<%BANNER%>

Method and Material Development for the Detection and Analysis of Cancer Cells

Permanent Link: http://ufdc.ufl.edu/UFE0021267/00001

Material Information

Title: Method and Material Development for the Detection and Analysis of Cancer Cells
Physical Description: 1 online resource (182 p.)
Language: english
Creator: Medley, Colin Donnell
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: cancer, nanotechnology
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: One of the most important aspects to cancer treatment is the early and accurate diagnosis of the disease. Early diagnosis enables current treatments to be much more effective and leads to greatly improved survival rates. In an effort to realize this, I developed two diagnostic assays based on aptamer conjugated nanoparticles. Aptamers are single stranded oligonucleotide chains that forms a three dimensional structure that can bind with high affinity and specificity to a targeted molecule. We developed a novel cell-based aptamer selection strategy called cell-SELEX to produce a group of aptamers for the specific recognition of individual cells without prior knowledge of the biomarkers on the cells. The cell-SELEX process uses whole cells as targets to select aptamers that can distinguish target from control cells. Once selected, the aptamer can be chemically synthesized and easily functionalized for bioconjugation to different nanomaterials, fluorophores, or therapeutic agents. The first assay is based on two types of silica nanoparticles, one where a fluorescent dye has been doped inside the particle while the other has a magnetic nanoparticle doped inside the silica. The aptamers allow the nanoparticles to bind to the cell surface. After the application of a magnetic field, the magnetic nanoparticles and anything bound to them are immobilized and the unbound materials can be washed away. This allows for the selective enrichment and detection of the target cells. The second assay uses gold nanoparticles instead of silica-based nanoparticles. The gold nanoparticles are in close proximity and their surface plasmons can interact. The interaction results in a red shift of the absorption of the particles and an increase in the extinction coefficient of the particles. Using these properties of the gold nanoparticles with the selectivity and affinity of the aptamers results in colorimetric assay where a solution containing the target cells changes color. However, detection is only one important criterion for cancer treatment. A better and more complete understanding of the disease at a biomolecular level is critical to developing more effective treatments. By microinjecting multiple molecular beacons with different fluorophores inside of single breast carcinoma cells and monitoring with advanced fluorescent microscopy, the expression of multiple genes can be simultaneously monitored inside of single living cells. The mRNA for B-actin, Manganese Superoxide Dismutase, and a control sequence were detected simultaneously using this method. Using ratiometric analysis as a basis for the measurements allows the different gene expression levels to be compared from cell to cell. Not only does this allow differentiation of individual mRNA expression levels between multiple single cells, but it also allows for mRNA expression trend analysis at the single cell level. This can be further coupled with in vivo ion monitoring experiments to allow a more complete understanding of cellular processes.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Colin Donnell Medley.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Tan, Weihong.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021267:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021267/00001

