hCysH+DAQ]+ m/z 255 [CySSCy+H]+

PAGE 123

123 Figure 5-11. Effect of applied EC cell voltage on oxidation products of (A) GSH and (B) hCySH, each in presence of DA (2.5 mM); Same conditions as in Figure 5-7. A [GSH+H]+[GSH+DAQ]+[GSSG+H]+[2GSH+H]+ B [hCysH+H]+[DA]+ [hCySSCyh+H]+[hCysH+DAQ]+[2DA-H]+ [2hCysH+DAQ]+

PAGE 124

124 1400120010008006004002000-200-400 -80 -60 -40 -20 0 20 Blank 0.5 mM Cys E (mV) i (nA) 1400120010008006004002000-200-400 -100 -80 -60 -40 -20 0 20 Blank 0.5 mM Cys 1 mM Fe(II) + 0.5 mM Cys E (mV) i (nA) 1400120010008006004002000-200-400 -80 -60 -40 -20 0 20 Blank 2.0 mM Cys E (mV) i (nA) 1400120010008006004002000-200-400 -100 -80 -60 -40 -20 0 20 Blank 2.0 mM Cys 2.0 mM Cys + 1 mM Fe(II) E (mV) i (nA) Figure 5-12. Cyclic voltmmograms ( = 50 mV/s) of cysteine (0.5 mM) alone (A) and with Fe2+ (1 mM) (B); cysteine (2.0 mM alone) (C) and with Fe2+ (1 mM) (D) on stainless steel electrode (r=50 m); Ref = SCE. Blank = 50/49/1 vol%, H2O/MeOH/HAc. A B C D

PAGE 125

125 081208 FeCys#1cm/z 700 600 500 400 300 200 100 Intensity 120 100 80 60 40 20 Figure 5-13. Positive mode ESI MS of CySH (0.5 mM) with Fe2+ (100 M). m/z 122 [CysH+H]+ Off scale ~m/z 241 [GSSG+H]+ m/z 538 [4CysH+Fe]+

PAGE 126

126 CHAPTER 6 OXIDATION OF PURINES DURING ESI MS AND EC/ESI MS Introduction Em il Fischer won the 1902 Nobel Prize in chemistry for synthesis of purines [ http://Nobelprize.org/nobel_prizes /chemistry/1902/fischer-lecture.pdf ], a group of heterocyclic organic compounds with fused pyrimidine and imidazole rings, including adenine, guanine, hypoxanthine, xanthine, and uric acid. Adenine and guanine bases are found incorporated in deoxyribonucleic acid (DNA) and ribonucleic acid (RNA) molecu les, whereas hypoxanthine and xanthine are formed as intermediates of nucleotid e catabolism which, in man, ends with uric acid as a final product [Waring et al., 2000] (see Chapter 1, Scheme 1-4). In DNA and RNA, adenine or guanine is attach ed to a sugar (2-deoxyribose or ribose) unit with an N-glycosidic bond between C1 of the sugar unit and N9 of the purine to form a nucleoside (see Chapter 1, Scheme 1-2, for numbe ring in the purine structure). The purine nucleosides are adenosine and guanosine, as well as inosine and xanthosine, which are not in DNA or RNA. Addition of one or more phosphate groups to the nucleoside (via a phosphate ester bond to C5 of the sugar or via an acid anhydride linkage if more than one phosphate is present) results in purine nucleotides : adenosine monophosphate (AMP), guanosine monophosphate (GMP), inosine monophosphate (IM P) and xanthosine monophosphate (XMP). Purines are produced from the br eakdown of adenosine triphosphate (ATP, Scheme 6-1) and guanosine triphosphate (GTP), the energy-ca rrying molecules in the body [Jencks and Wolfenden, 2000]. The ultimate source of ATP and GTP is catabolism of food. Various enzymes catalyze the different metabol ic reactions shown in Chapter 1, Scheme 14. [Simmonds et al., 1997]. These enzymes regulate purine levels within healthy limits in the

PAGE 127

127 Scheme 6-1. Structure of the nucleot ide, adenosine triphosphste (ATP). body via catalysis of biosynthesis (in case of purine deficiency) and salvage pathways (if excess of purines is produced). Abnormal conditions in cells, such as hypoxic stress and enzyme deficiency (or super-activity), can lead to accumulation of purines in various organs, such as the brain [Ford et al., 2007], heart and kidneys [Zvi et al., 2007; Ejaz et al., 2007], lead ing to disease [Simmonds et al., 1997; Kathiwala et al., 2008]. Manifestations of metabolic disorders in the purine pathway in man include i mmunological, haematological, neur ological and renal problems [Nyhan, 2005; Simmond s et al., 1997]. The analytical techniques used for purin e analysis include high performance liquid chromatography (HPLC) and capillar y electrophoresis (CE) using e ither ultraviolet (UV) light absorbance or electrospray ionization mass spectrometry (ESI MS) for detection [Ito et al., 2000; La Marca et al., 2006; Edwards et al., 2006] and cyclic voltamm etry (CV) [Cavalheiro and Brajter-Toth, 1999]. Selectivity di fficulties experienced with CV, long analysis times associated with HPLC are some of the limitations of these techniques in clinical analysis of purines and screening of biological samples. One of the st rategies for improving throughput involves direct injection ESI MS with high resolution mass analy zers, such as Fourier transform ion cyclotron resonance (FT-ICR), to identif y metabolites without need for chromatographic separation.

PAGE 128

128 However, electrochemical processes of the ES ion source can affect the sensitivity with which analytes of low oxidation potenti als such as purines are detect ed. Complete electrochemical oxidation of purines may involve oxygen, multiple el ectrons and protons, and is affected by pH [Cavalheiro and Brajter-Toth, 1999; Owens and Dryhurst, 1977]. It is apparent that electrochemical oxidation of purin es can be slow and complex, wh ich may affect their detection sensitivity. Literature peak oxidation potential (Ep) values [Dryhurst and Elving 1968; OlivieraBrett et al., 2002; Cavalheiro and Brajter-Toth, 1999] of purine bases (see structures in Scheme 6-2), at similar electrodes and pH values, can be arranged as uric acid < xanthine < guanine < hypoxanthine < adenine. This chap ter describes electrochemical ox idation reactions that purines may undergo during positive ion mode ESI MS. Scheme 6-2. Structures of purine bases, pKa values [Budavari, 1989; Rogstad et al., 2003] and one-electron oxidation potentials (E1 vs SHE) [Jovanovic and Simic, 1986]

