<%BANNER%>

Polymer Modification with Matrices for Matrix-Assisted Laser Desorption Ionization Mass Spectrometry and with Aptamers f...


PAGE 1

1 POLYMER MODIFICATION WITH MATRIC ES FOR MATRIX-ASSISTED LASER DESORPTION/IONIZATION MASS SPECTRO METRY AND WITH APTAMERS FOR SURFACE-ENHANCED LASER DESORP TION/IONIZATION MASS SPECTROMETRY By HONG YU 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 2007

PAGE 2

2 Copyright 2007 by Hong Yu

PAGE 3

3 To my daughter, Elaine Rogers

PAGE 4

4 ACKNOWLEDGMENTS I would like foremost to thank my parents, Shunyang Yu and Huaizhen Yan, who always put their childrens wellbeing first, and believed in me and encouraged me to be my best, as well as my sister, Qiong Yu for her friendship. I al so would like to thank my husband, James Rogers, for his guidance and help in my adjustment to life in a new country and in learning English, and for his companionship through my graduate career I would like to thank all my other family members, who gave me support and love throughout my life. Many thanks are given to the friends I have met along my graduate career, who have helped me in my personal life as well as my gr aduate life in chemistry: Dr. Rong Jiang, Dr. Chia Pooput, Dr. Yian Zhai, Dr. Lin Wang, and Dr. Hui Tao. I also sincerely thank Dr. James Yang and Yanrong Wu, who helped making part of my research possible by synthesizing aptamers whenev er I need them. I would like to thank Dr. Weihong Tan for encouraging the collaboration between our group and his group. I would like to thank Lin Yuan for sharing her hood with me for the carboxylation re action. I would like to thank Dr. Gary Cunningham and Nancy Li u for providing the access to the UV-vis spectrophotometer. I am very grateful for the help Michael Napolitano and Daniel Magparangalan gave me for correcting my dissertation. My thanks also are extended to other former and present group members: Dr. Timothy Garrett, Dr. Alisha Mitc hell-Roberts, Dr. Samaret Otero-Santos, Dr. Mike Belford, Christopher Hilton, Frank Ke ro, Dodge Baluya, and Rachelle Landgraf. I would like to thank Dr. David Powell and Dr Ken Wagener for their helpful discussion about my research project. I want to thank my advisor, Professor Rich ard Yost, for giving me the opportunity to obtain my doctorate under his direction. He o ffered me the freedom of choosing the project I

PAGE 5

5 was interested in, and gave me guidance and pers pective for the research along the way. I would also like to thank my first graduate advisor, Professor Timothy Patrick at Southern Illinois University, where I received a Master of Science in Chemistry. His belief in my abilities inspired me to continue my education at the Un iversity of Florida. There are many teachers and instructors I would like to thank over the years, all of whom have encouraged me and supported me to continue my stated goals.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES................................................................................................................ .........9 ABSTRACT....................................................................................................................... ............12 CHAPTER 1 INTRODUCTION..................................................................................................................13 Introduction to Mass Spectrometry: 3D And 2-D Quadrupole Ion Trap Mass Spectrometry................................................................................................................... ....14 Introduction to Mass Spectrometry.................................................................................14 Introduction to 3-D Quadrupole Ion Trap Mass Spectrometry.......................................15 Introduction to 2-D Quadrupole Ion Trap Mass Spectrometry.......................................16 Development of Ionization Methods for MS Analysis: from LD to MALDI and from MALDI to SELDI...............................................................................................................17 Introduction to Aptamers....................................................................................................... .22 Definition and Synthesis of Aptamers.............................................................................22 The Advantages of Using Aptamers vs Antibody and the Applications in Bioanalysis...................................................................................................................26 Overview of Dissertation....................................................................................................... .30 2 MATRIX-DERIVED RESO RCINOL-FORMALDEHYDE POLYMERS: SYNTHESIS VIA SOL-GEL METHOD AND USAGE IN MALDI-MS SAMPLE PREPARATION......47 Problems Associated with MALDI-M S Sample Preparation Methods..................................47 Background Interference from Ma trix Molecules in MALDI-MS..................................47 The Uneven Distribution of Analyte Molecules with the Traditional MALDI-MS Sample Preparation Methods.......................................................................................50 Introduction to the RF Polymer..............................................................................................52 Systematic Investigation in Preparation of the RF and Matrix-Derived RF Polymers..........54 Analysis of the Spiperone on the CHCA Derived RF Polymer for Lower Background Interference from Matrix Molecules...................................................................................59 Analysis of the Spiperone on the CHCA-Der ived RF Polymer vs. Analysis of the Spiperone on the Stainless Steel Using CH CA as Matrix for More Uniform Analyte Signal Intensity Across the Sample....................................................................................61 Conclusion..................................................................................................................... .........64 3 SURFACE MODIFIED RF POLYMER WITH APTAMERS: METHODOLOGY AND USAGE IN SELDI-MS SA MPLE PREPARATION.............................................................87

PAGE 7

7 Aptamers Used for RF Polymer Surface Modification..........................................................89 Direct Surface Modification of the RF Polymer with Aptamer.............................................92 Surface Modification of the RF Polymer with Aptamers via Surface Modification of the RF Polymer with Carboxylic Groups..................................................................................93 Characterization of the Carboxylic Group-Modified RF Polymer.........................................93 Affinity Capture of RG 19 with RG 19-Re taining Aptamer-Modified RF Polymer and Characterization of the Modified RF Polymer....................................................................95 Affinity Capture of Cocaine with Direct and Indirect CocaineRetaining AptamerModified RF Polymer.........................................................................................................96 Conclusion..................................................................................................................... .........99 4 CONCLUSION AND FUTURE WORK.............................................................................122 LIST OF REFERENCES.............................................................................................................127 BIOGRAPHICAL SKETCH.......................................................................................................138

PAGE 8

8 LIST OF TABLES Table page 2-1 RF polymers with ace tone as solvent and Na2CO3 as catalyst..........................................83 2-2 RF polymers with ethanol as solvent and Na2CO3 as catalyst.........................................84 2-3 RF polymers with water as solvent and Na2CO3 or HCl, or HClO4 as catalyst.................85 2-4 RF polymers with acetone as solvent and HCl as catalyst.................................................86

PAGE 9

9 LIST OF FIGURES Figure page 1-1 Parts of the 3-D quadrupole ion trap..................................................................................31 1-2 Stability diagram in (az qz) space for the region of simultaneous stability......................32 1-3 Basic design of the twodimensional linear ion trap..........................................................33 1-4 Application of DC, RF trapping, and AC excitation voltages for 2-D ion trap.................34 1-5 Electron ionization schematic............................................................................................35 1-6 Chemical ionization schematic..........................................................................................36 1-7 Illustration of FAB........................................................................................................ .....37 1-8 Electrospray ionization schematic.....................................................................................38 1-9 Most commonly used MALDI matrices............................................................................39 1-10 Mechanism of MALDI......................................................................................................40 1-11 The chemical and biochemical surfaces for SELDI..........................................................41 1-12 The bulge and stem structure of aptamers.........................................................................42 1-13 The hairpin structure of aptamer........................................................................................43 1-14 The pseduknot structure of aptamer...................................................................................44 1-15 The G-quartet structure of an aptamer...............................................................................45 1-16 Generalized scheme indicating th e key steps in the SELEX process................................46 2-1 The mechanism of th e TEOS sol-gel reaction...................................................................65 2-2 Illustration of incorporation of th e matrix molecules in the TEOS gel.............................66 2-3 Microscope pictures of sample su rfaces of manually prepared samples...........................67 2-4 MALDI ion images showing intensi ties for selected peptides using.................................68 2-5 Mechanism of the polymerization of RF polymer proposed by Lin and Ritter.................69 2-6 Base catalyzed RF polymerization.....................................................................................70 2-7 Cross-section of the ion trap mass spectrometer................................................................71

PAGE 10

10 2-8 Illustration of the path of la ser and ions in mass spectrometer..........................................72 2-9 Fragment path of spiperone...............................................................................................73 2-10 UV absorbances of the matrices-embedded RF polymers.................................................74 2-11 UV absorbances of the RF polymer, CHCA embedded RF polymer, and CHCA solution....................................................................................................................... ........75 2-12 Mass spectrum of the 100 ppm spiperone on CHCA embedded RF polymer and stainless steel................................................................................................................ ......76 2-13 CHCA embedded in the RF polymer structure..................................................................77 2-14 LTQ with vMALDI ion s ource from ThermoFinnigan.....................................................78 2-15 The microscopic image of the sample well of dry droplet sample preparation.................79 2-16 The distribution of the spiperone usi ng dry droplet method on stainless steel..................80 2-17 The microscopic image of the RF-16 polym er pellet after the de position of spiperone....81 2-18 The distribution of the spiperone us ing dry droplet method on RF-16 polymer...............82 3-1 The structure of the RF polymer......................................................................................100 3-2 Chemical structure of reactive green 19..........................................................................101 3-3 The 27-mer oligodeoxyribonucleotide sequence.............................................................102 3-4 The two cocaine isomers: a) pseudococaine and b) cocaine............................................103 3-5 Anti-cocaine aptamer MNS-4.1 bound to cocaine..........................................................104 3-6 The picture of a Applied Biosystems 3400 DNA Synthesizer........................................105 3-7 Scheme of the indirect surface modificati on of the RF polymer via the carboxylation..106 3-8 Chemical structure of PDAM..........................................................................................107 3-9 Fluorescent emission intensity of P DAM-bond and noncarboxylated RF polymer........108 3-10 Structure of Toluidine blue O..........................................................................................109 3-11 UV absorbance of toluidine blue O desorbed from the carboxylated and noncarboxylated RF polymer...........................................................................................110 3-12 UV of the toluidine blue O solu tions of different concentration.....................................111

PAGE 11

11 3-13 Calibration curve of the concentrat ion of the toluidine blue O solution..........................112 3-14 UV absorbance of reactive green 19 desorbed from aptamer attached RF polymer and RF polymer................................................................................................................113 3-15 Calibration curve of the concentratio n of the reactive green 19 solution........................114 3-16 Cocaine displaces diethylthiatricarbocyanine iodide complexed with aptamer MNS4.1............................................................................................................................ .........115 3-17 A microscopic image of the indirectly cocaine retaining-aptamer modified RF polymer........................................................................................................................ ....116 3-18 MS spectrum of the cocaine on the un modified RF polymer and on the cocaine retaining aptamer attached RF polymer...........................................................................117 3-19 The 2-D and 3-D image of the cocaine si gnal intensity across the indirect cocaine retaining-aptamer modified RF polymer.........................................................................118 3-20 The image of the direct cocaine re taining-aptamer modified RF polymer......................119 3-21 MS spectrum of the cocaine on the direct cocaine retaining-aptamer modified RF polymer........................................................................................................................ ....120 3-22 The 2-D and 3-D image of the cocaine signal intensity across the direct cocaine retaining-aptamer modified RF polymer.........................................................................121

PAGE 12

12 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 POLYMER MODIFICATION WITH MATRIC ES FOR MATRIX ASSISTED LASER DESORPTION IONIZATION MASS SPECTRO METRY AND WITH APTAMERS FOR SURFACE ENHANCED LASER DESORPTI ON IONIZATION MASS SPECTROMETRY By Hong Yu May 2007 Chair: Richard A. Yost Major Department: Chemistry A resorcinol-formaldehyde (RF) polymer and matrix-embedded RF polymers were developed to be used as substr ates for improved matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) analysis and for se lective analysis of biological samples. Matrix molecules can be embedded within this polymer, which renders excellent spectra without the interference imposed by the matrix molecules for low mo lecular weight analytes. By embedding the matrix molecules in this polymer more uniform spectra are obtained, solving the prevalent problem associated with the trad itional MALDI sample preparation methods. The phenol groups on the polymer made it possi ble to attach aptamers selective to reactive green 19 and cocaine onto it directly, and gave high specifi c affinity toward the reactive green 19 and cocaine, respectivel y. The phenol groups on the polym er also made it possible to attach carboxylic acid groups which could then be reacted with amine-modified aptamers. This indirect aptamer-modified RF pol ymer has higher reaction effici ency than the direct aptamermodified RF polymer, yield more aptamers at hi gher density on the polymer for affinity capture of analyte. This high specificity provided by the aptamers could make it a valuable tool to specifically retain the biomarkers of diseases and could be used for screen test for diseases.

PAGE 13

13 CHAPTER 1 INTRODUCTION In the history of analytical chemistry, new t echnologies of analysis and instruments have been developed to better determine the compositi ons and structures of substances in order to identify them in the pursuit of understanding the world around us. The analytical methods that have been used include some very rudimentary but still widely used techniques such as titration, density, melting point, boiling point, flame test, and combustion, which are methods still taught to new chemistry students. As the tasks for ch emical identification became more complicated and difficult, many new technologie s were developed to peek into the microscopic world of chemistry. These technologies include atom ic absorption, ultraviolet/visible (UV-vis) absorption, IR, fluorescence, flame, atomic emission, Raman, X-ray, X-ray crystallography, NMR, electrochemistry, gravimetric analysis, calorimetry, and thermogravimetric analysis (TGA), etc. Each of these more sophisticated technologies, based on a sp ecific theory, utilizes specific chemical and/or physical properties of th e substance, and requires a specific instrument to obtain specific informati on of the analyzed substance.1 A mass spectrometer is an instrument which wa s first invented at the beginning of the twentieth century, but took off in its application in the past thirty or forty years as a tool to identify various analytes. From gaseous molecule s to volatile small organic molecules, then to involatile large organic molecules, and eventually to proteins, DNA, cells, and bacteria, the range of the analytes than can be analyzed has e xpanded since then. The high sensitivity, speed, specificity, and reproducibility of mass spectrometry (MS) have made it an irreplaceable tool for its application in protein anal ysisincluding discovery, identific ation (i.e., peptide mapping and sequencing), and structural characterization.2

PAGE 14

14 Introduction to Mass Spectrometry: 3-D and 2-D Quadrupole Ion Trap Mass Spectrometry Introduction to Mass Spectrometry The first mass spectrometer, built in 1907 by J. J. Thomson, was used to detect positive rays based on the mass-to-charge ratio (m/z) of th e particles; this was the basis of modern mass spectrometry.3 The modern mass spectrometer is used to obtain the information on molecular mass according to the mass-to-charge ratio and ev en structures of molecules according to fragmentation patterns.4 The ionization device coupled to th e mass analyzer has evolved because of the demand of an expanded variety of analyt es. Electron ionization (EI) was the first used, followed by chemical ionization (CI), fast atom bombardment (FAB), electrospray ionization (ESI), and matrix-assisted laser desorption/ionization (MALDI). Analyte molecules are ionized by addition or loss of an electron or a proton or other ion. The mass spectrum consists of the molecular-type ions (M+, [M+H]+, [M-H]-, ect.) and often fragment ions, the pattern of which may be unique for each molecule. The mass spectrometer was first used to identify elements and their isotopes.5 Soon after, petroleum chemists used the mass spectrometer to identify organic molecules.6 Now, the applications of the mass spectrometry have broadened to include bioche mistry, explosives, fullerenes, toxics, and environmental pollutants.3 A mass spectrometer is named according to the mass analyzer which is its determining component. During the process of the developm ent of the mass spectrometer, many different types of mass analyzers have been invente d, which include magnetic-sector, quadrupole massfilter, 3-D quadrupole ion trap, 2-D quad rupole ion trap, and time-of-flight.3 Each of the mass analyzers has its forte and suitability for analysis of certain types of an alytes and is chosen according to the task and budget.

PAGE 15

15 Introduction to 3-D Quadrupole Ion Trap Mass Spectrometry The quadrupole ion trap was i nvented by German physicist Wolfgang Paul and co-workers in the early 1950s.7 The quadrupole ion trap can be used to confine ions in a small volume using just high-frequency electric fiel ds without a material container.8 This invention later earned Paul the 1989 Nobel Prize in Physics.9 Figure 1-1 illustrates a traditional 3-D quadrupole ion trap. This 3-D quadrupole ion trap has two hyperbolic e nd-cap electrodes with en trance and exit holes in the center of both end-caps. The hyperbolic ring electrode in-between the end-caps has a radius r0.10 The distance between the two end-caps is 2z0, where ideally, Z0=r0/ 2. Ions entering the trap are trapped at the center of the ion trap by a quadrupolar field wh ich is created by a radio frequency (RF) waveform applied to the ring electrode when the two end-caps are grounded. Equations 1-1 and 1-2 are derived from the Mathieu equation for a stretched ion trap, where e is the charge, U is the DC potential, typica lly kept at zero; V is the amplitude of the RF, and is the angular frequency of the applied RF. Solutions to these two equations define whether an ion with particular m/z will be stable inside the quad rupolar field according to their position in the Mathieu stabil ity diagram (Figure 1-2).10 az=-16eU/m(r0 2+2z0 2) 2 (1-1) qz=-8eV/m(r0 2+2z0 2) 2 (1-2) The quadrupole ion trap confines ions in th e small volume between the ring electrode and two endcap electrodes by appropriate oscillating (RF) electr ic fields. In modern ion trap, an AC waveform, termed the resonance ejection wave form, is applied acro ss the endcaps. The frequency is typically set to be sligh tly less than half of the RF frequency, Since the oscillating frequency, the ion motion in the z direction is given by Bz/2 (see Figure 2 for Bz values), this will correspond to a eject V value less than 0.908. When the amplitude of the

PAGE 16

16 applied RF is linearly ramped up, a mass spectrum is obtained via the massselective instability scan. In this instability scan, ions are ejected from the trap with increasing m/z with respect to time when they come into resonance with the resonance ejection waveform. Higher m/z ions are still trapped; the limitation of mass range is due to an inability to reach the needed RF voltage levels required to eject higher m/ z ions out of the trap. In normal operation, for the instrument used for this research, the maximum m/z detectable is 650. Lowering the AC resonance ejection frequency applied across the endcaps will allow ions to come into resonance earlier in the RF ramp, extending the mass range by lowering the qz according to the equation 1-3.11 ( m/z ) new = ( m/z ) old (q eject new/ q eject old) (1-3) A quadrupole ion trap can be connected with ex ternal ion sources such as matrix-assisted laser desorption/ionization (MALD I) and electrospray ionization (E SI). The range of ionization sources that can be used, in addition to its MSn capability, makes the qua drupole ion trap a very useful tool for compound identification. Introduction to 2-D Quadrupole Ion Trap Mass Spectrometry The 2-D ion trap, the linear variant of the 3-D ion trap are based on the four-rod 2-D quadrupole mass filter, omitting the quadrupolar-tra pping field along one of the three spatial axes.8 The 2-D ion trap shown in Figure 1-3 12 is composed of three s ections in which ions are mass selectively ejected radially through two slits in the two opposite rods in the center section. Ions are trapped in the z direc tion by more positive DC voltages a pplied to the two end sections for axial trapping of positive ions (Figure 1-4).12 The ions are trapped ra dially by the RF applied in two phases to opposite rod pairs, as shown in Figure 1-4.12 A supplementary AC voltage is applied across the x-rod pair in two phases for isolation, collis ion-induced dissociation (CID), and resonant ejection of ions (Figure 1-4)12.

PAGE 17

17 Ions confined within a linear trap can be mass-selectively ejected in a direction perpendicular to the central axis of the trap (radial ejection) or can be ejected along its axis (axial ejection). Having the detectors at both radial ex it slits doubles the number of detected ions in radial ejection design. The axiall y-ejected ions can be introduced into a second mass analyzer such as a Fourier-transform ion cyclotron re sonance (FT-ICR) analyzer to form a hybrid instrument.3 Relative to the well-established 3-D ion-trapping instruments with similar mass range, the 2-D linear trap with radial eject ion has significantly higher trappi ng efficiency and increased ion capacity that improves the detection sensitivity at least 5 to 10-fold. This linear trap can conduct all the typical scan modes intrinsic to 3-D trap s with comparable mass resolution but at higher overall scan rates.8 Development of Ionization Methods for MS Analysis: from LD to MALDI and from MALDI to SELDI The ionization method used is determined by th e type of the analyte and determines the characteristics of the mass spectrum obtained. The most common ionization methods include EI, CI, ESI, and MALDI. Electron ionization was the fi rst ionization method developed, and was used in the instrument invented by JJ Thompson. Electrons with an energy of about 70 eV collide with the vaporized analyte in vacuum (10-5-10-4 torr)3 to ionize the analyte molecules (Figure 1-5).13 The spectra obtained with EI are characterized by si gnificant fragmentation of the molecular ions ([M]+.). Later, when a softer ionization me thod was required for more fragile compounds chemical ionization (CI) was deve loped. With CI, electrons collid e with a reagent gas present in large excess compare to the analyte to produce reag ent ions which will later pass their charge to the analytes by collision without transf erring an excess of energy (Figure 1-6).14 The spectra

PAGE 18

18 obtained with CI are characterized by less fr agmentation of the analyte and more abundant molecular-type ions ([M+H]+) than with EI. For these two io nization methods, the analytes have to be vaporized before they can be ionized, which is a problem for less volatile compounds. Typically, the higher the molecular weight, the less volatile the mol ecules. With EI and CI, it is difficult to ionize most of the large organic mo lecules and is nearly impossible to ionize macromolecules. With increasing demand for analyzing large nonv olatile molecules, ionization methods that can ionize large, nonvolatile molecules were devel oped such as fast atom bombardment (FAB). With FAB, a beam of atoms bombard the analyt e, which is dissolved in a low volatile liquid matrix, to produce the analyte ions (Figure 1-7).15 Since the analyte molecules are not vaporized before they are ionized, FAB can be used to ioni ze polar, ionic, thermally labile, and relatively high molecular weight compounds that are not su itable for normal EI/CI analysis. Depending upon whether the cation or the anion is of in terest, positive-ion or negative-ion FAB/MS analyses can be performed. FAB can ionize molecules up to 2000 Da.3 Laser desorption (LD) ionization was also developed to tackle the problem associated with less volatile large compounds. In troduced in the early 1960s, LD irradiated low-mass organic molecules with a high-intensity laser pulse to form ions that could be successfully mass analyzed. With LD, nonvolatile la rge molecules with the ability to absorb the energy from the laser beam can be ionized. Although LD expanded the range of the analyte that can be ionized, there were some limitations that came along with LD. The waveleng th of the laser used in LD has to match the wavelength of the analyte mo lecules absorbance for the laser energy to be utilized for ionization. This may be difficult when a variety of analytes need to be analyzed. LD was also considered as an ioniza tion method which is too energetic for labile analyt es since laser

PAGE 19

19 energy is imparted directly into the molecule s, which my cause a large amount of thermal degradation and fragmentation. First developed in 1980s, ESI was develope d to ionize large a nd thermally labile molecules. By forcing a solution of the analyt e through a small capillary held at high voltage, the solution sprays into tiny droplets with ch arges on each droplet. After the solvent is evaporated, ions from nonvolatile compounds with one or more ch arges are generated, as shown in Figure 1-8.16 The existence of multiple charges on high mass ions lowers the mass-to-charge ratio, which expands the mass range of the molecules can analyzed.3 As a soft ionization method, ESI makes it is possible to investigate even intact noncovalent complexes. The invention is so important and revolutionary th at the inventor, John B. Fenn of the Virginia Commonwealth University, shared with Koichi Tanaka of the Shimadzu Corp. half of the 2002 Nobel Prize in Chemistry for his invention of ESI.17 Matrix-assisted laser desorption /ionization (MALDI) is a soft ionization method that rivals ESI for mass spectrometric analyses of bi ological macromolecules and the noncovalent complexes between peptides and metal ions,17, 18, 19, 20, 21, 22,23 single-stranded DNA, 21, 24, amino acids,25 drugs,26 and other peptides.27,28,2930 In 1985, Michael Karas and Franz Hillenkamp published the laser desorption result s of the mixture of aromatic amino acid which absorbs laser energy with aliphatic amino acid wh ich does not absorb laser energy.31 In 1987, Karas and Hillenkamp successfully used a matrix in LD to ionize and analyze high molecular weight molecules.32 Matrices are organic molecules with ar omatic groups that can absorb laser energy in UV. The most commonly used matrices are 2,5-dihydroxybenzoic acid (DHB), trans-3,5dimethoxy-4-hydroxycinnamic acid or sinapic acid (SA), nicotinic acid (NA), trans-3-methoxy4-hydroxycinnamic acid or ferulic acid (FA), 3-hydroxypicolinic acid (3HPA), 4-hydroxy-

PAGE 20

20 cyanocinnamic acid (4HCCA), and (in the infrared ) succinic acid and glycerol (Figure 1-9). With MALDI, a low concentration of the analyte is mixed with a large excess of organic matrix to completely isolate analyte molecules from each other to homogenous 'solid solution'. The laser beam is focused onto the surface of the matrix-analyte solid solution. The matrix molecules absorb the laser energy at the incident laser wa velength and then pass that energy on to analyte molecules, producing a pulse of i ons with each laser pulse. The cl usters ejected from the surface consist of analyte molecules surrounded by matr ix and salt ions. The matrix molecules evaporate away from the clusters to leave the free analyte in the gas-phase. The photo-excited matrix molecules are stabilized through proton transf er to the analyte. Cation attachment to the analyte is also encouraged during this process. It is in this way that the characteristic [M+X]+ (X= H, Na, K etc.) analyte ions are formed. These ionization reactions take place in the desorbed matrix-analyte cloud just above the surf ace. The ions are then extracted into the mass spectrometer for analysis. MALDI allows a larger variety of analytes to be ionized by the same laser than LD.33 Figure 1-1014 is an illustration of ionizat ion process of MALDI. The application of MALDI to bi ological macromolecules demonstrated by Koichi Tanaka 34 led him to receive a quarter of the 2002 Nobel Prize for chemistry.17 The matrix-to-analyte ratio ranges from 100 to 50,000.3 MALDI overcame the problem associated with the LD, but inevitably suffers fr om the interference from matrix molecules with analytes of low molecular weight. The excessive amount of matrix molecu les poses interference to the interpretation of the mass spectra of low mo lecular weight analytes. Some effort has been put into developing matrix-free laser desorption /ionization methods to eliminate the interference from matrix. Desorption/ionization on porous si licon (DIOS) is one of the methods developed which utilizes the large surface area of an UV absorbing semiconduc tor treated electrochemically

PAGE 21

21 to produce spectra without the interference of matrix clusters.35, 36, 37,38 DIOS certainly has its success, but suffers from the rela tively less than ideal sample-tosample reproducibility, stability of targets, molecular weight limitation, low analyt e signal intensity and is sensitive to how the samples are deposited on them-cr acks can form with an inappropr iate sample deposition method. Since MALDI opened the gate for the MS anal ysis of macromolecules, biological samples have been subject to the MS analysis for diseas e biomarker discovery and clinical diagnostics. Disease biomarkers are the characteristic amount or the existence of a certain substance or substances.39 Biological samples taken directly from humans or animals such as blood, serum, plasma, urine, and cellular secretion products also contain extremely large numbers of components which are not of interest a nd can interfere with the MS analysis.40 These unwanted components include biological molecules and orga nic and inorganic salt s which can make the MS analysis extremely difficult. Liquid chromatography (LC), high-pressure liquid chromatography (HPLC), membrane dial ysis, centrifugation, immunoprecipitation, gel electrophoresis and other separation techniques have been used to se parate and extract the target analyte from the biological sample prior to the MS analysis. Theses separation techniques are usually tedious and time-consuming and suffer loss of the analyte during the process, especially for the low abundant components.40 Surface-enhanced laser desorption/ionizati on (SELDI) was introduced by Hutchens and Yip in 1993 in order to achieve better sample preparation.41 SELDI is based on MALDI but with the surface of the substrate modified to have a spec ific affinity toward certain types of molecules to achieve an on-probe, one-step clean-up procedure before the sa mple is analyzed. There are several companies which provide SELDI chips, such as Ciphergen Biosystems (Palo Alto, CA). Affinity capturing for enrichment and cleanup of analyte is borrowed from affinity

PAGE 22

22 chromatography. Among them, immobilized me tal affinity capture (IMAC), antibody and protein affinity capture, lectin affinity capture, inherent hydrophobi c interaction bio-capture, and bio-molecular interaction capture are attractive one s, and have been used to retain biomolecules such as proteins and peptides.42 Figure 1-11 shows the biochemical and chemical modified affinity surfaces available from Ciphergen.42 The chemical modification of the substrate surface is well established. It retains a class of molecu les which is useful for certain applications. Biochemically modified surfaces have specific ity toward one specific molecule. Antibodies, DNA, enzymes, and receptors have been used in biochemical surface modification. Aptamers are pieces of synthetic DNA or R NA oligonucleotides which recently been applied to SELDI affinity capturing.73 Aptamers are screened from a randomly generated population of DNA/RNA sequences for their ability to bind with desired molecular targets such as peptides with high specifi city, and can be designed and synthesized easily with a DNA synthesizer. Aptamers have not yet been wi dely used for SELDI and are not available commercially. Introduction to Aptamers Aptamers have been employed in applica tions such as chromatography, capillary electrophoresis, biosensors, and signaling becaus e of its ability of affinity capture small molecules, peptides, and proteins.43 Aptamers has also been used for mass spectrometry sample preparation in surface enhanced la ser desorption/ionization (SELDI). Definition and Synthesis of Aptamers Aptamers are pieces of synthetic DNA or R NA with 3-D structures which provide the targets binding capability. The name aptamer is derived from the Latin word aptus which means to fit. Aptamers can be selected to fit small molecules, biopoly mers, surfaces or even whole cells.43 Because of the high specific affinity, aptamers can be used as recognition

PAGE 23

23 elements in heterogeneous a ssays to replace antibodies.44, 45 The selective binding affinity toward the target is due to spec ific interactions such as hydrogen bonding or association with the phosphate groups of the aptamer.46 The sequence-specific tertiary structure of the aptamer helps these interactions, which provides a rigid platform for the arrangeme nt of functionalities of the aptamer. The 3-D structures have been discove red to include the stem -loop/bulge (Figures 1-12 and 1-14),47, 48 the helix, the hairpin (Figure 1-13),47 the pseudoknot (Figure 1-14),49 and the Gquartet structure (Figure 1-15).50 Under certain condition such as the buffer with certain pH, certain components, and concentration; aptame r can form 3-D configuration which retains a specific protein or molecule which can fit in th is 3-D configuration. Because of its affinity capture capability, aptamers have been used in chromatography for separation and purification of mixtures.48, 51 An aptamer for a particular target is discovered and produced by the method called systematic evolution of ligands by exponential enrichment (SELEX), which utilizes in vitro selection and amplification procedures to isolat e aptamers for any small molecule or protein target.52, 53 This automatic method was developed i ndependently by Joyce, Szostak, and Gold.52, 54, 55 First, a library of oligonucleo tide molecules with the size of 1015 ~ 1018 is generated chemically in a standard oligonucleotide synthesizer.51 The greater the number of randomized nucleotide positions, the more complex a library is and the more likely the success of finding the molecules that interact with a target.44 Each member in a library has a unique sequence which is designed to contain a contiguous randomized region flanked by two fi xed sequence regions. Technological advances alre ady have made it possible to eliminate the requirement for the fixed regions in random sequence libraries used for the SELEX process, thereby producing short aptamer sequences. A mixture of phosphoramid ites containing all four bu ilding blocks: A, G, C,

PAGE 24

24 and T is delivered to the synthesizer to synt hesize each nucleotide position in the contiguous random region. The ratios of the four phosphoramidites in the mixture are adjusted based on the coupling efficiencies of individual monomers to obtain an unbiased library with equal representation of all four nucleotides. Because aptamers are identified from DNA as well as from RNA libraries, chemically synthesized DNA libraries are enzymatically converted to RNA libraries.56 The screening process is the next step, in which a random sequence oligonucleotide (RNA or DNA) library is incubated with a target of inte rest in a buffer of choi ce at a given temperature (see Figure 1-16).55 A very small fraction of individual sequences tend to interact with the target, and these sequences are separated from th e rest of the library by any one of the physical separation techniques. Typically, nitrocellulose filter partitioning is used with protein targets that are retained on the nitrocellulose. Small molecular targets are generally immobilized on a solid support to generate an affinity matrix in which case, sequences that do not interact with the target on the solid support can be removed easil y by a simple washing step. The mixture of aptamer candidates recovered from the target repr esents a mixture of sequences containing both highand low-affinity bindi ng molecules to the target.44 The sequences bound to the target are isolated and amplified to obtain an enriched library to be used for the selection/amplification cycle. The DNA aptamer candidate mixture is amplified directly by the polymerase chain reaction (PCR) while RNA sequences are amplified by PC R after being converted from DNA by reverse transcription (RT).55 The single-stranded DNA (ssD NA) population obtained by strand separation of PCR products is incubated with a fre sh sample of the target for the next round of selection. The RNA population is obtained by in vitro transcription. Se veral iterations of the

PAGE 25

25 selection process are carried ou t under increasingly stringent conditions to enrich the highaffinity sequences and eliminate the low-affinity binder. The enrichment efficiency of highaffinity binders is controlled by the stringency of selection at each round. Analysis of enriching populations against the target is carried out to determine the progr ess of the enrichment of highaffinity binders.44 The number of cycles requi red for aptamer identification is usually dependent on the degree of stringency imposed at each round as well as on the nature of the target. For most targets, affinity enrichment is reached within 8 cycles. In general, one cycle of SELEX takes two days. The enriched library is cloned and sequenced to obtain the sequence information of each member until the chosen sequences dominate the population.55 A typical SELEX experiment may take approximately 2 months including cloning and sequencing.58 Aptamers that come out of a SELEX experi ment are full-length sequences which are generally 70 nucleotides long, and can be truncat ed to eliminate nucleotide stretches that are not important for direct interacti on with a target or for folding in to the structure that facilitates target binding. The truncation of aptamers to the minimal ta rget-binding domain has been successfully carried out to obtain functiona l aptamers less than 40 nucleotides long.58, 59, 60, 61, 62, 63 Once the sequence is identified, an aptamer is produ ced by chemical synthesis. Aptamers have remarkable specificity and can discrimi nate targets on the basis of subtle structural differences such as the presence or absence of a methyl group64, 65 or a hydroxyl group66, 67 and the Dvs L-enantiomer of the target.66, 68 Because of the selective demand in the SELEX process that eliminates sequences that bind closely related an alogs of the target, sometimes the degree of specificity of apta mers is better than that of antibodies.64 Practically, elimination of the sequences that bind closely related analogs of the target is achieved by the process called "counter-SELEX" that effectively discards ligands that have an ability to bind the