Material Information

Title: Method and Material Development for the Detection and Analysis of Cancer Cells
Physical Description: 1 online resource (182 p.)
Language: english
Creator: Medley, Colin Donnell
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: cancer, nanotechnology
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: One of the most important aspects to cancer treatment is the early and accurate diagnosis of the disease. Early diagnosis enables current treatments to be much more effective and leads to greatly improved survival rates. In an effort to realize this, I developed two diagnostic assays based on aptamer conjugated nanoparticles. Aptamers are single stranded oligonucleotide chains that forms a three dimensional structure that can bind with high affinity and specificity to a targeted molecule. We developed a novel cell-based aptamer selection strategy called cell-SELEX to produce a group of aptamers for the specific recognition of individual cells without prior knowledge of the biomarkers on the cells. The cell-SELEX process uses whole cells as targets to select aptamers that can distinguish target from control cells. Once selected, the aptamer can be chemically synthesized and easily functionalized for bioconjugation to different nanomaterials, fluorophores, or therapeutic agents. The first assay is based on two types of silica nanoparticles, one where a fluorescent dye has been doped inside the particle while the other has a magnetic nanoparticle doped inside the silica. The aptamers allow the nanoparticles to bind to the cell surface. After the application of a magnetic field, the magnetic nanoparticles and anything bound to them are immobilized and the unbound materials can be washed away. This allows for the selective enrichment and detection of the target cells. The second assay uses gold nanoparticles instead of silica-based nanoparticles. The gold nanoparticles are in close proximity and their surface plasmons can interact. The interaction results in a red shift of the absorption of the particles and an increase in the extinction coefficient of the particles. Using these properties of the gold nanoparticles with the selectivity and affinity of the aptamers results in colorimetric assay where a solution containing the target cells changes color. However, detection is only one important criterion for cancer treatment. A better and more complete understanding of the disease at a biomolecular level is critical to developing more effective treatments. By microinjecting multiple molecular beacons with different fluorophores inside of single breast carcinoma cells and monitoring with advanced fluorescent microscopy, the expression of multiple genes can be simultaneously monitored inside of single living cells. The mRNA for B-actin, Manganese Superoxide Dismutase, and a control sequence were detected simultaneously using this method. Using ratiometric analysis as a basis for the measurements allows the different gene expression levels to be compared from cell to cell. Not only does this allow differentiation of individual mRNA expression levels between multiple single cells, but it also allows for mRNA expression trend analysis at the single cell level. This can be further coupled with in vivo ion monitoring experiments to allow a more complete understanding of cellular processes.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Colin Donnell Medley.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Tan, Weihong.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021267:00001


This item has the following downloads:


Full Text

PAGE 1

1 METHOD AND MATERIAL DEVELOPMENT FOR THE DETECTION AND ANALYSIS OF CANCER CELLS By COLIN D. MEDLEY 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 2007 Colin D. Medley

PAGE 3

3 To my family, Pingping and Nini

PAGE 4

4 ACKNOWLEDGMENTS I thank Dr. Weihong Tan for his support a nd encouragement during my studies at the University of Florida. I also thank the many past and current members of the Tan research for their support and encouragement. I appreciate Dr. Arup Sen for his insight ful discussions and assistance with my intracellular projects esp ecially in regard to molecular beaco n synthesis. I thank Dr. Richard Rogers for his expertise in cancer genomics wh ich provided great guidanc e and assistance to my work and Dr. Nico Omenetto for his interest and fruitful discussions regarding the SERS project. In particular, I thank Joshua Smith for his con tinuing insight, support, an d hard work in making many of our collaborative projects a success. I also thank Tim Drake for his assistance and guidance in learning about molecular beacons an d intracellular analysis and Hui Lin for her DNA synthesis mastery. None of my success would have been possible without the unconditional support and love of my family. Particularly, the love and support of my wife Pingping and my daughter Leilani made my achievements possible. I also sincerely thank my pa rents Donnell and Dianne Medley, my sisters Colleen and Shauna, Aunt Mari a, Grandma Jo and all of our extended family whose lover and prayers made this work possible. I owe a debt to all of them for their support.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES................................................................................................................ .......10 ABSTRACT....................................................................................................................... ............13 CHAPTER 1 INTRODUCTION................................................................................................................. .15 Challenges in Cancer Detection..............................................................................................15 Nanomaterials and Nanotechnology.......................................................................................16 Gold Nanoparticles..........................................................................................................17 Magnetic Nanoparticles...................................................................................................19 Fluorescent Nanoparticles...............................................................................................22 Semiconductor nanocrystals.....................................................................................24 Encapsulated nanoparticles......................................................................................26 Molecules for Bio-Recognition..............................................................................................29 Measuring and Analyzing Gene Expression...........................................................................33 Traditional Methods of Ge ne Expression Analysis.........................................................33 Molecular Beacons..........................................................................................................35 Intracellular Applications of Molecular Beacons............................................................37 Intracellular Ion Levels....................................................................................................41 2 DNA AND NANOPARTICLE SYNTHESIS........................................................................43 DNA Synthesis.................................................................................................................. .....43 Nanoparticle Synthesis......................................................................................................... ..48 Nanoparticle Bioconjugation to DNA....................................................................................51 Silica Nanoparticles.........................................................................................................51 Gold Nanoparticles..........................................................................................................53 Conclusion..................................................................................................................... ..54 3 TWO-PARTICLE ASSAY FOR THE COLLECTION AND ENRICHMENT OF TARG ETED CANCER CELLS...........................................................56 Introduction and Scheme........................................................................................................56 Methods and Materials.......................................................................................................... .57 DNA Aptamer Synthesis.................................................................................................57 Fluorescent Nanoparticle Synthesis................................................................................58 Magnetic Nanoparticle Synthesis....................................................................................60 Cell Culture................................................................................................................... ..61