PAGE 129

129 Results and Discussion The ESI MS and EC/ESI MS of Guanine (Gua) The ESI m ass spectrum of guanine prepared in 50/49/1 vol%, H2O/MeOH/HAc, pH 4.2, is shown in Figure 6-1. The peaks observed at m/z 152 and 302 are assigned to proton adducts, [Gua+H]+ and [2Gua+H]+, while those at m/z 174 and 325 are sodium adducts [Gua+Na]+ and [2Gua+Na]+ of guanine and guanine dimer, respectiv ely. The potassium adduct of guanine dimer [2Gua+K]+ (m/z 341) is also observed. Informed by the mechanism of oxidation of guanine proposed by Dryhurst, [1969], the small peak at m/z 168 was identified as 8-oxoguanine [OxoGua+H]+, a hydrolyzed 2e-, 2H+ oxidation product of guanine. A larger ESI MS signal is obtained for 8-oxoguanine [OxoGua+H]+ (m/z 168) peak when the EC cell is turned ON (Figure 6-2). Its intensity increases as applied EC cell voltage is increased (Figure 6-3), while the intensity of the guanine dimer decreas es. Intensities of guanine [Gua+H]+ (m/z 152) and the deaminated guanine [(Gua-NH3)+H]+ (m/z 135) increase with applied EC cell voltage, a behavior similar to that of DA discussed in Chapter 3. In 40/60 vol%,, H2O/MeOH, 1 mM NH4Ac, pH 6-3, peaks due to hydrogen bonded guanine tetramers (see Scheme 6-3) with Na+ and K+ metal ions in the center (m/z 627 and 643, respectively) [Manet et al., 2001] are observed (Figure 6-4), along with their adducts with Na+ and K+ ions. It should be noted that a high flow rate (50 L/h) was reached in this experiment to achieve a stable electrospray current. Based on flow rate vs intensity (of DA) data described in Chapter 3, more analyte ions, henc e high intensity, may be detected. It is also possible that the presence of a guanine dimer peak indicates th e tendency of the carrier solution composition to promote dimer formation, particularly in presence of alkali-metal ions [M anet et al., 2001] at concentrations that are relatively higher than that of protons (H+), at pH values below pKa.

PAGE 130

130 N N H N N N O NNH N N N O N H N NN N O N HN N N N O H H H H H H H H H H H H M+Guatetramer MW=604+M+ Scheme 6-3. The H-bonded guanine tetramer with a metal ion center (M+ = sodium ion, Na+ or potassium ion, K+) The proposed mechanism of formation of guanine dimer and 8-oxoguanine during positive ion mode ESI MS is shown in Scheme 6-4, where the initial loss of 1e-, 1H+ leads to a dimeric radical which is stabilized by hydrogen atom transfer through H-bonding with the third unoxidized guanine molecule (Chapt er 4, Scheme 4-4) for uric acid. A neutral guanine radical generated in this step can be oxidized further by loss of 1e-, 1H+, followed rapidly by hydrolysis to 8-oxoguanine. The instability of 2e-, 2H+ oxidation product observed here is not peculiar to guanine. The same level of instability is demonstrated by the 2e-, 2H+ oxidation product of ur ic acid, the uric

PAGE 131

131 Scheme 6-4. Proposed mechanism of oxi dation of guanine in 40/60 vol%, H2O/MeOH, 10-3 M NH4Ac, pH 6.3, during positive mode ESI MS. acid diimine, which is also hydrolyzed upon form ation [Volk et al., 1992]. The absence of 8oxoguanine in 50/49/1 vol%, H2O/MeOH/HAc, pH 4.2, (Figures 6-1 and 6-2) could be due to stable tetramers with a metal ion center formed by guanine in 40/60 vol%,, H2O/MeOH, 10-3 M

PAGE 132

132 NH4Ac, pH 6.3 carrier solution, whic h may be more difficult to ox idize than free guanine. The concentration of Na+ and K+ ions, which promotes aggregation of guanine [Manet et al., 2001], is relatively higher than that of th e protons in the 40/60 vol%, H2O/MeOH, 10-3 M NH4Ac, pH 6.3, carrier solution. Thus the formation of guanine tetramer is favored in this carrier solution than in 50/49/1 vol%, H2O/MeOH/HAc, pH 4.2. The ESI MS and EC/ESI MS of Adenine (Ad) Figure 6-5 shows the ESI MS of adenine in 50/49/1 vol%, H2O/MeOH/HAc, pH 4.2. The peaks observed at m/z 119, 136 and 158 are assigned to the proton adduct of the deaminated adenine [(Ad-NH3)+H]+, and the proton and sodium adducts of adenine [Ad+H]+ and [Ad+Na]+, respectively. None of the peaks observed in the ESI MS of adenine in th e same carrier solution (Figure 6-5) can be attributed to oxidation products of adenine (pKa (N7H) = 10.2) [Rogstad et al., 2003]. Neither did new peaks appear with applic ation of EC cell voltage (up to 4 V) to this solution of adenine. Use of the relativel y less acidic carrier solution 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, pH 6.3 (Figure 6-6) revealed some ne w peaks, including those at m/z 271 and 309, assigned to the proton and potassium a dducts of the adenine dimer [2Ad+H]+ and [2Ad+K]+. However, the application of EC cell voltage to the pH 6.3 solution produced multiple high mass peaks (data not shown) possibly due to adducts of multiple electron (up to 6e-) oxidation products of adenine with Na+ and K+ ions, made the mass spectrum too complex for meaningful analysis. The dimer peaks disappeared with in creasing EC cell voltage indicating oxidative decomposition. It has been reported by Dryhurst a nd Elving [1968] that adenine oxidation is an overall 6e-, 6H+ process, which they propos ed occurred by sequential 2e-, 2H+ oxidations. The ESI MS results described here support the 1e-, 1H+ oxidation of adenine, wh ich leads to transient 2e-, 2H+ oxidation products, similar to oxidation of uric acid (Chapter 4) and guanine (described in the preceding section).

PAGE 133

133 Fragmentation and further electrochemi cal reactions of the intermediate 2e-, 2H+ oxidation product of adenine are known to be numerous [D ryhurst and Elving 1968] and this is reflected here by the complex EC/ESI MS mass spectra. Ne vertheless, to the extent that adenine (pKa = 10.2) undergoes 1e-, 1H+ oxidation at pH~6.3, to form the detected adenine dimer [2Ad+H]+ (m/z 271) or [2Ad+K]+ (m/z 309), a mechanism is proposed (Scheme 6-5). Scheme 6-5. Proposed mechanism of oxidati on of adenine during positive mode ESI MS.