PAGE 26

26 target as well as closely related structur al analogs of the target.64 During counter-SELEX selection, the population of ap tamers bound to the target is subjected to affinity elution with structural analogs and the sequences eluted are discarded. The counter-SELEX strategy is a valuab le tool in identifying aptamers aimed at a specific target in a complex mixture, even without prior knowledge of the target. For example; in the search for aptamers that bind to an "epitope" only present on the surface of cancer cells but not in healthy cells, the cells from healthy tissue are used to remove sequences that bind to the background that does not contai n the epitope of interest before the library is challenged with cancer cells. The Advantages of Using Aptamers vs. Antibo dy and the Applications in Bioanalysis Both antibodies and aptamers are utilized in bioanalysis based on their molecular recognition capability. Antibodies are the most popular class of molecules providing molecular recognition.51 They have been used in SELDI becau se of its high specif icity and selective affinity towards a specific target. Antibodies started to get attention since the1950s and became very popular in the 1970s.51 Although antibodies have the advantages of having a very high selective affinity, they also have several limitations. First, antibody generation i nvolves untidy cell cultures and lab animals since antibodies must be selected and produced in a living or ganism. Antibodies of non-human origin have implications in dia gnostic applications;69 thus, the use of antibodies in therapeutic applications is limited since the generation of hybridomas is restri cted to rats and mice. Heterophilic antibodies (human antibodies that recognize antibodies of non-human origin) that exist in humans might lead to false-positive results be cause a capture antibody can be li nked with a detector antibody of non-human origin in the absence of the specific analyte.70

PAGE 27

27 Second, stocks of antibody-producing cells need to be stored at multiple sites to prevent the complete loss of the cell lines because of the possibility of accidental losses or the death of cell lines. Typically, high yields of monoc lonal antibodies are obt ained by growing the hybridomas in the peritoneal cavities of animals and purifyi ng the antibody from ascites (abdominal dropsy) fluid. Some hybridomas are di fficult to grow in vivo, thus restricting this route of antibody production.44 Third, molecules that are not well tolerated by an imals, such as toxins have difficulty being employed to produce antibodi es in living organisms.44 Furthermore, it is difficult to raise antibodies against inherently less immunogenic molecules.44 Fourth, the identification and production of m onoclonal antibodies are laborious and could become very expensive for searches of rare anti bodies that require scr eening of a large number of colonies.44 Fifth, the consistent performance of the same antibody from batch to batch is poor; thus, immunoassays need to be optimized with each new batch of antibodies.44 Sixth, since the antibodies are pr oduced in vivo and subject to in vivo variations, it is not practical to identify antibodies that could recogn ize targets under c onditions other than physiological conditions.44 The ability of aptamers to bind to various targ ets makes aptamers applicable as biosensors, imaging probes, MALDI targets, and drugs.51 First, the properties of aptamers can be changed on demand44 because aptamers are identified through an in vitro process that does not depend on in vivo conditions. Furthermore, selection conditions can be de signed to obtain aptamers with properties desirable for in vitro diagnostics.44

PAGE 28

28 Second, as opposed to antibodies, toxins as well as molecules that do not elicit a good immune response can be used to generate high-a ffinity aptamers because animals and cells are not involved in aptamer identification.44 Third, aptamers are stable during long-term storage and can be transported at ambient temperature. Aptamers can undergo reversible denaturation process and easily recover from exposure to undesirable conditions within minutes. The in vitro selection process for aptamers can be carried out under conditions akin to those us ed in the assay for whic h the aptamer is being developed. The aptamer will maintain its struct ure and function in the final assay and not fall apart as antibodies do.51 Fourth, aptamers with any sequences that can recognize any class of ta rget molecules with high affinity and specificity can be discovered with SELEX process. Once the sequence of a particular DNA aptamer is known, it is easy and inexpensive to synthesis the aptamer in a DNA synthesizer with high accuracy and reproducibility. They ar e then purified under denaturing conditions to achieve very a high degree of purity with little or no batch-to-batch variation in aptamer production.44, 51 Fifth, additional chemistries can be added on without a loss in function because aptamers are chemically synthesized, which gives aptamers superiority over antibodies. When chemical groups are attached to the ends of the aptamers, their life span in the bloodstream increases, they can be targeted to particular locations, or they can be immobilized onto a surface. Immobilizing aptamers without potential loss of f unction makes it superior to antibodies.51 Reporter molecules such as fluorescein and biotin can be attached to aptamers at exact locations designed by the user. Functional groups that allow subsequent de rivatization of aptamers with other molecules can also be attached during the chemical synthesis of aptamers.44

PAGE 29

29 Sixth, aptamers can also be used for the development of sandwich assays for small molecules or modified into a molecular-beacon format (a DNA strand with a fluorescent tag and a quencher) for reagentless assays, whic h is beyond the capab ility of antibodies.51 Seventh, aptamers have smaller molecular weight, which offer advantages such as ease to penetrate tissue, shortened residence time in bl ood, and a smaller footprint when being attached to a surface to get a higher binding density.51 Because of the advantages the aptamers posse s over antibodies, the aptamers can be used to replace antibodies to retain molecules, althoug h the recognition ability of aptamers is not as high as antibodies.51 A lot of effort has been put into research to explore the usability of aptamers in analytical chemistry. SomaLogic (Boulder, CO) is developi ng aptamer proteomic chips as diagnostic tools that screen for biomarkers in serum. Ellin gtons group from Department of Molecular Biology, Massachusetts General Hospital, Boston, has demonstrated that aptamers can be immobilized on beads, introduced onto a sensor array, and used for the detection and quantitation of proteins.71 Kennedys group from University of Florida showed that DNA aptamers immobilized on a chromatographic support can selectively bind an d separate adenosine monophosepahte (AMP), cyclic-AMP, adenosine diphosphate, adenosin e triphospahte, NAD+, and adenosine from mixtures as complex as tissue extracts.72 McGown from Duke University successfully used a thrombin-binding DNA aptamer for affinity capture for MALDI-MS.73 In her work, she attached aptamer pieces onto a treated fu sed-silica glass surface and then used the aptamer-attached glass to capture the thrombin from a pr otein mixture. The result showed that the thrombin-binding aptamer has high affinity toward the thrombin.

PAGE 30

30 Overview of Dissertation The importance of developing a new MALDI s ubstrate and the suitability of the RF polymer for MALDI-MS analysis was emphasized at the beginning of the dissertation. A new RF polymer and matrix-derived RF polymers have been developed and characterized as discussed in Chapter 2 for e limination of the background noises fro matrix molecules used in MALDI-MS analysis and for mo re even distribution of signal intensity of analyte across the sample spot. Surface modification of the RF polymer with carboxylate groups and then with aminemodified aptamers have been performed and characterized as discussed in Chapter 3 for attachment of functional groups and then attach ment of functionally-modified aptamers onto the RF polymer. With the affinity capture capability of this SELDI-MS substrate, it is possible to perform an on-probe one-step clean-up for the specific small molecules such as reactive green 19 and cocaine before the MS analysis of a biol ogical sample. The polymer and modification procedures are capable of being applied later on disease biomarkers. This is crucial for largescale cancer screening to be accessi ble for the vast majority of the population at lower cost and higher accuracy.

PAGE 31

31 Figure 1-1. Scheme of parts of the 3-D quadr upole ion trap. [Adopted from March, R.E. An introduction to quadrupole ion trap mass sp ectrometry Journal of Mass Spectrometry 1997, 32, 351]

PAGE 32

32 Figure 1-2. Stability diagram in (az qz) space for the region of simultaneous stability in both the rand z-directions near the origin for the three-dimensi onal quadrupole ion trap ; the isor and isoz lines are shown in the diagram. The qz-axis intersects the z=1 boundary at qz=0.908, which corresponds to qmax in the mass-selective instability mode. [Adopted from March, R.E. An introduction to quadr upole ion trap mass spectrometry Journal of Mass Spectrometry 1997 32, 351]

PAGE 33

33 Figure 1-3. Basic design of the two-dimensional linear ion trap. [Adopted from Schwartz, J.C.; Senko, M.W. A two-dimensional quadr upole ion trap mass spectrometer Journal of the American Society of Mass Spectrometr y 2002 13 659]

PAGE 34

34 Figure 1-4. Scheme for applica tion of DC, RF trapping, and AC excitation voltages necessary for operation of the 2-D ion trap. [Adopted from Schwartz, J.C.; Sento, M.W. A twodimensional quadrupole ion trap mass spectrometer Journal of the American Society of Mass Spectrometr y 2002 13 659]

PAGE 35

35 Figure 1-5. Electron ionization schematic. [Adapted from http://www.noble.org/plantb io/MS/iontech.ei.html]

PAGE 36

36 Figure 1-6. Chemical ionizati on schematic. [Adapted from http://www.noble.org/plantb io/MS/iontech.ci.html]

PAGE 37

37 Figure 1-7. Illustration of FAB. [Adapted from http://www.chm.bris.ac.uk/ms/theory/fabionisation.html ]

PAGE 38

38 Figure 1-8. Electrospray ionizati on schematic. [Adapted from http://www.noble.org/plantb io/MS/iontech.esi.html]

PAGE 39

39 COOH OH HO N N H3CO HO OCH3 COOH COOH H3CO COOH HO COOH HO HO COOH CN HOOC COOH CH2CH HO HO CH2 HO DHB SA NA FA 3HPA 4HCCA Succinic Acid Glycerol Figure 1-9. Most commonly used MALDI matrices

PAGE 40

40 Figure 1-10. A schematic diagram of the mechanism of MALDI. [Adapted from http://www.noble.org/plantb io/MS/iontech.ci.html]

PAGE 41

41 Figure 1-11. The chemical and biochemical surfaces available for SELDI from Ciphergen provides. [Adopted from Tang, N.; Tornatore, P.; Weinberger, S.R. Current developments in SELDI affinity technology Mass Spectrometry Reviews 2004 23, 34]

PAGE 42

42 Figure 1-12. The bulge and stem structure of aptamers. [Adopt ed from Biroccio, A.; Hamm, J.; Incitti, I.; Francesco, R.D.; Tomei, L. Selection of RN A aptamers that are specific and high-affinity ligands of the hepatiti s C virus RNA-dependent RNA polymerase Journal of Virology 2002 76 3688]

PAGE 43

43 Figure 1-13. The hairpin structure of aptamer. [Adopted from Biroccio, A.; Hamm, J.; Incitti, I.; Francesco, R.D.; Tomei, L. Selection of RNA ap tamers that are specific and highaffinity ligands of the hepatitis C virus RNA-dependent RNA polymerase Journal of Virology 2002 76 3688]

PAGE 44

44 Figure 1-14. The pseduknot structure of apta mer. [Adopted from Tuerk, C.; MacDougal, S.; Gold, L. RNA pseudoknots that inhibit hunman-immunodefficiency-virus type-1 reverse-transcriptase Proceedings of the National Academy of Sciences of the United States of America 1992, 89 6988]

PAGE 45

45 Figure 1-15. The G-quartet structure of an ap tamer. [Adopted from Wang, K.Y.; McCurdy, S.; Shea, R.G.; Swaminathan, S.; Bolton, P.H. A DNA aptamer which binds to and inhibits thrombin exhibits a new structural motif for DNA Biochemistry 1993 32 1899]

PAGE 46

46 Figure 1-16. Generalized scheme indicating th e key steps in the SELEX process. [Adopted from Klug, S.J.; Famulok, M. All you wanted to know about SELEX Molecular Biology Reports 1994 20 97]

PAGE 47

47 CHAPTER 2 MATRIX-DERIVED RESORC INOL-FORMALDEHYDE POLYMERS: SYNTHESIS VIA SOL-GEL METHOD AND USAGE IN MALDI-MS SAMPLE PREPARATION Since the invention of MALDI-M S, it has been used to analyze macro-biomolecules such as peptides mixtures,76 proteins,76 oligosaccharides,77 enzymatic protein digests,78 underivatized DNA, and oligomers.79 Although MALDI-MS is a powerful analysis method, it has problems associated with the sample pr eparation methods. Identifying and solving the problems can improve the quality of MALDI-MS spectra. Problems Associated with MALDIMS Sample Preparation Methods As a widely used ionization method for mass spectrometric analysis for macromolecules, MALDI has shortcomings associated with th e methodology of sample preparation. These problems include background interference from matr ix molecules when the molecular weight of the analyte is less than 2-3 times the mo lecular weight of the matrix molecule74 and poor signal uniformity of the signal intensity of analyt e across the sample from the inhomogeneous distribution of the matr ix-analyte co-crystals. 93 These problems can eith er interfere with the interpretation of MS spectra of low-mass analyt es or require significant effort to locate the sweet spot both of which ro utine analysis with MALDI-MS more difficult. Therefore, to eliminate the interference from matrix molecules and at the same time to increase the uniformity of the analyte signals intensities in MS spectru m, development of a polymeric substrate which has the physical properties amenable to MALDI-MS analysis and has the chemical properties to incorporate matrix molecules into it s structure is hi ghly desirable. Background Interference from Ma trix Molecules in MALDI-MS MALDI ionization coupled w ith mass spectrometry made the analysis of nonvolatile molecules possible. In MALDI, an excessive amount of matrix molecules, with the matrix-toanalyte ratio typica lly in the range of 100 to 50,000, is mixed with analyte in order to ionize the

PAGE 48

48 analyte particles. The excessive matrix mo lecules are not anchored to any support, and, inevitably, the matrix molecules will be releas ed and ionized upon ablation of the laser beam, and appear in the mass spectrum together with the analyte. Signals produced by the matrix molecules present in the mass spectrum along with the signals from analyte pose interference in analysis of low molecular weight analyte. Much effort has been devoted into finding th e mechanism of MALDI which can lead to the development of matrix suppression methods. Co mpetition model, in which competition between matrix and analyte for free prot ons based on the proton affinity,80, 81 is one of them. According to this model, matrix suppression takes place when adjusting the matrix-to-analyte ratio for every protonated matrix molecule to interact with at l east one analyte with high er proton affinity than the matrix molecule during desorption. In a nother proposed mechanism, analyte molecules are ionized by non-ionic matrix precursor species wh ich are vibrationally a nd electronically excited matrix molecules.82 Knochenmuss and Dubois proposed a model combining the previous two models. In this model excited, but not ionize d, matrix molecules are the common precursor for all subsequent ion products and the simultaneous ne ighboring presence of tw o such excitations is required for ionization of each analyte molecule.83 This model explained matrix suppression phenomena observed in a medium matrix-to-anal yte ratio with relatively low analyte signal intensity and less matrix suppres sion in a high matrix-to-analyt e ratio. According to this mechanism, suppression of the matrix signals ha s to be compromised to get high enough analyte signal intensity, or vice versa, when only the matrix-to-analyt e ratio is involved. Modifications other than adjusting the ma trix-to-analyte ratio in MALDI sample preparation have been made in order to reduc e the matrix interference in mass spectrometric analyses. These modifications include adding iodi ne to the matrix to su ppress the signals from

PAGE 49

49 the matrix-related ions and increa se the signals from analyte ions,84 using alkali dihydroxybenzoic acid salts as cationizing agents to suppr ess the DHB related ions,85 mixing graphite powder with glycer ol to reduce the interference,86, 90 and using nitrocellulose as a substrate together with matrix to eliminate the matrix signals.88 In each case the modification can reduce the signal intensities from matrix mo lecules but cannot completely eliminate them. In all those methods, the matrix molecules are not covalently attached to any support but merely interact with the support by dipol e-dipole interaction and van der Waals force; which are at the magnitude of one-thousandth and one-ten thousa ndth of the strength of the covalent bond, respectively. Therefore, are ea sy to be ablated and ionized when the energy of the laser beam shot on the target is much higher than the energy required to break these interactions. In an effort to fundamentally solve the matrix related problem, Lin and Chen incorporated matrix molecules into the gel formed from th e tetraethoxysilane (TEOS) precursor via sol-gel method.75 Total elimination of the matrix molecule signals was observed when small proteins, peptides, amino acids, and small organics were an alyzed. In the experiment, the matrix (DHB) was mixed with the monomer precursor to form the sol-gel solution and was claimed to become part of the polymer structure after the solu tion is polymerized. The mechanism of the polymerization reaction was proposed as shown in Figure 2-1. The hydroxyl groups on the matrix molecules can form covalent bonds with the hydroxyl groups in the silicic acid and thus covalently incorporated into the TEO S polymer as shown in Figure 2-2. Since the energy required to br eak the covalent b ond between the matrix molecule and the polymer is greater than the laser energy used in MALDI, trapping the matrix molecules covalently into the polymeric structure make it possible to totally eliminate the matrix ions background interferences in a MS spectrum. One problem associated with the TEOS gel, which

PAGE 50

50 was not mentioned in the paper, is that cracks formed when an aqueous solution was applied onto it since the structure of the gel is not st rong and can be easily broken. Developing a different matrix-embedded polymer which has be tter physical properties than the TEOS polymer is valuable for the polymeric substrate to be useful for MALDI-MS analysis with analytes in aqueous solution. The Uneven Distribution of Analyte Molecule s with the Traditional MALDI-MS Sample Preparation Methods There have been many investigations on the e ffect of MALDI sample preparation methods on the quality of mass spectrometric analyses MALDI-MS analyses results have strong dependence on the sample preparation method. Ever y single choice such as the solvent, matrix, and pH in sample preparation procedure can a ffect the outcome of the MALDI measurements.89, 90, 91, 92, 93, 94, 95, 96 The different combination of matrices, concentration of matrix, and solvent produce different distribution of matrix molecu les across the sample as shown in Figure 2-3.97 The problem which concerns the mass spectrometry analysts the most is the poor spot-to-spot and sample-to-sample reproducibility caused by inhomogeneous distribution of the matrixanalyte co-crystal.93 The methods used for the appl ication of the analyte samp le includes the rudimentary dried-droplet method which is to deposit matrix and sample mixture on the target with the matrix-to-analyte ratio as 103 to 105 and let it dry.98 This method is also called the air dried method.99 The formation of small co-crystals which contain matrix molecules and analyte molecules in the dried-droplet me thod is crucial for the ionization of the analyte molecules. The crystals formed in the sample prepared from dried-droplet method are not evenly distributed throughout the sample plate. The analyte-matrix co-crystals are found mostly at the edge of the droplet when DHB is used as matrix100, 101 as shown in Figure 2-4102 according to the partitioning

PAGE 51

51 process when solvent slowly evaporates. When the solvent slowly ev aporates the analyte molecules are excluded from the matrix crystal.99 An analyst would, therefore, to find sweet spot by scanning the sample surface. A thin-layer sample preparation method ha s been developed based on the dried-droplet method100 using a volatile solvent to create homogeneous matrix microcrystals because quick evaporation of the solvent can minimize the pa rtition process which occurs under the slow evaporation process. A drop of matrix solution in volatile solvent was deposited onto a target plate and was allowed to spread and dry quickly to form a thin layer of homogeneous matrix microcrystals. Then a drop of analyte soluti on was deposited on the top of the matrix layer.92 The spectra obtained from the thin-layer met hod produce more uniformed distribution of the analyte-matrix crystals than from the dry dr oplet method, therefore mo re uniformed analyte signal intensity across the sample (Figure 2-4). Sandwich sample preparation method is a co mbination of the thin layer and the drieddroplet method.98 It is called sandwich because sample so lution is applied on one layer of matrix crystals, then another matrix solution is added on the sample spot. This method has better mapping results than the dried-droplet method. To utilize the quick evaporation of the solven t further, electrospray sample preparation method was used to further improve homogeneity of the components distribution in MALDI-ToF MS than the use of volatile solvents and/or ra pid evaporation under vacuum or heating and the small and evenly distributed droplets can.99 This method produced much better spot-to-spot and sample-to-sample reproducibility in th e mass spectra than dried-droplet method. Since the selection of the matrix is very im portance in some cases, experiments using more than one matrix for the sample pr eparation have been also conducted.94 These improvements are

PAGE 52

52 made on a case-by-case basis. Limiting the size of the droplet is another t echnique that has been tested to improve the MALDI-MS analysis quality.103 Over all, there is not a universal sample preparation method that produces high quality mass spectra for variety of analytes. Mass sp ectra quality can be im proved by modifications made to the sample preparation methods. Elimina tion of the poor spot-to-spot reproducibility of the analyte molecules (the uneve n distribution of analyte across the sample) is impossible with the efforts have been done, because this problem is related to the proce ss of the crystallization. In the matrix-embedded polymeric MALDI-MS s ubstrate, the matrix molecules are evenly distributed throughout the polymer structure on the molecular level and cannot be changed during the sample preparation process and no anal yte-matrix co-crystals are formed, thus, the matrix-embedded polymer has the capa bility of solving this problem. Introduction to the RF Polymer According to the discussion above, a matrix-e mbedded polymeric substrate is the solution to the background noise from matrix molecules in mass spectra and the poor spot-to-spot and sample-to-sample reproducibility of MALDI ma ss spectra. Matrix-derived TEOS polymer was used by Lin and Chen75 and was synthesized and tested in our lab. The matrix-derived TEOS polymer was able to significantly lower the matrix signals when using MALDI-MS. Although the TEOS gel has the advantage of lowering the chemical noise from matrix molecules, the structure of the gel is fragile and is easy to break apart, cracks when aqueous solutions were deposited on it. This shortcoming made it a less than ideal substr ate for MALDI-MS since aqueous solutions are usually used and made it impossible for easy manipulation and further chemical modification. Resorcinol-formaldehyde (RF) polymer was finally chosen because of its tough physical structure and abundant pheno l groups on it which render the capability of

PAGE 53

53 being chemically modified. Resorcinol-forma ldehyde resins have been used as adhesion promoters to increase adhesion of rubber to fabric or metal. RF sol-gels were first synthesized by Peka la and co-workers acco rding to a hydrolysiscondensation reaction mechanism (Figure 2-5).104 Figure 2-6 shows the base catalyzed polymerization reaction of RF polym er. The R in the RF polyme r is referred to resorcinol, the F in the RF polymer is re ferred to formaldehyde. A patent was filed by Pekala on the synthesis of RF polymer.104 Resorcinol-formaldehyde resins have been used as adhesion promoters to increase adhesion of rubber to fabric or metal. A lot of research has been conducted since then on the effect of different reaction conditions on the final structures of the pol ymers which affect the physical, chemical, and electrochemical properties of RF aerogels and xerogels.105 Many patents have been filed and approved for RF polymers with different reactio n conditions. Those reac tion conditions include the ratio between the resorcinol and formaldehyde, the solvent and catalyst, the initial pH of the reaction, the ratio between the ca talyst and the resorcinol, the temperature of the reaction, and initial ratio between the reactants and the solvent.106 With these variable s there are tremendous possibilities of combinati ons of the conditions which make it a very versatile reaction that can be tailored for the desired properties. The RF polymers are categorized according to the solvent used. The gels with water solvent are called hydrogels or aquagels; the ge ls with organic solven ts are called lyogels.107 The initial reactants and the solvent ratios of the r eaction affect the final density of the gels size.108 Using higher concentrations of solvent can result in dilute effect which results in increased particle size.108 Using the higher concentra tions of the reactants can re sult in higher density of the formation of the RF crosslinked clusters.

PAGE 54

54 The most commonly used stoichiometric resorc inol / formaldehyde (R/F) molar ratio is 1:2 .104 Using more formaldehyde than twice the amount of the resorcinol as the starting material would cause a dilute effectthe effect caused by higher solv ent concentration.108 The most commonly used catalys t is sodium carbonate. The mo lar resorcinol-to-catalyst (R/C) ratio is usually between 50 and 300, or can be as high as 1500. The lower the R/C ratio, the smaller the polymer particles and the higher dens ity of the gels. In some cases acids such as HClO4 107 or HNO3 109 were used as catalyst, respec tively, with reduced gelation time.110 Using an acid catalyst combined with low concentration of reactants can result in small, smooth, fractal aggregates of gel particles.107 The pH should be controlled between 5.4 and 7.6111 since high pH can hinder the polymerization-condensation reaction112 and reactants precipitate at low pH113. Diluted acid such as HNO3, HClO4 and bases such as NH4OH can be used to adjust the initial pH of the reaction solution. Systematic Investigation in Preparation of the RF and Matrix-Derived RF Polymers To address the problems facing analysts working on MALDI-MS, RF polymers were synthesized to be suitable as substrate for MADLI-MS. The RF polymers developed were wet gels without any further drying processes. Thes e polymers are made via the quick and easy solgel method which makes it easy to be customized with any desired composition, on any surface, and any shape and size. The solutions can be stor ed in freezer for about 6 months. The polymer has a very strong structure due to the 1,3 bonding position on the aromatic rings in the polymer. The polymer did not crack upon th e deposition of aqueous soluti on and can only be broken by a severe impact. Usually the resorcinol formaldehyde ratio is 1: 2, both acid and ba se were used as a catalyst with different molar ratio of resorci nol/catalyst (R/C). Water, acetone, methanol,

PAGE 55

55 ethanol, n -propanol, and iso-propanol can all be used as the solvent. The catalyst used, ratio of R/C, the solvent, and concentration of the star ting materials determine the final structure and properties of the gel. Eventually, acid (H Cl) was chosen as the catalyst to provide H+ ions which are necessary for MALDI; acetone was chosen as solvent for quick polymerization; different matrices were used to investigate the ionizati on capability. The matrix molecules used for MALDI with similar structure as the RF polymer can be embedded into the polymer via covalent bonding. The ratio of the resorcinol and formaldehyde is chosen as 1:2 since the ratio of the resorcinol and formaldehyde in th e structure is 1:2. Water, alc ohol, and acetone have all been used in the literature and suita ble as the solvent of the RF pol ymerization reaction. Different solvents have been tested and eventually acetone was chosen as the solvent throughout the rest of the synthesis. With acetone as solvent, the temperature of the curing stage can be lowered significantly (from 80 C to 40 C ) with reasonable gelation time.104 The initial concentration of the reactants was chosen as (R+F )/solvent 30% w/w. The sodium carbonate was first chosen as the catalyst because it is the most commonly used in the previous literature. The higher the ratio, the smaller the polymer particles and the higher dens ity of the gel. The R/C ratio can be as high as 200:1 was chosen as the R/C ratio. Since an acidic polymer can provide H+ which is necessary in MALDI to ionize anal yte, acids were tested. HClO4 was used as first acid catalyst because it was used in the litera ture. Later, HCl was used thr oughout the rest of the synthesizing because it is a commonly used acid in lab, although there is no evidence in the literature about using HCl as catalyst in RF polymer synthesis. The initial synthesis of the RF polymer was tested with the following procedure: 4.00 g Resorcinol (0.036 mol), 5.00 mL formalde hyde (0.072 mol), 12.00 mL acetone, and 0.0078 g

PAGE 56

56 Na2CO3 (0.036 mmol) were added to the flask which was sealed and immersed in a 45C water bath. The formaldehyde-to-resorcinol ratio is 2: 1 and the resorcinol-to-Ca talyst ratio is 1000:1. Every half hour several drops of solution were taken out of the flask and deposited on a glass slide. The solutions taken out at one hour and after one hour all successfu lly polymerized. Thus, the reaction time was then determined as one hour. The solution is stored in a vial in the freezer with -20C. The solution can be used for six months after synthesis. This RF polymer was given the serial number as RF-1. There were thirty nine experime nts that have been conducted in an effort to systematically investigate the properties of the RF polymerizat ion process. The tests were conducted using a combination of different solvents such as acetone, ethanol, an d water; different resorcinol/catalyst ratios; different catalysts; di fferent matrices with varying concentration; and addition of different doping agents. Every RF solution was named with RF" followed by the serial number assigned from one to thirty nine. Thirteen RF polymers have been synthesized with Na2CO3 as a catalyst and acetone as solvent as shown in Table 2-1. These polymers have a different re sorcinol-to-catalyst ratio, with the addition of different matrices with varying concentration and the addi tion of doping agents in order to bring functional groups into the polymer. The resorcinol-to-catalyst was changed to 500 and 100 which are assigned the serial numbers RF-2 and RF-39, respectively. These tw o polymers, together with the RF-1 polymer, all polymerized. After the successful synthesis of the RF-1 and RF-2 polymers, RF-3 polymer was synthesized with 1000 ppm of DHB which also polymerized as expected. The RF-20 polymer was able to polymerize when the con centration of the DHB was raised to 40,000 ppm with the high catalyst concentration (r esorcinol-to-catalyst ratio as 100).

PAGE 57

57 After DHB was successfully incorporated into the RF polymers, CHCA was tested for incorporation into the RF polymer s with different concentration, di fferent resorcinol-to-catalyst ratio, and different concentration of 4-(imidazole -1-yl)phenole with or without being cued with HCl which were assigned the serial numbers from RF-27 to RF-35. RF-29, RF-30, and RF-31 polymerized, while others did not. The R/C can be as high as 1000 for the sol to polymerize as long as there is no imidazole added even when th ere is matrix added. When there is imidazole added, the R/C can not be higher than 50. The c oncentration of the matrix can be as high as 10,000 ppm when CHCA was used. Ethanol was tested as a solvent for RF-4 w ith 1000 resorcinol-to-ca talyst ration and 1000 ppm DHB, and RF-5 with 500 resorcinol-to-cata lyst ration and 1000 ppm DHB Both of the sol crystallized instead of polymerized (Table 2-2). Nine polymers (RF-6 to RF-12, RF 15, and RF 19) were test ed in the water as solvent with varying concentration of Na2CO3 as catalyst, varying concentr ation of DHB, and addition of Fe2Cl3H2O in order to add metal chelating capability as shown in Table 2-3. All of these polymers polymerized with varying reaction time. HClO4 was the first acid catalyst used in the synthesis of RF-12 polymer with water as solvent. The necessity to test an acid catalyst is based on the need of the protons in order to ionize analyte molecules in MALDI. HClO4 is documented as an acid catalyst for the synthesis of RF polymer, thus was chosen.104 Since storing HClO4 requiring extra caution, HCl was test as catalyst in the synthesis of RF polymer. Alt hough there is no evidence in the literature about using HCl as a catalyst for RF synthesis, the experiment was a success. There are thirteen polymers (RF-14, RF-16 to RF-18, RF-21 to RF 26, and RF-36 to RF 38) that were synthesized with HCl as catalyst for its proton; with acetone as solvent for the fast

PAGE 58

58 evaporation during the gel process; and with different matrices and varying concentration, with addition of imidazole, EDTA, and vinylimidazole; as shown in Table 2-4. Only seven of tem polymerized (RF-14, RF-16, RF-21 to RF23, RF-25, and RF-38). Among the thirty nine tests twenty four of them were able to polymerize. Nineteen of the polymerized solutions contain matrix. For each matrix-containing polymer, mass spectrometry was conducted with the sol-gel so lution deposited on a stainless st eel probe. After the solution was polymerized, the polymer was taped onto the st ainless steel with double-sided tap, and then 100 ppm spiperone in methanol was deposited on the polymer. After the spiperone solution dried the sample was analyzed in a custom built MALDI ion trap mass spectrometer. Among the twenty four matrix-embedded RF polymers, only the CHCA-embedded RF polymers produced MS signals with 100 ppm spip erone. As the concentration of the CHCA increases, the intensity of the analyte signal in creases under the same analysis condition. There is an up limit of the concentration of the matrix in the solution for the matrix-embedded resorcinol formaldehyde solution to polymeri ze which is 20,000 ppm. When 40,000 ppm of the matrix is added, the polymer (RF-24) forms a lo t of cracks which is not suitable as a MALDI-MS substrate. RF-16, which contai ned 20,000 ppm CHCA, produced sati sfactory MS analysis result and is used throughout the rest of the MS analysis. The solution is colorless while polymer polymerized from the solution is dark red. The color change indicates the formation of bonds on th e conjugated structure which is the aromatic rings in this reaction. The mechanism of the resorcinol formaldehyde polymerization shown in Figure 2-5 is proposed by Lin and Ritter111 which has an addition and a condensation step. Base was used as the catalyst in the addition step. In the second step the acid was used as catalyst to cure the solution. In the experiment reported here only base, only acid, and base first then acid

PAGE 59

59 as catalyst(s). In all cases th ere are successful examples, with the base catalyzed solution took longer time than the acid catal yst solution to poly merize. Figure 2-6 is the mechanism I proposed for the base-only catalyzed resorc inol formaldehyde polymerization reaction. Analysis of the Spiperone on the CHCA Derived RF Polymer for Lower Background Interference from Matrix Molecules This comparative analysis was performed with a microprobe quadrupole ion trap mass spectrometer (QITMS) which was built in the Yost laboratory by Christopher Reddick in 1997.11 The UV laser (Laser Science Inc. model VSL-337 ND) used for all MALDI MS analysis has 20 Hz pulsed nitrogen laser with a wavele ngth of 337.1 nm and a 3 ns pulse width.11 The maximum energy output of the laser is >250 J/ pulse with a peak power of 85 kW.11 Since the laser is near-diffraction limited, the beam can be focused to a diameter within a few times the lasers wavelength; for the studies reporte d here, the laser spot size range d from a diameter of 25 to 50 m. The laser is focused into the mass spect rometer chamber by a sing le fused silica lens (Melles-Griot) with a focal length of 25.4 cm as illustrated in Figure 2-7.11 The laser is externally triggered after a 1 ms delay by a Wavetek model function generator so the ion source gate has fully opened before ionization.11 The position of the sample plate can be adjusted manually along x and y axes.11 Photosensitive paper is used to determine the position of the laser beam with respect to the sample plate to align the in cident laser light.11 The software used to control the mass scan and data acquisition has been developed in our research group which is used to control auxiliary modulation fre quency and amplitude to extend mass range.11 As shown in Figure 2-8, ions are produced during the MALDI process, and then bend 90 into the ion trap by using a DC quadrupole deflector. This is to reduce the possibility of neutral collisions occurring within the ion trap caused by neutral molecules. After being ejected by the QIT from the exit end, ions are detected using a conve rsion dynode and electron multiplier which is behind

PAGE 60

60 the exit end electrode of the QIT.11 The data taken by this instrume nt was converted to excel file to generate the spectrum. The analyte used here, spiperone, is an antidep ressant. Figure 2-9 shows the fragmentation path ways of spiperone. Different matrices su ch as DHB, sinapinic acid, caffeic acid, and CHCA have been embedded into the RF polymer and th e matrices-embedded polymers have been tested for the ability to ionize spiperone. MS analysis showed that the RF polymer with CHCA was the only combination that generated a spiperone MS signal (Table 2-1 to 2-4). This phenomenon can be explained by the result from the UVvis spectra of the various matrix-embedded RF polymers (Figure 2-10). The discontinuity of spectrum at 350 nm is caused by the changing of light source from UV to visible. The CHCA ha d the highest absorption at 337 nm at the same concentration compare with other matrices; th e absorbance of the other matrices may render them less effective MALDI matrices. Figure 2-11 shows the UV absorbance of CHCA in acetone (purple), the pure RF polymer (blue), and the RF polymer embedded with CHCA (orange) which were analyzed on Varian Cary-100 Conc UV-visible spectrometer from Varian (Palo Alto, CA). The CHCA-embedded RF polym er showed UV absorption at 337 nm which is not the case in the pure RF polymer. Sin ce the solution of CHCA in acetone shows high absorbance around 337 nm range, it is obvious th at the CHCA-embedded RF polymer obtained its high UV absorbance at 337 nm from embedding CHCA into it. A polymerized RF 16 (20,000 ppm CHCA embe dded RF polymer) pellet was attached onto the stainless steel MALDI micropr obe with double-sided tape. A 0.2 L volume of 100 ppm spiperone was deposited onto each polymer dire ctly with pipette. After five minutes the sample is ready for MALDI-MS analysis.