PAGE 6

6 Two Particle Assay Methodology...................................................................................61 Fluorescence Imaging......................................................................................................62 Flow Cytometry...............................................................................................................62 Plate Reader Measurements............................................................................................62 Two-Particle Assay Proof of Concept....................................................................................63 Characterization of the Two Particle Assay...........................................................................65 Dye and Nanoparticle Fluorescent Intensity Comparison...............................................65 Collection Efficiency.......................................................................................................66 Enrichment Validation.....................................................................................................67 Limit of Detection...........................................................................................................68 Improving LOD with Multiple Aptamers........................................................................70 Separations from Complex Samples.......................................................................................71 Mixed Cell Sample Assays..............................................................................................71 Bone Marrow Samples....................................................................................................73 Whole Blood Sample Assays..........................................................................................74 Collection and Detection of Multiple Cancer Cells................................................................75 Instrumental Validation...................................................................................................75 Single Cell Type Extractions...........................................................................................77 Multiple Cell Type Extraction Method...........................................................................80 Mixed Cell Samples........................................................................................................81 Serum Samples................................................................................................................83 Small Cell Lung Cancer Cell Extractions.......................................................................85 Conclusion..................................................................................................................... ..86 4 COLORIMETRIC ASSAY FOR DIRECT DETECTION OF CANCER CELLS................88 Introduction................................................................................................................... ..........88 Experimental Methods........................................................................................................... .89 DNA Aptamer Synthesis.................................................................................................89 Aptamer Conjugated Gold Nanoparticle Synthesis.........................................................90 Cells.......................................................................................................................... .......90 Assay Protocol.................................................................................................................91 Results and Discussion......................................................................................................... ..91 Nanoparticle Size Effect..................................................................................................93 Assay Response...............................................................................................................94 5 MOLECULAR BEACON DESIGN, DELIVERY, AND EVALUATION.........................100 Introduction................................................................................................................... ........100 Methods and Materials.........................................................................................................101 Equipment......................................................................................................................101 Fluorescence Imaging....................................................................................................102 Plate Reader Experiments..............................................................................................102 Cell Culture...................................................................................................................103 Fluorescent Probes.........................................................................................................103 Molecular Beacon Design.....................................................................................................104 Molecular Beacon Delivery..................................................................................................107

PAGE 7

7 Electroporation..............................................................................................................109 Reversible Permeabilization..........................................................................................111 Liposome Delivery........................................................................................................112 Viral Vector Delivery....................................................................................................114 Microinjection...............................................................................................................116 Incubation..................................................................................................................... .117 Delivery Method Conclusions.......................................................................................118 Ratiometric Analysis.....................................................................................................119 In Vitro Testing of Molecular Beacons.........................................................................121 Conclusions...................................................................................................................122 6 STUDY OF CANCER CELLS THR OUGH MULTIPLE GENE MONITORING.............124 Introduction................................................................................................................... ........124 Methods and Materials.........................................................................................................125 Cell Culture...................................................................................................................125 RNA Isolation and Northern Blot Analysis...................................................................125 Equipment......................................................................................................................126 Fluorescence Imaging....................................................................................................126 Molecular Beacon Design for Multiple Gene Monitoring...................................................127 Multiple Gene Monitoring....................................................................................................128 Monitoring Stimulated Gene Expression..............................................................................130 Gene Expression Pattern Comparison...........................................................................134 Conclusions...................................................................................................................136 7 STUDY OF CANCER CELLS BY MONITORING MULTIPLE ANALYTES................139 Methods and Materials.........................................................................................................139 Fluorescent Probes.........................................................................................................139 Cells.......................................................................................................................... .....140 Protein Isolation and Western Analysis........................................................................140 Equipment......................................................................................................................141 Fluorescence Imaging....................................................................................................141 Data Analysis.................................................................................................................142 Simultaneous Detection of Ca2+ and Gene Expression Inside Single Cells.........................142 Studying the Effects of Trichostatin A on Ca2+ Levels and Gene Expression.....................145 8 INTRACELLULAR APPLICATIONS OF LOCK ED NUCLEIC ACID MOLECULAR BEACONS.......................................................................................153 Introduction................................................................................................................... ........153 Methods and Materials.........................................................................................................155 Equipment......................................................................................................................155 Molecular Beacon Synthesis.........................................................................................156 Hybridization Study.......................................................................................................157 DNase I Se nsitivity........................................................................................................157