PAGE 134

134 The ESI MS and EC/ESI MS of Hypoxanthine (hXan) Because hypoxanthine is m uch less soluble in 50/49/1 vol%, H2O/MeOH/HAc, pH 4.2 than in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, pH 6.3, this latter carri er solution was chosen for ESI MS and EC/ESI MS experiments. When prepared in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, pH 6.3 carrier solution, hypoxanthine (pKa1 = 8.6) [Budavari, 1989] gives two prominent peaks in the ESI MS mass spectrum (Figure 6-7). The peaks at m/z 137 and 273 are due to proton adducts of hypoxanthine [hXan+H]+ and hypoxanthine dimer [2hXan+H]+. Only a small [hXan+H]+ (m/z 137) peak, which could not be reproduced well enough for meaningful analysis, was observed when using 50/49/1 vol%, H2O/MeOH/HAc, pH 4.2 carrier solution, and the hypoxanthine dimer [2hXan+H]+ (m/z 273) peak was not observed. Like the other purines analyzed thus far, hypoxanthine appears to undergo 1e-, 1H+ oxidation during positive ion mode ESI (proposed mechanism shown in Scheme 6-6), although no 2e-, 2H+ oxidation products of hypoxanthine were dete cted even when voltage was applied to the EC cell. With increasing EC cell voltage, hypoxa nthine is efficiently ionized and remains the base peak. New (unassigned) peaks appear in the mass spectrum at m/z values between 137 and 550, while the relative intensity of hypoxanthine dimer [2hXan+H]+ (m/z 273) decreases (Figure 6-8) in the same manner as shown for dopamine, ur ic acid and guanine (C hapters 3, 4, and 6). It is possible that the later 1e-, 1H+ oxidation steps of hypoxanthine are kinetically more facile than hydrolysis or hydroxylation steps which form xanthine, and highly reactive species such as diimine are formed.

PAGE 135

135 Scheme 6-6. Proposed mechanism of oxidati on of hypoxanthine during positive mode ESI MS. The ESI MS and EC/ESI MS of Xanthine (Xan) Like hypoxanthine, xan thine is poorly soluble in low pH solutions and undergoes multiple electron (up to 4e-) oxidation overall [kathiwa la et al., 2008]. Xanthine (50 M) was prepared in

PAGE 136

136 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, pH 6.3, the same carri er solution used for hypoxanthine, and the mass spectrum shown in Fi gure 6-9 was obtained. In contrast to other purines, the ESI MS mass sp ectrum of xanthine indicates that it forms a dimeric radical which is detected as [(2Xan-H).+Na]+ (m/z 326). The presence of xanthi ne radicals supports a radical pathway of electrochemical oxidation of xa nthine, recently proposed by Kathiwala et al [kathiwala et al., 2008] based on cyclic voltammetry. The dimeric radical intensity of xanthine increases slowly as applied EC cell voltage is increased (Figure 6-10). Similar to guanine, xanthine also forms an H-bonded tetramer with a sodium ion [4Xan+Na]+ (m/z 631] at the center (Scheme 6-7). N N H N H O N O NN NH O N O N N H NO N O N N HN O N O H H H H M+H H H H Xantetramer MW=608+M+ Scheme 6-7. The H-bonded xanthine te tramer with a metal ion center (M+ = sodium ion, Na+) The signal intensity of the H-bonded te tramer/sodium ion dduct [4Xan+Na]+ decreases in intensity as the EC cell voltage increases (Fig ure 6-10), probably because it is oxidized. The proposed mechanism of formation of the dimeri c xanthine radical is shown in Scheme 6-8.

PAGE 137

137 Scheme 6-8. Proposed oxidation mechanism of xa nthine to xanthine ra dicals [Adapted from Kathiwala et al., 2008]. The final product (2 Xan-H) is detected as a [(2Xan-H).+Na]+ (m/z 326). Conclusions The ESI MS and EC/ESI MS data presented in th is chapter indicate that oxidation of purines does occur during positive ion mode ESI by stepwise 1e-, 1H+ losses. With the exception of xanthine, oxidations of purines appear to follow a general pa th illustrated in Scheme 6-9,

PAGE 138

138 H-atom transfer Final Oxidation Products [2m] Dimer [m.]Electrospray Capillary Wall m m m [m.+m] Scheme 6-9. Proposed general pa th of oxidation of purines duri ng positive mode ESI MS (m = purine metabolite). where the radical gene rated by the initial 1e-, 1H+ oxidation step attacks an unoxidized purine to stabilize and by so doing forms the dimeric radical. Through hydrogen bonding with a third unoxidized purine, a hydrogen atom is transfered to the dimeric radical thereby fo rming a neutral purine dimer. Th e neutral purine radical resulting from H-atom transfer to the dimeric radical either combines with another unoxidized purine molecule in a cycle (Scheme 69), or it undergoes the second 1e-, 1H+ oxidation step to form a 2e-, 2H+ oxidation product and subse quent products, if the 2e-, 2H+ oxidation product is not stable.

PAGE 139

139 Hydrogen bonding between two purine molecules is apparently not favored because the two molecules are required to be perpendicu lar to one another. The favorable H-bonding arrangement is one where four purine molecules fo rm a square. Higher stability is reached with a metal ion in the center. The tetramer arrangemen t observed in the guanine and xanthine cases is evidence of this peculiar Hbonding. Antecedent also supports th e proposition that the observed purine dimers are electrochemically generated, wh ich is also indicated by the positive response when voltage is applied to the EC cell leading to the observed 2e-, 2H+ products.

PAGE 140

140 080907 Gua#1bm/z 700 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 6-1. The ESI MS of guanine (50 M) in 50/49/1 vol%, H2O/MeOH/HAc, pH~4.2; HV 3 kV; Flow rate 30 L/h. 080907 Gua#4bm/z 700 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 6-2. The EC/ESI MS of gua nine (50 M) in 50/49/1 vol%, H2O/MeOH/HAc, pH~4.2; HV 3 kV; Flow rate 30 L/h; EC cell voltage 1.5V. m/z 152 [Gua+H]+ m/z 303 [2Gua+H]+ m/z 135 [(Gua-NH3) +H]+ m/z 325 [2Gua + Na] + m/z 168 [OxoGua+H]+ m/z 341 [2Gua+K]+ m/z 152 [Gua+H]+ m/z 303 [2Gua+H]+ m/z 174 [Gua+Na]+ m/z 325 [2Gua+Na]+ m/z 168 [OxoGua+H]+ m/z 341 [2Gua+K]+

PAGE 141

141 0 20 40 60 80 100 120 00.511.522.533.5 EC Cell Voltage (V)Intensity (%) m/z 135 m/z 152 m/z 168 m/z 303 Figure 6-3. The EC/ESI MS of guanine (50 M). Other conditions as in Figure 6.2. 050708 Gua #1m/z 700 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 6-4. The ESI MS of guani ne (50 M) in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, pH~6.3; HV 3 kV; Flow rate 50 L/h. m/z 152 [Gua+H]+ m/z 190 [Gua+K]+ m/z 303 [2Gua+H]+ m/z 627 [ 4Gua+Na ] + m/z 643 [4Gua+K]+ m/z 325/341 [2Gua+Na/K]+ m/z 174 [Gua+Na]+ m/z 135 [(Gua-NH3) +H]+ [(Gua-NH3)+H]+ [(Gua+H]+ [OxoGua+H]+ [2Gua+H]+

PAGE 142

142 072607 Ad#1am/z 700 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 6-5. The ESI MS of adenine (50 M) in 50/49/1 vol%, H2O/MeOH/HAc, pH 4.2; HV 3 kV; Flow rate 30 L/h; EC cell voltage 1.5V. 041808 Ad #1bm/z 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 6-6. The ESI MS of adenine (50 M) in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, pH 6.3. Other conditions as in Figure 6-4. m/z 136 [Ad+H]+ m/z 309 [2Ad+K]+ m/z 271 [2Ad+H]+ m/z 174 [Ad+K]+ m/z 136 [Ad+H]+ m/z 119 [(AdNH3)+H]+ m/z 158 [Ad+Na]+