PAGE 61

61 A 0.2 L volume of 2mM CHCA was also depos ited onto the stainless steel microprobe with pipette; it was dry after 5 min past, then 0.2 L of 100 ppm spiperone was deposited onto the matrix spot with pipette. The results are shown in Figure 2-12 a and b, eac h showing the average of six spectra. The spectrum of 100 ppm spiperone with 20,000 ppm CHCA on the stai nless steel showed [M+H]+ ( m/z 396) peak together with fragments from CHCA ( m/z 123.2 and m/z 172.5) which pose interference to the interpretation of the spectru m while the spectrum of 100 ppm spiperone on 2% CHCA-embedded RF polymer showed high [M+H]+ ( m/z 396) peak and no noticeable fragment from CHCA. By embe dding matrix molecules into RF polymer, the matrix molecules form covalent bonds with the polymer and are fixed into the polymer. Figure 2-13 is the illustration of one of the several possible bindings. Since the covalent bond between the polymer structure and the matrix molecules are much stronger than the hydrogen bond or the van der Waals force and thus requires much more en ergy to break the bond, the polymer keeps the matrix from being ablated while the matrix still ha s the ability to absorb laser energy to transfer to spiperone molecules and provide H+ for ionization. Moreover, while using the same laser power for both analyses, the spectrum of sp iperone on CHCA-embedded RF polymer showed lower intensities of daughter ions of spiperone (m/z 291.5, 238; m/z 165.5, 885) than the spectrum of spiperone with CHCA on stainless steel (m/z 291.5, 700; m/z 165.5, 3760). Comparing these two spectra, the spectrum obt ained with spiperone on the CHCA-embedded RF polymer showed improved quality. Analysis of the Spiperone on the CHCA-Derive d RF Polymer vs. Analysis of the Spiperone on the Stainless Steel Using CHCA as Matrix for More Uniform Analyte Signal Intensity Across the Sample The LTQ linear ion trap with vMALDI ion source from ThermoFinnigan (Figure 2-14) which is capable of imaging was used in this experiment. The laser power is 250 J/ pulse. The

PAGE 62

62 laser beam was set up to move in 120 m steps across the sample spot (3mm in diameter) and the data are collected automatically by the Xcalibur data system to generate the image of the signal intensity of the analyte throughout the analyzed area. The dried-droplet method was chosen for the comparison experiment because this method is still widely used for MALDI-MS analysis despite the fact that the matrix-analyte co-crystals are unevenly distributed across the sample whic h causes uneven distribution of signal intensity of analyte across the sample s pot. It would be exciting to see the improvement the CHCAembedded RF polymer can make by simply using dry droplet method, and make the using drieddroplet method by mass spectrometry analyst with less effort needed for search of sweet spots. For comparison of the dry droplet on stainless steel analysis vs spiperone solution on CHCA embedded RF polymer analysis; 0.2 L of 0.5 M CHCA in MeOH was applied on the stainless steel; after the matrix solution was dried, 0.2 L of 100 ppm spiperone solution in MeOH was applied on the dried matrix spot and on the CHCA embedded RF polymer, respectively. The microscopic image of the sa mple spot prepared with dried-droplet method before MS analysis is shown in Figure 2-15. After the sample plate was inserted into the instrument, the image of the sample plate was taken by the CCD camera and a circle was drawn by hand. The instrument started to run MALDI analysis throughout the circled area with each analysis spot 120 m apart both in x direct and y directi on. The position parameters (xand yaxis) were recorded as attachment of each sp ectrum for each analysis spot by the vMALDI software to be used to generate the image of the sample spots with the information of mass spectra after the analysis. Both image of the di stribution of intensity of TIC (total ion counts) and image of the distribution of in tensity of the ions within certain range can be generated. RSD, relative standard deviation which is calculated by dividing the st andard deviation by the average

PAGE 63

63 value, was calculated for m/z 304 (for cocaine) by surfer 8 based on all the spectra taken throughout the circled area and reported in gri d report. Typical resu lts from the sample prepared with dry droplet method on the stainle ss steel sample plate ar e shown in Figure 2-16 with relative standard deviation (RSD) as 228%. Figure 2-17 is the microscopic image of the CHCA-embedded RF polym er pellet after the deposition of 0.2 L of 100 ppm spiperone. Same mass spectrometry analysis as for the drieddroplet of spiperone on stainl ess steel was run for dried-dr oplet of spiperone on CHCAembedded RF polymer. Both image of the distribu tion of intensity of TIC (total ion counts) and image of the distribution of intens ity of the ions within certain ra nge can be generated. RSD was calculated for m/z 304 (for cocaine) by surfer 8 based on all the spectra taken throughout the circled area and reported in grid report. Typi cal results from the sample prepared with dry droplet method on the CHCA-embedded RF polymer are shown in Figure 2-18 with relative standard deviation (RSD) as 103%. Compared to the results obtained from the dr y droplet method on the stainless steel (Figure 2-16), the results from the dried-droplet on the CHCA-embedded RF polymer (Figure 2-17) produced more even distribution of the analyte across the sample with the RSD of the sample prepared on the CHCA-embedded RF polymer is about the half of that with the sample prepared with the dried-droplet method on stainless steel; thus, analyte signals can be detected easily without searching for the sweet spot. The more even distribution of the analyte signal resulted from the incorporation of matrix molecules into the rigid polymeric stru cture of RF polymer, since the matrix molecules are evenly distri buted throughout the whole polymeric structure on the molecular level. The morphology of the anal yte crystals only depend s on the distribution of the analyte across the sample and no concern of distribution of matrix molecules is necessary.

PAGE 64

64 Conclusion With the 1, 4 polymerization position on the ar omatic ring, RF polymers can withstand the harsh condition of chemical modification and washing with aqueous solution which is the condition for most biological sample. With the ar omatic groups in the RF polymers and reaction sites on the aromatic rings it is possible to incorporate aromatic groups containing matrix molecules in the polymeric network with covale nt bonds. By trapping the matrix molecules in the polymeric network with covalent bonds it stil l provides the assistant to ionization but can reduce the amount of matrix mol ecules been ablated by laser whic h can interfere with the MS signals from analyte with low molecular weight because it needs more energy to break a covalent bond than to break a inte rmolecular noncovalent bond.75 By incorporating the matrix molecules into the polymeric network structure, it can help solve the problem of unevenly distributed signal throughout the sample plate which is always a problem in MALDI-M S analysis. Many different sample preparation methods were developed to minimize the problem. The new resorcinolformaldehyde (RF) polymers developed have ph ysical and chemical properties suitable for embedding matrix molecules in the substrate fo r lowering the background noise from matrix molecules. With the nature of the matrix-embe dded polymer the matrix molecules distributed in the polymer evenly at the molecular level, thus provide the possibility to produce uniform signal intensity of analyte across the sample. This property also provides the possibility for quantization of the analyte in MALDI which is an important issue related to MALDI sample preparation.

PAGE 65

65 Figure 2-1. The mechanism of the TEOS sol-ge l reaction. [Adopted from Lin, Y.; Chen, Y. Laser desorption/ionization time-of-flight mass spectrometry on sol-gel-derived 2,5dihydroxybenzoic acid film Analytical Chemistry 2002 74 5793]

PAGE 66

66 Figure 2-2. Illustration of inco rporation of DHB molecules in the TEOS gel. [Adopted from Lin, Y.; Chen, Y. Laser desorption/ionizat ion time-of-flight mass spectrometry on sol-gel-derived 2,5-dihydroxybenzoic acid film Analytical Chemistry 2002 74 5793]

PAGE 67

67 Figure 2-3. Microscope pictures showing the sample surfac es of manually prepared samples with the dried-droplet (DD) or the seed layer (SL) method using (a) -cyanohydroxycinnamic acid (CHCA) DD, (b) CHCA SL, (c) sinapinic acid (SA) DD, (d) SA SL, (e) ferulic acid (FA) DD, (f) FA SL. [Adopted from Onnerfjord, P ,.; Ekstrom, S ,.; Bergquist, J .; Nilsson, J .; Laurell, T ,.; Marko-Varga, G Homogeneous sample preparation for automated high throughp ut analysis with matrix-assisted laser desorption/ionization time-of -flight mass spectrometry Rapid Communications in Mass Spectrometry 1999 13 315]

PAGE 68

68 Figure 2-4. MALDI ion images show ing intensities for selected peptides using (A) the thin-film, -CHC sample and (B) the dried-droplet, DHB sample. Approximately 250 fmol of 1-8, 250 fmol of AP, and 1 pmol of ELH were prepared with each matrix. The blackto-red color map corresponds to the arbitrar y intensity values specified for each peptide, while the grid lines correspond to 50 m increments within each image. [Adopted from Garden, R.W.; Sweedler, J. V. Heterogeneity within MALDI samples as revealed by mass spectrometric imaging Analytical Chemistry 2000 72 30]

PAGE 69

69 OH OH Na2CO3 O OH +CO H H 2 OH OH CH2OH CH2OH 1. Addition Reaction 2. Condensation Reaction OH OH CH2OH CH2OH H OH OH CH2 CH2 OH OH CH2OH CH2OH + OH OH H2C CH2OH HO CH2 OH CH2OH O CH2 CH2OH OH CH2OH H OH OH H2C CH2 CH2 OH O CH2 C H2 OH OH O CH2 OH OH C H2 CH2OH OH HO HO OH CH2 O CH2 OH OH OH CH2OH HO Figure 2-5. Mechanism of the polymerization of RF polymer proposed by Lin and Ritter. [Adopted from AL-Muhtaseb, S.A.; Ritter, J.A. Preparation and properties of resorcinol-formaldehyde organic and carbon gels Advanced Materials 2003 15 101]

PAGE 70

70 OH OH B O OH O OH O O OH OH CO H H O OH H CH2O O OH CH2 O OH O OH CH2 O OH OH OH CH2 O OH O OH CH2 O OH OH O OH CH2 O OH HO HO OH O OH Figure 2-6. Base-catalyz ed RF polymerization.

PAGE 71

71 Figure 2-7. Cross-section of the ion trap mass spectrometer. [Adopted from Reddick, C.D. The detection of pharmaceutical drug compounds fr om intact biological tissue by matrixassisted laser desorption ionization me thod (MALDI) quadrupole ion trap mass spectrometry 1997 PhD dissertation. University of Florida, Gainesville, Florida.]

PAGE 72

72 Figure 2-8. Illustration of the path of laser and ions in ma ss spectrometer. [Adopted from Reddick, C.D. The detection of pharmaceutical drug compounds from intact biological tissue by matrix-assisted lase r desorption ioniza tion method (MALDI) quadrupole ion trap mass spectrometry 1997 PhD dissertation. Univ ersity of Florida, Gainesville, Florida.] Laser Beam in Ions out

PAGE 73

73 N N N H H O O F C N O F N O H H N N N H H O H C O F N N N H H O [Spiperone + H]+ m/z 396.49 m/z 165.19 m/z 291.35 m/z 232.31 m/z 260.36 Figure 2-9. Fragment path ways of spiperone.

PAGE 74

74 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 250300350400450500 Wavelength (nm)Absorbance 337 10,000 ppm CHCA in RF gel 10,000 ppm Sinapinic acid in RF gel 10,000 ppm caffeic acid RF gel 10,000 ppm DHB in RF gel Figure 2-10. UV absorbances of the va rious matrix-embedded RF polymers.

PAGE 75

75 0 1 2 3 4 5 6 7 8 9 10 250300350400450500 Wavelength (nm)Absorbance 337 RF polymer CHCA in acetone solution 10,000 ppm CHCA in RF polymer Figure 2-11. UV absorbances of the RF polymer (blue), CHCA-embedded RF polymer (orange), and CHCA solution (pink). The spike at 320 and 350 is caused by the instrument used for analysis.

PAGE 76

76 0 1000 2000 3000 4000 5000 6000 7000 8000 50100150200250300350400450500550600650 m/zIntensity396.5 165.5 291.5 0 1000 2000 3000 4000 5000 6000 7000 8000 50150250350450550650 m/zIntensity396.6 165.3 123.2 291.5 172.5 Figure 2-12. Mass spectrum of the 100 ppm spiperone on A) 20,000 ppm CHCA embedded RF polymer, and B) stainless st eel with CHCA as matrix. [ s p i p erone + H ] + [ s p i p erone + H ] + C1 3 H1 7 N 3 O [spiperone + H]+ C7H7 [ s p i p erone + H ] + C13H17 N 3O [ s p i p erone + H ] +-C 7 H 7 [ s p i p erone + H ] + [ CHCA + H ] + C 3 H2 N O [ CHCA +H ] + H2O A B

PAGE 77

77 O CH2 O OH O CH2 O HO HO OH HO OH O NC O OH Figure 2-13. CHCA embedded in the RF polymer structure.

PAGE 78

78 Figure 2-14. LTQ with vMALDI ion source from ThermoFinnigan.

PAGE 79

79 Figure 2-15. The microscopic image of the sample well after the dry droplet sample preparation before MS analysis (3 mm in diameter).

PAGE 80

80 Figure 2-16. The distribution of the spiperone intensity ( m/z 396) across the sample using dry droplet method on stainless steel. A) 2-D image. B) 3-D image. X ( micron ) X ( micron ) 0 0 500 500 1000 1000 1500 1500 2000 2000 2500 2500 3000 300 0 d anc e d ance 1.000e+007 1.000e+007 5.000e+006 5.000e+006 0.000e+000 0.000e+000 Y ( micron ) Y ( micron ) 250 0 2500 2000 2000 1500 1500 1000 1000 500 500 0 0 Intensit y A B

PAGE 81

81 Figure 2-17. The microscopic image of the RF -16 polymer pellet after the deposition of 0.3 L of 100 ppm spiperone before MS analysis.

PAGE 82

82 Figure 2-18. The distribution of the spiperone intensity ( m/z 396) across the sample using dry droplet method on RF-16 polymer pellet. A) 2-D image. B) 3-D image. X ( micron ) X ( micron ) 0 0 500 500 1000 1000 1500 1500 2000 2000 2500 2500 3000 300 0 d ance d anc e 1.000e+007 1.000e+007 5.000e+006 5.000e+006 0.000e+000 0.000e+000 Y ( micron ) Y ( micron ) 250 0 2500 2000 2000 1500 1500 1000 1000 500 500 0 0 Intensit y A B

PAGE 83

83 Table 2-1. RF polymers with acetone as solvent and Na2CO3 as catalyst: 4.00 g Resorcinol + 5.00 mL formaldehyde in 12.00 mL acet one with Na2CO3 as catalyst Catalyst Matrix Polymerized Signal RF-1 0.0078 g Na2CO3 (R/C 1000) no yes no RF-2 0.0156 g Na2CO3 (R/C 500) no yes no RF-39 0.078 g Na2CO3 (R/C 100) no yes no RF-3 0.0156 g Na2CO3 (R/C 500) 0.0168 g DHB (1000 ppm) yes no RF-20 0.039 g Na2CO3 (R/C 100) 0.672 DHB (40,000 ppm) yes no RF-27 0.0156 g Na2CO3 (R/C 500) 0.4146 g CHCA (20,000 ppm) + 0.3492 g 4(imidazole-1-yl)phenole (20,000 ppm) no N/A RF-28 0.0156 g Na2CO3 (R/C 500) 0.4146 g CHCA (20,000 ppm) + 0.3492 g 4(imidazole-1-yl)phenole (20,000 ppm) 2 h later add 1.6 mL 0.1 M HCl to cue no N/A RF-29 0.0156 g Na2CO3 (R/C 500) 0.2073 g CHCA (10,000 ppm) + 0.2746 g 4(imidazole-1-yl)phenole (16,000 ppm) ), 2 h later add 1.6 mL 0.1 M HCl to cue yes no RF-30 0.156 g Na2CO3 (R/C 50) 0.2073 g CHCA (10,000 ppm) + 0.2746 g 4(imidazole-1-yl)phenole (16,000 ppm) yes no RF-31 0.156 g Na2CO3 (R/C 50) 0.2073 g CHCA (10,000 ppm) + 0.2746 g 4(imidazole-1-yl)phenole (16,000 ppm), 2 h later add 1.6 mL 0.1 M HCl to cue yes no RF-32 0.078 g Na2CO3 (R/C 100) 0.2073 g CHCA (10,000 ppm) + 0.2746 g 4(imidazole-1-yl)phenole (16,000 ppm) no N/A RF-33 0.078 g Na2CO3 (R/C 100) 0.2073 g CHCA (10,000 ppm) + 0.2746 g 4(imidazole-1-yl)phenole (16,000 ppm), 2 h later add 1.6 mL 0.1 M HCl to cue no N/A RF-34 0.034 g Na2CO3 (R/C 200) 0.2073 g CHCA (10,000 ppm) + 0.2746 g 4(imidazole-1-yl)phenole (16,000 ppm) no N/A RF-35 0.034 g Na2CO3 (R/C 200) 0.2073 g CHCA (10,000 ppm) + 0.2746 g 4(imidazole-1-yl)phenole (16,000 ppm), 2 h later add 1.6 mL 0.1 M HCl to cue no N/A

PAGE 84

84 Table 2-2. RF polymers with ethanol as solvent and Na2CO3 as catalyst: 4.00 g Resorcinol + 5.00 mL formaldehyde in 12.00 mL ethanol Catalyst Matrix Polymerized Signal RF-4 0.0078 g Na2CO3 (R/C 1000) 0.0168 g DHB (1000 ppm) Crystalline no RF-5 0.0156 g Na2CO3 (R/C 500) 0.0168 g DHB (1000 ppm) Crystalline no

PAGE 85

85 Table 2-3. RF polymers with water as solvent and Na2CO3, or HCl, or HClO4 as catalyst: 4.00 g Resorcinol + 5.00 mL formaldehyde in 12.00 mL H2O Catalyst Matrix Polymerized Signal RF-19 0.0156 g Na2CO3 (R/C 50) no too quick no RF-6 0.0156 g Na2CO3 (R/C 500) 0.0168 g DHB (1000 ppm) yes (5 hrs) no RF-7 0.0156 g Na2CO3 (R/C 500) 0.0168 g DHB (1000 ppm) + 0.266 g of FeCl3H2O (1000 ppm) too quick (15 mins) no RF-8 0.0156 g Na2CO3 (R/C 500) 0.0168 g DHB (1000 ppm) + 0.0266 g FeCl3H2O (100 ppm) yes (2 hrs) no RF-9 0.0156 g Na2CO3 (R/C 500) 0.0336 g DHB (2000 ppm) + 0.0266 g FeCl3.6H2O (100 ppm) yes (3 hrs) no RF-10 0.0156 g Na2CO3 (R/C 500) 0.0336 g DHB (2000 ppm) yes (2 hrs) no RF-11 0.0156 g Na2CO3 (R/C 500) 0.168 g DHB (10,000 ppm) yes (2 hrs) no RF-12 0.1737 mL1 M HClO4 (R/C 200) 0.168 g DHB (10,000 ppm) yes (1 hr) no RF-15 1.6 mL 0.1 M HCl (R/C 200) 0.336 g DHB (20,000 ppm) yes (< 0.5 hr) no

PAGE 86

86 Table 2-4. RF polymers with ace tone as solvent and HCl as cat alyst: 4.00 g Resorcinol + 5.00 mL formaldehyde in 12.00 mL acetone with HCl as catalyst Catalyst Matrix Polymerized Signal RF-38 1.6 mL of 0.1 M HCl (R/C 200) no yes no RF-17 1.6 mL of 0.1 M HCl (R/C 200) 0.49 mL 1-vinylimidazole (100,000 ppm) no no RF-18 1.6 mL of 0.1 M HCl (R/C 200) 0.1593 g EDTA (10,000 ppm) no no RF-26 1.6 mL of 0.1 M HCl (R/C 200) 0.3492 g 4-(imidazole-1-yl)phenole (20,000 ppm) no no RF-14 1.6 mL of 0.1 M HCl (R/C 200) 0.336 g DHB (20,000 ppm) yes (< 0.5 hr) sol became milky in freezer no RF-16 1.6 mL of 0.1 M HCl (R/C 200) 0.4146 g CHCA (20,000 ppm) yes (1 hr) yes RF-24 1.6 mL of 0.1 M HCl (R/C 200) 0.8292 g CHCA (40,000 ppm) no no RF-25 1.6 mL of 0.1 M HCl (R/C 200) 0.2073 g CHCA (10,000 ppm) yes yes RF-21 1.6 mL of 0.1 M HCl (R/C 200) 0.488 g SA (20,000 ppm) yes no RF-22 1.6 mL of 0.1 M HCl (R/C 200) 0.390 g caffeic acid (20,000 ppm) yes no RF-23 1.6 mL of 0.1 M HCl (R/C 200) 0.312 g 3-amino-qumiline (20,000 ppm) yes no RF-36 3.2 mL of 0.1 M HCl (R/C 100) 0.2073 g CHCA (10,000 ppm) + 0.2746 g 4(imidazole-1-yl)phenole (16,000 ppm) no no RF-37 0.16 mL of 1 M HCl (R/C 200) 0.2073 g CHCA (10,000 ppm) + 0.2746 g 4(imidazole-1-yl)phenole (16,000 ppm) no no

PAGE 87

87 CHAPTER 3 SURFACE MODIFIED RF POLYMER WITH APTAMERS: METHODOLOGY AND USAGE IN SELDI-MS SAMPLE PREPARATION Because of the complexity of biological sa mples obtained from i ndividuals for disease screening tests and the chemicals used during the sample handling pr ocess, one or more cleaning or separation steps is usually employed to remove most of the unwanted species and preconcentrate the biomarkers before MALDI-MS analyses.114 Those separation and enrichment methods are usually tedious, time-consumi ng, and require large amounts of sample. Many investigations have been performe d to reduce the inte rferences from the uninformative components in MALDI analyses wit hout a cleaning step. Se veral polymers were used to achieve on-probe reducti on of contaminations. High-den sity polyethylene membrane, high-density polypropylene, octyl (C8) and octadecyl (C18) extraction disk from 3M (St. Paul, MN) were used successfully to remove salt, dete rgent, and glycerol contaminants from peptide and protein sample solutions.117 Immobilization of proteins onto nylon-66 and positive chargemodified nylon (Zetabind) membranes removed the water soluble contaminates before MS analysis.116 Perfluorosulfonated ionomer film were used to analyze bi ological mixtures such as chemical digests of proteins, proteins in milk and egg white, cell lysate, and oligonucleotide without pre-purification with enhanced results.118 On-probe purification also achieved by using digests nitrocellulose for MALDI-MS analysis of PCR (Polymerase chain reaction) products and DNA fragment.118 Polyethylene membrane was used for high-mass proteins (100,000 Da) MALDI-MS analysis with e nhanced spectral quality.119 Hydrophilic poly(acr ylic acid) (PAA) was used for its selective adsorption of partially digested myoglobin to remove contaminants by rinsing with water.120 Metals such as patterned gold with poly(acrylic acid)121 as analysis surfaces were tested to reduce the contaminations in the samples.

PAGE 88

88 Surface modifications have been performed in several cases with polymeric materials in order to covalently bind a certain class of analyte for purification from solution for enhanced performance (SELDI). A surface of poly(4-vinylpyridine) which was extended with spacer arms (N -ethylsuccinamyl-) and termin ated with leaving groups ( N -hydroxysulfosuccinimide) for covalent attachment of an 18-resi due peptide (N terminus of human -casein) was used for onprobe purification for pre-analysis purification.122 Poly-lysine-immobilized poly-2-hydroxyethyl methacrylate membrane was used to sele ctively absorb DNA from aqueous solution.123 Affinity adsorption without the formation of covalent bonds allows relatively easy mass spectrometric analysis. Surface modification, whic h adds specific affinity to the substrate surface for affinity extraction, was developed. The extremely strong avidin glycoprotein-biotin interaction (kd= 10-15) was utilized to retain the biotinylat ed analyte onto the avidin immobilized agarose beads124, 125 or polymer thin films.126 Lectins have been immobilized onto gold foils to affinity capture bacteria from th e contents in the sample solution.127 Several dextrans were immobilized onto the fresh gold support as a self-assembled monolayer to retain binding molecules which capture analytes of interest.128 DNA has been immobilized onto the poly(ethylene terephthalate) microfib er to capture anti-DNA antibodies.129 Although surface modification with affinity agents is superior to simply using the polymers as substrates for MALDI-MS analysis, each of the methods has its limitation. The avidin modified substrate requires the analyte to be biotinylated; lectins can be used to capture bacteria but lacks specificity; the dextran-modi fied gold surface can work the same way as IMAC does and both lack specificity; DNA has hi gher specificity but w ith the drawback of difficulty in obtaining and handli ng DNA, plus the large size of DNA causes less amount to be attached onto the solid support. A universal one-step on-probe clean -up method which has a

PAGE 89

89 high specificity towards biomarkers, high enough de nsity of recognizing site s, with low cost and yet easy to manipulate for biological sample anal ysis has been developed. This method is a surface modification of a RF polymer with aptame rs for affinity capture. Antibodies have the highest specific affinity among th e biological receptors used for modifications, though is one of the most expensive. Aptamers have numerous advantages over antibodies, yet, have not been wildly tested. As shown in Figure 3-1, there are two phenol groups on each aromatic ring which make up most of the RF polymer structure. These phenol groups provide plenty of reaction sites for direct surface modification with aptamer and surface modi fication with aptamer via carboxylation, i.e., indirect surface modification. The phenol group s on the RF polymer have a pKa of 9.86 and the hydroxyl groups on the silica partic le surface have a pKa of 9.82; thus, the reactivity of the phenol groups should be very close to the hydr oxyl groups on the gla ss nanoparticles. The surface modification of the RF polymer can follow th e procedures used in the research lab of Dr. Weihong Tan (University of Florida) for the surface modification of silica particles. Aptamers Used for RF Polymer Surface Modification Aptamers have been used as molecular beacons in Weihong Tans lab ( University of Florida) for in vitro fluorescent detection. Molecu lar beacons detect the ex istence of an analyte according to the changing fluorescent signal inte nsity when the beacon changes the conformation as the result of the affinity capture of the anal yte. Fluorescent signal detection for analyte with molecular beacon works very well if the analyte of detection in known to exist in the tested solution, otherwise, there can be a false positive re sult and a definite identif ication is impossible. Tandem mass spectrometry analysis (MSn) has the capability for definite identification of a substance based on its fragmentation pattern. C oupling the high specific recognition capability of molecular beacon with the in forming power of mass spectrometry would be the answer for the

PAGE 90

90 limitations of molecular beacons. Thrombin-bindi ng aptamers were covalently attached to the surface of a glass slide and were used successfully for affinity capture of thrombin in MALDI targets by McGown at Duke University.73 The RF polymer devel oped in our lab has enough reaction sites, with simi lar reactivity as silica s lides, to be used as solid support for aptamer attachment and with the flexib ility in the size and shape th e sol-gel polymerization method renders. There is a long list of aptamers has been generated by SELEX process; moreover, an aptamer can be produced with affinity towards any specific target molecule by the SELEX process. The choice of the aptamer to be att ached onto the RF polymer is based on the concern that the successful attachment of the aptamers onto the RF polymer should be verified before the affinity capture of a specific molecule is prove d by mass spectrometric analysis. Verification of the attachment of the aptamers on the RF with MS would include MS-related variables into the verification process; thus, includ e more uncertainty in the detect ion. Since the task in this project is to develop a methodol ogy for affinity capture of anal yte of interest prior to MS analysis, choosing an exis ting aptamer with high affinity of a specific molecule would save a lot of effort in generating a new aptamer. An aptame r with high affinity towards a molecule that can be detected by the naked eye and be measured by UVvis was chosen. The molecule chosen is reactive green 19 and belongs to a group of reactive dyes used in textile industry.130, 131 This group of dyes has been used in dye-ligand affinity chromatography for the purification of proteins because of the high selectivity and the reversibility of the affinity binding between the dye and the protein.132 These synthetic dyes interact with the active sites of many proteins by mimicking the structure of the s ubstrates, cofactors, or binding agents for these proteins. The aromatic triazine dye structure re sembles the nucleotide structure of nicotinamide

PAGE 91

91 adenine dinucleotide and the dye interacts with the dinucleotide fold in these proteins. These reactive dyes can bind proteins by electostatic an d hydrophobic interactions and by more specific pseudoaffinity interactions with ligand-bi nding sites.131 The degree of purification achieved with dye-ligand chromatography is generally be tter than that obtained with less specific techniques such as ion-exchange or gel f iltration chromatography. The reactive dyes are relatively inert and unaffected by en zymes in crude cellular extracts.131 Figure 3-2133 shows the structure of the reactive green 19 molecule. Th is dye molecule has absorbance peak at around 630 nm wavelength and can be detected by the naked eye and UVvis spectrometry. The sequence of the aptamer with high speci fic affinity toward reactive green 19 is ACCCG GCGTT CGGGG GGTAC CG GGT which was discovered by Ellinton and Szostak in 1992134, 135 and can be synthesized in a DNA synthesizer.132 In 20 mM Tris buffer at pH 7.6 with 0.5 M LiCl and 1 mM MgCl2, the aptamer folds into a stem-loop structure as shown in Figure 33133 and can be used in a bioassay when labeled with fluorescein.136 After the reactive green 19-retaining aptamer was successfully attached onto the RF polymer, a cocaine-retaining aptamer was chosen to be attached onto the RF polymer for the cocaine capturing and MS analysis. Cocaine is an alkaloid found in leaves of the South American shrub Erythroxylon coca The active ingredient of the coca plant was first isolated in the West by the German chemist Friedrich G aedcke in 1855. Albert Niemann described an improved purification process for his PhD whic h he named as "cocai ne". Cocaine is a powerfully addictive stimulant drug which wa s tried by approximately 33.7 million Americans ages 12 and older, representing 13.8% of the populat ion, at least once in th eir lifetimes according to the 2005 National Survey on Drug Use and Health.137 Although detection of cocaine by GCMS is well established, it will be helpful if cocai ne can be captured from biological sample such

PAGE 92

92 as blood and urine and to be an alyzed by MALDI-MS. Actually, after this aptamer affinity capture procedure is establishe d, any disease biomarker could be captured and analyzed by MS the same way as reactive green 19 and cocaine. Aptamer affinity capture of cocaine can be achieved by attaching a cocaine-retaining aptamer onto the RF polymer. An aptamer with th e affinity towards cocaine (Figure 3-4 b), one of the two cocaine isomers, was constructe d by Landry et al., as shown in Figure 3-545 which forms a three-way junction in 20 mM Tris bu ffer with pH 7.4, 140 mM NaCl, and 2 mM MgCl2 to accommodate a cocaine molecule at the center of the aptamer tertiary structure. With dissociation constant of 0.4 10 M between cocaine and the cocaine-retaining aptamer,45 cocaine-retaining aptamer was used as a fluorescent sensor for cocaine45 and colorimetric probe for the detection of cocaine.133 Direct Surface Modification of the RF Polymer with Aptamer The reactive green 19-retaining aptamer and cocaine-retaining aptamer were directly attached onto the RF polymer in the nucleic acid synthesizer (App lied Biosystems 3400 DNA Synthesizer, Figure 3-6). The Applied Bi osystems 3400 DNA Synthesizer from Applied Biosystems (Foster City, CA) is a fully progr ammable instrument that provides four-column simultaneous synthesis, and features automatic base dilution and analysis of coupling efficiency. The RF polymer pellets were first polymerized by depositing RF solution onto the top of a stainless steel microprobe and allowing several hours for the solution to polymerize. These pellets were then hammered to small pieces to fit in the columns of th e DNA synthesizer. These RF polymer pieces were loaded into the reaction column as solid support for aptamer attachment. The sequence of the aptamer was set up using the so ftware and the synthesis of the aptamers was controlled by the computer. In each step, the solution is pumped through the column which is attached to the reagent delivery lines and the nuc leic acid synthesizer. Ea ch base is added via

PAGE 93

93 computer control of the reagent delivery. After the synthesis was finished the RF polymer pieces were taken out to be immersed in NH3OH / CH3NH2 ratio as one at room temperature for three hours for de-protecti on of the aptamers. Surface Modification of the RF Polymer with Aptamers via Surface Modification of the RF Polymer with Carboxylic Groups Research has been conducted in the re search group of Dr. Weihong Tan group on carboxylation of silica nanoparticles and reaction of the carboxylic acid groups with amine modified aptamers for indirect surface modifi cation. According to th e procedure, the RF polymer will be carried out by immersion in freshl y prepared 1% (v/v) solution of distilled (3(trimethoxysilyl)propyl)diethylene triamine (DETA) and 1 mM acet ic acid for 30 min at room temperature. The DETA modified RF polymer w ill be rinsed with deionized water to thoroughly remove excess DETA. The amine-functionalized polymer was then treated with succinic anhydride in dry tetrahydrofuran (THF) in presence of an argon atmosphere for 6 h. Figure 3-7 shows the scheme of the surface modification. After the RF polymer was carboxylated amine-m odified cocaine aptamers were attached onto the RF polymer via the formation of an amide bond following the procedure used by Weihong Tans lab.136 Characterization of the Carboxy lic Group-Modified RF Polymer The verification of the carboxylation was pe rformed by labeling the RF polymer with 1pyrenyldiazomethane (PDAM) (Figure 3-8), a new fluorescent labeling r eagent from Molecular Probes (Eugene, OR) with an amine group on it.139 The covalent attachment of the PDAM to the carboxylated RF polymer was performed according to the procedure described by Gaber et al.139 The carboxylated RF polymer pieces (sample) and non-carboxylated RF polymer pieces (control) were placed in two se parate vials and immersed in an aliquot of 10 mM PDAM solution