PAGE 8

8 Cell Culture...................................................................................................................157 Fluorescence Imaging....................................................................................................157 Results and Discussion.........................................................................................................158 LNA MB Optimization.........................................................................................................161 Solution Experiments....................................................................................................161 Neuron Imaging.............................................................................................................164 LIST OF REFERENCES.............................................................................................................167 BIOGRAPHICAL SKETCH.......................................................................................................182

PAGE 9

9 LIST OF TABLES Table page 3-1 Single cell type extraction................................................................................................ ..79 3-2 Multiple cell type extraction..............................................................................................83 5-1 Molecular beacon sequences............................................................................................104

PAGE 10

10 LIST OF FIGURES Figure Page 1-1 Typical Jablonksi diagram illustrating the fluorescence process.......................................23 1-2 Cell-based aptamer selection (cell-SELEX)......................................................................32 2-1 DNA chemical synthesis three primary st eps, Detritylation, Coupling, and Oxidation are repeated until the desired DNA sequence has been synthesized.................................44 2-2 The coupling step in DNA synthesis depi cting the activiation of the phosphoramidite derivative and the subsequent c oupling of the two nucleotides.........................................46 2-3 Typical chromatograph of a molecula r beacon purification using reverse phase HPLC using absorbance for detection...............................................................................48 2-4 TEM images of iron oxide nanoparticle and silica coated iron oxide nanoparticles.........51 3-1 Two-particle assay......................................................................................................... ....57 3-2 Fluorescence images of samples anal yzed using the two-particle assay...........................64 3-3 Fluorescence images of extracted samples after five minute incubation with different labels......................................................................................................................... .........66 3-4 Flow cytometric determination of magne tic nanoparticle collection and separation efficiencies between target and control cells.....................................................................67 3-5 Study of the effects of magnetic enrichment.....................................................................68 3-6 Limit of detection for the step wise addition of the MNP and FNP...................................69 3-7 Comparison of magnetic nanoparticle with multiple aptamer sequences conjugated to the surface.................................................................................................................... ......71 3-8 Fluorescence images of mi xed cell samples extractions....................................................72 3-9 Fluorescence images of cells ex tracted from bone marrow samples.................................74 3-10 Confocal images of extractions from whole blood............................................................75 3-11 Plate reader measurements of multiple dyes to test spectral compatibility. ....................76 3-12 Fluorescence images of pure cell samples in buffer after magnetic extraction and washes......................................................................................................................... .......78 3-13 Multiple extraction procedur e of analyzing cell samples..................................................81