PAGE 143

143 081208 hXan #1bm/z 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 6-7. The ESI MS of hypoxant hine (50 M) in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, pH 6.3. Other conditions as in Figure 6-4. Figure 6-8. The EC/ESI MS of hypoxanthine (50 M). Other conditions as in Figure 6-4. [hXan+H]+ [2hXan+H]+ m/z 137 [hXan+H]+ m/z 273 [2hXan+H]+

PAGE 144

144 081208 Xan #1m/z 700 600 500 400 300 200 100 Intensity (%) 100 90 80 70 60 50 40 30 20 10 Figure 6-9. The ESI MS of xant hine (50 M) in 40/60 vol%, H2O/MeOH, 1 mM NH4Ac, Other conditions as in Figure 6-4. Figure 6-10. The EC/ESI MS of xanthine (50 M). Other conditions as in Figure 6-4. m/z 153 [Xan+H]+ m/z 326 [(2Xan-H).+Na] + m/z 631 [ 4Xan+Na ] + m/z 175 [Xan+Na]+ [Xan+H]+ [(2Xan-H).+Na]+ [4Xan+Na]+

PAGE 145

145 CHAPTER 7 CONCLUSIONS A novel, integrated on-line EC/ESI MS syst em with adjustable EC cell voltage was developed in this work. It is a robust system for detection of many types of analytes, including neutrals and negatively charged an alytes, which were previously regarded as outside the domain of positive ion mode ESI MS. The on-line elec trochemical cell was integrated into the electrospray needle by dividing th e needle into two sections and joining them with plastic tubing as described in Chapter 2. The EC cell voltage made it possible to augment the electrochemical processes that are inhere nt to the ESI operation for sensitiv ity enhancement and elucidation of electrochemical reactions. As i ndicated by the cited literature, others have sought to suppress electrochemistry of the electrospray ion sour ce to avoid oxidation of analytes with limited success. The approach adopted in the present work was to augment this inherent electrochemistry and actually understand how it affects analytes in order to use ESI MS more effectively for bioanalytical work and chemical analysis in general. The EC/ESI MS results of dopamine discussed in Chapter 3 showed th at applied EC cell voltage allows electrochemical reactions at the back-end of the ESI needle, in addition to th ose occurring at the tip. This increases the active surface area of the electr ode (i.e. the ES needle), as refl ected by up to 100% increase in ES current and improved ionization efficiency. While substantial, the extent to which the ESI MS signal was enhan ced by application of EC cell voltage in EC/ESI MS was a factor of eight less than the signal enhancement achieved by using a cone-shaped MS capillary inlet instead of a standard cylindrical inlet. The large, coneshaped orifice (about 2.5 cm dept h) allows 100% collection of the electrospray vapor from an 80 m internal diameter (i.d.) electrospray needle positioned 0mm away. The ES needle can be 0 mm away from the MS inlet and still allow an air gap for electrospray, because the needle

PAGE 146

146 diameter (~ 100 m) is much smaller than the 6.1 mm orifice diameter of the MS inlet. With appropriate carrier soluti on flow rate and conductivity, it is believed that this arrangement can ensure 100% collection of the electrospray pl ume. However, collection efficiency does not exactly translate into ion transmission efficienc y, because of electrostatic attraction of positively charged ions to the relatively negative MS inlet and subsequent neutralization. More precisely, the MS inlet is held at a positive potential which is less than the applied needle voltage. It is more appropriate then, to talk of electrostatic repulsion from the ES needle tip and migration of ions in the electric field gradient across the inte rface, rather than electrostatic attraction to the MS inlet. If the latter were the predominant cas e, most ions would simply be pulled from the axial center to the MS in let surface/wall and not ente r the mass spectrometer. Ion formation mechanisms have been elucidat ed to varying degrees in this work. The validity of assignments of various ions was verified by close matc hes of measured and throretical isotope distributions, as well as by hydrogen/deuterium (H/D) excha nge experiments. In the case of dopamine and uric acid, infra-red multiple photon dissociation (IRMPD) followed by tandem mass spectrometric (MS/MS) analyses was also performed. By itself, MS/MS evidence was not conclusive but did not disprove the covalent nature of the elec trochemically generated dopamine and uric acid dimers. The numbers of excha ngeable hydrogens, determined by H/D exchange experiments, were in perfect agreement with th ose in proposed covalently linked dimers which differ by at least two hydrogens (i.e. 2 mass units ) from their hydrogen bonded counterparts. In the case of dopamine, a hydrogen-bonded dimer woul d be doubly charged in the carrier solution used, considering the pH of the solution and the pKa values of dopamine. The clear effect of applied EC cell voltage reflected in the signal inte nsity vs applied EC cell voltage curves, as well as the appearance of both covalently bonded di sulfide dimers and hydrogen bonded thiol dimers,

PAGE 147

147 are evidence of electrochemical activity and the presence of electrochemically generated covalently bonded dimers. Mechanisms of formation of ions of the diffe rent analytes observed in their ESI MS mass spectra were proposed in this work, both indivi dually and in generic te rms. The ESI MS data indicate, as reflected by the proposed mechanisms, that oxidation of all th e analytes considered in this work proceeds by discrete 1e-, 1H+ losses. Positive ion mode ESI is a radiation-free technique to generate radicals and study their reactions, par ticularly those involving biological molecules in aqueous milieu. Analytes with pe ndant groups, such as dopamine and glutathione, undergo fragmentation with clea vage of the pendant groups. More fragments are observed for larger molecules such as glutat hione. Products from the second 1e-, 1H+ losses, where applicable, were observed as minor peaks, which could be an indication that this step is kinetically slow during ESI. Analytes known to involve more than 2e-, 2H+, which have relatively high oxidation potentials, such as adenine and xanthine, te nd to produce complex mass spectra, and further work is required to elucidate a ll the oxidation steps using ESI MS. Future work could include the application of the EC/ESI MS system developed in this work to analysis of real biological samples such as cells. With better und erstanding of all factors affecting analyte detectability in ESI MS, particularly ion intensity/ES current as a function of concentration, calibration curves could be constructed for analytical determinations. Understanding is a mental process. If mental pr ocesses are electrochemical in nature, then the answer to the question, Is electrochemistry essent ial to the unde rstanding of electrospray ionization? would be Yes, irre spective of any relation, or lack thereof, between electrochemistry and the electrospra y process. John Fenn [in Electrochemical processes in electrospray ionization mass spectrometry Special Feature: Discussion, J. Mass Spectrom. 2000, 35, 939 952]