PAGE 94

94 in ethanol. The two suspensions were mixed vigorously on a vortex mixer for two hours at room temperature before the polymer pieces were thor oughly rinsed with ethanol. The polymer pieces were transferred to two separate vials a nd were rinsed three times with ethanol. The excitation wavelength of PDAM is 351 nm and the fluorescent emission wavelength is 392 nm. Fluorescence emission intensity spectra of the sample and control in ethanol were obtained by readings from Tecan Microplat e Reader from Tecan (Mannedorf/Zurich, Switzerland) to verify the succe ss of surface modification of the RF polymer (Figure 3-9). The PDAM-bond polymer pieces has a much higher emi ssion at 392 nm wavelength after excited at 351 nm than the control which proved the exis tence of carboxylic acid groups on the RF polymer. The determination of the amount of th e carboxylic groups on the RF polymer was evaluated from the uptake of a basic dye mo lecule, toluidine blue O (Figure 3-10),140 by the carboxylated RF polymer. Toluidine blue O co mplexes to equivalent moles of carboxylate groups which is the base for calculating the am ount of the carboxylic acid groups according to the amount of the toluidine blue O ab sorbed by the carboxylated RF polymer.140 The carboxylated RF polymer pieces were immersed in 500 m toluidine blue O of pH 10 at 30 C for five hours before the RF polymer pieces were rinsed thoroughly with 5.0 x 10-4 N NaOH aqueous solution. The polymer pieces were rinsed three times after they were transferred to another vial. The dye molecules were desorb ed from the RF polymer by immersing the RF polymer pieces in 50% acetic acid solution. All the UV spectra in this project were re corded on Varian Cary-100 Conc UV-visible spectrometer (Palo Alto, CA). Toluidine blue O has UV absorbance at 285 and 623 nm. The UV absorbance of the toluidine blue O desorbed from the carboxylated and noncarboxylated RF

PAGE 95

95 polymer served as a second proof of the exis tence of the carboxylic acid groups on the RF polymer (Figure 3-11). The UV absorbance of the toluidine blue O solu tions with different c oncentrations (Figure 3-12) was used to construct a calibration curve at 285 nm (Figure 3-13). Bases on the linear regression fit shown, the concen tration of the toluidine blue O desorbed from twelve round pieces (3.5 mm diameter) of carboxylated RF polymer and non-carboxylated RF polymer in 9 mL of the 50% acetic acid solution was 7.8x10-6 M and 4.8x 10-6 M, respectively. The difference of the concentration of toluidine bl ue O desorbed between the carboxylated RF and non-carboxylated RF polymer pieces was 3.0x10-6 M mol by subtracting the concentration of toluidine blue O desorbed from the noncarboxylated RF polymer from the concentration of the toluidine blue O desorbed from the carboxylated RF polymer. The difference of the amount of toluidine blue O in 9 mL of solutions betw een the carboxylated RF and non-carboxylated RF polymer pieces was calculated by timing the difference of the con centration of toluidine blue O between the carboxylated RF and non-carboxylated RF polymer pieces with 9 mL to get 2.7x108 M. The density of the carboxylic groups on the polymer was cal culated by dividing 2.7x10-8 mol by the total surface area of the twelve polymer pieces with two sides (231 mm2) to get 1.2x10-10 mol/mm2. The number of carboxylic groups av ailable to be ablated by a 100 micrometer radius laser pulse is thus 3.7x 10-12 mol, which is more than adequate to yield a strong MALDI signal. Affinity Capture of RG 19 with RG 19-Re taining Aptamer-Modified RF Polymer and Characterization of the Modified RF Polymer The affinity capture of the RG 19 by the RG 19-retaining aptamer was demonstrated by desorption of this dye as detected by UVvis spectrometry. The aptamer-modified RF polymer pieces and non-modified RF polymer pieces were immersed in the 20 mM Tris-Cl buffer with

PAGE 96

96 RG 19 in two separate plastic vials for four hour s. The polymer pieces were rinsed thoroughly with the Tris buffer before they were transferred to two different vials in order to avoid the nonspecific capture of the RG 19 on the plastic vial wall. The polymers were rinsed three times again before deionized (DI) water was added to th e vial to desorb the RG 19. The DI water in the vial containing the aptamer-attached RF polymer turned blue instantly, while the DI water in the vial contained the non-modifi ed RF polymer did not change color. UV spectrum was taken for the desorption solution fr om both vials (Figure 3-14). In order to determine the amount of aptamer on the polymer surface, the concentration of RG 19 desorbed into solution was measured by its UV absorbance. A calibration curve was constructed (Figure 3-15) according to the UV absorbance at 634 nm of the reactive green 19 solutions different concentration. The equa tion of the calibration curve as y=0.00061x + 0.0019. According to the calibration curv e the amount of the RG 19 was calculated to be 0.20 mM from the 0.123 UV absorbance of the sample. The amount of the RG 19 in the 5 mL solution was 7.8x10-7 mol. Two round pieces of the RF pol ymer have a total surface area of 38.5 mm2. The density of the reactive green 19 absorbed on the aptamer-attached polymer was calculated as 2.03x10-8 mol/mm2. The number of reactive green 19 can be ablated by laser (the radius of the laser shot is estimated to be 100 m) is 6.0x10-10 mol /spot, which is plenty for MALDI-MS analysis. Affinity Capture of Cocaine with Direct and Indirect CocaineRetaining AptamerModified RF Polymer The affinity capture of cocaine from a cocaine solution by the direct aptamer modified RF polymer was first detected using a visual detection method, conducte d according to the procedure described by Landry et al.53 Diethylthiatricarbocyanine iodide, a cyanine dye, can be captured by the cocaine-retaining aptamer. Th e addition of the cocaine solution into the

PAGE 97

97 diethylthiatricarbocyanine iodide-aptamer complex containing would release the dye molecule to capture the cocaine molecule (Figure 3-16)53 which serves as an easy visual detection of the existence of the cocaine-retaining aptamers on the RF polymer. The pieces of the cocaine retaining-aptamer modified RF polymer and pieces of non-modified RF polymer were put into two separate vials with 500 L of 20 mM Tris buffer. A 50 L of 7 M of diethylthiatricarbocyanine iodide in methanol was added into the vial followed by vigorous mixing on the Vortex for several seconds. Afte r five minutes, both suspensions were thoroughly rinsed with buffer solution and transferred to tw o vials. The polymer pieces were rinsed with buffer three times before 5 L of cocaine solution (1 mg/mL acetonitrile) was added into both vials with 500 L Tris buffer. The solution in the vial with non-modified RF polymer pieces did not change color, while the solution in the vial with the cocaine retaining-aptamer modified RF polymer changed from colorless to blue. The co lor of the solution faded several hours later due to the hydrolysis of the dye molecules in the slightly basic buffer.51 MALDI-MS spectra of the cocaine-retaini ng-aptamer modified RF polymer and unmodified RF polymer were taken on the LTQ w ith vMALDI source. Three pieces of cocaine retaining-aptamer modified RF polymer and th ree pieces of non-modified RF polymer were immersed in 500 L of 20 mM Tris buffer for the aptamers to form the affinity capturing configuration. Then 5 L of cocaine solution (1 mg/mL acet onitrile, 3mM) was added into the vials, waited for five minutes before they were thoroughly rinsed with 20 mM Tris buffer solution. The polymer pieces were transferred to another two vials to be rinsed three times before they were taken out and inspecte d under the microscope (Figure 3-17). The polymers were taped onto the sample slide with double-sided tape and inserted into the instrument. Sinapinic acid (SA) 2,5-dihydroxybenzoic acid (DHB), and -cyano-4-

PAGE 98

98 hydroxycinnamic acid (CHCA) were used as MALDI matrices on three of the modified polymer respectively and on three non-modified respectivel y. The cocaine retain ing-aptamer modified RF polymer showed cocaine signal (the [M+H]+ ion at m/z 304) when SA was used as the matrix, meanwhile there was no cocaine signal when DHB or CHCA was used as matrices, which is consistent with the results McGown observed with SA for her aptamer-captured thrombin sample.73 The polymer without the cocaine reta ining aptamer (Figure 3-18a) showed no cocaine ( m/z 304) compared to the cocaine retaini ng aptamer attached polymer (Figure 318b). The 2-D and 3-D images of the intensity of the m/z at 304 ions of cocaine on the RF polymer is shown in Figure 3-19. An indirect cocaine retainingaptamer modification of the RF polymer was also conducted for the in Weihong Tans lab after the polymer pellets were carboxylated in our lab. Three pieces of cocaine retaining-aptamer modi fied RF polymer were immersed in 500 L of 20 mM Tris buffer for the aptamers to form the affinity capturing co nfiguration. Then 5 L of cocaine solution (1 mg/mL acetonitrile, 3mM) was added into the vial, waited for five minutes before they were thoroughly rinsed with 20 mM Tris buffer solution. The polymer pieces were transferred to another vial to be rinsed three times before they were taken out and examed under the microscope (Figure 3-20). The polymer pell ets were taped onto the sample slide with double sided tape and insert into the instrument. Sina pinic acid (SA) was used on the polymer as the matrix. The cocaine retaining-aptamer modi fied RF polymer showed cocaine signal ( m/z 304) with SA as matrix (Figure 3-21). The 2-D a nd 3-D image of the intensity of the cocaine ( m/z 304) on the RF polymer is shown in Figure 3-22. The amount of the cocaine detected on the aptamer-indirect modified polymer by the MS is substantially higher than that on the direct

PAGE 99

99 modified polymer (more spots with cocaine) due to the expected higher reac tion efficiency of the indirect modification. Conclusion With the phenol groups on the RF polymer, it is relatively straight-forward to follow the established procedure to attach aptamer directly onto the RF polymer to surface modify it. The up-take of reactive green 19 and cocaine proved the exis tence of the aptamers on the RF polymer and the affinity capture capability of the aptamers. With the phenol groups on the RF polymer, it is easy to follow the established procedure to attach carboxylic acid groups onto the RF polymer, then react with amine modified aptamers to surface modify the RF polymer with aptamers. The carboxylation of the RF polymer was proved by the detection of the up taki ng of toluidine blue O and PDAM The amount of the cocaine detected on the aptamer indirect modified polymer by the MS is substantially higher than that on the direct modified polymer due to the expect ed higher reaction effici ency of the indirect modification. With the procedure of attaching the aptame rs onto the RF polymer established in the research project, other disease biomarker retain ing-aptamers could be attached onto the RF polymer following the same procedure, which can be used for pre-analysis clean-up for MALDIMS analysis for disease screening.

PAGE 100

100 OH OH CH2 CH2 CH2 OH O CH2 CH2 OH OH O CH2 OH OH CH2 CH2OH OH OH OH OH CH2 O CH2 OH OH OH CH2OH OH Figure 3-1. The structure of the RF polymer.

PAGE 101

101 Figure 3-2. Chemical structure of reactive green 19. [Adapted from McGettrick, A.F.; Worrall, D.M. Protein purificatio n protocols, 2nd edition Methods in molecular biology vol 244, p 151 Edited by Cutler, P. Humana Press Inc., Totowa, NJ.]

PAGE 102

102 Figure 3-3. The 27-mer oligodeoxy ribonucleotide sequence, from a to b is designed the consensus sequence. [Adapted from Yilmaz, M; Bayramo lu, G.; Arica, M.Y. Separation and purification of lysozyme by Reactive Green 19 immobilised membrane affinity chromatography Food Chemistry 2005 89, 11]

PAGE 103

103 Figure 3-4. The two cocaine isomers: a) pseudococaine and b) cocaine. a) b)

PAGE 104

104 Figure 3-5. Anti-cocaine aptamer MNS-4.1 bound to cocaine 1 (black elipsoid). The bold region was cut out to generate MNS-7.9 Also shown are cocaine metabolites benzoyl-ecgonine 2 and ecgonine methyl ester 3 [Adapted from Hermann, T.; Patel, D.J. Biochemistry Adaptive r ecognition by nuclei c acid aptamers Science 2000 287 820]

PAGE 105

105 Figure 3-6. The picture of a App lied Biosystems 3400 DNA Synthesizer.

PAGE 106

106 OH OH DETA O O Si H H Si O N H H N N H H N NH2 NH2 Succinic Anhydride DMF/Ar H N H N C C O O COOH COOH AptamerH2N + C C O O H N N H Aptamer Aptamer Figure 3-7. Scheme of the indirect surf ace modification of the RF polymer via the carboxylation.

PAGE 107

107 CHN2 Figure 3-8. Chemical structure of PDAM.

PAGE 108

108 0 10000 20000 30000 40000 50000 60000 70000 80000 365415465515565615 Wavelength (nm)Fluorescent emission Sample Control Figure 3-9. Fluorescent emission intensity of PDAM-bond (blue) and noncarboxylated (red) RF polymer.

PAGE 109

109 S N N CH3 NH2 H3C CH3 Figure 3-10. Structure of Toluidine blue O.

PAGE 110

110 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 250300350400450500550600650700 Wavelength (nm)Abs 623 nm 285 nm Sample Control Figure 3-11. UV absorbance of the toluidine bl ue O desorbed from th e carboxylated (red) and noncarboxylated (blue) RF polymer.

PAGE 111

111 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 250300350400450500550600650700 Wavelength (nm)Abs 285 nm 623 nm Figure 3-12. UV of the toluidine blue O so lutions of different concentrations (1.0 M, 2.0 M, 3.0 M, 4.0 M, 5.5 M, 7.3 M, 9.1 M, 14.8 M).

PAGE 112

112 y = 0.0319x + 0.0374 R2 = 0.9913 0 0.1 0.2 0.3 0.4 0.5 0.6 0.002.004.006.008.0010.0012.0014.0016.00 Concentration ( M)Abs Figure 3-13. Calibration curve of the concentrati on of the toluidine blue O solution at 285 nm. The imprecision of the concentrations of the toluidine blue O solutions caused the calibration curve not to pass the (0, 0) point.

PAGE 113

113 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 300350400450500550600650700750800 Wavelength (nm)Abs Sample Control 634 nm Figure 3-14. UV absorbance of reactive green 19 desorbed from aptamer attached RF polymer (blue) and RF (red) polymer.

PAGE 114

114 y = 0.00061x + 0.00187 R2 = 0.97734 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 020406080100120 Concentration (uM)Abs Figure 3-15. Calibration curve of the concentra tion of the reactive green 19 solution at 634 nm. The imprecision of the concentrations of the toluidine blue O solutions caused the calibration curve not to pass the (0, 0) point.

PAGE 115

115 Figure 3-16. Cocaine (1) displaces diethylthiatr icarbocyanine iodide complexed with aptamer MNS-4.1, causing an immediate attenuation of absorbance and eventual precipitation of dye. [Adapted from Stojanovic, M.N. ; Landry, D.W. Aptamer-Based Colorimeter probe for cocaine Journal of the American Chemical Society 2002 124 9678]

PAGE 116

116 Figure 3-17. A microscopic image of the indire ctly cocaine retaining-aptamer modified RF polymer.

PAGE 117

117 Figure 3-18. MS spectrum of the cocaine on A) the unmodified RF polymer. B) on the cocaine retaining-aptamer attached RF polymer. Cocaine soaked RF (control)_TI # 1-379 RT: 0.00-5.10 AV: 379 NL: 3.62E2 T: ITMS + c MALDI Full ms [150.00-2000.00] 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 600 m/z 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100Relative Abundance 190.31 172.33 417.24 240.35 379.16 212.29 359.21 180.29 224.32 168.30 550.72 388.24 312.42 522.70 268.43 254.39 204.32 284.43 575.29 296.61 361.28 447.35 324.30 391.25 458.35 427.33 164.41 335.38 480.48 567.33 540.56 586.58 494.10 Cocaine soaked aptamer attached RF_SA # 1-400 RT: 0.00-5.38 AV: 400 NL: 1.34E5 T: ITMS + c MALDI Full ms [150.00-2000.00] 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 600 m/z 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100Relative Abundance 304.39 177.36 199.33 154.41 290.47 551.03 172.34 375.18 182.51 361.44 312.35 273.38 326.34 573.07 397.18 245.38 449.53 223.38 427.34 589.07 257.37 353.15 511.14537.13 488.21 465.31 B Cocaine [M+H]+ CHCA [ M+H ] + CHCA [M+H]+ H2O A

PAGE 118

118 Figure 3-19. The 2-D and 3-D image of the cocaine ( m/z at 304) signal intensity across the indirect cocaine retaining-aptamer modified RF polymer. A) 2-D image. B) 3-D image. X ( micron ) X ( micron ) 0 0 500 500 1000 1000 1500 1500 2000 2000 d anc e d anc e 2.000e+007 2.000e+007 1.000e+007 1.000e+007 0.000e+000 0.000e+000 Y ( micron ) Y ( micron ) 2000 2000 1500 1500 1000 1000 500 500 0 0 Intensit y A B

PAGE 119

119 Figure 3-20. The image of the di rect cocaine retaining-aptamer m odified RF polymer (3 mm in diameter).

PAGE 120

120 Figure 3-21. MS spectrum of the cocaine on the direct cocaine retaining-aptamer modified RF polymer. Coacine_aptamer_09_25_2006_TI # 899 RT: 12.71 AV: 1 NL: 2.77E6 T: ITMS + c MALDI Full ms [50.00-1000.00] 50 100 150 200 250 300 350 400 450 500 550 600 m/z 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100Relative Abundance 207.57 304.50 225.57 122.70 471.37 387.50 247.55 431.39 328.48 449.44 493.27 401.44 515.28 269.57 358.47 179.64 537.23 473.40 295.32 152.64 263.51 118.75 559.25 595.28cocaine [M+H]+ sinapinic acid [M+H]+ sinapinic acid [M+H]+ H2O sinapinic acid [M+H]+ C4H6O3 sinapinic acid [M+Na]+

PAGE 121

121 Figure 3-22. The 2-D and 3-D image of the cocai ne signal intensity across the direct cocaine retaining-aptamer modified RF polymer A) 2-D image. B) 3-D image. X ( micron ) X ( micron ) 0 0 500 500 1000 1000 1500 1500 2000 2000 2500 2500 3000 3000 350 0 3500 d anc e d ance 1.500e+006 1.500e+006 1.000e+006 1.000e+006 5.000e+005 5.000e+005 0.000e+000 0.000e+000 Y ( micron ) Y ( micron ) 2500 2500 2000 2000 1500 1500 1000 1000 500 500 0 0 Intensit y A B

PAGE 122

122 CHAPTER 4 CONCLUSION AND FUTURE WORK As a relatively new technology developed for th e analysis of biomolecu les, MALDI is very powerful; nevertheless, there is problems associat ed with the sample preparation methods and still room for the improvement of the spect ral quality. The problems include the background noise from the matrix molecules which are used to assist in the ionizati on of the analytes in MALDI, and the uneven distribution of the an alyte signal across the sample due to the inhomogeneous morphology of the matrix-analyte co -crystals associated with the traditional MADLI sample preparation methods. The complex ity of biological samples also poses problem for the detection of the analyte of interest wi th low concentration. To address the problems facing MALDI-MS, an RF polymer and matrix-e mbedded RF polymers were synthesized and a direct and an indirect surface modification of the RF polymer with aptamers were developed. An RF polymer and matrix-embedded RF pol ymers were synthesized via the sol-gel method. With the 1, 4 polymerization position on the aromatic ring, RF polymers can withstand the washing with aqueous solution, which make th em suitable for analysis of biological sample and harsh condition of the surface modification wi th functional groups and aptamers; thus, they are a perfect substrate for MALDI and SELDI. With the aromatic groups in the RF polymers and the reaction sites on the aromatic rings, it is possible to incorporate aromatic group-contai ning matrix molecules in the polymeric network with covalent bonds. Being trapped by the polym eric network with covalent bonds, the matrix molecules still provided the assistance for ioni zation, but dramatically reduced the amount of matrix molecule ablated and ionized by the lase r due to the higher energy required to break the covalent bond than intermolecu lar interaction such as hydroge n bond, dipole-dipole interaction,

PAGE 123

123 and van der Waals force. With much less matr ix released and ionized, there is much less interference from matrix molecules in sp ectrum for low molecular weight analytes. By incorporating the matrix molecules into the polymeric network structure, the problem of unevenly distributed signal across the sample due to the inhomogeneous morphology of the matrix-analyte co-crystals was solved. Many mo difications of the traditional MALDI sample preparation methods have been developed to minimize the problem. Those modifications can more or less improve the evenness of the distri bution of the analyte signa l across the sample, but cannot fundamentally solve this problem. Th e new matrix-embedded resorcinol-formaldehyde (RF) polymers developed have th e matrix molecules distributed in the polymer evenly at the molecular level, and trap the matrix molecules inside the polymer structure, which makes the formation of the matrix-analyte co-crystals unn ecessary; thus, the eve nness of the signal distribution only depends on the ev enness of the analyte distributi on. By using the most widely used and most problematic MALDI sample prep aration method, dried-droplet, for both analytematrix solution on the stainless steel and anal yte solution on the CHCA-embedded RF polymer, the reproducibility of analyte signal distributi on was improved from 228% to 103% in RSD. This property also provides the possibility for qu antization of the analyte in MALDI, which is to be evaluated in future. With the phenol groups on the RF polymer, it is easy to follow established procedures to attach aptamer directly onto the RF polymer in a nucleic acid synthesizer. The aptamers attached onto the RF polymer are the reactive green 19retaining aptamer and the cocaine-retaining aptamer. The uptake of reactive 19 by the ap tamer modified RF polymer was detected by UV vis spectrometry for the reactive green 19 deso rbed from the reactive green 19-absorbed RF polymer. The amount of the reactive green 19 captured on the modified RF polymer was

PAGE 124

124 determined according to the calibration curve co nstructed with the UV absorbance of toluidine blue O solutions with different concentrations. The uptake of cocaine by the aptamer-modified RF polymer was visually detected by re placing the aptamer affinity captured diethylthiatricarbocyanine iodide with cocaine. The cocaine captured RF polymer was analyzed in the LTQ MS with the dete ction of the cocaine signal ( m/z 304). With the phenol groups on the RF polymer, it is also easy to follow established procedures to attach carboxylic acid groups onto the RF pol ymer. The carboxylation of the RF polymer was proved by the UVvis detection of the uptake of toluidine blue O and PDAM. The amount of the carboxylic acid groups on the RF polymer was determined by the amount of the toluidine blue O desorbed from the toluidine blue O-absorbed RF polymer according to the calibration curve constructed with the UV absorbance of tolu idine blue O with different concentrations. After the RF polymer was carboxylated, it was react ed with amine-modified cocaine retainingaptamers to form the aptamer-modified RF polymer via the formation of an amide bond. The cocaine captured RF polymer was analyzed in th e LTQ MS with the detection of the cocaine signal ( m/z 304) as done with the direct cocaine retaining-aptamer modi fied polymer. The amount of the cocaine detected on the aptame r indirect modified polymer by the MS is substantially higher than that on the direct modified polymer due to the expected higher reaction efficiency of the indirect modification. The research conducted in this project has pr oved that the RF polymer is a versatile substrate that can both be derived with matrix molecules to provide more even analyte signal distribution across the sample a nd be surface modified with highly selective affinity binding aptamers for affinity capture prior to MS anal ysis. To fully asses the potential of the RF polymer, more research should be performed bot h on providing more even signal distribution

PAGE 125

125 across the sample with matrix-embedded RF polym er and affinity capturing capability of the aptamer-modified polymer towards cocaine and expanding the the ranges of aptamers can be attached for affinity capture. Other sample preparation methods such as airb rush or electrospray can be used to produce more even distribution of the analyte on the RF polymer to obtain more improved results and asses the detection limit of the analyte on the CHCA-embedded RF polymer. The assessment of the selectivity of the cocaine aptamer towards co caine can be performed by affinity capture of cocaine from a mixture of cocaine and its metabo lites solution and from biological samples such as blood and urine from a cocaine user. The lower limit of the affinity capturing ability of the specific cocaine-retaining aptamer also needs to be tested, since the concentration of cocaine solution employed here for affinity capturing is 30 ppm. The concentration of unmetabolized cocaine extracted in urine is mu ch lower; indeed, the concentrat ion of extracted cocaine is one hundredth of the concentration of the cocaine metabolites (1 ppm).141 The affinity capturing of the cocaine with the aptamer used in this expe riment may not possess the detection limit needed for the analysis of urine samples from individu als using cocaine because of the inherent high dissociation constant (0.4-10 M) between the aptamer and cocaine. The dissociation constant between an aptamer and the target should be around 10-15 M.143 A different cocaine aptamer with lower dissociation constant should be s ynthesized by SELEX proce ss for higher selectivity and then be attached onto the RF polymer followi ng the procedure established in this research. Aptamers with high affinity toward disease biomarkers can also be synthesized by SELEX process and be attached onto the RF polymer following the same indirect modification procedure. Then these disease biomarker-retaining ap tamer modified RF polymers can be used

PAGE 126

126 to affinity capture the disease biomarkers fr om biological samples taken from individuals for pre-analysis clean-up for MALDI-MS analysis for disease screening.

PAGE 127

127 LIST OF REFERENCES 1. Wikipedia, the Free Encyclopedia [Internet]. Wikimedia Foundation, Inc. (US); [updated 2006 Dec 7; cited 2006 December 8]. Available from http://en.wikipedia.org/w iki/Analytical_chemistry 2. Vlahou, A.; Schellhammer, P.F.; Mendrinos, S.; Patel, K.; Kondylis, F.I.; Gong L.; Nasim, S.; Wright, G.L. Jr. Development of a novel proteomic approach for the detection of transitional cell carcinoma of the bladder in urine The American Journal of Pathology 2001 158, 1491-1502 3. Watson J.T. Introduction to Mass Spectrome try. Third Edition. Lippincott-Raven: Philadelphia, NY, 1997 4. Burlingame, A.L.; Boyd, R.K.; Gaskell, S.J. Mass spectrometry. Analytical Chemistry 1996 68 599R 5. Aston, F.W. Mass Spectra and Isotopes. Edition 3. Edward Arnold and Co: London, 1942 6. Biemann, K. Mass spectrometry: Applications to organic chemistry. McGraw-Hill: New York, 1962 7. Paul, W.; Steinwedel, H: Patent 2,939, 952 Germany, 1960 8. Drug Discovery and Development [Internet]. Advantage Business Media (US); [updated 2003 Sep 29; cited 2006 Dec 8]. Ava ilable from: http://www.dddmag.com 9. The Nobel Prizes in Physics 1989 NobelPri ze.org [Internet]. Nobel Prize Foundation (Sweden); [updated 1989 Oct 18; cited 2006 Dec 8]. Available from http://nobelprize.or g/nobel_prizes/physi cs/laureates/1989/ 10. March, R.E. An introduction to quad rupole ion trap mass spectrometry Journal of Mass Spectrometry 1997 32, 351 11. (Reddick, C.D. The detection of pharmaceuti cal drug compounds from intact biological tissue by matrix-assisted laser desorpti on ionization method (MALDI) quadrupole ion trap mass spectrometry 1997 PhD dissertation) 12. (Schwartz, J.C.; Senko, M.W. A two-dimensi onal quadrupole ion trap mass spectrometer Journal of the American Society of Mass Spectrometr y 2002 13 659) 13. Electron Ionization (EI) Anal ysis [Internet]. The Samuel Roberts Noble Foundation (US); [updated 2006 Nov 27; cited 2006 Dec 8]. Available from http://www.noble.org/plantb io/MS/iontech.ei.html 14. Chemical Ionization (CI) Analysis [Intern et]. The Samuel Roberts Noble Foundation (US); [updated 2006 Nov 27; cited 2006 Dec 8]. Available from http://www.noble.org/plantb io/MS/iontech.ci.html

PAGE 128

128 15. Mass Spectrometry Source [Internet]. Univer sity of Bristol, School of Chemistry (England); [updated 2004 Jan 27; cited 2006 Dec 8]. Available from http://www.chm.bris.ac.uk/ms/theory/fab-ionisation.html 16. Mass Spectrometry Source [Internet]. Univer sity of Bristol, School of Chemistry (England); [updated 2004 Jan 27; cited 2006 Dec 8]. Available from http://www.chm.bris.ac.uk/ms/theory/esi-ionisation.html 17. The Nobel Prizes in Chemistry 2002 NobelPri ze.org [Internet]. Nobel Prize Foundation (Sweden); [updated 2002 Oct 9; ci ted 2006 Dec 8]. Available from http://nobelprize.org/chemistr y/laureates/2002/press.html 18. Salih, B.; Masselon, C.; Zenobi, R. Matrix-assisted la ser desorption/ionization mass spectrometry of noncovalent protei n transition metal ion complexes Journal of Mass Spectrometry 1998, 3,3 994 19. Nelson, R.W.; Hutchens, T.W. Mass-spectrometric anal ysis of a transition-metal-binding peptide using matrix-assisted laser desorption time-of-flight mass-spectrometrya demonstration of probe tip chemistry Rapid Communications in Mass Spectrometry 1992, 6, 4 20. Woods, A.S.; Buchsbaum, J.C.; Worrall T.A.; Berg, J.M.; Cotter, R.J. Matrix-assisted laser desorption/ionization of noncovalent bound compounds Analytical Chemistry 1995, 67 4462 21. Lehmann, E.; Zenobi, R. Detection of specific noncovalent zinc finger peptideoligodeoxynucleotide complexes by matrix-a ssisted laser desorption/ionization mass spectrometry Angewandte Chemie-International Edition 1998, 37, 3430 22. Masselon, C.; Salih, B.; Zenobi, R. Matrix-assisted lase r desorption/ionization Fourier transform mass spectrometry of luteiniz ing hormone releasing hormone-metal ion complexes Journal of the American Soc iety for Mass Spectrometry 1999, 10, 19 23. Lehmann, E.; Zenobi, R.; Vetter, S. Matrix-assisted laser desorption/ionization mass spectra reflect solution-phase zi nc finger peptide complexation Journal of the American Society for Mass Spectrometry 1999, 10, 27 24. Lin, S.H.; Long, S.R.; Ramirez, S.M.; Cotter, R.J.; Woods, A.S. Characterization of the "helix clamp" motif of HIV-1 reverse tran scriptase using MALDI-TOF MS and surface plasmon resonance Analytical Chemistry 2000, 72, 2635 25. Woods, A.; Zangen, A. A direct chemical intera ction between dynorphi n and excitatory amino acids Neurochemical Research 2001, 26, 395 26. Woods, A.S.; Moyer, S.C.; Wang, H.Y.J,.; Wise, R.A. Interaction of chlorisondamine with the neuronal nicotinic acetylcholine receptor Journal of Proteome Research 2003, 2, 207

PAGE 129

129 27. Glocker, M.O.; Bauer, S.H.J.; Kast, J.; Volz, J.; Przybylski, M. Characterization of specific noncovalent protein complexes by UV matrix-assisted laser desorption ionization mass spectrometry Journal of Mass Spectrometry 1996, 31, 1221 28. Woods, A.S.; Huestis, M.A. A study of peptide-pep tide interaction by matrix-assisted laser desorption/ionization Journal of the American Society for Mass Spectrometry 2001, 12, 88 29. Woods, A.S.; Koomen, J.M.; Ruotolo, B.T.; Gillig, K.J. Russell, D.H.; Fuhrer, K.; Gonin, M.; Egan, T.F.; Schultz, J.A. A study of peptide-pe ptide using MALDI ion mobility oTOF and ESI mass spectrometry Journal of the American Socie ty for Mass Spectrometry 2002, 13, 166 30. Moyer, S.C.; Marzilli, L.A.; Woods, A.S.; Laiko, V.V.; Doroshenko, V.M.; Cotter, R.J. Atmospheric pressure matrix-assisted la ser desorption/ionization (AP MALDI) on a quadrupole ion trap mass spectrometer International Journal of Mass Spectrometry 2003, 226, 133 31. Karas, M.; Hillenkamp, F. Influence of the wa velength in high-irradiance ultraviolet laser desorption mass spectrometry of organic molecules Analytical Chemistry 1985 57 2935 32. Karas, M.; Bachmann, D.; Bahr, U.; Hillenka mp, F. Matrix-assisted ultraviolet-laser desorption of nonvolatile compounds International Journal of Mass Spectrometry and Ion Processes 1987 78 53 33. Karas, M; Hillenkamp, F Laser desorption ionization of proteins with molecular masses exceeding 10000 daltons Analytical Chemistry 1988 60 2299 34. Tanaka, K.; Waki, H.; Ido, Y.; Akita, S.; Yosh ida, Y.; Yoshida, T. Protein and polymer analyses up to m/z 100,000 by laser ionizat ion time-of-flight mass spectrometry Rapid Communications in Mass Spectrometry 1988 2 ; 151 35. Wei, J.; Buriak, J.M.; Siuzdak, G. Desorption-ionization mass spectrometry on porous silicon Nature 1999 399 243) (Kruse, R.A.; Li, X.; Bohn, P.W.; Sweedler, J.V. Experimental factors control ling analyte ion generation in laser desorption/ionization mass spectrometry on porous silicon Analytical Chemistry 2001 73 3639 36. Thomas, J.J.; Shen, Z.; Blackledge, R.; Siuz dak, G. Desorption-ioni zation on silicon mass spectrometry: an application in forensics Analytica Chimica Acta 2001 442 183 37. Shen, Z.; Thomas, J.J.; Averbuj, C.; Broo, K. M.; Engelhard, M.; Crowell, J.E.; Finn, M.G.; Siuzdak, G. Porous silicon as a versat ile platform for lase r desorption/ionization mass spectrometry Analytical Chemistry 2001 73 612 38. Bhattacharya, S.; Raiford, T.; Murray, K.K. Infrared laser desorption/ionization on silicon Analytical Chemistry 2002 74 2228

PAGE 130

130 39. Diamandis, E.P.; Yousel, G.M. Human tissue kallikrein gene family: a rich source of novel disease biomarkers Future Drugs 2001 1 182 40. Merchchant, M.; Weinberger, S.R. Recent advancements in surface-enhanced laser desorption/ionization-time if flight-mass spectrometry Electrophoresis 2000 21 1164 41. Hutchens, T.W.; Yip, T.T. New desorption stra tegies for the mass-spectrometric analysis of macromolecules Rapid Communication in Mass Spectrometry 1993 7 576 42. Tang, N.; Tornatore, P.; Weinberger, S.R. Current developments in SELDI affinity technology Mass Spectrometry Reviews 2004 23, 34 43. Hermann, T.; Patel, D.J. Biochemistry Adaptive recognition by nucleic acid aptamers Science 2000 287 820 44. Jayasena, S.D. Aptamers: an emerging class of molecules that rival antibodies in diagnostics Clinical Chemistry 1999 45 1628 45. Stojanovic, M.N.; Prada, P., Landry, D.W. Aptamer-based Folding Fluorescent Sensor for Cocaine Journal of the American Chemical Society 2001 123 4928 46. McGown, L.B.; Joseph, M.J. The Nucleic Acid Ligand: A New Tool for Molecular Recognition Analytical Chemistry 1995 67 665A 47. Biroccio, A.; Hamm, J.; Incitti, I.; Francesco, R.D.; Tomei, L. Selection of RNA aptamers that are specific and high-affinity ligands of the hepatitis C virus RNA-dependent RNA polymerase Journal of Virology 2002 76 3688 48. Gan, H.H.; Pasquali, S.; Schlick, T. Explori ng the repertoire of RNA secondary motifs using graph theory; implications for RNA design Nucleic Acids Research 2003 31 2926 49. Tuerk, C.; MacDougal, S.; Gold, L. RNA pseudoknots that inhibit hunmanimmunodefficiency-virus type -1 reverse-transcriptase Proceedings of the National Academy of Sciences of th e United States of America 1992, 89 6988 50. Wang, K.Y.; McCurdy, S.; Shea, R.G.; Swaminathan, S.; Bolton, P.H. A DNA aptamer which binds to and inhibits thrombin e xhibits a new structural motif for DNA Biochemistry 1993 32 1899 51. Mukhopadhyay, R. Aptamers are ready for the spotlight Analytical Chemistry 2005 77 115A 52. Tuerk, C.; Gold, L. Systematic evolution of ligands by exponential enrichment RNA ligands to bacteriophage-T4 DNA-polymerase Science 1990 249 505 53. Stojanovic, M.N.; Landry, D.W. Aptamer-B ased Colorimeter probe for cocaine Journal of the American Chemical Society 2002 124 9678