PAGE 148

148 LIST OF REFERENCES Aebersold, R.; Mann, M. Nature 2003, 422, 198. Alderm an, M. Kala, J. V. A. Curr. Med. Res. Opinion 2004, 20, 369. Allen, R. N.; Shukla, M. K.; Leszczynski, J. Int. J. Quant. Chem. 2004, 100, 801. Ames, B. N.; Cathcart, R.; Schwiers, E.; Hochstein, P. Proc. Nat. Acad. Sci.USA 1981, 78, 6858. Anderson, R. F.; Harris, T. A.; Free Rad. Res. 2002, 37, 1131. Aplin, K. L. Rev. Sci. Instrum. 2005, 76, 104501-1. Arakawa, R. Yamaguchi, M. Hotta, H. Osakai, T. Kimoto, T. J. Am. Soc. Mass Spectrom. 2004, 15, 1228. Arnobe, M. G.; Luth, M. S.; Roitzsch, M.; Cerda, M. M.; Lax, P.; Kampf, G.; Sigel, H.; Lippert, B. Chem. Eur. J. 2004, 10, 1046. Bald, E. B. J. Chromatogr. A 2004, 1032, 109. Bard, A. J.; Faulkner, R. L. Electrochemical Methods, 2nd ed, John Wiley & Sons, 2001. Baykut, G.; Eyler, J. R. Trends Anal. Chem. 1986 5, 44. Becerra, A.; Lazcano, A. Orig. Life Evol. Bios. 1998, 28, 539. Benavente, F. van der Heijden, R. Tjaden, U. R. van der Greef, J. Hankemeier, T. Electrophoresis 2006, 27 4570. Binyamin, G. Chen, T. Heller, A. J. Electroanal. Chem. 2001 500, 604. Blackley, C. R.; Vestal, M. L. Anal. Chem. 1983, 55, 750. Blades, T.; Ikonomou, M. G.; Kebarle, P. Anal. Chem. 1991, 64, 2109. Blakney, G. T.; Robinson, D. E.; Ly, N. V.; Kelleher, N. L., Hendrickson, C. L.; Marshall, A. G. Predator: PCI Data Station fo r FT-ICR Mass spectrometry, 53rd Amer. Soc. Mass Spectrom. Ann. Conf. on Mass Spectrom. & Allied Topics, San Antonio, TX, June 4-9 (2005) (Poster) Blank, C. L.; McCreery, R. L.; Wightman, R. M.; Chey, W.; Adams, R. N.; Reid, J. R. E. E. Smissman, J. Med. Chem. 1976, 19, 178. Bogdanov, B.; Smith, R. D. Mass Spectrom. Rev. 2005, 24, 168. Bond, A. M.; Colton, R.; DAgostino, A.; Downard, A. J.; Traeger, J.C. Anal. Chem. 1995, 67, 1691.

PAGE 149

149 Bouligand, J.; Deroussent, A.; Paci A.; Morizet, J.; Vassal, G. J. Chromatogr. B 2006, 832, 67. Bravo, R.; Hsueh, C.; Jaramillo, A.; Brajter-Toth, A. Analyst 1998, 123, 1625. Brockman, T. J.; Anderson, L. B. Anal. Chem. 1984, 88, 2. Bruckenstein, S.; Gadde, R. R. J. Am. Chem. Soc. 1971, 93, 793. Budavari, S.; ONeil, M. J.; Smith, A.; Heckelman, P. E., Eds. The Merck Index 11th ed. Merck & Co. Inc., New Jersey, 1989. Buettner, G. R.; Jerkiewicz, A. A. Radiation Res. 1996, 145, 532. Camera, E.; Picardo, M. J. Chromatogr. B 2002, 781, 181. Cappiello A.; Bruner, F. Anal. Chem. 1993, 65, 1281. Cavalheiro, E, T. G.; Brajter-Toth, A. J. Pharm. Biomed. Anal. 1999, 19, 217. Cech, N. B.; Enke, C. G. Mass Spectrom. Rev. 2001, 20, 362. Constentin, C.; Robert M.; Saveant, J. J. Am. Chem. Soc. 2007, 129, 5870. Cotton, F. A.; Wilkinson, G.; Gaus, P. L. Basic Inorganic Chemistry 2nd Ed. John Wiley & Sons, New York 1987, p672. Creaser, C. S.; Stygal, J. W. Analysis 1993, 118, 1467. Curello, S.; Ceconi, C.; Cargnoni, A; Cornacch iari, A.; Ferrari, R.; Albertini, A. Clin. Chem. 1987, 33, 1448. de la Mora J. F. J. Fluid Mech. 1992, 243, 561. Demuth, K.; Drunat, S.; Girerd, X.; Moatti, N; Paul, J-. L.; Safar, M; Boutouyrie, P. Atherosclerosis 2002, 165 167. Deng, H.; Van Berkel, G. J. Electroanalysis 1999 11, 857. Diehl, G.; Karst, U. Anal. Bioanal. Chem. 2002, 373, 390. Dole, M.; Mack, L. L.; Hines, R. L.; Mobley, R. C.; Ferguson, L. D.; Alice, M. B. J. Chem. Phys. 1968, 49; 2240. Dryhurst, G. J. Electrochem. Soc. 1969, 116, 1411. Dryhurst, G.; Elving, P. J. Electrochem. Soc. 1968, 115, 1014. Dutt, J. S. N.; Cardosi, M. F.; Davis, J. Analyst 2003, 128, 811.

PAGE 150

150 Edwards, J. L.; Chisolm, C. N.; Shackman, J. G.; Kennedy, R. T. J. Chromatogr. A 2006, 1106, 80. Ejaz, A. A.; Mu. W.; Kang, D. H.; Roncal, C.; Sautin, Y. Y.; Henderson, G.; Tabah-Fisch, I.; Keller, B.; Beaver, T. M.; Nakagawa, T.; Johnson, R. J. Clin. J. Am. Soc. Nephrol. 2007, 2, 16. Fenn, J. B. Angew. Chem. Int. Ed. 2003, 42, 3871. Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Mass Spectrom. Rev. 1990, 9, 37. Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science 1989, 246 64. Ford, E. S.; Li, C. Y.; Cook, S.; Choi, H. K. Circulation 2007 115, 2526. Forman, H. J.; Dickinson, D. A. Biofactors 2003, 17, 1. Gamache, P. H. Meyer, D. F. Granger, M. C. Acworth, I. N. J. Am. Soc. Mass. Spectrom. 2004, 15, 1717. Garcia, S. C.; Schott, K.; Charao, M.; Moro, A.; Bu lcao, R.; Grotto, D.; Valentini, J.; Bohrer, D.; Cardoso, S.; Pomblum, V. Biomed. Chromatogr. 2008, 22, 460. Gerhardt, R.; Adams, N. Anal. Chem. 1982, 54, 2618. Getek, T. A.; Korfmacher W. A.; McRae, T. A.; Hinson, J. A. J. Chromatogr. A 1989, 474, 245. Gilbert, H. F. Methods Enzymol. 1995, 251, 8. Giustrarini, D.; Milzani, A.; Dalle-Donne, I.; Rossi, R. Blood Cells, Molecules & Diseases 2008, 40, 174. Glish, G. L.; Vachet, R. W.; Nature, 2003, 2, 140. Grosshans, P. B.; Marsh all, A. G. Int. Mass Spectrom. 1991, 107, 49. Gucek, M.; Makuc, S.; Mlakar, A.; Be rincnik-Vrbovsek, J.; Marsel, J. Rapid Commun. Mass Spectrom. 2002, 16, 1186. Hambitzer, G.; Heitbaum, J. Anal. Chem. 1986, 58, 1067. Harris, D. C. Quantitative Chemical Analysis 4th ed, W.H. Freeman & Co., New York 2007. Hartman, S. Okun, J. G. Schmidt, C. Langhans, C. Carbade, S. F. Burgard, P. Haas, D. Sass, J. O. Nyhan, W. Ho ffmann, G. F. Clin. Chem. 2006 526, 1127. Hawley, M. D.; Tatawadi, S. V.; Piekarski, S.; Adams, R. N. J. Am. Chem. Soc. 1967, 89, 447.

PAGE 151

151 Hayden, M. R.; Tyagi, S. C. Nutrition & Metabolism 2004, 1:10, http://www.nutritionandmetabolism. com /content/pdf/1743-7075-1-10.pdf Hicks, M. Wong, L. S. Day, R. O. Free Rad. Res. Comms. 1993, 18, 337. Himmelfarb, J.; McMenamin, E.; McMongle, E. Kidney Int. 2002 61, 705. Hooper, D. C. Scott, G. S. Zborek, A. Mikheeva, T. Kean, R. B. Koprowski, H. Spitsin, S. V. FASEB 2000 14, 691. Ikonomou, M. G.; Blades, A. T.; Kerbale, P. Anal. Chem. 1991, 63, 1989. Iribane, J. V.; Thomson, B. A. J. Chem. Phys. 1976, 64, 2287. Ito, T. van Kuilenburg, A. B. P. Bootsma, A. H. Haasnoot, A. J. van Cruchten, A. Wada, van Y. Gennip, A. H. Clin. Chem. 2000, 46, 445. Griffiths, M. J. Biol. Chem. 1952, 197, 399. Jackson, G. S.; Enke, C. G. Anal. Chem. 1999, 71, 3777. Jencks, W. P.; Wolfenden, R. V. Biogr. Mems Fell. R. Soc. 2000, 46, 335. Jiang, H.; Dolan, A. R.; Li, M.; Moy, M. A.; Gong, B.; Wood, T. D. Int. J. Mass Spectrom. 2007, 261, 13. Jovanovic, S. C.; Simic, M. G. J. Phys. Chem. 1986, 90, 974. Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60, 2299. Karst, U. Angew. Chem. Int. Ed., 2004 43, 2476. Kathiwala, M.; Affum, A. O.; Perry, J.; Brajter-Toth, A. Analyst 2008, 133, 810. Kebarle, P. J. Mass. Spectrom. 1997, 32, 922. Kelly, R.; P age, J.; Tang, K.; Smith, R. Anal. Chem. 2007, 79, 4192. Kemp, M.; Go, Y. -M. Jones D. P. Free Radical Biol. Med. 2008, 44, 921. Kertesz, V.; Van Berkel, G. J. J. Am. Soc. Mass Spectrom. 2006, 17, 953. Kertesz, V.; Van Berkel, G. J. J. Solid State Electrochem. 2005, 9, 390. Kertesz, V.; Van Berkel, G. J.; J. Mass Spectrom. 2001, 36, 204. Kim, T.; Udseth, H.; Smith, R.; Anal. Chem. 2000 72, 5014. Konermann, L.; Silva, E. A.; Sogbein, O. F. Anal. Chem. 2001 73, 4836.

PAGE 152

152 Kupeli, S. Yurdakok, M. Kilinc, G. Bilgetekin, E. Pediatr. Nephrol. 2005, 20, 1361. la Marca, G. B. Casetta, S. Malvagia, E. Pasqui ni, M. Innocenti, M. A. Donati, E. Zammarchi, J. Mass Spectrom. 2006, 41, 1442. LaVoie, M. J.; Hastings, T. G. J. Neurosci. 1999 19, 1484. Li, Y.; Pozniak, B. P.; Cole, R. B. Anal. Chem. 2003, 75, 6987. Liao, C. -Y.; Zen, J. M. Sens. Actuators B 2008, 129, 896. Liljegren, G.; Dahlin, A.; Zettersten, C.; Bergquist, J.; Nyholm, L. Lab Chip 2005, 5, 1008. Linert, W.; Herlinger, E.; Jameson, R. F.; Ki enzl, E.; Jellinger, K.; Youdim, M. B. H. Biochim. Biophys. Acta 1996, 1316, 160. Liu, D. Q.; Wu, L.; Sun, M.; MacGregor, P. A. J. Pharm. Biomed. Anal. 2007, 44, 320. Liu, M.; Li, P.; Cheng, Y.; Xian, Y.; Jin, L. Anal. Bioanal. Chem. 2004, 380, 742. Lohman, W.; Karst, U. Anal. Bioanal. Chem. 2008 391, 79. Lu, W.; Xu, X.; Cole, R. B. Anal. Chem. 1997, 69, 2478. Mack, L. L.; Kraklik, P.; Rheude, A.; Dole, M. J. Chem. Phys. 1970, 52; 4977. Maleknia, S. D.; Chance, M. R.; Downard, K. M. Rapid Commun. Mass Spectrom. 1999, 13, 2352. Manet, I.; Francini, L.; Masiero, S.; Pierac cini, S.; Spada, G. P.; Gottarelli, G. Helvetica Chim. Acta 2001, 84, 2096. Maples, K. R. Mason R. P. J. Biol. Chem. 1988, 263, 1709. Marshall, A. G.; Co misarow, M. Chem. Phys. Lett. 1974, 25, 282. Marshall, A. G.; Hendrickson, C. L.; Jackson, G. S. Mass Spectrom. Rev. 1998, 17, 1. Mautjana, N. A. Estes, J. Eyler, J. R. Brajter-Toth, A. Electroanalysis 2008a, 20, 1959. Mautjana, N. A.; Estes, J.; Eyler, J. R.; Brajter-Toth, A. Electroanalysis 2008b In Press Meister, A. Science 1983 220, 472. Mentasti, E. Rinaudo, C. Boistelle, R. J. Chem. Eng. Data 1983, 28, 247. Moini, M.; Cao, P.; Bard, A. J.; Anal. Chem. 1999, 71, 1658. Mutoh, M.; Kieda, S.; Kmimura, K. J. Appl. Phys. 1979, 50, 3174.