PAGE 131

131 54. Ellington, A.D.; Szostak, J.W. In vitro sele ction of RNA molecules that bind specific ligands Nature 1990 346 818 55. Klug, S.J.; Famulok, M. All you wanted to know about SELEX Molecular Biology Reports 1994 20 97 56. Self, C.H.; Dessi, J.L.; Winger, L.A. Highperformance assays of small molecules: enhanced sensitivity, rapidity, and conve nience demonstrated with a noncompetitive immunometric ant-immune comp lex assay system for digoxin Clinical Chemistry 1994 40 2035 57. Cox, J.C., Rudolph, P.; Ellington, A.D. Automated RNA selection Biotechnology Progress 1998 14, 845 58. Tuerk, C.; Eddy, S.; Parma, D.; Gold, L. Aut ogenous translational operator recognized by bacteriophage T4 DNA polymerase Journal of Molecular Biology 1990 213 749 59. Jellinek, D.; Green, L.S.; Bell, C.; Ly nott, C.K.; Gill, N.; Vargeese, C.; Kirschenheuter, G.; McGee, D.P.C.; Abesinghe, P.; Pieken, W.A.; Shapiro, R.; Rifkin, D.B.; Moscatelli, D.; Janjic, N. Potent 2-amino-2 deoxypyrimid ine RNA inhibitors of basic fibroblast growth factor Biochemistry 1995 34 11363 60. Tasset, D.M.; Kublk, M.F.; Steiner, W. Oli gonucleotide inhibitors of human thrombin that bind distinct epitopes Journal of Molecular Biology 1997 272 688 61. Jenison, R.D.; Jennings, S.D.; Walker, D.W.; Bargatze R.F.; Parma D. Oligonucleotide inhibitors of P-selectin-depe ndent neutrophilplatelet adhesion Antisense Nucleic acid Drug Development 1998 8 265 62. Kublk M.F.; Stephens, A.W.; Schneider, D.; Marlar, R.A.; Tasset, D. High-affinity RNA ligands to human -thrombin Nucleic Acid Research 1994 22 2619 63. Lin, Y.; Padmaprlya, A.; Morden, K.M.; Jayasena, S.D. Peptide conjugation to an in vitro-selected DNA ligand improves enzyme inhibition Proceedings of the National Academy of Sciences of th e United States of America 1995, 92 11044 64. Jenison, R.D.; Gill, S.C.; Pardi, A.; Polisky, B. High-resolution molecular discrimination by RNA Science 1994 263, 1425 65. Haller, A.A.; Sarnow, P. In vitro selecti on of a 7-methyl-guanosine binding RNA that inhibits translation of capped mRNA molecules Proceedings of the National Academy of Sciences of the United States of America 1997 94, 8521 66. Sassanfar, M.; Szostak, J.W. An RNA motif that binds ATP Nature 1993 364 550 67. Mannironi, C.; Nardo, A.D.; Fruscoloni, P.; Tocch ini-Valentini, G.P. In vitro selection of dopamine RNA ligands Biochemistry 1997 36, 9726

PAGE 132

132 68. Geiger, A.; Burgstaller, P.; von der Eltz, H.; Roeder, A.; Famulok, M. RNA aptamers that bind L-arginine with sub-micromolar dissociation constants and high enantioselectivity Nucleic Acids Research 1996 24 1029 69. Clark, P.P.; Price, C.P. Removal of inte rference by immunoglobulins in an enzymeamplified immunoassay for thyrotropin in serum Clinical Chemistry 1987 33 414 70. Cook, D.B.; Self, C.H. Monocl onal antibodies in diagnosti c immunoassays. In: Ritter MA, Ladyman HM eds. Monoclonal antibodie s. Cambridge, UK, Cambridge University press, 1995, 180-208 71. Savran, C.A.; Knudsen, S.M.; Ellington, A.D.; Manalis, S.R. Micromechanical detection of proteins using aptamer-based receptor molecules Analytical Chemistry 2004 76 3194 72. Deng, Q.; German, I.; Buchanan, D.; Kennedy, R.T. Retention and separation of adenosine and analogues by affinity chromat ography with an aptamer stationary phase Analytical Chemistry 2001 73 5415 73. Dick, L.W.; McGown, L.B. Aptamer-enhanced la ser desorption/ ionization for affinity mass spectrometry Analytical Chemistry 2004 76 3037 74. Ueda, E.K.M; Gout, P.W.; Morganti, L. Curre nt and prospective applications of metal ion-protein binding Journal of chromatography A 2003 988, 1 75. Lin, Y.; Chen, Y. Laser desorption/ionizati on time-of-flight mass spectrometry on solgel-derived 2,5-dihydroxybenzoic acid film Analytical Chemistry 2002 74 5793 76. Vertes, A.; Gijbels, R. Laser Ionization Mass Analysis; Vertes, A.; Gijbels, R.; Adams F.; John Wiley & Sons: New York, 1993 p127 77. Stahl, B.; Steup, M.; Karas, M.; Hillenkamp, F. Analysis of neutral oligosacchrides by matrix-asssited laser desorption-ionization mass-spectrometry Analytical Chemistry, 1991 63 1463 78. Schar, M.; Bornsen, K.; Gassmann, E. Fa st protein-sequence determination and ionization mass-spectrometry Rapid Communications in Mass Spectrometry 1991 5 319 79. Wu, K.J.; Steding, A.; Becker, C.H. Matrix -assisted laser desorp tion time-of-flight massspectrometry of oligonucleotides using 3-hydroxypicolinic acid as an ultraviolet-sensitive matrix Rapid Communications in Mass Spectrometry 1993 7 142 80. Chan, T.W.D.; Colburn, A.W.; Derrick, P. J. Matrix-assisted UV laser desorptionsuppression of the matrix peaks Organic Mass Spectrometry 1991 26 342 81. Chan, T.W.D.; Colburn, A.W.; Derrick, P.J.; Ga rdiner, D.J.; Bowden, M. Suppression of matrix ions in ultraviolet-laser desorpti on scanning electronmicroscopy and RAMANspectroscopy of the solid samples Organic Mass Spectrometry 1992 27 188

PAGE 133

133 82. Juhasz, P.; Wang, B.H.; Biemann, K. Experime ntal studies on the mechanism of the matrix-assisted laser desorption Proceedings of the 40th ASMS Conference on Mass Spectrometry and Allied Topics 1992 p372 83. Knochenmuss, R.; Dubois, D.M. The ma trix suppression effect and ionization mechanisms in matrix-assisted laser desorption/ionization Rapid communications in Mass Spectrometry 1996 10 871 84. Bashir, S.; Burkitt, W.I.; Derrick, P.J.; Gi annakopulos, A.E. Iodine-assisted matrixassisted laser desorption/ionization International Journal of Mass Spectrometry 2002 219 697 85. Ryuichi, A.; Naoki, M.; Shoji, O.; Hiroshi, Y. Signal enhancement on laser desorption/ionization using alkali dihydroxybe nzoic acid salts as cationizing agents Rapid Communications in Mass Spectrometry 2006 20 2063 86. Sunner, J.; Dratz, E.; Chen, Y.C. Graphite surface assisted lase r desorption/ionization time-of-flight mass-spectrometry of peptid es and proteins from liquid solutions Analytical Chemistry 1995 67 4335 87. Dale, M.; Knochenmuss, R.; Zenobi, R. Graphite/liquid mixed matrices for laser desorption/ionization mass spectrometry Analytical Chemistry 1996 68 3321 88. Jackson, S.N.; Dutta, S.M.; Murray, K.K. A nitrocellulose matrix for infrared matrixassisted laser desorption/ionization of polycyclic aromatic hydrocarbons Rapid Communications in Mass Spectrometry 2004 18 228 89. Dai, Y.Q.; Whittal, R.M.; Li, L. Two-layer sample preparation: A method for MALDIMS analysis of complex pe ptide and protein mixtures Analytical Chemistry 1999 71 1087 90. Dai, Y.Q.; Whittal, R.M.; Li, L. Confocal fluorescence microscopic imaging for investigating the analyte dist ribution in MALDI matrices Analytical Chemistry 1996 68 2494 91. Kampmeier, J.; Dreisewerd, K.; Sch renberg, M.; Strupat, K. Investigation of 2,5-DHB and succinic acid as matrices for IR and UV MA LDI. Part I: UV and IR laser ablation in the MALDI process International Journal of Mass Spectrometry and Ion Process 1997 169/170 31 92. Vorm, O.; ORM O, Mann, M. Improved mass accuracy in matrix-a ssisted laser desorption/ionization time-of-flight mass-spectrometry of peptides Journal of the American Society for Mass Spectrometry 1994 5 955 93. Hung, K.C.;, Ding, H.; Guo, B.C. Use of poly(tetrafluor oethylene)s as a sample support for the MALDI-TOF analysis of DNA and proteins Analytical Chemistry 1999 71 518

PAGE 134

134 94. Gusev, A.I.; Wilkinson, W.R.; Proctor, A.; Hercules, D.M.; Improvement of signal reproducibility and matrix/c omatrix effects in MALDI Analytical Chemistry 1995 67 1034 95. Sadeghi, M.; Vertes, A. Crystallite size depe ndence of volatilizati on in matrix-assisted laser desorption ionization Applied Surface Science 1998 127-129 226 96. Allwood, D.A,.; Perera, I.K.; Perkins, J.; Dyer, P.E.; Oldershaw, G.A. Preparation of 'near' homogeneous samples for the analysis of matrix-assisted laser desorption ionization processes Applied Surface Science 1996 103 231 97. Onnerfjord, P,.; Ekstrom, S,.; Bergquist, J.; Nilsson, J.; Laurell, T,.; Marko-Varga, G. Homogeneous sample preparation for automa ted high throughput analysis with matrixassisted laser desorption/ionization time-of-flight mass spectrometry Rapid Communications in Mass Spectrometry 1999 13 315 98. Kussmann, M.; Nordhoff, E.; Rahbek-Nielsen, H.; Haebel, S.; Rossel-Larsen, M.; Jakobsen, L.; Gobom, J.; Mirgorodskaya, E.; Kroll-Kristensen, A.; Palm, L.; Roepstorff, P. Matrix-assisted laser desorption/ionizat ion mass spectrometry sample preparation techniques designed for various peptide and protein analytes Journal of Mass Spectrometry 1997 32 593 99. Hensel, R.R.; King, R.C.; Owens, K.G. Elect rospray Sample Preparation for Improved Quanttitation in Matrix-assisted Laser Desorption/Ionization Time-of-flight Mass Spectrometry Rapid Communications in Mass Spectrometry 1997 11 1785 100. Strupat, K.; Karas, M.; Hillenkamp, E. 2,5-Dihydroxybenzoic acid a new matrix for laser desorption ionizat ion mass-spectrometry International Journal of Mass Spectrometry Ion Processes 1991 111 89 101. King, R.C.; Sepa, J.; Owens, K.G. Protein and polymer analysis by matrix-assisted laser desorption/ionization using 2,5-duhy droxybenzoic acid as a matrix Proceedings of 42nd ASMS Conference on Mass Spectrometry and Allied Topics 1994 p978 102. Garden, R.W.; Sweedler, J.V. Heterogeneity within MALDI samples as revealed by mass spectrometric imaging Analytical Chemistry 2000 72 30 103. Nordhoff, E.; Sch renberg, M.; Thieles, G.; L bbert, C.; Kloeppel, K.-D.; Theiss, D. Sample preparation protocols for MALDI-MS of peptides and o ligonucleotides using prestructured sample support International Journal of Mass Spectrometry 2003 226 163 104. Pekala, RW US patent 4873218, 1989 105. Garziella, A; Pilato, LA; Knop, A; Phenolic Resins, 2nd ed., Springer, New York 2000) (AL-Muhtaseb, S.A.; Ritter, J.A. Preparati on and properties of resorcinol-formaldehyde organic and carbon gels Advanced Materials 2003 15 101

PAGE 135

135 106. Al-Muhtaseb, S.A.; Ritter, J.A. Preparation and properties of resorcinol-formaldehyde organic and carbon gels Advanced Materials 2003 15 101 107. Berthon, S; Barbieri, O; Ehrbuge r-Dolle, F; Geissler, E; Achard P; Bley, F; Hecht, A; Livet, F; Pajonk G; Pinto, N; Riggacci, A; Rochas, C. DLS and SAXS investigations of organic gels and aerogels Journal of Noncrystal Solids 2001 285 154 108. Petricevic, R.; Glora, M.; Fricke, J. Plan ar fiber reinforced carbon aerogels for application in PEM fuel cells Carbon 2001 39 857 109. Merzbacher, C.I.; Meier, S.R.; Pierce J.R. ; Korwin, M.L. Carbon aerogels as broadband non-reflective materials Journal of Noncrystal Solids 2001 285 210 110. Barbieri, O.; Ehrburger-Dolle, F.; Rieker T.P.; Pajonk, G.M.; Pinto, N.; Rao, A.V. Small-angle X-ray sca ttering of a new series of organic aerogels Journal of Noncrystal Solids 2001 285 109 111. Zanto, E.J.; Al-Muhtaseb, S.A.; Ritter, J.A. Sol-gel-derived carbon ae rogels and xerogels: Design of experiments appro ach to materials synthesis Industrial & Engineering Chemistry Research 2002 41 3151 112. Lin, C.; Ritter, J.A. Effect of synthesi s pH on the structur e of carbon xerogels Carbon 1997 35 1271 113. Pekala, R.W. Organic aerogels from th e polycondensation of resorcinol with formaldehyde Journal of Material Sciences 1989 24 3221 114. Sturgeon, C.M. Tumor markers in the laborat ory: closing the guideline-practice gap Clinical Biochemsitry 2001 34 353 115. Worrall, T.A.; Cotter, R.J.; Woods, A.S. Pu rification of contaminated peptides and proteins on synthetic membrane surfaces for matrix-assisted laser desorption/ionization mass spectrometry Analytical Chemistry 1998 70 750 116. Zaluzec, E.J.; Gage, D.A.; Allison, J.; Wa ston, J.T. Direct matrix-assisted laser desorption ionization mass spectrometric anal ysis of proteins immobilized on nylonbased membranes Journal of the American Society for Mass Spectrometry 1994 5 230 117. Bai, J.; Liu, Y.H.; Cain, C.C.; Lubman, D.M. Matrix-assisted laser desorption/ionization using an active perfluorosulf onatedd ionomer film substrate Analytical Chemistry 1994 66 3423 118. Liu, Y.H.; Bai, J.; Liang, X.; Lubman, D.M. Use of a nitrocellulose film substrate in matrix-assisted laser deso rption/ionization mass spectrometry for DNA mapping and screening Analytical Chemistry 1995 67 3482

PAGE 136

136 119. Blackledge, J.A.; Alexander, A.J. Polyethylen e membrane as a sample support for direct matrix-assisted laser desorption/ionization mass spectrometric analysis of high mass proteins Analytical Chemistry 1995 67 843 120. Xu, Y.; Bruening, M.L.; Waston, J.T. Use of polymer-modified MALDI-MS probes to improve analyses of protein digests and DNA Analytical Chemistry 2004 76 3106 121. Xu, Y.; Waston, J.T.; Bruening, M.L. Patterned monolayer/polymer films for analysis of dilute or salt-contaminated protein samples by MALDI-MS Analytical Chemistry 2003 75 185 122. Ching, J.; Voivodov, K.I.; Hutchens, T.W. Po lymers as surfacebased tethers with photonlytic triggers enabling laser-induced release/desorption of covalently bound molecules Bioconjugate Chemistry 1996 7 525 123. Senel, S.; Bayramo lu, G.; Arica, M.Y. DNA adsorption on a polylysine-immobilized poly 2-hydroxyethyl methacrylate membrane Polymer International 2003 52 1169 124. Schriemer, D.C.; Li, L. Combining avidin-biotin chemistry with matrix-assisted laser desorption/ionization mass spectrometry Analytical Chemistry 1996 68 3382 125. Schriemer, D.C.; Yalcin, T.; Li, L. MALD I mass spectrometry combined with avidinbiotin chemistry for analysis of protein modifications Analytical Chemistry 1998 70 1569 126. Wang, H.; Tseng, K.; Lebrilla, C .B. A general method for producing bioaffinity MALDI probes Analytical Chemistry 1999 71 2014 127. Bundy, J.; Fenselau, C. Lectin-based affinity capture for MALDI-MS analysis of bacteria Analytical Chemistry 1999, 71 1460 128. Brockman, A.H.; Orlando, R. New immobiliza tion chemistry for probe affinity mass spectrometry Rapid Communications in Mass Spectrometry 1996 10 1688 129. Kato, K.; Ikada, Y. Immobilization of DNA onto a polymer support and its potentiality as immunoadsorbent Biotechnology and Bioengineering 1996 51 581 130. McGettrick, A.F.; Worrall, D.M. Protei n purification protocols, 2nd edition Methods in molecular biology Edited by Cutler, P. Humana Press Inc.: Totowa, NJ. vol 244, p 151 131. Denizli, A.; Piskin, E. Dye-ligand affinity systems Journal of Biophysical and Biophysical Methods 2001 49 391 132. Yilmaz, M; Bayramo lu, G.; Arica, M.Y. Separation and purification of lysozyme by Reactive Green 19 immobilised membrane affinity chromatography Food Chemistry 2005 89, 11

PAGE 137

137 133. Ellinton, A.D.; Szostak, J.W. Selection in vitro of single-stranded-DNA molecules that fold into specific ligand-binding structures Nature 1992 355 850 134. Alberghina, G.; Fisichella, S. ; Renda, E. Separation of G structures formed by a 27-mer guanosine-rich oligodeoxyr ibonucleotide by dye-ligand affinity chromatography Journal of Chromatography A 1999 840 51 135. Kawazoe, N.; Ito, Y.; Imanishi, Y. Bioassay using a labeled oligonucleotide obtained by in vitro selection Biotechnology Progress 1997 13 873 136. Drug Facts Office of National Dr ug Control Policy [Internet]. Executive Office of the President (US); [updated 2006 Oct 30; cite d 2006 December 8]. Available from http://www.whitehousedrugpolicy.gov/ drugfact/cocaine/index.html 137. Herr, J.K.; Smith, J.E.; Medley, C.D.; Shangguan, D.; Tan, W. Aptamer-conjugated nanoparticles for selec tive collection and dete ction of cancer cells Analytical Chemistry 2006 78 2918 138. Nimura, N; Kinoshita, T; Yoshida, T; Uetake A.; Nakai, C. 1-Pyrenyldiazomethane as a fluorescent labeling reagent for liquid chromat ographic determination of carboxylic acids Analytical Chemistry 1988 60 2067 139. Deng, G.; Markowitz, M.A.; Kust, P.R.; Gaber, B.P. Control of surface expression of functional groups on silica particles Materials Science and Engi neering C-Biomimetic and Supramolecular Systems 2000 11 165 140. Sano, S.; Kato, K.; Ikada, Y. Introduction of functional groups onto the surface of polyethylene for protein immobilization Biomaterials 1993 14 817 141. Preston, K.L.; Silverman, K.; Schuster, C.R.; Cone, E.J. Assessment of cocaine use with quantitative urinalysis an d estimation of new uses Addiction 1997 92 717

PAGE 138

138 BIOGRAPHICAL SKETCH Hong Yu was born in Shanghai, Peoples Republic of China, on November 6th, 1967, to Shunyang Yu and Huaizhen Yan. Hong graduate d from Shanghai Middle School in 1986 and went to East China University of Science and Technology, where she majored in inorganic nonmetallic materials. After receiving a Bachel or of Science degree in inorganic nonmetallic materials at age 23, Hong began work at Shanghai TV Glass Tube Plant as a ceramic engineer, and stayed there for ten years ti ll the plant was shut down. Hong then started a job at Shanghai Omniworks Electronics Ltd. as a QA engineer. In August 2000, Hong came to United States of America to attend the masters program in ch emistry at Southern Illinois University at Edwardsville to work under the tutelage of Prof essor Timothy Patrick. She received a Master of Science in chemistry in August of 2003; she gr aduated with high honors a nd received an award as an outstanding graduate student Hong also met her future husba nd at Southern Illinois, James Rogers, who also received a Master of Science in Chemistry in the summer of 2003. In the summer of 2002, Hong and James began their PhD st udy at the University of Florida and were married in 2003. Hong is working under the me ntorship of Professor Richard Yost. James obtained his PhD. degree in December 2005 and t ook on the job as a patent examiner at US Patent and Trademark Office at Washington DC in January 2006. Hong suspended her study in the PhD program and moved to Washington DC w ith James and their daughter, Elaine Rogers, who was born on May 22, 2005. In August 2006 H ong brought their daughter along with her back to the University of Florida to con tinue her study in PhD program in Chemistry.


Permanent Link: http://ufdc.ufl.edu/UFE0018104/00001

Material Information

Title: Polymer Modification with Matrices for Matrix-Assisted Laser Desorption Ionization Mass Spectrometry and with Aptamers for Surface-Enhanced Laser Desorption Ionization Mass Spectrometry
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0018104:00001

Permanent Link: http://ufdc.ufl.edu/UFE0018104/00001

Material Information

Title: Polymer Modification with Matrices for Matrix-Assisted Laser Desorption Ionization Mass Spectrometry and with Aptamers for Surface-Enhanced Laser Desorption Ionization Mass Spectrometry
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0018104:00001


This item has the following downloads:


Full Text





POLYMER MODIFICATION WITH MATRICES FOR MATRIX-ASSISTED LASER
DESORPTION/IONIZATION MASS SPECTROMETRY AND WITH APTAMERS FOR
SURFACE-ENHANCED LASER DESORPTION/IONIZATION MASS SPECTROMETRY























By

HONG YU


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2007
































Copyright 2007

by

Hong Yu


































To my daughter, Elaine Rogers









ACKNOWLEDGMENTS

I would like foremost to thank my parents, Shunyang Yu and Huaizhen Yan, who always

put their children's wellbeing first, and believed in me and encouraged me to be my best, as well

as my sister, Qiong Yu for her friendship. I also would like to thank my husband, James Rogers,

for his guidance and help in my adjustment to life in a new country and in learning English, and

for his companionship through my graduate career. I would like to thank all my other family

members, who gave me support and love throughout my life.

Many thanks are given to the friends I have met along my graduate career, who have

helped me in my personal life as well as my graduate life in chemistry: Dr. Rong Jiang, Dr. Chia

Pooput, Dr. Yi'an Zhai, Dr. Lin Wang, and Dr. Hui Tao.

I also sincerely thank Dr. James Yang and Yanrong Wu, who helped making part of my

research possible by synthesizing aptamers whenever I need them. I would like to thank Dr.

Weihong Tan for encouraging the collaboration between our group and his group. I would like

to thank Lin Yuan for sharing her hood with me for the carboxylation reaction. I would like to

thank Dr. Gary Cunningham and Nancy Liu for providing the access to the UV-vis

spectrophotometer.

I am very grateful for the help Michael Napolitano and Daniel Magparangalan gave me

for correcting my dissertation. My thanks also are extended to other former and present group

members: Dr. Timothy Garrett, Dr. Alisha Mitchell-Roberts, Dr. Samaret Otero-Santos, Dr.

Mike Belford, Christopher Hilton, Frank Kero, Dodge Baluya, and Rachelle Landgraf.

I would like to thank Dr. David Powell and Dr. Ken Wagener for their helpful discussion

about my research proj ect.

I want to thank my advisor, Professor Richard Yost, for giving me the opportunity to

obtain my doctorate under his direction. He offered me the freedom of choosing the proj ect I









was interested in, and gave me guidance and perspective for the research along the way. I would

also like to thank my first graduate advisor, Professor Timothy Patrick at Southern Illinois

University, where I received a Master of Science in Chemistry. His belief in my abilities

inspired me to continue my education at the University of Florida. There are many teachers and

instructors I would like to thank over the years, all of whom have encouraged me and supported

me to continue my stated goals.











TABLE OF CONTENTS


page

ACKNOWLEDGMENTS .............. ...............4.....

LIST OF TABLES ....._ ................ ...............8.......

LIST OF FIGURES .............. ...............9.....

AB S TRAC T ........._. ............ ..............._ 12...

CHAPTER

1 INTRODUCTION ................. ...............13.......... ......

Introduction to Mass Spectrometry: 3-D And 2-D Quadrupole lon Trap Mass
Spectrometry .............. ........................1
Introduction to Mass Spectrometry .............. ......... .............1
Introduction to 3-D Quadrupole lon Trap Mass Spectrometry ................... ...............15
Introduction to 2-D Quadrupole lon Trap Mass Spectrometry ................. .. .................16
Development of Ionization Methods for MS Analysis: from LD to MALDI and from
MALDI to SELDI ............ ...... .__ ...............17...
Introduction to Aptamers ........._._.. ..... .... ...............22...
Definition and Synthesis of Aptamers ........._................ ._.. .. ........ ...........2
The Advantages of Using Aptamers vs. Antibody and the Applications in
Bioanalysis............... ..............2
Overview of Dissertation ............ ...... .. ...............30.

2 MATRIX-DERIVED RESORCINOL-FORMALDEHYDE POLYMERS: SYNTHESIS
VIA SOL-GEL METHOD AND USAGE IN MALDI-MS SAMPLE PREPARATION......47

Problems Associated with MALDI-MS Sample Preparation Methods ............... .... ........._..47
Background Interference from Matrix Molecules in MALDI-MS...................... ........._..47
The Uneven Distribution of Analyte Molecules with the Traditional MALDI-MS
Sample Preparation Methods .............. ...............50....
Introduction to the RF Polym er ..................... ..... .. ......... ... .. ..... ... .......5
Systematic Investigation in Preparation of the RF and Matrix-Derived RF Polymers ..........54
Analysis of the Spiperone on the CHCA Derived RF Polymer for Lower Background
Interference from Matrix Molecules ............. ..... .....__ ... ..._ ..... ........5
Analysis of the Spiperone on the CHCA-Derived RF Polymer vs. Analysis of the
Spiperone on the Stainless Steel Using CHCA as Matrix for More Uniform Analyte
Signal Intensity Across the Sample .............. ...............61....
Conclusion ............ ..... ._ ...............64...

3 SURFACE MODIFIED RF POLYMER WITH APTAMERS: METHODOLOGY AND
USAGE IN SELDI-MS SAMPLE PREPARATION ................. ..............................87











Aptamers Used for RF Polymer Surface Modification .............. ...............89....
Direct Surface Modification of the RF Polymer with Aptamer ................. ..........___.......92
Surface Modification of the RF Polymer with Aptamers via Surface Modification of the
RF Polymer with Carboxylic Groups...................... ... ...............9
Characterization of the Carboxylic Group-Modified RF Polymer .............. .. ..........___...93
Affinity Capture of RG 19 with RG 19-Retaining Aptamer-Modified RF Polymer and
Characterization of the Modified RF Polymer .............. ....... ._ ...... ..___..........9
Affinity Capture of Cocaine with Direct and Indirect Cocaine-Retaining Aptamer-
M odified RF Polymer .............. ...............96....
Conclusion ............ ............ ...............99...

4 CONCLUSION AND FUTURE WORK ................. ...............122......__....

LIST OF REFERENCES ............_........... ...............127...

BIOGRAPHICAL SKETCH ............ ............ ...............138...










LIST OF TABLES


Table page

2-1 RF polymers with acetone as solvent and Na2CO3 aS catalyst .............. ....................8

2-2 RF polymers with ethanol as solvent and Na2CO3 as catalyst .............. ....................84

2-3 RF polymers with water as solvent and Na2CO3 Or HC1, or HCIO4 aS catalyst.................85

2-4 RF polymers with acetone as solvent and HCI as catalyst ................. ......._ ..........86











LIST OF FIGURES


Figure page

1-1 Parts of the 3 -D quadrupole ion trap ............ ..... ._ ...............31

1-2 Stability diagram in (az qz) space for the region of simultaneous stability. .....................32

1-3 Basic design of the two-dimensional linear ion trap ................. ................ ......... .3 3

1-4 Application of DC, RF trapping, and AC excitation voltages for 2-D ion trap. ................34

1-5 Electron ionization schematic ................. ...............35................

1-6 Chemical ionization schematic .............. ...............36....

1-7 Illustration of FAB ............ ..... ._ ...............37...

1-8 Electrospray ionization schematic .............. ...............38....

1-9 Most commonly used MALDI matrices .............. ...............39....

1-10 Mechanism of MALDI ................. ...............40................

1-11 The chemical and biochemical surfaces for SELDI .............. ...............41....

1-12 The bulge and stem structure of aptamers .............. ...............42....

1-13 The hairpin structure of aptamer ................. ...............43...............

1-14 The pseduknot structure of aptamer ................. ...............44...............

1-15 The G-quartet structure of an aptamer ................. ...............45........... ..

1-16 Generalized scheme indicating the key steps in the SELEX process .............. .... ........._..46

2-1 The mechanism of the TEOS sol-gel reaction ......... ................ .............. ....6

2-2 Illustration of incorporation of the matrix molecules in the TEOS gel ................. ...........66

2-3 Microscope pictures of sample surfaces of manually prepared samples .........._................67

2-4 MALDI ion images showing intensities for selected peptides using............... ................68

2-5 Mechanism of the polymerization of RF polymer proposed by Lin and Ritter.. ...............69

2-6 Base catalyzed RF polymerization............... .............7

2-7 Cross-section of the ion trap mass spectrometer ................. ...............71..............











2-8 Illustration of the path of laser and ions in mass spectrometer............_ .. ......._.._.. ....72

2-9 Fragment path of spiperone .............. ...............73....

2-10 UV absorbances of the matrices-embedded RF polymers ......____ .... ... ..__............74

2-11 UV absorbances of the RF polymer, CHCA embedded RF polymer, and CHCA
soluti on ........... __..... ._ ...............75....

2-12 Mass spectrum of the 100 ppm spiperone on CHCA embedded RF polymer and
stainless steel............... ...............76.

2-13 CHCA embedded in the RF polymer structure ................. ...............77..............

2-14 LTQ with vMALDI ion source from ThermoFinnigan .............. ...............78....

2-15 The microscopic image of the sample well of dry droplet sample preparation ................79

2-16 The distribution of the spiperone using dry droplet method on stainless steel. .................80

2-17 The microscopic image of the RF-16 polymer pellet after the deposition of spiperone....81

2-18 The distribution of the spiperone using dry droplet method on RF-16 polymer. .............82

3-1 The structure of the RF polymer. ........... ..... ._ ...............100

3-2 Chemical structure of reactive green 19. ................ ................ ......... ........ 101

3-3 The 27-mer oligodeoxyribonucleotide sequence. .............. ...............102....

3-4 The two cocaine isomers: a) pseudococaine and b) cocaine............... .................0

3-5 Anti-cocaine aptamer MNS-4.1 bound to cocaine .............. ...............104....

3-6 The picture of a Applied Biosystems 3400 DNA Synthesizer .............. .....................105

3-7 Scheme of the indirect surface modification of the RF polymer via the carboxylation ..106

3-8 Chemical structure of PDAM .....___................ ................. ..............._107

3-9 Fluorescent emission intensity of PDAM-bond and noncarboxylated RF polymer........108

3-10 Structure of Toluidine blue O .............. .....................109

3-11 UV absorbance of toluidine blue O desorbed from the carboxylated and
noncarboxylated RF polymer. ..........._ ..... ._ ...............110..

3-12 UV of the toluidine blue O solutions of different concentration ................ ...............11 1










3-13 Calibration curve of the concentration of the toluidine blue O solution ................... .......1 12

3-14 UV absorbance of reactive green 19 desorbed from aptamer attached RF polymer
and RF polymer ................. ...............113................

3-15 Calibration curve of the concentration of the reactive green 19 solution ................... .....1 14

3-16 Cocaine displaces diethylthiatricarbocyanine iodide completed with aptamer MNS-
4.1............... ...............115..

3-17 A microscopic image of the indirectly cocaine retaining-aptamer modified RF
polym er. .............. .. ...............116......... ......

3-18 MS spectrum of the cocaine on the unmodified RF polymer and on the cocaine
retaining aptamer attached RF polymer ................. ...............117........... ...

3-19 The 2-D and 3-D image of the cocaine signal intensity across the indirect cocaine
retaining-aptamer modified RF polymer ................. ...............118...............

3-20 The image of the direct cocaine retaining-aptamer modified RF polymer ................... ...1 19

3-21 MS spectrum of the cocaine on the direct cocaine retaining-aptamer modified RF
polym er ................. ...............120......... ......

3 -22 The 2-D and 3-D image of the cocaine signal intensity across the direct cocaine
retaining-aptamer modified RF polymer .............. ...............121....









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

POLYMER MODIFICATION WITH MATRICES FOR MATRIX ASSISTED LASER
DESORPTION IONIZATION MASS SPECTROMETRY AND WITH APTAMERS FOR
SURFACE ENHANCED LASER DESORPTION IONIZATION MASS SPECTROMETRY

By

Hong Yu

May 2007

Chair: Richard A. Yost
Maj or Department: Chemistry

A resorcinol-formaldehyde (RF) polymer and matrix-embedded RF polymers were

developed to be used as substrates for improved matrix-assisted laser desorption/ionization mass

spectrometry (MALDI-MS) analysis and for selective analysis of biological samples.

Matrix molecules can be embedded within this polymer, which renders excellent spectra

without the interference imposed by the matrix molecules for low molecular weight analytes. By

embedding the matrix molecules in this polymer more uniform spectra are obtained, solving the

prevalent problem associated with the traditional MALDI sample preparation methods.