PAGE 153

153 Nekrassova, O.; Lawrence, N. S.; Compton, R. G. Talanta 2003, 60, 1085. Nemes, P.; Marginean, I.; Vertes, A. Anal. Chem. 2007, 79, 3105. Nolin, T. D.; McMenamin, M. E.; Himmelfarb, J. J. Chromatogr. B 2007, 852, 554. Nyhan, W. L.; Mol. Genet. Metab ., 2005, 86, 25. Ochran, R. A.; Konermann, L. J. Am. Soc. Mass Spectrom. 2004, 15, 1747. Ogasawara, Y.; Mukai, Y.; Togawa, T.; Suzuki, T.; Tanabe, S.; Ishii, K. J. Chromatogr. B 2007, 845, 157. Okumura, L. L.; Stradiotto, N. R.; Rees, N. V.; Compton, R. G. Electroanalysis 2008 8, 916. Oliviera-Brett, A. M.; Dicule scu, V.; Piedade, J. A. P. Bioelectrochemistry 2002, 55, 61. Owens, J. L.; Dryhurst, G. Anal. Chim. Acta 1977, 89, 609. Pacsial-Ong, E. J.; McCarley, R. L.; Wang, W.; Strongin, R. M. Anal. Chem. 2006, 78, 7577. Pascual, E.; Batlle-Gualda, E.; Martnez, A.; Rosas, J.; Vela, P. Ann. Internal Med. 1999; 131, 756. Permentier, H.; Bruins, A. P.; Bischoff, R. Mini-Rev.Med. Chem. 2008, 8, 46. Phillips, W. D.; Poe, M.; Weiher, J. F.; McDonald, C. C.; Lovenberg, W. Nature, 1970, 227, 574. Pozniak, B. P.; Cole, R. B. J. Am. Soc. Mass Spectrom. 2004, 15, 1737. Pozniak, B. P.; Cole, R. B. J. Am. Soc. Mass Spectrom. 2007, 18, 737. Price, W D.; Schnier, P. D.; Williams, E. R. Anal. Chem. 1996 68, 859. Rahman, I.; Biswas, S. K.; Jimenez, Luis, A.; Torres, M.; Forman, H. J. Antioxid. Oxidation Signal. 2005 7, 42. Regino, M. C. S.; Brajter-Toth, A. Anal. Chem. 1997, 69, 5067. Rogstad, K. N.; Jang, Y. H.; Sowers, L. C.; Goddard, W. A. Chem. Res. Toxicol. 2003 16, 1455. Rohner, T. C.; Lion, N.; Girault, H. H. Phys. Chem. Chem. Phys. 2004, 6, 3056. Rose, K.; Shndle, S. E.; Eidsness, M. K.; Kurtz, D. M. Jr; Scott, R.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1998, 120, 10743. Roussel, C.; Dayon, L.; Lion, N.; Rohner, T.; Jossera nd, J.; Rossier, J. S.; Je nsen, H.; Girault, H. H. J. Am. Soc. Mass Spectrom. 2004, 15, 1767.

PAGE 154

154 Rubino, F. M.; Pitton, M.; Brambilla, G.; Colombi, A. J. Mass Spectrom. 2006, 41, 1578. Rubino, F. M.; Verduci, C.; Giampiccolo, R.; Pulvirenti, S.; Brambilla, G.; Colombi, A. J. Am. Soc. Spectrom. 2004, 15, 288. Sahlin, E.; Beisler, A. T.; Webber, S. G.; Sandberg, M. Electrochemical Detection of Peptides p367 398 in Electroanalytical Methods for Biological materials Marcel Dekker, New York, 2002, Editors: Brajter-Toth, A.; Chambers, J. Q. Samiec, P. S.; Drews-Botsch, C.; Flagg, E. W.; Kurt z, J. C.; Sternberg, P.; Reed, R. L.; Jones, D. P. Free Radic. Biol. Med. 1998, 24, 699. Sanchez-Rivera, A. E.; Corona-Avendano, S.; Alarcon-Angeles, G.; Rojas-Hernandez, A.; Ramirez-Silva, M. T.; Romero-Romo, M. A. Spectrochim. Acta. A 2003, 59, 3193. Schafer, F. Q.; Buettner, G. R. Free Radical Biol. Med. 2001, 30, 1191. Schneider, B. B.; Douglas, D. J.; Chen, D. D. Y. Rapid Commun. Mass Spectrom. 2001, 15, 2168. Seiwert, B.; U. Karst, Anal. Chem. 2007, 79, 7131. Senko, M. W. Canterbury, J. D. Guan, S. Marshall, A. G. Rapid Commun. Mass Spectrom. 1996, 10, 1839. Shen, X. M.; Dryhurst, G. Tetrahedron 2001, 57, 393. Shen, X.; Dryhurst, G. Bioorg. Chem. 1996a, 24, 340. Shen, X.; Dryhurst, G. J. Med. Chem. 1996b 39, 2018. Shen, X.; Xia, B.; Wrona, Z.; Dryhurst, G.; Chem. Res. Toxicol 1996, 9, 1117. Shen, Y.; Smith, R. D.; Unger, K. K.; Kumar, D.; Lubda, D. Anal. Chem. 2005, 77, 6692. Shi, Y. Evans, J. E. Rock, K. L. Natu re 2003, 425, 516. Sica, D. A.; Schoolwerth, A. C. Curr. Opin. Nephrol. Hyperten. 2002, 11, 475. Siegbahn, P. E. M. Blomberg, M. R. A. Crabtree, R. H. Theor. Chem. Acc. 1997, 97, 289. Simic, M. G. Jovanovic, S. V. J. Am. Chem. Soc. 1989, 111, 5778. Simmonds, H. A.; Duley, J. A. Fairbanks, L. D.; McBride, M. B. J. Inher. Metab. Dis. 1997, 20, 214. Skinner, K. A.; Whites, C. R.; Patel, R.; Tan, S.; Barnes, S.; Kirk, M.; Darley-Usmar, V.; Parks D. A. J. Biol. Chem. 1998 273, 24491.