The phenol groups on the polymer made it possible to attach aptamers selective to

reactive green 19 and cocaine onto it directly, and gave high specific affinity toward the reactive

green 19 and cocaine, respectively. The phenol groups on the polymer also made it possible to

attach carboxylic acid groups which could then be reacted with amine-modified aptamers. This

indirect aptamer-modified RF polymer has higher reaction efficiency than the direct aptamer-

modified RF polymer, yield more aptamers at higher density on the polymer for affinity capture

of analyte. This high specificity provided by the aptamers could make it a valuable tool to

specifically retain the biomarkers of diseases and could be used for screen test for diseases.









CHAPTER 1
INTTRODUCTION

In the history of analytical chemistry, new technologies of analysis and instruments have

been developed to better determine the compositions and structures of substances in order to

identify them in the pursuit of understanding the world around us. The analytical methods that

have been used include some very rudimentary but still widely used techniques such as titration,

density, melting point, boiling point, flame test, and combustion, which are methods still taught

to new chemistry students. As the tasks for chemical identification became more complicated

and difficult, many new technologies were developed to peek into the microscopic world of

chemistry. These technologies include atomic absorption, ultraviolet/visible (UV-vis)

absorption, IR, fluorescence, flame, atomic emission, Raman, X-ray, X-ray crystallography,

NMR, electrochemistry, gravimetric analysis, calorimetry, and thermogravimetric analysis

(TGA), etc. Each of these more sophisticated technologies, based on a specific theory, utilizes

specific chemical and/or physical properties of the substance, and requires a specific instrument

to obtain specific information of the analyzed substance.'

A mass spectrometer is an instrument which was first invented at the beginning of the

twentieth century, but took off in its application in the past thirty or forty years as a tool to

identify various analytes. From gaseous molecules to volatile small organic molecules, then to

involatile large organic molecules, and eventually to proteins, DNA, cells, and bacteria, the range

of the analytes than can be analyzed has expanded since then. The high sensitivity, speed,

specificity, and reproducibility of mass spectrometry (MS) have made it an irreplaceable tool for

its application in protein analysis--including discovery, identification (i.e., peptide mapping and

sequencing), and structural characterization.2









Introduction to Mass Spectrometry: 3-D and 2-D Quadrupole lon Trap Mass Spectrometry

Introduction to Mass Spectrometry

The first mass spectrometer, built in 1907 by J.J. Thomson, was used to detect positive

rays based on the mass-to-charge ratio (m/z) of the particles; this was the basis of modern mass

spectrometry.3 The modern mass spectrometer is used to obtain the information on molecular

mass according to the mass-to-charge ratio and even structures of molecules according to

fragmentation patterns.4 The ionization device coupled to the mass analyzer has evolved because

of the demand of an expanded variety of analytes. Electron ionization (EI) was the first used,

followed by chemical ionization (CI), fast atom bombardment (FAB), electrospray ionization

(ESI), and matrix-assisted laser desorption/ionization (MALDI).

Analyte molecules are ionized by addition or loss of an electron or a proton or other ion.

The mass spectrum consists of the molecular-type ions (M [M+H] [M-H]-, ect.) and often

fragment ions, the pattern of which may be unique for each molecule. The mass spectrometer

was first used to identify elements and their isotopes.' Soon after, petroleum chemists used the

mass spectrometer to identify organic molecules.6 Now, the applications of the mass

spectrometry have broadened to include biochemistry, explosives, fullerenes, toxics, and

environmental pollutants.3

A mass spectrometer is named according to the mass analyzer which is its determining

component. During the process of the development of the mass spectrometer, many different

types of mass analyzers have been invented, which include magnetic-sector, quadrupole mass-

Eilter, 3-D quadrupole ion trap, 2-D quadrupole ion trap, and time-of-flight.3 Each of the mass

analyzers has its forte and suitability for analysis of certain types of analytes and is chosen

according to the task and budget.









Introduction to 3-D Quadrupole lon Trap Mass Spectrometry

The quadrupole ion trap was invented by German physicist Wolfgang Paul and co-workers

in the early 1950s.7 The quadrupole ion trap can be used to confine ions in a small volume using

just high-frequency electric fields without a material container.8 This invention later earned Paul

the 1989 Nobel Prize in Physics.9 Figure 1-1 illustrates a traditional 3-D quadrupole ion trap.

This 3-D quadrupole ion trap has two hyperbolic end-cap electrodes with entrance and exit holes

in the center of both end-caps. The hyperbolic ring electrode in-between the end-caps has a

radius ro.10 The distance between the two end-caps is 2zo, where ideally, Zo=ro/92. Ions entering

the trap are trapped at the center of the ion trap by a quadrupolar field which is created by a radio

frequency (RF) waveform applied to the ring electrode when the two end-caps are grounded.

Equations 1-1 and 1-2 are derived from the Mathieu equation for a "stretched" ion trap,

where e is the charge, U is the DC potential, typically kept at zero; V is the amplitude of the RF,

and 0Z is the angular frequency of the applied RF. Solutions to these two equations define

whether an ion with particular m/z will be stable inside the quadrupolar field according to their

position in the Mathieu stability diagram (Figure 1-2).10

az=-1 6eU/m(r02+2z02) ~22 1

qz=-8eV/m(r02+2z02) 22 (1-2)

The quadrupole ion trap confines ions in the small volume between the ring electrode and

two endcap electrodes by appropriate oscillating (RF) electric fields. In modern ion trap, an AC

waveform, termed the resonance ejection waveform, is applied across the endcaps. The

frequency is typically set to be slightly less than half of the RF frequency, G2. Since the

oscillating frequency, co, the ion motion in the z direction is given by DZBz/2 (see Figure 2 for Bz

values), this will correspond to a ej ect V value less than 0.908. When the amplitude of the









applied RF is linearly ramped up, a mass spectrum is obtained via the mass-selective instability

scan. In this instability scan, ions are ej ected from the trap with increasing na : with respect to

time when they come into resonance with the resonance ej section waveform. Higher na: ions are

still trapped; the limitation of mass range is due to an inability to reach the needed RF voltage

levels required to ej ect higher m/z ions out of the trap. In normal operation, for the instrument

used for this research, the maximum na: detectable is 650. Lowering the AC resonance ej section

frequency applied across the endcaps will allow ions to come into resonance earlier in the RF

ramp, extending the mass range by lowering the qz according to the equation 1-3.11

(nz z) new = (nz z) old q eject new/ q eject old) (1-3)

A quadrupole ion trap can be connected with external ion sources such as matrix-assisted

laser desorption/ionization (MALDI) and electrospray ionization (ESI). The range of ionization

sources that can be used, in addition to its MSn capability, makes the quadrupole ion trap a very

useful tool for compound identification.

Introduction to 2-D Quadrupole lon Trap Mass Spectrometry

The 2-D ion trap, the linear variant of the 3-D ion trap are based on the four-rod 2-D

quadrupole mass filter, omitting the quadrupolar-trapping field along one of the three spatial

axes.8

The 2-D ion trap shown in Figure 1-3 12 is composed of three sections in which ions are

mass selectively ej ected radially through two slits in the two opposite rods in the center section.

Ions are trapped in the z direction by more positive DC voltages applied to the two end sections

for axial trapping of positive ions (Figure 1-4).12 The ions are trapped radially by the RF applied

in two phases to opposite rod pairs, as shown in Figure 1-4.12 A supplementary AC voltage is

applied across the x-rod pair in two phases for isolation, collision-induced dissociation (CID),

and resonant ej section of ions (Figure 1-4)12









Ions confined within a linear trap can be mass-selectively ej ected in a direction

perpendicular to the central axis of the trap (radial ej section) or can be ej ected along its axis (axial

ej section Having the detectors at both radial exit slits doubles the number of detected ions in

radial ejection design. The axially-ejected ions can be introduced into a second mass analyzer

such as a Fourier-transform ion cyclotron resonance (FT-ICR) analyzer to form a hybrid

instrument.3

Relative to the well-established 3-D ion-trapping instruments with similar mass range, the

2-D linear trap with radial ej section has significantly higher trapping efficiency and increased ion

capacity that improves the detection sensitivity at least 5 to 10-fold. This linear trap can conduct

all the typical scan modes intrinsic to 3-D traps with comparable mass resolution but at higher

overall scan rates.8

Development of Ionization Methods for MS Analysis: from LD to 1MALDI and from
MALDI to SELDI

The ionization method used is determined by the type of the analyte and determines the

characteristics of the mass spectrum obtained. The most common ionization methods include EI,

CI, ESI, and MALDI.

Electron ionization was the first ionization method developed, and was used in the

instrument invented by JJ Thompson. Electrons with an energy of about 70 eV collide with the

vaporized analyte in vacuum (10- -10-4 torr)3 to ionize the analyte molecules (Figure 1-5).13 The

spectra obtained with El are characterized by significant fragmentation of the molecular ions

([M] ). Later, when a softer ionization method was required for more fragile compounds

chemical ionization (CI) was developed. With CI, electrons collide with a reagent gas present in

large excess compare to the analyte to produce reagent ions which will later pass their charge to

the analytes by collision without transferring an excess of energy (Figure 1-6).14 The spectra









obtained with CI are characterized by less fragmentation of the analyte and more abundant

molecular-type ions ([M+H] ) than with EI. For these two ionization methods, the analytes have

to be vaporized before they can be ionized, which is a problem for less volatile compounds.

Typically, the higher the molecular weight, the less volatile the molecules. With El and CI, it is

difficult to ionize most of the large organic molecules and is nearly impossible to ionize

macromolecules.

With increasing demand for analyzing large nonvolatile molecules, ionization methods that

can ionize large, nonvolatile molecules were developed such as fast atom bombardment (FAB).

With FAB, a beam of atoms bombard the analyte, which is dissolved in a low volatile liquid

matrix, to produce the analyte ions (Figure 1-7).15 Since the analyte molecules are not vaporized

before they are ionized, FAB can be used to ionize polar, ionic, thermally labile, and relatively

high molecular weight compounds that are not suitable for normal EI/CI analysis. Depending

upon whether the cation or the anion is of interest, positive-ion or negative-ion FAB/MS

analyses can be performed. FAB can ionize molecules up to 2000-3000 Da.3

Laser desorption (LD) ionization was also developed to tackle the problem associated with

less volatile large compounds. Introduced in the early 1960s, LD irradiated low-mass organic

molecules with a high-intensity laser pulse to form ions that could be successfully mass

analyzed. With LD, nonvolatile large molecules with the ability to absorb the energy from the

laser beam can be ionized. Although LD expanded the range of the analyte that can be ionized,

there were some limitations that came along with LD. The wavelength of the laser used in LD

has to match the wavelength of the analyte molecule' s absorbance for the laser energy to be

utilized for ionization. This may be difficult when a variety of analytes need to be analyzed. LD

was also considered as an ionization method which is too energetic for labile analytes since laser









energy is imparted directly into the molecules, which my cause a large amount of thermal

degradation and fragmentation.

First developed in 1980s, ESI was developed to ionize large and thermally labile

molecules. By forcing a solution of the analyte through a small capillary held at high voltage,

the solution sprays into tiny droplets with charges on each droplet. After the solvent is

evaporated, ions from nonvolatile compounds with one or more charges are generated, as shown

in Figure 1-8.16 The existence of multiple charges on high mass ions lowers the mass-to-charge

ratio, which expands the mass range of the molecules can analyzed.3 As a soft ionization

method, ESI makes it is possible to investigate even intact noncovalent complexes. The

invention is so important and revolutionary that the inventor, John B. Fenn of the Virginia

Commonwealth University, shared with Koichi Tanaka of the Shimadzu Corp. half of the 2002

Nobel Prize in Chemistry for his invention of ESI.1

Matrix-assisted laser desorption/ionization (MALDI) is a soft ionization method that rivals

ESI for mass spectrometric analyses of biological macromolecules and the noncovalent

complexes between peptides and metal ions,17, 18, 19' 20' 21, 22,23 Single-stranded DNA, 21, 24, amino

acids,25 drugs,26 and other peptides.27,28,2930 In 1985, Michael Karas and Franz Hillenkamp

published the laser desorption results of the mixture of aromatic amino acid which absorbs laser

energy with aliphatic amino acid which does not absorb laser energy.31 In 1987, Karas and

Hillenkamp successfully used a matrix in LD to ionize and analyze high molecular weight

molecules.32 Matrices are organic molecules with aromatic groups that can absorb laser energy

in UV. The most commonly used matrices are 2,5-dihydroxybenzoic acid (DHB), trans-3,5-

dimethoxy-4-hydroxycinnamic acid or sinapic acid (SA), nicotinic acid (NA), trans-3-methoxy-

4-hydroxycinnamic acid or ferulic acid (FA), 3-hydroxypicolinic acid (3HPA), 4-hydroxy-a-









cyanocinnamic acid (4HCCA), and (in the infrared) succinic acid and glycerol (Figure 1-9).

With MALDI, a low concentration of the analyte is mixed with a large excess of organic matrix

to completely isolate analyte molecules from each other to homogenous 'solid solution'. The

laser beam is focused onto the surface of the matrix-analyte solid solution. The matrix molecules

absorb the laser energy at the incident laser wavelength and then pass that energy on to analyte

molecules, producing a pulse of ions with each laser pulse. The clusters ejected from the surface

consist of analyte molecules surrounded by matrix and salt ions. The matrix molecules

evaporate away from the clusters to leave the free analyte in the gas-phase. The photo-excited

matrix molecules are stabilized through proton transfer to the analyte. Cation attachment to the

analyte is also encouraged during this process. It is in this way that the characteristic [M+X]

(X= H, Na, K etc.) analyte ions are formed. These ionization reactions take place in the

desorbed matrix-analyte cloud just above the surface. The ions are then extracted into the mass

spectrometer for analysis. MALDI allows a larger variety of analytes to be ionized by the same

laser than LD.33 Figure 1-1014 iS an illustration of ionization process of MALDI. The

application of MALDI to biological macromolecules demonstrated by Koichi Tanaka 34 led him

to receive a quarter of the 2002 Nobel Prize for chemistry."

The matrix-to-analyte ratio ranges from 100 to 50,000.3 MALDI overcame the problem

associated with the LD, but inevitably suffers from the interference from matrix molecules with

analytes of low molecular weight. The excessive amount of matrix molecules poses interference

to the interpretation of the mass spectra of low molecular weight analytes. Some effort has been

put into developing matrix-free laser desorption/ionization methods to eliminate the interference

from matrix. Desorption/ionization on porous silicon (DIOS) is one of the methods developed

which utilizes the large surface area of an UV absorbing semiconductor treated electrochemically









to produce spectra without the interference of matrix clusters.35, 36, 37,38 DIOS certainly has its

success, but suffers from the relatively less than ideal sample-to-sample reproducibility, stability

of targets, molecular weight limitation, low analyte signal intensity and is sensitive to how the

samples are deposited on them-cracks can form with an inappropriate sample deposition method.

Since MALDI opened the gate for the MS analysis of macromolecules, biological samples

have been subj ect to the MS analysis for disease biomarker discovery and clinical diagnostics.

Disease biomarkers are the characteristic amount or the existence of a certain substance or

substances.39 Biological samples taken directly from humans or animals such as blood, serum,

plasma, urine, and cellular secretion products also contain extremely large numbers of

components which are not of interest and can interfere with the MS analysis.40 These unwanted

components include biological molecules and organic and inorganic salts which can make the

MS analysis extremely difficult. Liquid chromatography (LC), high-pressure liquid

chromatography (HPLC), membrane dialysis, centrifugation, immunoprecipitation, gel

electrophoresis and other separation techniques have been used to separate and extract the target

analyte from the biological sample prior to the MS analysis. Theses separation techniques are

usually tedious and time-consuming and suffer loss of the analyte during the process, especially

for the low abundant components.40

Surface-enhanced laser desorption/ionization (SELDI) was introduced by Hutchens and

Yip in 1993 in order to achieve better sample preparation.41 SELDI is based on MALDI but with

the surface of the substrate modified to have a specific affinity toward certain types of molecules

to achieve an on-probe, one-step clean-up procedure before the sample is analyzed. There are

several companies which provide SELDI chips, such as Ciphergen Biosystems (Palo Alto, CA).

Affinity capturing for enrichment and clean-up of analyte is borrowed from affinity









chromatography. Among them, immobilized metal affinity capture (IMAC), antibody and

protein affinity capture, lectin affinity capture, inherent hydrophobic interaction bio-capture, and

bio-molecular interaction capture are attractive ones, and have been used to retain biomolecules

such as proteins and peptides.42 Figure 1-11 shows the biochemical and chemical modified

affinity surfaces available from Ciphergen.42 The chemical modification of the substrate surface

is well established. It retains a class of molecules which is useful for certain applications.

Biochemically modified surfaces have specificity toward one specific molecule. Antibodies,

DNA, enzymes, and receptors have been used in biochemical surface modification.

Aptamers are pieces of synthetic DNA or RNA oligonucleotides which recently been

applied to SELDI affinity capturing.73 Aptamers are screened from a randomly generated

population of DNA/RNA sequences for their ability to bind with desired molecular targets such

as peptides with high specificity, and can be designed and synthesized easily with a DNA

synthesizer. Aptamers have not yet been widely used for SELDI and are not available

commercially.

Introduction to Aptamers

Aptamers have been employed in applications such as chromatography, capillary

electrophoresis, biosensors, and signaling because of its ability of affinity capture small

molecules, peptides, and proteins.43 Aptamers has also been used for mass spectrometry sample

preparation in surface enhanced laser desorption/ionization (SELDI).

Definition and Synthesis of Aptamers

Aptamers are pieces of synthetic DNA or RNA with 3-D structures which provide the

targets binding capability. The name "aptamer" is derived from the Latin word "aptus" which

means "to fit". Aptamers can be selected "to fit" small molecules, biopolymers, surfaces or even

whole cells.43 Because of the high specific affinity, aptamers can be used as recognition









elements in heterogeneous assays to replace antibodies.44, 45 The selective binding affinity

toward the target is due to specific interactions such as hydrogen bonding or association with the

phosphate groups of the aptamer.46 The sequence-specific tertiary structure of the aptamer helps

these interactions, which provides a rigid platform for the arrangement of functionalities of the

aptamer. The 3-D structures have been discovered to include the stem-loop/bulge (Figures 1-12

and 1-14),47, 48 the helix, the hairpin (Figure 1-13),47 the pseudoknot (Figure 1-14),49 and the G-

quartet structure (Figure 1-15).5o Under certain condition such as the buffer with certain pH,

certain components, and concentration; aptamer can form 3-D configuration which retains a

specific protein or molecule which can fit in this 3-D configuration. Because of its affinity

capture capability, aptamers have been used in chromatography for separation and purification of

mixtures.48, 51

An aptamer for a particular target is discovered and produced by the method called

systematic evolution of ligands by exponential enrichment (SELEX), which utilizes in vitro

selection and amplification procedures to isolate aptamers for any small molecule or protein

target.52, 53 This automatic method was developed independently by Joyce, Szostak, and Gold.52,

54, 55 First, a library of oligonucleotide molecules with the size of 101 ~ 101 is generated

chemically in a standard oligonucleotide synthesizer." The greater the number of randomized

nucleotide positions, the more complex a library is and the more likely the success of finding the

molecules that interact with a target.44 Each member in a library has a unique sequence which is

designed to contain a contiguous randomized region flanked by two fixed sequence regions.

Technological advances already have made it possible to eliminate the requirement for the fixed

regions in random sequence libraries used for the SELEX process, thereby producing short

aptamer sequences. A mixture of phosphoramidites containing all four building blocks: A, G, C,









and T is delivered to the synthesizer to synthesize each nucleotide position in the contiguous

random region. The ratios of the four phosphoramidites in the mixture are adjusted based on the

coupling efficiencies of individual monomers to obtain an unbiased library with equal

representation of all four nucleotides. Because aptamers are identified from DNA as well as

from RNA libraries, chemically synthesized DNA libraries are enzymatically converted to RNA

libraries.56

The screening process is the next step, in which a random sequence oligonucleotide (RNA

or DNA) library is incubated with a target of interest in a buffer of choice at a given temperature

(see Figure 1-16).5 A very small fraction of individual sequences tend to interact with the

target, and these sequences are separated from the rest of the library by any one of the physical

separation techniques. Typically, nitrocellulose filter partitioning is used with protein targets

that are retained on the nitrocellulose. Small molecular targets are generally immobilized on a

solid support to generate an affinity matrix in which case, sequences that do not interact with the

target on the solid support can be removed easily by a simple washing step. The mixture of

aptamer candidates recovered from the target represents a mixture of sequences containing both

high- and low-affinity binding molecules to the target.44 The sequences bound to the target are

isolated and amplified to obtain an enriched library to be used for the selection/amplifi cati on

cycle.

The DNA aptamer candidate mixture is amplified directly by the polymerase chain

reaction (PCR), while RNA sequences are amplified by PCR after being converted from DNA by

reverse transcription (RT).55 The single-stranded DNA (ssDNA) population obtained by strand

separation of PCR products is incubated with a fresh sample of the target for the next round of

selection. The RNA population is obtained by in vitro transcription. Several iterations of the









selection process are carried out under increasingly stringent conditions to enrich the high-

affinity sequences and eliminate the low-affinity binder. The enrichment efficiency of high-

affinity binders is controlled by the stringency of selection at each round. Analysis of enriching

populations against the target is carried out to determine the progress of the enrichment of high-

affinity binders.44 The number of cycles required for aptamer identification is usually dependent

on the degree of stringency imposed at each round as well as on the nature of the target. For most

targets, affinity enrichment is reached within 8-15 cycles. In general, one cycle of SELEX takes

two days. The enriched library is cloned and sequenced to obtain the sequence information of

each member until the chosen sequences dominate the population.'" A typical SELEX

experiment may take approximately 2-3 months including cloning and sequencing.'

Aptamers that come out of a SELEX experiment are full-length sequences which are

generally 70-80 nucleotides long, and can be truncated to eliminate nucleotide stretches that are

not important for direct interaction with a target or for folding into the structure that facilitates

target binding. The truncation of aptamers to the minimal target-binding domain has been

successfully carried out to obtain functional aptamers less than 40 nucleotides long.58 59, 60' 61, 62,

63 Once the sequence is identified, an aptamer is produced by chemical synthesis.

Aptamers have remarkable specificity and can discriminate targets on the basis of subtle

structural differences such as the presence or absence of a methyl group64, 65 Or a hydroxyl

group66' 67 and the D- vs. L-enantiomer of the target.66, 68 Because of the selective demand in the

SELEX process that eliminates sequences that bind closely related analogs of the target,

sometimes the degree of specificity of aptamers is better than that of antibodies.64 Practically,

elimination of the sequences that bind closely related analogs of the target is achieved by the

process called "counter-SELEX" that effectively discards ligands that have an ability to bind the









target as well as closely related structural analogs of the target.64 During counter-SELEX

selection, the population of aptamers bound to the target is subj ected to affinity elution with

structural analogs and the sequences eluted are discarded.

The counter-SELEX strategy is a valuable tool in identifying aptamers aimed at a specific

target in a complex mixture, even without prior knowledge of the target. For example; in the

search for aptamers that bind to an "epitope" only present on the surface of cancer cells but not in

healthy cells, the cells from healthy tissue are used to remove sequences that bind to the

background that does not contain the epitope of interest before the library is challenged with

cancer cells.

The Advantages of Using Aptamers vs. Antibody and the Applications in Bioanalysis

Both antibodies and aptamers are utilized in bioanalysis based on their molecular

recognition capability. Antibodies are the most popular class of molecules providing molecular

recognition." They have been used in SELDI because of its high specificity and selective

affinity towards a specific target.

Antibodies started to get attention since thel950s and became very popular in the 1970s.5

Although antibodies have the advantages of having a very high selective affinity, they also have

several limitations.

First, antibody generation involves untidy cell cultures and lab animals since antibodies

must be selected and produced in a living organism. Antibodies of non-human origin have

implications in diagnostic applications;69 thus, the use of antibodies in therapeutic applications is

limited since the generation of hybridomas is restricted to rats and mice. Heterophilic antibodies

(human antibodies that recognize antibodies of non-human origin) that exist in humans might

lead to false-positive results because a capture antibody can be linked with a detector antibody of

non-human origin in the absence of the specific analyte.70









Second, stocks of antibody-producing cells need to be stored at multiple sites to prevent

the complete loss of the cell lines because of the possibility of accidental losses or the death of

cell lines. Typically, high yields of monoclonal antibodies are obtained by growing the

hybridomas in the peritoneal cavities of animals and purifying the antibody from ascites

(abdominal dropsy) fluid. Some hybridomas are difficult to grow in vivo, thus restricting this

route of antibody production.44

Third, molecules that are not well tolerated by animals, such as toxins have difficulty being

employed to produce antibodies in living organisms.44 Furthermore, it is difficult to raise

antibodies against inherently less immunogenic molecules.44

Fourth, the identification and production of monoclonal antibodies are laborious and could

become very expensive for searches of rare antibodies that require screening of a large number

of co omies.44

Fifth, the consistent performance of the same antibody from batch to batch is poor; thus,

immunoassays need to be optimized with each new batch of antibodies.44

Sixth, since the antibodies are produced in vivo and subj ect to in vivo variations, it is not

practical to identify antibodies that could recognize targets under conditions other than

physiological conditions.4

The ability of aptamers to bind to various targets makes aptamers applicable as biosensors,

imaging probes, MALDI targets, and drugs.51 First, the properties of aptamers can be changed

on demand44 because aptamers are identified through an in vitro process that does not depend on

in vivo conditions. Furthermore, selection conditions can be designed to obtain aptamers with

properties desirable for in vitro diagnostics.44









Second, as opposed to antibodies, toxins as well as molecules that do not elicit a good

immune response can be used to generate high-affinity aptamers because animals and cells are

not involved in aptamer identification.44

Third, aptamers are stable during long-term storage and can be transported at ambient

temperature. Aptamers can undergo reversible denaturation process and easily recover from

exposure to undesirable conditions within minutes. The in vitro selection process for aptamers

can be carried out under conditions akin to those used in the assay for which the aptamer is being

developed. The aptamer will maintain its structure and function in the final assay and not fall

apart as antibodies do."

Fourth, aptamers with any sequences that can recognize any class of target molecules with

high affinity and specificity can be discovered with SELEX process. Once the sequence of a

particular DNA aptamer is known, it is easy and inexpensive to synthesis the aptamer in a DNA

synthesizer with high accuracy and reproducibility. They are then purified under denaturing

conditions to achieve very a high degree of purity with little or no batch-to-batch variation in

aptamer production.44, 51

Fifth, additional chemistries can be added on without a loss in function because aptamers

are chemically synthesized, which gives aptamers superiority over antibodies. When chemical

groups are attached to the ends of the aptamers, their life span in the bloodstream increases, they

can be targeted to particular locations, or they can be immobilized onto a surface. Immobilizing

aptamers without potential loss of function makes it superior to antibodies." Reporter molecules

such as fluorescein and biotin can be attached to aptamers at exact locations designed by the

user. Functional groups that allow subsequent derivatization of aptamers with other molecules

can also be attached during the chemical synthesis of aptamers.44









Sixth, aptamers can also be used for the development of sandwich assays for small

molecules or modified into a molecular-beacon format (a DNA strand with a fluorescent tag and

a quencher) for reagentless assays, which is beyond the capability of antibodies."

Seventh, aptamers have smaller molecular weight, which offer advantages such as ease to

penetrate tissue, shortened residence time in blood, and a smaller footprint when being attached

to a surface to get a higher binding density."

Because of the advantages the aptamers posses over antibodies, the aptamers can be used

to replace antibodies to retain molecules, although the recognition ability of aptamers is not as

high as antibodies."

A lot of effort has been put into research to explore the usability of aptamers in analytical

chemistry. SomaLogic (Boulder, CO) is developing aptamer proteomic chips as diagnostic tools

that screen for biomarkers in serum. Ellington's group from Department of Molecular Biology,

Massachusetts General Hospital, Boston, has demonstrated that aptamers can be immobilized on

beads, introduced onto a sensor array, and used for the detection and quantitation of proteins."

Kennedy's group from University of Florida showed that DNA aptamers immobilized on a

chromatographic support can selectively bind and separate adenosine monophosepahte (AMP),

cyclic-AMP, adenosine diphosphate, adenosine triphospahte, NAD+, and adenosine from

mixtures as complex as tissue extracts.72 McGown from Duke University successfully used a

thrombin-binding DNA aptamer for affinity capture for MALDI-MS.73 In her work, she attached

aptamer pieces onto a treated fused-silica glass surface and then used the aptamer-attached glass

to capture the thrombin from a protein mixture. The result showed that the thrombin-binding

aptamer has high affinity toward the thrombin.









Overview of Dissertation

The importance of developing a new MALDI substrate and the suitability of the RF

polymer for MALDI-MS analysis was emphasized at the beginning of the dissertation.

A new RF polymer and matrix-derived RF polymers have been developed and

characterized as discussed in Chapter 2 for elimination of the background noises fro matrix

molecules used in MALDI-MS analysis and for more even distribution of signal intensity of

analyte across the sample spot.

Surface modification of the RF polymer with carboxylate groups and then with amine-

modified aptamers have been performed and characterized as discussed in Chapter 3 for

attachment of functional groups and then attachment of functionally-modified aptamers onto the

RF polymer.

With the affinity capture capability of this SELDI-MS substrate, it is possible to perform

an on-probe one-step clean-up for the specific small molecules such as reactive green 19 and

cocaine before the MS analysis of a biological sample. The polymer and modification

procedures are capable of being applied later on disease biomarkers. This is crucial for large-

scale cancer screening to be accessible for the vast maj ority of the population at lower cost and

higher accuracy.








































EXIT LENS
/SLEEVE






IONS IONS
IN OUT
Z





SECOND EXIT
OCTAPOLE LN

EXIT ENDCAP
ELECTRODE

It RING
ELECTRODE

SPACER NUT
RINGS
ENTRANCE ENDCAP SPRING
ELECTRODE POST WASHER


Figure 1-1. Scheme of parts of the 3-D quadrupole ion trap. [Adopted from March, R.E. An
introduction to quadrupole ion trap mass spectrometry Journal of Mass Spectrometry
1997, 32, 351]


Entrance Ring Exit
Endcap Electrode Endcap


ANALYZER























































.Stability diagram in (a, q, ) space for the region of simultaneous stability in both the
r- and z-directions near the origin for the three-dimensional quadrupole ion trap ; the
iso-B, and iso-BZ lines are shown in the diagram. The q-axis intersects the jL=1
boundary at q,=0.908, which corresponds to qmax in the mass-selective instability
mode. [Adopted from March, R.E. An introduction to quadrupole ion trap mass
spectrometry Journal of~a~ss Spectrometry 1997, 32, 351]


0.2


1.0 0
Pq o0.8,C 0.1 .~, q,=-0.908
0-' Z 07
0.6
0.4


-


0.1




0-





--0.2 -



--0.3-



--0.4 -i



--0.5



--0.6



--0.7




Figure 1-2

















Back
Secrtlang


Front
S~ctlo


Figure 1-3. Basic design of the two-dimensional linear ion trap. [Adopted from Schwartz, J.C.;
Senko, M.W. A two-dimensional quadrupole ion trap mass spectrometer Journal of
the American Society of~a~ss Spectrometry 2002, 13, 659]

























RF+

Radial Quadrupolar Tra3pping


GND

AC+AC-


DC 1 DC 2 IDC 3



DC 1 DC 2 DC 3

Axial Trapping


Y


t,


Y


t.x


Radial Dipola Excitation

Figure 1-4. Scheme for application of DC, RF trapping, and AC excitation voltages necessary
for operation of the 2-D ion trap. [Adopted from Schwartz, J.C.; Sento, M.W. A two-
dimensional quadrupole ion trap mass spectrometer Journal of the American Society
of~a~ss Spectrometry 2002, 13, 659]





FI~l~m ~it.it Ilri


rT


F1n~i l~I~ TO I~~~t 1 31


Figure 1-5. Electron ionization schematic. [Adapted from
http://www.noble.org/plantbio/MS/iontech~e~tl





F' ?lniti~lit IIi~li~t


1111 111111. 1 CTT~~I: I










~ I
I ~JI rd ;r-]


FI i I1ii nl r iri oi~ -;


I1~ t
cnal ~ -er


Figure 1-6. Chemical ionization schematic. [Adapted from
http://www.noble.org/plantbio/MS/iontech~c~tl









twt ppltt~~~~tttt~~~~tttr~~~c
balln


ush rlatts twl


I

I 1







rid


Irv.-


e;r~p~e~ 8a


Figure 1-7. Illustration of FAB. [Adapted from http://www.chm.bris. ac.uk/ms/theory/fab-
ionisation.html]









lon Formation By ESI


SAdapted frorn Kebarle and Tang, Aral Chern. 65, 2, 972A-985A (1993)

Figure 1-8. Electrospray ionization schematic. [Adapted from
http ://www.noble.org/plantbio/MS/iontech. esi.html]


Taylor Cone










;OOH


.COOH


OCH3


DHB




COOH


COOH


.COOH


4HCCA


3HPA





HOOCCOOH


Succimic Acid


HO CH2

HO-CH


HO CH2


Glycerol


Figure 1-9. Most commonly used MALDI matrices









MAL D I De rorption~llon lmilan


hrOpJ cal fbcuslrg lene


hig
h


Figure 1-10. A schematic diagram of the mechanism of MALDI. [Adapted from
http://www.noble.org/plantbio/MS/iontech~c~tl













DNA


Anc~ibrl


,~T~t~


i lr


Hydro~phoic


IMA~C


Iniuc


Figure 1-11. The chemical and biochemical surfaces available for SELDI from Ciphergen
provides. [Adopted from Tang, N.; Tornatore, P.; Weinberger, S.R. Current
developments in SELDI affinity technology Ma~ss Spectrometry Reviews 2004, 23,
34]


BiiochemicalT Surfalces


Reepu


Enyzme


Chemical Surfaces














muth1, Kd~l l
AC
G C
G A
U-0
A-U


C ~
G mutA3, Kd= >> 200 nM
C-G
C-G
G-C m K,K= 10~t0.7 nM
A-U GA
A-U C
G-C C-G
C-G CGA GU

SLII, K = 3 & 0.5 nM


AC
G C
GU-GA
A-U

U-G


C-G
C-G
G-C
A-U
A-u SLII(S13)
G.C
c-a Kd=> 0n


G C
SLH(G) A
Kd=>>200nM U-G
-GG SLHGC)

C-GC Kd=>>200nM
G-C
C-G
C-G
SLH(-4)-- IC
Kd=410.3nM A-U
G.C- SLH(-2)
C-G Kd 210.4 nM


Figure 1-12. The bulge and stem structure of aptamers. [Adopted from Biroccio, A.; Hamm, J.;
Incitti, I.; Francesco, R.D.; Tomei, L. Selection of RNA aptamers that are specific and
high-affinity ligands of the hepatitis C virus RNA-dependent RNA polymerase
Journal of Virology 2002, 76, 3688]











Juncilns; Internal loops Bullge HI rpln loop





Two-stern ~ ~ ~ ~ ~ ~ ~~9 The-tm Furse smeti yie
-~ ~ 551 AU, UA G 0 Uo G --AA C G U
AU ,orGAG A C U rU

Fiur 1-3 h ari tutr fatmr A ote rmBrciA;H m ,J;Icti .
Fraceco R.D. ToeL eeto fN paer htaeseii n ih




Virlog 2002, 76,, 3688] ~I










C-oneeQeu Pisalndama (o 18 isolum)


Loop 1


Stem 1


lil 18 111 la


X -XX--X-X-X
11 1 15 1 12


XX'- x- X-~ .X`- X- 31


A


Coop 2

Figure 1-14. The pseduknot structure of aptamer. [Adopted from Tuerk, C.; MacDougal, S.;
Gold, L. RNA pseudoknots that inhibit hunman-immunodefficiency-virus type-1
reverse-transcriptase Proceedings of the National Academy of Sciences of the Chrited
States of America 1992, 89, 6988]


GG Ge-ic-A-


AX























15 r-_~;c3


Fiur -1.Th -qate trcur f a pae.[dpe frro Wag K.. M~rdS





Figue 1-in hibt Gqatthrmi exhibitsr ane sptmrucua mAoptif fror DNAn Bichmity. 1993r, 32,

1899]









~.~W.I............lrr ~[I~CC
IRNl\poiFsla
I~R.. ~6~Rldd~L


Random sePquce9
olgonrucleatkle I~bray


Tarlge~lkotdrt EI


ubnaunnd
oligowrwcsoicie


aptarner
candidBalso


ras ougolwidotide
oibray


Clone & Seqluence
Figure 1-16. Generalized scheme indicating the key steps in the SELEX process. [Adopted
from Klug, S.J.; Famulok, M. All you wanted to know about SELEX M~olecular
Biology Reports 1994, 20, 97]









CHAPTER 2
MATRIX-DERIVED RESORCINOL-FORMALDEHYDE POLYMERS: SYNTHESIS VIA
SOL-GEL METHOD AND USAGE IN MALDI-MS SAMPLE PREPARATION

Since the invention of MALDI-MS, it has been used to analyze macro-biomolecules such

as peptides mixtures,76 prOteins,76 Oligosaccharides,7 enzymatic protein digests,78 underivatized

DNA, and oligomers.79 Although MALDI-MS is a powerful analysis method, it has problems

associated with the sample preparation methods. Identifying and solving the problems can

improve the quality of MALDI-MS spectra.