PAGE 155

155 Sours, R. E.; Swift, J. A. Am. Crystallogr. Assoc. Trans. 2004, 39, 83. Spataru, N.; Sarada, B. V.; Popa, E.; Tryk, D. A.; Fujishima, A. Anal. Chem. 2001, 73, 514. Spencer, J. P. E.; Jenner, P.; Daniel, S. E.; Lees, A. J.; Marsden, D. C.; Halliwell, B. J. Neurochem. 1998, 71, 2112. Spina, M. B.; Cohen, G. Proc. Natl. Acad. Sci. USA 1989, 86, 1398. Staal, F. J. T.; Ela, S. W.; Roederer, M.; Anderson, M. T.; Herzenberg, L. A., Lancet 1992, 339 909. Starkey, J. A.; Mechref, Y.; Muzikar, J.; McBride, W. J.; Novotny, M. V. Anal. Chem. 2006, 78 3342. Subramanian, P. Tyagi, S. K. Dryhurst, G. Nucleosides & Nucleotides 1987, 6, 25. Tanaka, N.; Kolthoff, I. M.; Stricks, W. J. Am. Chem. Soc. 1955, 77, 1996. Tanaka, N.; Kolthoff, I. M.; Stricks, W. J. Am. Chem. Soc. 1955, 77, 2004. Tang, K.; Lin, Y.; Matson, D. W.; Kim, T.; Smith, R. D.; Anal. Chem. 2001 73, 1658. Taylor, G. I. Proc. R. Soc. London, A 1964, 280, 383. Taylor, M. J. Richardson, T. J. Food Sci 1980, 45 1223. Terashima, C.; Rao, T. N.; Sarada, B. V.; Kubota, Y.; Fujishima, A. Anal. Chem. 2003, 75, 1564. Thompson, J. J. Rays of Positive Electricity and Thei r Application to Chemical Analyses Longmans, Green and Co.: London, 1913. Tse, D. C. S .; McCreery, R. L. R.; Adams, N. J. Med. Chem. 1976, 19, 37. Tyapochkin, E.; Kozliak, E. J. Mol. Catal. A 2005 242, 1. Uchiyama, S.; Sekioka, N. Electroanalysis 2005 17, 2052. Van Berkel, G. J.; Asano, K. G.; Granger, M. C. Anal. Chem. 2004, 76, 1493. Van Berkel, G. J.; Asano, K. G.; Granger, M. C. Anal. Chem. 2005, 77, 4366. Van Berkel, G. J.; Enke, C. G.; Cole, R. B.; Martinez-Sanchez, M.; Fenn, J. B. J. Mass Spectrom. 2000, 35, 939. Van Berkel, G. J.; J. Am. Soc. Mass Spectrom. 2004 15, 1691. Van Berkel, G. J.; Kertesz, V.; Anal. Chem. 2007, 79, 5510.

PAGE 156

156 Van Berkel, G. J.; Kertz, V. J. Mass Spectrom. 2001, 36, 1125. Van Berkel, G. J.; McLuckey, S. A.; Glish, G. L. Anal. Chem. 1992, 64; 1586. Van Berkel, G. J.; Zhou, F. Anal. Chem. 1995a, 67, 2916. Van Berkel, G. J.; Zhou, F. Anal. Chem. 1995b 67 3958. Van Berkel, G. J.; Zhou, F.; Aaronson, J. T. Int. J. Mass Spectrom. Ion Processes 1997 162, 55. Van den Brandhof, W. E.; Haks, K. ; Schouten, E. G.; Verhoef, P Atherosclerosis 2001 157, 403. Vanhoutte, K.; Van Dongen, W.; Esmans, E. L. Rapid Commun. Mass Spectrom. 1998, 12, 15. Vestal, M. L. Chem. Rev. 2001, 101, 361. Volk, K.J.; Yost, R. A.; Brajter-Toth, A. Anal. Chem. 1986, 60, 720. Volk, K.J.; Yost, R. A.; Brajter-Toth, A. J. Electrochem. Soc. 1990, 137, 1764. Volk, K. J.; Yost, R. A.; Brajter-Toth, A. Anal. Chem. 1992, 64, 21A. Volk, K. J.; Yost, R. A.; Brajter-Toth, A. Anal. Chem. 1999, 61, 1709. Wang, X.; Stanbury, D. M.; Inorg. Chem. 2008, 47, 1224. Wang, Y. H.; Goddard, W.; Noyes, K. T.; Sowers, L. C.; Hwang, S; Chung, D. S. J. Phys. Chem. B 2003, 107, 344. Wang, Z.; Konigsberger, E. Thermochim. Acta 1998, 310, 237. Waring, W S. Webb, D. J. Maxwell, S. R. J. Q. J. Med. 2000, 93, 707. Watson, J. T. Introduction to Mass Spectrometry 3rd ed, Lippincot Raven, New York 1997. White, P. C.; Lawrence, N. S.; Davis, J.; Compton, R. G. Anal. Chim. Acta 2001a, 447, 1. White, P. C.; Lawrence, N. S.; Tsai, Y. C.; Davis, J.; Compton, R. G. Mikrochim. Acta 2001b 137, 87. White, P. C.; Lawrence, N. S.; Davis, J.; Compton, R. G. Electroanalysis 2002, 14, 89. Whitehead, R. E.; Ferrer, J. V.; Javitch, J. A.; Justice, J. B. J. Neuron. Chem. 2001, 76, 1242. Whitehouse, C. M.; Dreyer, R. N.; Yamashita, M.; Fenn, J. B. Anal. Chem. 1985, 57, 675. Willard, H. H.; Merritt, L. L.; Dean, J. A.; Settle, F. A. Instrumental Methods of Analaysis, 7th ed, Wadsworth Publishing Co.: California 1988.

PAGE 157

157 Willoughby, R. C.; Browner, R. F. Anal. Chem. 1984, 56, 2626. Willoughby, R. C.; Browner, R. F. Anal. Chem. 1988, 60, 489. Winters, R. A.; Zukowski, J.; Ercal, N.; Matthews, R. H.; Spitz, D. R. Anal. Biochem. 1995, 227, 14. Wu, S.; Zhang, K.; Kaiser, N. K.; Bruce, J. E.; Prior, D. C.; Anderson, G. A. J. Am. Soc. Mass Spectrom. 2006, 17, 772. Xu, X.; Lu, W.; Cole, R. B. Anal. Chem. 1996, 68, 4244. Xu, X.; Nolan, S. P.; Cole, R. B. Anal. Chem. 1994, 66, 119. Yoshida, T. J. Chromatogr. B 1996, 678, 157. Yue-Dong, Y. Biomed. Chromatogr. 1998, 12, 47. Zenely, J. J. Phys. Rev. 1917, 10, 1. Zhang, F.; Dryhurst, G. J. Med. Chem. 1994, 37, 1084. Zhang, T.; Palii, S. P.; Eyler, J. R.; Brajter-Toth, A. Anal. Chem. 2002, 74, 1097. Zhang, T; Brajter-Toth, A, Anal. Chem. 2000, 72, 2533. Zhou, F.; Van Berkel, G. J. Anal. Chem. 1995, 67, 3643. Zhou, L.; Zhai, B.; Yue, E.; Lee, L. M. Anal. Bioanal. Chem. 2006, 385, 1087. Zhou, M.; Ding, J.; Shang, Q. Anal. Chem. 2007, 79, 5328. Zvi, I.; Green, Y.; Nakhoul, Y.; Kanter, Y.; Nagler, R. M.; Nephron. Clin. Pract. 2007 105, 114.

PAGE 158

158 BIOGRAPHICAL SKETCH N. Alpheus Mautjana obtained a Bachelor of Technology (B.Tech.) degree in analytical chem istry in 1997 from Vaal Triangle Tec hnikon, now called Vaal Triangle Technical University in South Africa. He accepted a schola rship from African Explosives and Chemicals Industry (AECI), South Africa, to study for Master of Science (M.Sc.) de gree in analytical chemistry at the University of Cape Town, wh ich he completed in December 2000. He accepted a position with African Explosives Limited (AEL), a subsidiary of AECI, as a Team Manager in January 2001 and was later promoted to Plant Technical Services Manager which was his exit position in November 2003. N. Alpheus Mautjana jo ined the PhD program at the University of Florida in spring 2004 and at this, the completion of his Ph D, he returns to South Africa.