Problems Associated with MALDI-MS Sample Preparation Methods

As a widely used ionization method for mass spectrometric analysis for macromolecules,

MALDI has shortcomings associated with the methodology of sample preparation. These

problems include background interference from matrix molecules when the molecular weight of

the analyte is less than 2-3 times the molecular weight of the matrix molecule74 and poor signal

uniformity of the signal intensity of analyte across the sample from the inhomogeneous

distribution of the matrix-analyte co-crystals. 93 These problems can either interfere with the

interpretation of MS spectra of low-mass analytes or require significant effort to locate the

"sweet spot" both of which routine analysis with MALDI-MS more difficult. Therefore, to

eliminate the interference from matrix molecules and at the same time to increase the uniformity

of the analyte signals intensities in MS spectrum, development of a polymeric substrate which

has the physical properties amenable to MALDI-MS analysis and has the chemical properties to

incorporate matrix molecules into its structure is highly desirable.

Background Interference from Matrix Molecules in MALDI-MS

MALDI ionization coupled with mass spectrometry made the analysis of nonvolatile

molecules possible. In MALDI, an excessive amount of matrix molecules, with the matrix-to-

analyte ratio typically in the range of 100 to 50,000, is mixed with analyte in order to ionize the









analyte particles. The excessive matrix molecules are not anchored to any support, and,

inevitably, the matrix molecules will be released and ionized upon ablation of the laser beam,

and appear in the mass spectrum together with the analyte. Signals produced by the matrix

molecules present in the mass spectrum along with the signals from analyte pose interference in

analysis of low molecular weight analyte.

Much effort has been devoted into Einding the mechanism of MALDI which can lead to the

development of matrix suppression methods. Competition model, in which competition between

matrix and analyte for free protons based on the proton affinity,so, sl is one of them. According

to this model, matrix suppression takes place when adjusting the matrix-to-analyte ratio for every

protonated matrix molecule to interact with at least one analyte with higher proton affinity than

the matrix molecule during desorption. In another proposed mechanism, analyte molecules are

ionized by non-ionic matrix precursor species which are vibrationally and electronically excited

matrix molecules.82 Knochenmuss and Dubois proposed a model combining the previous two

models. In this model excited, but not ionized, matrix molecules are the common precursor for

all subsequent ion products and the simultaneous neighboring presence of two such excitations is

required for ionization of each analyte molecule.83 This model explained matrix suppression

phenomena observed in a medium matrix-to-analyte ratio with relatively low analyte signal

intensity and less matrix suppression in a high matrix-to-analyte ratio. According to this

mechanism, suppression of the matrix signals has to be compromised to get high enough analyte

signal intensity, or vice versa, when only the matrix-to-analyte ratio is involved.

Modiaications other than adjusting the matrix-to-analyte ratio in MALDI sample

preparation have been made in order to reduce the matrix interference in mass spectrometric

analyses. These modifications include adding iodine to the matrix to suppress the signals from









the matrix-related ions and increase the signals from analyte ions,84 USing alkali

dihydroxybenzoic acid salts as cationizing agents to suppress the DHB related ions, mixing

graphite powder with glycerol to reduce the interference,86, 90 and using nitrocellulose as a

substrate together with matrix to eliminate the matrix signals.8 In each case the modification

can reduce the signal intensities from matrix molecules but cannot completely eliminate them.

In all those methods, the matrix molecules are not covalently attached to any support but merely

interact with the support by dipole-dipole interaction and van der Waals force; which are at the

magnitude of one-thousandth and one-ten thousandth of the strength of the covalent bond,

respectively. Therefore, are easy to be ablated and ionized when the energy of the laser beam

shot on the target is much higher than the energy required to break these interactions.

In an effort to fundamentally solve the matrix related problem, Lin and Chen incorporated

matrix molecules into the gel formed from the tetraethoxysilane (TEOS) precursor via sol-gel

method." Total elimination of the matrix molecule signals was observed when small proteins,

peptides, amino acids, and small organic were analyzed. In the experiment, the matrix (DHB)

was mixed with the monomer precursor to form the sol-gel solution and was claimed to become

part of the polymer structure after the solution is polymerized. The mechanism of the

polymerization reaction was proposed as shown in Figure 2-1. The hydroxyl groups on the

matrix molecules can form covalent bonds with the hydroxyl groups in the silicic acid and thus

covalently incorporated into the TEOS polymer as shown in Figure 2-2.

Since the energy required to break the covalent bond between the matrix molecule and the

polymer is greater than the laser energy used in MALDI, trapping the matrix molecules

covalently into the polymeric structure make it possible to totally eliminate the matrix ions'

background interference in a MS spectrum. One problem associated with the TEOS gel, which









was not mentioned in the paper, is that cracks formed when an aqueous solution was applied

onto it since the structure of the gel is not strong and can be easily broken. Developing a

different matrix-embedded polymer which has better physical properties than the TEOS polymer

is valuable for the polymeric substrate to be useful for MALDI-MS analysis with analytes in

aqueous solution.

The Uneven Distribution of Analyte Molecules with the Traditional MALDI-MS Sample
Preparation Methods

There have been many investigations on the effect of MALDI sample preparation methods

on the quality of mass spectrometric analyses. MALDI-MS analyses results have strong

dependence on the sample preparation method. Every single choice such as the solvent, matrix,

and pH in sample preparation procedure can affect the outcome of the MALDI measurements.89,

90, 91, 92, 93, 94, 95, 96 The different combination of matrices, concentration of matrix, and solvent

produce different distribution of matrix molecules across the sample as shown in Figure 2-3.97

The problem which concerns the mass spectrometry analysts the most is the poor spot-to-spot

and sample-to-sample reproducibility caused by inhomogeneous distribution of the matrix-

analyte co-crystal.93

The methods used for the application of the analyte sample includes the rudimentary

"dried-droplet" method which is to deposit matrix and sample mixture on the target with the

matrix-to-analyte ratio as 103 to 105 and let it dry.98 This method is also called the "air dried"

method.99 The formation of small co-crystals which contain matrix molecules and analyte

molecules in the dried-droplet method is crucial for the ionization of the analyte molecules. The

crystals formed in the sample prepared from dried-droplet method are not evenly distributed

throughout the sample plate. The analyte-matrix co-crystals are found mostly at the edge of the

droplet when DHB is used as matrix1oo, 101 as shown in Figure 2-4102 according to the partitioning









process when solvent slowly evaporates. When the solvent slowly evaporates the analyte

molecules are excluded from the matrix crystal.99 An analyst would, therefore, to find "sweet

spot" by scanning the sample surface.

A "thin-layer" sample preparation method has been developed based on the dried-droplet

method1oo using a volatile solvent to create homogeneous matrix microcrystals because quick

evaporation of the solvent can minimize the partition process which occurs under the slow

evaporation process. A drop of matrix solution in volatile solvent was deposited onto a target

plate and was allowed to spread and dry quickly to form a thin layer of homogeneous matrix

microcrystals. Then a drop of analyte solution was deposited on the top of the matrix layer.92

The spectra obtained from the thin-layer method produce more uniformed distribution of the

analyte-matrix crystals than from the dry droplet method, therefore more uniformed analyte

signal intensity across the sample (Figure 2-4).

"Sandwich" sample preparation method is a combination of the thin layer and the dried-

droplet method.98 It is called sandwich because sample solution is applied on one layer of matrix

crystals, then another matrix solution is added on the sample spot. This method has better

mapping results than the dried-droplet method.

To utilize the quick evaporation of the solvent further, electrospray sample preparation

method was used to further improve homogeneity of the components distribution in MALDI-ToF

MS than the use of volatile solvents and/or rapid evaporation under vacuum or heating and the

small and evenly distributed droplets can.99 This method produced much better spot-to-spot and

sample-to-sample reproducibility in the mass spectra than dried-droplet method.

Since the selection of the matrix is very importance in some cases, experiments using more

than one matrix for the sample preparation have been also conducted.94 These improvements are









made on a case-by-case basis. Limiting the size of the droplet is another technique that has been

tested to improve the MALDI-MS analysis quality.103

Over all, there is not a universal sample preparation method that produces high quality

mass spectra for variety of analytes. Mass spectra quality can be improved by modifications

made to the sample preparation methods. Elimination of the poor spot-to-spot reproducibility of

the analyte molecules (the uneven distribution of analyte across the sample) is impossible with

the efforts have been done, because this problem is related to the process of the crystallization.

In the matrix-embedded polymeric MALDI-MS substrate, the matrix molecules are evenly

distributed throughout the polymer structure on the molecular level and cannot be changed

during the sample preparation process and no analyte-matrix co-crystals are formed, thus, the

matrix-embedded polymer has the capability of solving this problem.

Introduction to the RF Polymer

According to the discussion above, a matrix-embedded polymeric substrate is the solution

to the background noise from matrix molecules in mass spectra and the poor spot-to-spot and

sample-to-sample reproducibility of MALDI mass spectra. Matrix-derived TEOS polymer was

used by Lin and Chen75 and was synthesized and tested in our lab. The matrix-derived TEOS

polymer was able to significantly lower the matrix signals when using MALDI-MS. Although

the TEOS gel has the advantage of lowering the chemical noise from matrix molecules, the

structure of the gel is fragile and is easy to break apart, cracks when aqueous solutions were

deposited on it. This shortcoming made it a less than ideal substrate for MALDI-MS since

aqueous solutions are usually used and made it impossible for easy manipulation and further

chemical modification. Resorcinol-formaldehyde (RF) polymer was finally chosen because of

its tough physical structure and abundant phenol groups on it which render the capability of









being chemically modified. Resorcinol-formaldehyde resins have been used as adhesion

promoters to increase adhesion of rubber to fabric or metal.

RF sol-gels were first synthesized by Pekala and co-workers according to a hydrolysis-

condensation reaction mechanism (Figure 2-5).104 Figure 2-6 shows the base catalyzed

polymerization reaction of RF polymer. The "R" in the RF polymer is referred to "resorcinol",

the "F" in the RF polymer is referred to "formaldehyde". A patent was filed by Pekala on the

synthesis of RF polymer.104 Resorcinol-formaldehyde resins have been used as adhesion

promoters to increase adhesion of rubber to fabric or metal.

A lot of research has been conducted since then on the effect of different reaction

conditions on the final structures of the polymers which affect the physical, chemical, and

electrochemical properties ofRF aerogels and xerogels.los Many patents have been filed and

approved for RF polymers with different reaction conditions. Those reaction conditions include

the ratio between the resorcinol and formaldehyde, the solvent and catalyst, the initial pH of the

reaction, the ratio between the catalyst and the resorcinol, the temperature of the reaction, and

initial ratio between the reactants and the solvent.106 With these variables there are tremendous

possibilities of combinations of the conditions which make it a very versatile reaction that can be

tailored for the desired properties.

The RF polymers are categorized according to the solvent used. The gels with water

solvent are called hydrogels or aquagels; the gels with organic solvents are called lyogels.107 The

initial reactants and the solvent ratios of the reaction affect the final density of the gel's size.10s

Using higher concentrations of solvent can result in "dilute effect" which results in increased

particle size.10s Using the higher concentrations of the reactants can result in higher density of

the formation of the RF crosslinked clusters.









The most commonly used stoichiometric resorcinol / formaldehyde (R/F) molar ratio is 1:2

104 Using more formaldehyde than twice the amount of the resorcinol as the starting material

would cause a "dilute effect"-the effect caused by higher solvent concentration.'os

The most commonly used catalyst is sodium carbonate. The molar resorcinol -to-cataly st

(R/C) ratio is usually between 50 and 300, or can be as high as 1500. The lower the R/C ratio,

the smaller the polymer particles and the higher density of the gels. In some cases acids such as

HCIO4107 Or HNO3109 were used as catalyst, respectively, with reduced gelation time.110 Using

an acid catalyst combined with low concentration of reactants can result in small, smooth, fractal

aggregates of gel particles.10

The pH should be controlled between 5.4 and 7.6 "1 since high pH can hinder the

polymerization-condensation reactionl12 and reactants precipitate at low pH113. Diluted acid

such as HNO3, HCIO4 and bases such as NH40H can be used to adjust the initial pH of the

reaction solution.

Systematic Investigation in Preparation of the RF and Matrix-Derived RF Polymers

To address the problems facing analysts working on MALDI-MS, RF polymers were

synthesized to be suitable as substrate for MADLI-MS. The RF polymers developed were wet

gels without any further drying processes. These polymers are made via the quick and easy sol-

gel method which makes it easy to be customized with any desired composition, on any surface,

and any shape and size. The solutions can be stored in freezer for about 6 months. The polymer

has a very strong structure due to the 1,3 bonding position on the aromatic rings in the polymer.

The polymer did not crack upon the deposition of aqueous solution and can only be broken by a

severe impact.

Usually the resorcinol formaldehyde ratio is 1: 2, both acid and base were used as a

catalyst with different molar ratio of resorcinol/catalyst (R/C). Water, acetone, methanol,









ethanol, n-propanol, and iso-propanol can all be used as the solvent. The catalyst used, ratio of

R/C, the solvent, and concentration of the starting materials determine the final structure and

properties of the gel. Eventually, acid (HC1) was chosen as the catalyst to provide H+ ions which

are necessary for MALDI; acetone was chosen as solvent for quick polymerization; different

matrices were used to investigate the ionization capability. The matrix molecules used for

MALDI with similar structure as the RF polymer can be embedded into the polymer via covalent

bonding.

The ratio of the resorcinol and formaldehyde is chosen as 1:2 since the ratio of the

resorcinol and formaldehyde in the structure is 1:2. Water, alcohol, and acetone have all been

used in the literature and suitable as the solvent of the RF polymerization reaction. Different

solvents have been tested and eventually acetone was chosen as the solvent throughout the rest of

the synthesis. With acetone as solvent, the temperature of the curing stage can be lowered

significantly (from 80 oC to 40 OC) with reasonable gelation time.104 The initial concentration of

the reactants was chosen as (R+F)/solvent 30% w/w. The sodium carbonate was first chosen as

the catalyst because it is the most commonly used in the previous literature. The higher the ratio,

the smaller the polymer particles and the higher density of the gel. The R/C ratio can be as high

as 200:1 was chosen as the R/C ratio. Since an acidic polymer can provide H+ which is

necessary in MALDI to ionize analyte, acids were tested. HCIO4 WAS used as first acid catalyst

because it was used in the literature. Later, HCI was used throughout the rest of the synthesizing

because it is a commonly used acid in lab, although there is no evidence in the literature about

using HCI as catalyst in RF polymer synthesis.

The initial synthesis of the RF polymer was tested with the following procedure: 4.00 g

Resorcinol (0.036 mol), 5.00 mL formaldehyde (0.072 mol), 12.00 mL acetone, and 0.0078 g










Na2CO3 (0.036 mmol) were added to the flask which was sealed and immersed in a 450C water

bath. The formaldehyde-to-resorcinol ratio is 2:1 and the resorcinol-to-Catalyst ratio is 1000:1.

Every half hour several drops of solution were taken out of the flask and deposited on a glass

slide. The solutions taken out at one hour and after one hour all successfully polymerized. Thus,

the reaction time was then determined as one hour. The solution is stored in a vial in the freezer

with -200C. The solution can be used for six months after synthesis. This RF polymer was

given the serial number as RF-1.

There were thirty nine experiments that have been conducted in an effort to systematically

investigate the properties of the RF polymerization process. The tests were conducted using a

combination of different solvents such as acetone, ethanol, and water; different

resorcinol/catalyst ratios; different catalysts; different matrices with varying concentration; and

addition of different doping agents. Every RF solution was named with "RF" followed by the

serial number assigned from one to thirty nine.

Thirteen RF polymers have been synthesized with Na2CO3 aS a catalyst and acetone as

solvent as shown in Table 2-1. These polymers have a different resorcinol-to-catalyst ratio, with

the addition of different matrices with varying concentration and the addition of doping agents in

order to bring functional groups into the polymer.

The resorcinol-to-catalyst was changed to 500 and 100 which are assigned the serial

numbers RF-2 and RF-39, respectively. These two polymers, together with the RF-1 polymer,

all polymerized. After the successful synthesis of the RF-1 and RF-2 polymers, RF-3 polymer

was synthesized with 1000 ppm of DHB which also polymerized as expected. The RF-20

polymer was able to polymerize when the concentration of the DHB was raised to 40,000 ppm

with the high catalyst concentration (resorcinol-to-catalyst ratio as 100).









After DHB was successfully incorporated into the RF polymers, CHCA was tested for

incorporation into the RF polymers with different concentrate on, different resorcinol -to-cataly st

ratio, and different concentration of 4-(imidazole-1 -yl)phenole with or without being cued with

HCI which were assigned the serial numbers from RF-27 to RF-35. RF-29, RF-30, and RF-31

polymerized, while others did not. The R/C can be as high as 1000 for the sol to polymerize as

long as there is no imidazole added even when there is matrix added. When there is imidazole

added, the R/C can not be higher than 50. The concentration of the matrix can be as high as

10,000 ppm when CHCA was used.

Ethanol was tested as a solvent for RF-4 with 1000 resorcinol-to-catalyst ration and 1000

ppm DHB, and RF-5 with 500 resorcinol-to-catalyst ration and 1000 ppm DHB. Both of the sol

crystallized instead of polymerized (Table 2-2).

Nine polymers (RF-6 to RF-12, RF 15, and RF 19) were tested in the water as solvent

with varying concentration of Na2CO3 aS catalyst, varying concentration of DHB, and addition of

Fe2 13-H20 in order to add metal chelating capability as shown in Table 2-3. All of these

polymers polymerized with varying reaction time.

HCIO4 WAS the first acid catalyst used in the synthesis ofRF-12 polymer with water as

solvent. The necessity to test an acid catalyst is based on the need of the protons in order to

ionize analyte molecules in MALDI. HCIO4 is documented as an acid catalyst for the synthesis

of RF polymer, thus was chosen.104 Since storing HCIO4 requiring extra caution, HCI was test as

catalyst in the synthesis of RF polymer. Although there is no evidence in the literature about

using HCI as a catalyst for RF synthesis, the experiment was a success.

There are thirteen polymers (RF-14, RF-16 to RF-18, RF-21 to RF 26, and RF-36 to RF

38) that were synthesized with HCI as catalyst for its proton; with acetone as solvent for the fast









evaporation during the gel process; and with different matrices and varying concentration, with

addition of imidazole, EDTA, and vinylimidazole; as shown in Table 2-4. Only seven of tem

polymerized (RF-14, RF-16, RF-21 to RF23, RF-25, and RF-3 8).

Among the thirty nine tests twenty four of them were able to polymerize. Nineteen of the

polymerized solutions contain matrix. For each matrix-containing polymer, mass spectrometry

was conducted with the sol-gel solution deposited on a stainless steel probe. After the solution

was polymerized, the polymer was taped onto the stainless steel with double-sided tap, and then

100 ppm spiperone in methanol was deposited on the polymer. After the spiperone solution

dried the sample was analyzed in a custom built MALDI ion trap mass spectrometer.

Among the twenty four matrix-embedded RF polymers, only the CHCA-embedded RF

polymers produced MS signals with 100 ppm spiperone. As the concentration of the CHCA

increases, the intensity of the analyte signal increases under the same analysis condition. There

is an up limit of the concentration of the matrix in the solution for the matrix-embedded

resorcinol formaldehyde solution to polymerize which is 20,000 ppm. When 40,000 ppm of the

matrix is added, the polymer (RF-24) forms a lot of cracks which is not suitable as a MALDI-MS

substrate. RF-16, which contained 20,000 ppm CHCA, produced satisfactory MS analysis result

and is used throughout the rest of the MS analysis.

The solution is colorless while polymer polymerized from the solution is dark red. The

color change indicates the formation of bonds on the conjugated structure which is the aromatic

rings in this reaction. The mechanism of the resorcinol formaldehyde polymerization shown in

Figure 2-5 is proposed by Lin and Ritter "l which has an addition and a condensation step. Base

was used as the catalyst in the addition step. In the second step the acid was used as catalyst to

cure the solution. In the experiment reported here only base, only acid, and base first then acid









as catalyst(s). In all cases there are successful examples, with the base catalyzed solution took

longer time than the acid catalyst solution to polymerize. Figure 2-6 is the mechanism I

proposed for the base-only catalyzed resorcinol formaldehyde polymerization reaction.

Analysis of the Spiperone on the CHCA Derived RF Polymer for Lower Background
Interference from Matrix Molecules

This comparative analysis was performed with a microprobe quadrupole ion trap mass

spectrometer (QITMS) which was built in the Yost laboratory by Christopher Reddick in 1997.11

The UV laser (Laser Science Inc. model VSL-337ND) used for all MALDI MS analysis has 20

Hz pulsed nitrogen laser with a wavelength of 337. 1 nm and a 3 ns pulse width."l The maximum

energy output of the laser is >250 CIJ/ pulse with a peak power of 85 kW."1 Since the laser is

near-diffraction limited, the beam can be focused to a diameter within a few times the laser' s

wavelength; for the studies reported here, the laser spot size ranged from a diameter of 25 to 50

Clm. The laser is focused into the mass spectrometer chamber by a single fused silica lens

(Melles-Griot) with a focal length of 25.4 cm as illustrated in Figure 2-7.11 The laser is

externally triggered after a 1 ms delay by a Wavetek model function generator so the ion source

gate has fully opened before ionization."l The position of the sample plate can be adjusted

manually along x and y axes." Photosensitive paper is used to determine the position of the

laser beam with respect to the sample plate to align the incident laser light." The software used

to control the mass scan and data acquisition has been developed in our research group which is

used to control auxiliary modulation frequency and amplitude to extend mass range.ll As shown

in Figure 2-8, ions are produced during the MALDI process, and then bend 900 into the ion trap

by using a DC quadrupole deflector. This is to reduce the possibility of neutral collisions

occurring within the ion trap caused by neutral molecules. After being ej ected by the QIT from

the exit end, ions are detected using a conversion dynode and electron multiplier which is behind









the exit end electrode of the QIT.11 The data taken by this instrument was converted to excel file

to generate the spectrum.

The analyte used here, spiperone, is an antidepressant. Figure 2-9 shows the fragmentation

path ways of spiperone. Different matrices such as DHB, sinapinic acid, caffeic acid, and CHCA

have been embedded into the RF polymer and the matrices-embedded polymers have been tested

for the ability to ionize spiperone. MS analysis showed that the RF polymer with CHCA was the

only combination that generated a spiperone MS signal (Table 2-1 to 2-4). This phenomenon

can be explained by the result from the UV-vis spectra of the various matrix-embedded RF

polymers (Figure 2-10). The discontinuity of spectrum at 350 nm is caused by the changing of

light source from UV to visible. The CHCA had the highest absorption at 337 nm at the same

concentration compare with other matrices; the absorbance of the other matrices may render

them less effective MALDI matrices. Figure 2-11 shows the UV absorbance of CHCA in

acetone (purple), the pure RF polymer (blue), and the RF polymer embedded with CHCA

(orange) which were analyzed on Varian Cary-100 Conc UV-visible spectrometer from Varian

(Palo Alto, CA). The CHCA-embedded RF polymer showed UV absorption at 337 nm which is

not the case in the pure RF polymer. Since the solution of CHCA in acetone shows high

absorbance around 337 nm range, it is obvious that the CHCA-embedded RF polymer obtained

its high UV absorbance at 337 nm from embedding CHCA into it.

A polymerized RF 16 (20,000 ppm CHCA embedded RF polymer) pellet was attached

onto the stainless steel MALDI microprobe with double-sided tape. A 0.2 CLL volume of 100

ppm spiperone was deposited onto each polymer directly with pipette. After five minutes the

sample is ready for MALDI-MS analysis.









A 0.2 CIL volume of 2mM CHCA was also deposited onto the stainless steel microprobe

with pipette; it was dry after 5 min past, then 0.2 CIL of 100 ppm spiperone was deposited onto

the matrix spot with pipette.

The results are shown in Figure 2-12 a and b, each showing the average of six spectra. The

spectrum of 100 ppm spiperone with 20,000 ppm CHCA on the stainless steel showed [M+H]+

(m/z 396) peak together with fragments from CHCA (m z 123.2 and m/z 172.5) which pose

interference to the interpretation of the spectrum while the spectrum of 100 ppm spiperone on

2% CHCA-embedded RF polymer showed high [M+H]' (m/z 396) peak and no noticeable

fragment from CHCA. By embedding matrix molecules into RF polymer, the matrix molecules

form covalent bonds with the polymer and are fixed into the polymer. Figure 2-13 is the

illustration of one of the several possible bindings. Since the covalent bond between the polymer

structure and the matrix molecules are much stronger than the hydrogen bond or the van der

Waals force and thus requires much more energy to break the bond, the polymer keeps the

matrix from being ablated while the matrix still has the ability to absorb laser energy to transfer

to spiperone molecules and provide H+ for ionization. Moreover, while using the same laser

power for both analyses, the spectrum of spiperone on CHCA-embedded RF polymer showed

lower intensities of daughter ions of spiperone (m/z 291.5, 238; m/z 165.5, 885) than the

spectrum of spiperone with CHCA on stainless steel (m/z 291.5, 700; m/z 165.5, 3760).

Comparing these two spectra, the spectrum obtained with spiperone on the CHCA-embedded RF

polymer showed improved quality.

Analysis of the Spiperone on the CHCA-Derived RF Polymer vs. Analysis of the Spiperone
on the Stainless Steel Using CHCA as Matrix for More Uniform Analyte Signal Intensity
Across the Sample

The LTQ linear ion trap with vMALDITM ion source from ThermoFinnigan (Figure 2-14)

which is capable of imaging was used in this experiment. The laser power is 250 CLJ/ pulse. The









laser beam was set up to move in 120 Cpm steps across the sample spot (3mm in diameter) and the

data are collected automatically by the XcaliburTM data system to generate the image of the

signal intensity of the analyte throughout the analyzed area.

The dried-droplet method was chosen for the comparison experiment because this method

is still widely used for MALDI-MS analysis despite the fact that the matrix-analyte co-crystals

are unevenly distributed across the sample which causes uneven distribution of signal intensity

of analyte across the sample spot. It would be exciting to see the improvement the CHCA-

embedded RF polymer can make by simply using dry droplet method, and make the using dried-

droplet method by mass spectrometry analyst with less effort needed for search of "sweet spots".

For comparison of the dry droplet on stainless steel analysis vs. spiperone solution on

CHCA embedded RF polymer analysis; 0.2 CLL of 0.5 M CHCA in MeOH was applied on the

stainless steel; after the matrix solution was dried, 0.2 pIL of 100 ppm spiperone solution in

MeOH was applied on the dried matrix spot and on the CHCA embedded RF polymer,

respectively. The microscopic image of the sample spot prepared with dried-droplet method

before MS analysis is shown in Figure 2-15. After the sample plate was inserted into the

instrument, the image of the sample plate was taken by the CCD camera and a circle was drawn

by hand. The instrument started to run MALDI analysis throughout the circled area with each

analysis spot 120 Cpm apart both in x direct and y direction. The position parameters (x- and y-

axis) were recorded as attachment of each spectrum for each analysis spot by the vMALDI

software to be used to generate the image of the sample spots with the information of mass

spectra after the analysis. Both image of the distribution of intensity of TIC (total ion counts)

and image of the distribution of intensity of the ions within certain range can be generated. RSD,

relative standard deviation which is calculated by dividing the standard deviation by the average









value, was calculated for m/z 304 (for cocaine) by surfer 8 based on all the spectra taken

throughout the circled area and reported in "grid report". Typical results from the sample

prepared with dry droplet method on the stainless steel sample plate are shown in Figure 2-16

with relative standard deviation (RSD) as 228%.

Figure 2-17 is the microscopic image of the CHCA-embedded RF polymer pellet after the

deposition of 0.2 CLL of 100 ppm spiperone. Same mass spectrometry analysis as for the dried-

droplet of spiperone on stainless steel was run for dried-droplet of spiperone on CHCA-

embedded RF polymer. Both image of the distribution of intensity of TIC (total ion counts) and

image of the distribution of intensity of the ions within certain range can be generated. RSD was

calculated for m/z 304 (for cocaine) by surfer 8 based on all the spectra taken throughout the

circled area and reported in "grid report". Typical results from the sample prepared with dry

droplet method on the CHCA-embedded RF polymer are shown in Figure 2-18 with relative

standard deviation (RSD) as 103%.

Compared to the results obtained from the dry droplet method on the stainless steel (Figure

2-16), the results from the dried-droplet on the CHCA-embedded RF polymer (Figure 2-17)

produced more even distribution of the analyte across the sample with the RSD of the sample

prepared on the CHCA-embedded RF polymer is about the half of that with the sample prepared

with the dried-droplet method on stainless steel; thus, analyte signals can be detected easily

without searching for the "sweet spot". The more even distribution of the analyte signal resulted

from the incorporation of matrix molecules into the rigid polymeric structure of RF polymer,

since the matrix molecules are evenly distributed throughout the whole polymeric structure on

the molecular level. The morphology of the analyte crystals only depends on the distribution of

the analyte across the sample and no concern of distribution of matrix molecules is necessary.









Conclusion

With the 1, 4 polymerization position on the aromatic ring, RF polymers can withstand the

harsh condition of chemical modification and washing with aqueous solution which is the

condition for most biological sample. With the aromatic groups in the RF polymers and reaction

sites on the aromatic rings it is possible to incorporate aromatic groups containing matrix

molecules in the polymeric network with covalent bonds. By trapping the matrix molecules in

the polymeric network with covalent bonds it still provides the assistant to ionization but can

reduce the amount of matrix molecules been ablated by laser which can interfere with the MS

signals from analyte with low molecular weight because it needs more energy to break a covalent

bond than to break a intermolecular noncovalent bond.75 By incorporating the matrix molecules

into the polymeric network structure, it can help solve the problem of unevenly distributed signal

throughout the sample plate which is always a problem in MALDI-MS analysis. Many different

sample preparation methods were developed to minimize the problem. The new resorcinol-

formaldehyde (RF) polymers developed have physical and chemical properties suitable for

embedding matrix molecules in the substrate for lowering the background noise from matrix

molecules. With the nature of the matrix-embedded polymer the matrix molecules distributed in

the polymer evenly at the molecular level, thus provide the possibility to produce uniform signal

intensity of analyte across the sample. This property also provides the possibility for

quantization of the analyte in MALDI which is an important issue related to MALDI sample

preparation.










(a) Mydes~esis




Br~relrrn inn .2H cm A
HO-~ 5 f+t I-l K-i--G~-5RGH


DHl CFUH ONr I- emp


Figure 2-1. The mechanism of the TEOS sol-gel reaction. [Adopted from Lin, Y.; Chen, Y.
Laser desorption/ionization time-of-flight mass spectrometry on sol-gel-derived 2,5-
dihydroxybenzoic acid film Analytical Chemistry 2002, 74, 5793]


~L~lm



B


































Figure 2-2. Illustration of incorporation of DHB molecules in the TEOS gel. [Adopted from
Lin, Y.; Chen, Y. Laser desorption/ionization time-of-flight mass spectrometry on
sol-gel-derived 2,5-dihydroxybenzoic acid film Analytical Chemistry 2002, 74, 5793]





















TCWT


1


t .
'1


II
i.


Figure 2-3. Microscope pictures showing the sample surfaces of manually prepared samples
with the dried-droplet (DD) or the seed layer (SL) method using (a) a-cyano-
hydroxycinnamic acid (CHCA) DD, (b) CHCA SL, (c) sinapinic acid (SA) DD, (d)
SA SL, (e) ferulic acid (FA) DD, (f) FA SL. [Adopted from Onnerfjord, P,.;
Ekstrom, S,.; Bergquist, J.; Nilsson, J.; Laurell, T,.; Marko-Varga, G. Homogeneous
sample preparation for automated high throughput analysis with matrix-assisted laser
desorption/ionization time-of-flight mass spectrometry Rapid' Communications in
M~ass Spectrometry 1999, 13, 315]























AP


A


11-8~
0-70000


Figure 2-4. MALDI ion images showing intensities for selected peptides using (A) the thin-film,
a-CHC sample and (B) the dried-droplet, DHB sample. Approximately 250 fmol of
al.3, 250 fmol of AP, and 1 pmol of ELH were prepared with each matrix. The black-
to-red color map corresponds to the arbitrary intensity values specified for each
peptide, while the grid lines correspond to 50 Cpm increments within each image.
[Adopted from Garden, R.W.; Sweedler, J.V. Heterogeneity within MALDI samples
as revealed by mass spectrometric imaging Analytical Chemistry 2000, 72, 30]











1. Addition Reaction


OH OHC


2. Condensation Reaction

OH OH


CH20H H2


2 1
H/CH


I::HOH


CH20H


Figure 2-5. Mechanism of the polymerization of RF polymer proposed by Lin and Ritter.
[Adopted from AL-Muhtaseb, S.A.; Ritter, J.A. Preparation and properties of
resorcinol-formaldehyde organic and carbon gels Advanced Materials 2003, 15, 101]
















O O


OH, _OH


OH
B

aiOH


OI OO


H HH,





CH2CH2

OH OH,



O O


Figure 2-6. Base-catalyzed RF polymerization.






























Laser Microprobe
lon Trap Mass Spectrom eter


Figure 2-7. Cross-section of the ion trap mass spectrometer. [Adopted from Reddick, C.D. The
detection of pharmaceutical drug compounds from intact biological tissue by matrix-
assisted laser desorption ionization method (MALDI) quadrupole ion trap mass
spectrometry 1997 PhD dissertation. University of Florida, Gainesville, Florida.]













Laser Beam in Ions out











Figure 2-8. Illustration of the path of laser and ions in mass spectrometer. [Adopted from
Reddick, C.D. The detection of pharmaceutical drug compounds from intact
biological tissue by matrix-assisted laser desorption ionization method (MALDI)
quadrupole ion trap mass spectrometry 1997 PhD dissertation. University of Florida,
Gainesville, Florida.]










[Spiperone + H]' m/z 396.49


OF


m/z 165.19


H
N
H


m z 291.35


m/z 232.31 m/z 260.36


Figure 2-9. Fragment path ways of spiperone.


HO















4.5 -1 I~,nlu,uuu ppm cl-los In re- gel

4.0

3.5
10,000 ppm Sinapinic acid in RF gel
a 3.0

E2.5 -1nl 10,000 ppm caffeic acid

<(2.0 -r/ 10,000 ppm DHB in RF gel

1.5

1.0 RF gel

0.5

0.0
250 300 350 400 450 500
Wavelength (nm)

Figure 2-10. UV absorbances of the various matrix-embedded RF polymers.












10




CHCA in acetone solution


337
S6-
r RF polymer
-85

<4-

3 / 10,000 ppm CHCA in RF polymer







250 300 350 400 450 500
Wavelength (nm)
Figure 2-11i. UV absorbances of the RF polymer (blue), CHCA-embedded RF polymer (orange),
and CHCA solution (pink). The spike at 320 and 350 is caused by the instrument
used for analysis.












8000


7000


6000


5000


4000


3000


2000


1000



5




8000


7000


6000


5000


4000


3000


2000


1000


[spiperone + H1

A 396.5


















[spiperone + H]+ C13H17N30
155 [spiperone + H]+ C7H,


0O 100 150 200 250 300


350 400
m/z


450 500 550 600 650


[spiperone + Hi
396.6




[CHCA + H]+ C3H2NO



[spiperone + H]+ C13H17N30

165.3


[CHCA +H1' H20




I spiperone + Hi' C7H7
123.2
172.529.


50 150 250 350 450 550 650
m/z

Figure 2-12. Mass spectrum of the 100 ppm spiperone on A) 20,000 ppm CHCA embedded RF
polymer, and B) stainless steel with CHCA as matrix.





























Figure 2-13. CHCA embedded in the RF polymer structure.





































Figure 2-14. LTQ with vMALDI ion source from ThermoFinnigan.





































Figure 2-15. The microscopic image of the sample well after the dry droplet sample preparation
before MS analysis (3 mm in diameter).










































Y (micron) 2000,
B 1500,
1000
500,









r 5.000e+006\





0.000e0000"

1000
1500
.2000
X (micron)
2500
3000

Figure 2-16. The distribution of the spiperone intensity (m z 396) across the sample using dry
droplet method on stainless steel. A) 2-D image. B) 3-D image.




































Figure 2-17. The microscopic image of the RF-16 polymer pellet after the deposition of 0.3 CLL
of 100 ppm spiperone before MS analysis.














































400 1500 1600 1700 1800 1900 2000


01 1


zouu0


Y (micron) 2000
1500J1
loo
soo0 ,


II ~1


1.000e+007


5.000e+006


)1
c,
V)
r

r


Figure 2-18. The distribution of the spiperone intensity (m z 396) across the sample using dry
droplet method on RF-16 polymer pellet. A) 2-D image. B) 3-D image.










Table 2-1. RF polymers with acetone as solvent and Na2CO3 aS catalyst: 4.00 g Resorcinol +
5.00 mL formaldehyde in 12.00 mL acetone with Na2CO3 as catalyst


Catalyst
0.0078 g
Na2CO3 (/
RF-1 1000)
0.0156 g
Na2CO3 (/
RF-2 500)

0.078 g Na2CO3
RF-39 (R/C 100)
0.0156 g
Na2CO3 (/
RF-3 500)

0.039 g Na2CO3
RF-20 (R/C 100)
0.0156 g
Na2CO3 (R/C
RF-27 500)
0.0156 g
Na2CO3 (R/C
RF-28 500)
0.0156 g
Na2CO3 (R/C
RF-29 500)

0.156 g Na2CO3
RF-30 (R/C 50)

0.156 g Na2CO3
RF-31 (R/C 50)

0.078 g Na2CO3
RF-32 (R/C 100)

0.078 g Na2CO3
RF-33 (R/C 100)

0.034 g Na2CO3
RF-34 (R/C 200)

0.034 g Na2CO3
RF-35 (R/C 200)


Matrix


no


no


no


0.0168 g DHB (1000 ppm)


Polymerized


Signal


0.672 DHB (40,000 ppm)

0.4146 g CHCA (20,000 ppm) + 0.3492 g 4-
(imidazole-1 -yl)phenole (20,000 ppm)
0.4146 g CHCA (20,000 ppm) + 0.3492 g 4-
(imidazole-1 -yl)phenole (20,000 ppm) 2 h later add 1.6
mL 0.1 M HCI to cue
0.2073 g CHCA (10,000 ppm) + 0.2746 g 4-
(imidazole-1 -yl)phenole (16,000 ppm) ), 2 h later add
1.6 mL 0.1 M HCI to cue

0.2073 g CHCA (10,000 ppm) + 0.2746 g 4-
(imidazole-1 -yl)phenole (16,000 ppm)
0.2073 g CHCA (10,000 ppm) + 0.2746 g 4-
(imidazole-1 -yl)phenole (16,000 ppm), 2 h later add
1.6 mL 0.1 M HCI to cue

0.2073 g CHCA (10,000 ppm) + 0.2746 g 4-
(imidazole-1 -yl)phenole (16,000 ppm)
0.2073 g CHCA (10,000 ppm) + 0.2746 g 4-
(imidazole-1 -yl)phenole (16,000 ppm), 2 h later add
1.6 mL 0.1 M HCI to cue

0.2073 g CHCA (10,000 ppm) + 0.2746 g 4-
(imidazole-1 -yl)phenole (16,000 ppm)
0.2073 g CHCA (10,000 ppm) + 0.2746 g 4-
(imidazole-1 -yl)phenole (16,000 ppm), 2 h later add
1.6 mL 0.1 M HCI to cue


N/A



N/A










Table 2-2. RF polymers with ethanol as solvent and Na2CO3 aS catalyst: 4.00 g Resorcinol + 5.00
mL formaldehyde in 12.00 mL ethanol


Catalyst
0.0078 g Na2CO3
RF-4 (R/C 1000)
0.0156 g Na2CO3
RF-5 (R/C 500)


Matrix


Polymerized


Signal


0.0168 g DHB (1000 ppm)

0.0168 g DHB (1000 ppm)


Crystalline no

Crystalline no










Table 2-3. RF polymers with water as solvent and Na2CO3, Or HC1, or HCIO4 aS catalyst: 4.00 g
Resorcinol + 5.00 mL formaldehyde in 12.00 mL H20
Catalyst Matrix Polymerized Signal
0.0156 g Na2CO3
RF-19 (R/C 50) no too quick no
0.0156 g Na2CO3
RF-6 (R/C 500) 0.0168 g DHB (1000 ppm) yes (5 hrs) no
0.0156 g Na2CO3 0.0168 g DHB (1000 ppm) + 0.266 g of
RF-7 (R/C 500) FeCl3-6H20 (1000 ppm) too quick (15 mins) no
0.0156 g Na2CO3 0.0168 g DHB (1000 ppm) + 0.0266 g
RF-8 (R/C 500) FeCl3-6H20 (100 ppm) yes (2 hrs) no
0.0156 g Na2CO3 0.0336 g DHB (2000 ppm) + 0.0266 g
RF-9 (R/C 500) FeCl3.6H20 (100 ppm) yes (3 hrs) no
0.0156 g Na2CO3
RF-10 (R/C 500) 0.0336 g DHB (2000 ppm) yes (2 hrs) no
0.0156 g Na2CO3
RF-11 (R/C 500) 0.168 g DHB (10,000 ppm) yes (2 hrs) no
0.1737 mL1 M
RF-12 HCIO4 (R/C 200) 0.168 g DHB (10,000 ppm) yes (1 hr) no
1.6 mL 0.1 M HCI
RF-15 (R/C 200) 0.336 g DHB (20,000 ppm) yes (< 0.5 hr) no










Table 2-4. RF polymers with acetone as solvent and HCI as catalyst: 4.00 g Resorcinol + 5.00
mL formaldehyde in 12.00 mL acetone with HCI as catalyst


Catalyst
1.6 mL of 0.1 M
RF-38 HCI (R/C 200)
1.6 mL of 0.1 M
RF-17 HCI (R/C 200)
1.6 mL of 0.1 M
RF-18 HCI (R/C 200)
1.6 mL of 0.1 M
RF-26 HCI (R/C 200)

1.6 mL of 0.1 M
RF-14 HCI (R/C 200)
1.6 mL of 0.1 M
RF-16 HCI (R/C 200)
1.6 mL of 0.1 M
RF-24 HCI (R/C 200)
1.6 mL of 0.1 M
RF-25 HCI (R/C 200)

1.6 mL of 0.1 M
RF-21 HCI (R/C 200)
1.6 mL of 0.1 M
RF-22 HCI (R/C 200)

1.6 mL of 0.1 M
RF-23 HCI (R/C 200)

3.2 mL of 0.1 M
RF-36 HCI (R/C 100)

0.16 mL of 1 M
RF-37 HCI (R/C 200)


Matrix


Polymerized


Signal


0.49 mL 1-vinylimidazole (100,000 ppm)

0.1593 g EDTA (10,000 ppm)

0.3492 g 4-(imidazole-1-yl)phenole (20,000 ppm)


no
yes (< 0.5 hr) sol
became milky in
freezer


0.336 g DHB (20,000 ppm)


0.4146 g CHCA (20,000 ppm)

0.8292 g CHCA (40,000 ppm)


0.2073 g CHCA (10,000 ppm)


0.488 g SA (20,000 ppm)


0.390 g caffeic acid (20,000 ppm)


0.312 g 3-amino-qumiline (20,000 ppm)

0.2073 g CHCA (10,000 ppm) + 0.2746 g 4-
(imidazole-1-yl)phenole (16,000 ppm)
0.2073 g CHCA (10,000 ppm) + 0.2746 g 4-
(imidazole-1-yl)phenole (16,000 ppm)


yes (1 hr)









CHAPTER 3
SURFACE MODIFIED RF POLYMER WITH APTAMERS: METHODOLOGY AND USAGE
IN SELDI-MS SAMPLE PREPARATION

Because of the complexity of biological samples obtained from individuals for disease

screening tests and the chemicals used during the sample handling process, one or more cleaning

or separation steps is usually employed to remove most of the unwanted species and pre-

concentrate the biomarkers before MALDI-MS analyses.114 Those separation and enrichment

methods are usually tedious, time-consuming, and require large amounts of sample.

Many investigations have been performed to reduce the interference from the

uninformative components in MALDI analyses without a cleaning step. Several polymers were

used to achieve on-probe reduction of contaminations. High-density polyethylene membrane,

high-density polypropylene, octyl (Cs) and octadecyl (Cls) extraction disk from 3M (St. Paul,

MN) were used successfully to remove salt, detergent, and glycerol contaminants from peptide

and protein sample solutions. Immobilization of proteins onto nylon-66 and positive charge-

modified nylon (Zetabind) membranes removed the water soluble contaminates before MS

analysis.116 Perfluorosulfonated ionomer film were used to analyze biological mixtures such as

chemical digests of proteins, proteins in milk and egg white, cell lysate, and oligonucleotide

without pre-purification with enhanced results.ll On-probe purification also achieved by using

digests nitrocellulose for MALDI-MS analysis of PCR (Polymerase chain reaction) products and

DNA fragment.' Polyethylene membrane was used for high-mass proteins (100,000 Da)

MALDI-MS analysis with enhanced spectral quality.119 Hydrophilic poly(acrylic acid) (PAA)

was used for its selective adsorption of partially digested myoglobin to remove contaminants by

rinsing with water.120 Metals such as patterned gold with poly(acrylic acid)121 as analysis

surfaces were tested to reduce the contaminations in the samples.









Surface modifications have been performed in several cases with polymeric materials in

order to covalently bind a certain class of analyte for purification from solution for enhanced

performance (SELDI). A surface of poly(4-vinylpyridine) which was extended with spacer arms

(-N-ethyl succinamyl-) and terminated with leaving groups (N-hydroxysulfosuccinimi de) for

covalent attachment of an 18-residue peptide (N terminus of human ci-casein) was used for on-

probe purification for pre-analysis purification.122 Poly-lysine-immobilized poly-2-hydroxyethyl

methacrylate membrane was used to selectively absorb DNA from aqueous solution.123

Affinity adsorption without the formation of covalent bonds allows relatively easy mass

spectrometric analysis. Surface modification, which adds specific affinity to the substrate

surface for affinity extraction, was developed. The extremely strong avidin glycoprotein-biotin

interaction (kd= 10 i) was utilized to retain the biotinylated analyte onto the avidin immobilized

agarose beadsl24, 125 Or polymer thin films.126 Lectins have been immobilized onto gold foils to

affinity capture bacteria from the contents in the sample solution.12 Several dextrans were

immobilized onto the fresh gold support as a self-assembled monolayer to retain binding

molecules which capture analytes of interest.12 DNA has been immobilized onto the

poly(ethylene terephthalate) microfiber to capture anti-DNA antibodies.129

Although surface modification with affinity agents is superior to simply using the

polymers as substrates for MALDI-MS analysis, each of the methods has its limitation. The

avidin modified substrate requires the analyte to be biotinylated; lectins can be used to capture

bacteria but lacks specificity; the dextran-modified gold surface can work the same way as

IMAC does and both lack specificity; DNA has higher specificity but with the drawback of

difficulty in obtaining and handling DNA, plus the large size of DNA causes less amount to be

attached onto the solid support. A universal one-step on-probe clean-up method which has a









high specificity towards biomarkers, high enough density of recognizing sites, with low cost and

yet easy to manipulate for biological sample analysis has been developed. This method is a

surface modification of a RF polymer with aptamers for affinity capture. Antibodies have the

highest specific affinity among the biological receptors used for modifications, though is one of

the most expensive. Aptamers have numerous advantages over antibodies, yet, have not been

wildly tested.

As shown in Figure 3-1, there are two phenol groups on each aromatic ring which make up

most of the RF polymer structure. These phenol groups provide plenty of reaction sites for direct

surface modification with aptamer and surface modification with aptamer via carboxylation, i.e.,

indirect surface modification. The phenol groups on the RF polymer have a pKa of 9.86 and the

hydroxyl groups on the silica particle surface have a pKa of 9.82; thus, the reactivity of the

phenol groups should be very close to the hydroxyl groups on the glass nanoparticles. The

surface modification of the RF polymer can follow the procedures used in the research lab of Dr.

Weihong Tan (University of Florida) for the surface modification of silica particles.

Aptamers Used for RF Polymer Surface Modification

Aptamers have been used as molecular beacons in Weihong Tan's lab ( University of

Florida) for in vitro fluorescent detection. Molecular beacons detect the existence of an analyte

according to the changing fluorescent signal intensity when the beacon changes the conformation

as the result of the affinity capture of the analyte. Fluorescent signal detection for analyte with

molecular beacon works very well if the analyte of detection in known to exist in the tested

solution, otherwise, there can be a false positive result and a definite identification is impossible.

Tandem mass spectrometry analysis (MSn) has the capability for definite identification of a

substance based on its fragmentation pattern. Coupling the high specific recognition capability

of molecular beacon with the informing power of mass spectrometry would be the answer for the









limitations of molecular beacons. Thrombin-binding aptamers were covalently attached to the

surface of a glass slide and were used successfully for affinity capture of thrombin in MALDI

targets by McGown at Duke University.73 The RF polymer developed in our lab has enough

reaction sites, with similar reactivity as silica slides, to be used as solid support for aptamer

attachment and with the flexibility in the size and shape the sol-gel polymerization method

renders.

There is a long list of aptamers has been generated by SELEX process; moreover, an

aptamer can be produced with affinity towards any specific target molecule by the SELEX

process. The choice of the aptamer to be attached onto the RF polymer is based on the concern

that the successful attachment of the aptamers onto the RF polymer should be verified before the

affinity capture of a specific molecule is proved by mass spectrometric analysis. Verification of

the attachment of the aptamers on the RF with MS would include MS-related variables into the

verification process; thus, include more uncertainty in the detection. Since the task in this

proj ect is to develop a methodology for affinity capture of analyte of interest prior to MS

analysis, choosing an existing aptamer with high affinity of a specific molecule would save a lot

of effort in generating a new aptamer. An aptamer with high affinity towards a molecule that can

be detected by the naked eye and be measured by UV-vis was chosen.

The molecule chosen is reactive green 19 and belongs to a group of reactive dyes used in

textile industry.130, 131 This group of dyes has been used in dye-ligand affinity chromatography

for the purification of proteins because of the high selectivity and the reversibility of the affinity

binding between the dye and the protein.132 These synthetic dyes interact with the active sites of

many proteins by mimicking the structure of the substrates, cofactors, or binding agents for these

proteins. The aromatic triazine dye structure resembles the nucleotide structure of nicotinamide









adenine dinucleotide and the dye interacts with the dinucleotide fold in these proteins. These

reactive dyes can bind proteins by electostatic and hydrophobic interactions and by more specific

"pseudoaffinity" interactions with ligand-binding sites.131 The degree of purification achieved

with dye-ligand chromatography is generally better than that obtained with less specific

techniques such as ion-exchange or gel filtration chromatography. The reactive dyes are

relatively inert and unaffected by enzymes in crude cellular extracts.131 Figure 3-2133 Shows the

structure of the reactive green 19 molecule. This dye molecule has absorbance peak at around

630 nm wavelength and can be detected by the naked eye and UV-vis spectrometry.

The sequence of the aptamer with high specific affinity toward reactive green 19 is

ACCCG GCGTT CGGGG GGTAC CGGGTwhich was discovered by Ellinton and Szostak in

1992134, 135 and can be synthesized in a DNA synthesizer.132 In 20 mM Tris buffer at pH 7.6 with

0.5 M LiCl and 1 mM MgCl2, the aptamer folds into a stem-loop structure as shown in Figure 3-

3133 and can be used in a bioassay when labeled with fluorescein.136

After the reactive green 19-retaining aptamer was successfully attached onto the RF

polymer, a cocaine-retaining aptamer was chosen to be attached onto the RF polymer for the

cocaine capturing and MS analysis. Cocaine is an alkaloid found in leaves of the South

American shrub Erythroxylon coca. The active ingredient of the coca plant was first isolated in

the West by the German chemist Friedrich Gaedcke in 1855. Albert Niemann described an

improved purification process for his PhD which he named as "cocaine". Cocaine is a

powerfully addictive stimulant drug which was tried by approximately 33.7 million Americans

ages 12 and older, representing 13.8% of the population, at least once in their lifetimes according

to the 2005 National Survey on Drug Use and Health.137 Although detection of cocaine by GC-

MS is well established, it will be helpful if cocaine can be captured from biological sample such









as blood and urine and to be analyzed by MALDI-MS. Actually, after this aptamer affinity

capture procedure is established, any disease biomarker could be captured and analyzed by MS

the same way as reactive green 19 and cocaine.

Aptamer affinity capture of cocaine can be achieved by attaching a cocaine-retaining

aptamer onto the RF polymer. An aptamer with the affinity towards cocaine (Figure 3-4 b), one

of the two cocaine isomers, was constructed by Landry et al., as shown in Figure 3-545 which

forms a three-way junction in 20 mM Tris buffer with pH 7.4, 140 mM NaC1, and 2 mM MgCl2

to accommodate a cocaine molecule at the center of the aptamer tertiary structure. With

dissociation constant of 0.4 10 pLM between cocaine and the cocaine-retaining aptamer,45

cocaine-retaining aptamer was used as a fluorescent sensor for cocaine45 and colorimetric probe

for the detection of cocaine.133

Direct Surface Modification of the RF Polymer with Aptamer

The reactive green 19-retaining aptamer and cocaine-retaining aptamer were directly

attached onto the RF polymer in the nucleic acid synthesizer (Applied Biosystems 3400 DNA

Synthesizer, Figure 3-6). The Applied Biosystems 3400 DNA Synthesizer from Applied

Biosystems (Foster City, CA) is a fully programmable instrument that provides four-column

simultaneous synthesis, and features automatic base dilution and analysis of coupling efficiency.

The RF polymer pellets were first polymerized by depositing RF solution onto the top of a

stainless steel microprobe and allowing several hours for the solution to polymerize. These

pellets were then hammered to small pieces to fit in the columns of the DNA synthesizer. These

RF polymer pieces were loaded into the reaction column as solid support for aptamer attachment.

The sequence of the aptamer was set up using the software and the synthesis of the aptamers was

controlled by the computer. In each step, the solution is pumped through the column which is

attached to the reagent delivery lines and the nucleic acid synthesizer. Each base is added via









computer control of the reagent delivery. After the synthesis was finished the RF polymer pieces

were taken out to be immersed in NH30H / CH3 H2 ratio as one at room temperature for three

hours for de-protection of the aptamers.

Surface Modification of the RF Polymer with Aptamers via Surface Modification of the RF
Polymer with Carboxylic Groups

Research has been conducted in the research group of Dr. Weihong Tan group on

carboxylation of silica nanoparticles and reaction of the carboxylic acid groups with amine

modified aptamers for indirect surface modification. According to the procedure, the RF

polymer will be carried out by immersion in freshly prepared 1% (v/v) solution of distilled (3-

(trimethoxy silyl)propyl)di ethyl enetri amine (DETA) and 1 mM aceti c aci d for 30O min at room

temperature. The DETA modified RF polymer will be rinsed with deionized water to thoroughly

remove excess DETA. The amine-functionalized polymer was then treated with succinic

anhydride in dry tetrahydrofuran (THF) in presence of an argon atmosphere for 6 h. Figure 3-7

shows the scheme of the surface modification.

After the RF polymer was carboxylated amine-modified cocaine aptamers were attached

onto the RF polymer via the formation of an amide bond following the procedure used by

Weihong Tan's lab.136

Characterization of the Carboxylic Group-Modified RF Polymer

The verification of the carboxylation was performed by labeling the RF polymer with 1-

pyrenyldiazomethane (PDAM) (Figure 3-8), a new fluorescent labeling reagent from Molecular

Probes (Eugene, OR) with an amine group on it.139 The covalent attachment of the PDAM to the

carboxylated RF polymer was performed according to the procedure described by Gaber et al.139

The carboxylated RF polymer pieces (sample) and non-carboxylated RF polymer pieces

(control) were placed in two separate vials and immersed in an aliquot of 10 mM PDAM solution









in ethanol. The two suspensions were mixed vigorously on a vortex mixer for two hours at room

temperature before the polymer pieces were thoroughly rinsed with ethanol. The polymer pieces

were transferred to two separate vials and were rinsed three times with ethanol.

The excitation wavelength of PDAM is 3 51 nm and the fluorescent emission wavelength is

392 nm. Fluorescence emission intensity spectra of the sample and control in ethanol were

obtained by readings from Tecan Microplate Reader from Tecan (Mannedorf/Zurich,

Switzerland) to verify the success of surface modification of the RF polymer (Figure 3-9). The

PDAM-bond polymer pieces has a much higher emission at 392 nm wavelength after excited at

351 nm than the control which proved the existence of carboxylic acid groups on the RF

polymer.

The determination of the amount of the carboxylic groups on the RF polymer was

evaluated from the uptake of a basic dye molecule, toluidine blue O (Figure 3-10),140 by the

carboxylated RF polymer. Toluidine blue O complexes to equivalent moles of carboxylate

groups which is the base for calculating the amount of the carboxylic acid groups according to

the amount of the toluidine blue O absorbed by the carboxylated RF polymer.140 The

carboxylated RF polymer pieces were immersed in 500 Cpm toluidine blue O of pH 10 at 30 OC

for five hours before the RF polymer pieces were rinsed thoroughly with 5.0 x 10-4 N NaOH

aqueous solution. The polymer pieces were rinsed three times after they were transferred to

another vial. The dye molecules were desorbed from the RF polymer by immersing the RF

polymer pieces in 50% acetic acid solution.

All the UV spectra in this proj ect were recorded on Varian Cary-100 Conc UV-visible

spectrometer (Palo Alto, CA). Toluidine blue O has UV absorbance at 285 and 623 nm. The

UV absorbance of the toluidine blue O desorbed from the carboxylated and noncarboxylated RF









polymer served as a second proof of the existence of the carboxylic acid groups on the RF

polymer (Figure 3-11).

The UV absorbance of the toluidine blue O solutions with different concentrations (Figure

3-12) was used to construct a calibration curve at 285 nm (Figure 3-13). Bases on the linear

regression fit shown, the concentration of the toluidine blue O desorbed from twelve round

pieces (3.5 mm diameter) of carboxylated RF polymer and non-carboxylated RF polymer in 9

mL of the 50% acetic acid solution was 7.8x10-6 M and 4.8x 10-6 M, respectively. The

difference of the concentration of toluidine blue O desorbed between the carboxylated RF and

non-carboxylated RF polymer pieces was 3.0x10-6 M mol by subtracting the concentration of

toluidine blue O desorbed from the noncarboxylated RF polymer from the concentration of the

toluidine blue O desorbed from the carboxylated RF polymer. The difference of the amount of

toluidine blue O in 9 mL of solutions between the carboxylated RF and non-carboxylated RF

polymer pieces was calculated by timing the difference of the concentration of toluidine blue O

between the carboxylated RF and non-carboxylated RF polymer pieces with 9 mL to get 2.7x10~

SM. The density of the carboxylic groups on the polymer was calculated by dividing 2.7x10-s

mol by the total surface area of the twelve polymer pieces with two sides (23 1 mm2) to get

1.2x10-10 mol/mm2. The number of carboxylic groups available to be ablated by a 100

micrometer radius laser pulse is thus 3.7x 10-12 mel, which is more than adequate to yield a

strong MALDI signal.

Affinity Capture of RG 19 with RG 19-Retaining Aptamer-Modified RF Polymer and
Characterization of the Modified RF Polymer

The affinity capture of the RG 19 by the RG 19-retaining aptamer was demonstrated by

desorption of this dye as detected by UV-vis spectrometry. The aptamer-modified RF polymer

pieces and non-modified RF polymer pieces were immersed in the 20 mM Tris-Cl buffer with









RG 19 in two separate plastic vials for four hours. The polymer pieces were rinsed thoroughly

with the Tris buffer before they were transferred to two different vials in order to avoid the non-

specific capture of the RG 19 on the plastic vial wall. The polymers were rinsed three times

again before deionized (DI) water was added to the vial to desorb the RG 19. The DI water in

the vial containing the aptamer-attached RF polymer turned blue instantly, while the DI water in

the vial contained the non-modified RF polymer did not change color. UV spectrum was taken

for the desorption solution from both vials (Figure 3-14).

In order to determine the amount of aptamer on the polymer surface, the concentration of

RG 19 desorbed into solution was measured by its UV absorbance. A calibration curve was

constructed (Figure 3-15) according to the UV absorbance at 634 nm of the reactive green 19

solutions different concentration. The equation of the calibration curve as y=0.00061x + 0.0019.

According to the calibration curve the amount of the RG 19 was calculated to be 0.20 mM from

the 0.123 UV absorbance of the sample. The amount of the RG 19 in the 5 mL solution was

7.8x10-7 mol. Two round pieces of the RF polymer have a total surface area of 38.5 mm2. The

density of the reactive green 19 absorbed on the aptamer-attached polymer was calculated as

2.03x10-s mol/mm2. The number of reactive green 19 can be ablated by laser (the radius of the

laser shot is estimated to be 100 Cpm) is 6.0x10-10 mol /spot, which is plenty for MALDI-MS

analy si s.

Affinity Capture of Cocaine with Direct and Indirect Cocaine-Retaining Aptamer-
Modified RF Polymer

The affinity capture of cocaine from a cocaine solution by the direct aptamer modified RF

polymer was first detected using a visual detection method, conducted according to the

procedure described by Landry et al.53 Diethylthiatricarbocyanine iodide, a cyanine dye, can be

captured by the cocaine-retaining aptamer. The addition of the cocaine solution into the









diethylthiatricarbocyanine iodide-aptamer complex containing would release the dye molecule to

capture the cocaine molecule (Figure 3-16)53 which serves as an easy visual detection of the

existence of the cocaine-retaining aptamers on the RF polymer. The pieces of the cocaine

retaining-aptamer modified RF polymer and pieces of non-modified RF polymer were put into

two separate vials with 500 CIL of 20 mM Tris buffer. A 50 CLL of 7 CLM of

diethylthiatricarbocyanine iodide in methanol was added into the vial followed by vigorous

mixing on the Vortex for several seconds. After five minutes, both suspensions were thoroughly

rinsed with buffer solution and transferred to two vials. The polymer pieces were rinsed with

buffer three times before 5 CIL of cocaine solution (1 mg/mL acetonitrile) was added into both

vials with 500 CLL Tris buffer. The solution in the vial with non-modified RF polymer pieces did

not change color, while the solution in the vial with the cocaine retaining-aptamer modified RF

polymer changed from colorless to blue. The color of the solution faded several hours later due

to the hydrolysis of the dye molecules in the slightly basic buffer."

MALDI-MS spectra of the cocaine-retaining-aptamer modified RF polymer and

unmodified RF polymer were taken on the LTQ with vMALDI source. Three pieces of cocaine

retaining-aptamer modified RF polymer and three pieces of non-modified RF polymer were

immersed in 500 CIL of 20 mM Tris buffer for the aptamers to form the affinity capturing

configuration. Then 5 CIL of cocaine solution (1 mg/mL acetonitrile, 3mM) was added into the

vials, waited for five minutes before they were thoroughly rinsed with 20 mM Tris buffer

solution. The polymer pieces were transferred to another two vials to be rinsed three times

before they were taken out and inspected under the microscope (Figure 3-17).

The polymers were taped onto the sample slide with double-sided tape and inserted into

the instrument. Sinapinic acid (SA), 2,5-dihydroxybenzoic acid (DHB), and a-cyano-4-









hydroxycinnamic acid (CHCA) were used as MALDI matrices on three of the modified polymer

respectively and on three non-modified respectively. The cocaine retaining-aptamer modified

RF polymer showed cocaine signal (the [M+H]+ ion at m/z 304) when SA was used as the

matrix, meanwhile there was no cocaine signal when DHB or CHCA was used as matrices,

which is consistent with the results McGown observed with SA for her aptamer-captured

thrombin sample.73 The polymer without the cocaine retaining aptamer (Figure 3-18a) showed

no cocaine ( m/z 304) compared to the cocaine retaining aptamer attached polymer (Figure 3-

18b). The 2-D and 3-D images of the intensity of the m/z at 304 ions of cocaine on the RF

polymer is shown in Figure 3-19.

An indirect cocaine retaining-aptamer modification of the RF polymer was also conducted

for the in Weihong Tan's lab after the polymer pellets were carboxylated in our lab. Three

pieces of cocaine retaining-aptamer modified RF polymer were immersed in 500 CLL of 20 mM

Tris buffer for the aptamers to form the affinity capturing configuration. Then 5 CLL of cocaine

solution (1 mg/mL acetonitrile, 3mM) was added into the vial, waited for five minutes before

they were thoroughly rinsed with 20 mM Tris buffer solution. The polymer pieces were

transferred to another vial to be rinsed three times before they were taken out and examed under

the microscope (Figure 3-20). The polymer pellets were taped onto the sample slide with double

sided tape and insert into the instrument. Sinapinic acid (SA) was used on the polymer as the

matrix. The cocaine retaining-aptamer modified RF polymer showed cocaine signal (m/z 304)

with SA as matrix (Figure 3-21). The 2-D and 3-D image of the intensity of the cocaine (m/z

304) on the RF polymer is shown in Figure 3-22. The amount of the cocaine detected on the

aptamer-indirect modified polymer by the MS is substantially higher than that on the direct









modified polymer (more spots with cocaine) due to the expected higher reaction efficiency of the

indirect modification.

Conclusion

With the phenol groups on the RF polymer, it is relatively straight-forward to follow the

established procedure to attach aptamer directly onto the RF polymer to surface modify it. The

up-take of reactive green 19 and cocaine proved the existence of the aptamers on the RF polymer

and the affinity capture capability of the aptamers.

With the phenol groups on the RF polymer, it is easy to follow the established procedure to

attach carboxylic acid groups onto the RF polymer, then react with amine modified aptamers to

surface modify the RF polymer with aptamers. The carboxylation of the RF polymer was proved

by the detection of the up taking of toluidine blue O and PDAM. The amount of the cocaine

detected on the aptamer indirect modified polymer by the MS is substantially higher than that on

the direct modified polymer due to the expected higher reaction efficiency of the indirect

modification.

With the procedure of attaching the aptamers onto the RF polymer established in the

research proj ect, other disease biomarker retaining-aptamers could be attached onto the RF

polymer following the same procedure, which can be used for pre-analysis clean-up for MALDI-

MS analysis for disease screening.





























CH2OH


Figure 3-1. The structure of the RF polymer.