PAGE 11

11 3-14 Fluorescence images of buffer extracted mixed cell samples using the multiple extraction procedure...........................................................................................................82 3-15 Fluorescence images of target and control SCLC cells extracted......................................86 4-1 Plots depicting the absorption spectra obt ained for various samples analyzed using ACGNPs......................................................................................................................... ...92 4-2 Spectra of the different sizes of Aptame r Conjugated Gold Nanoparticles with target cells.......................................................................................................................... ..........93 4-3 Images and spectra of cell samples analyzed with ACGNPS............................................95 4-4 Calibration curve and response of the ACGNPs in complex samples...............................96 5-1 Comparison of the enhancements of molecular beacons with FAM/Dabcyl and higher quality fluorophore quencher pairs.......................................................................107 5-2 Fluorescence images and data plot of cells after elec troporation in a 5 M DNA spiked medium.................................................................................................................110 5-3 Fluorescence images and data plot of cells using reversible permeabilization...............112 5-4 Fluorescence images and data plot of ce lls using a lipid based transfection reagent......114 5-5 Fluorescence images and data plot of ce lls after incubation with FITC-streptavidin linked viral vectors...........................................................................................................115 5-6 Fluorescence images and data plot of cells microinjected with the fluorophore labeled DNA sequence.....................................................................................................117 5-7 Fluorescence images and data plot of cells incubated with the fluorophore labeled DNA............................................................................................................................ .....118 5-8 Average intensities from the cells for each DNA delivery method tested.......................119 5-9 Fluorescent intensitie s and ratiometric values over time for five cells injected with different amounts of fluorophore labeled DNA...............................................................121 5-10 Signal enhancements in buffer solution fo r each MB utilized in the intracellular experiments.................................................................................................................... ..122 6-1 Time elapsed fluorescent images of each MB inside of a single MDA-MB-231 cell.....129 6-2 Northern analysis of LPS-inducib le MnSOD mRNA expre ssion in MDA-MB-231 cells.......................................................................................................................... ........131 6-3 Ratiometric analysis of the real time monitoring of gene expression..............................132

PAGE 12

12 6-4 Histograms showing distribution of ra tiometric responses for control and LPS induced MDA-MB-231 cells...........................................................................................134 6-5 Plots showing the relati onship between the relative -actin mRNA expression and the relative MnSOD mRNA expression in both the LPS-induced and non-induced cells.......................................................................................................................... ........135 7-1 Time resolved fluorescence images a nd plots of cells with MBs and Fluo-4.................144 7-2 Calcium and MnSOD control experiments......................................................................146 7-3 Fluorescence images and plots s howing the effect of TSA on cells................................148 7-4 Histograms showing the dist ribution of intensities from each probe in the basal cell group and the TSA exposed cell group............................................................................149 7-5 Correlation plots comparing the levels of MnSOD to the control and Fluo-4 levels......150 8-1 Locked Nucleic Acid bases..............................................................................................155 8-2 Fluorescence image of cells injected with LNA MB.......................................................159 8-3 Comparison of the LNA and DNA MBs with and without target inside of living cells.......................................................................................................................... ........159 8-4 Monitoring of the ratiometric values of the probes inside of cells depicting the degradation of the probes.................................................................................................160 8-5 Hybridization rates of the different molecular beacon compositions..............................162 8-6 The response of the different MB co mpositions to the addition of DNase I...................164 8-7 Representative images for neurons inject ed with the negative control sequence and the B-tubulin sequence.....................................................................................................165 8-8 Fluorescence intensity versus time for each neuron injected w ith the alternating LNA/DNA molecular beacon..........................................................................................166

PAGE 13

13 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 METHOD AND MATERIAL DEVELOPMENT FOR THE DETECTION AND ANALYSIS OF CANCER CELLS By Colin D. Medley August 2007 Chair: Weihong Tan Major: Chemistry One of the most important aspects to cancer treatment is the early and accurate diagnosis of the disease. Early diagnosis enables current treatments to mu ch more effective and leads to greatly improved survival rates. In an effort to realize this, I developed two diagnostic assays based on aptamer conjugated nanoparticles. Apta mers are single stranded oligonucleotide chains that forms a three dimensional structure that can bind with high affinity and specificity to a targeted molecule. We developed a novel ce ll-based aptamer selection strategy called cellSELEX to produce a group of aptamers for the specific recognition of individual cells without prior knowledge of the biomarkers on the cells. The cell-SELEX process uses whole cells as targets to select aptamers that can distinguish target from control cells. Once selected, the aptamer can be chemically synthesized and easil y functionalized for bioc onjugation to different nanomaterials, fluorophores, or therapeutic agents. The first assay is based on two types of s ilica nanoparticles, one where a fluorescent dye has been doped inside the particle while the other has a magnetic nanoparticle doped inside the silica. The aptamers allow the nanoparticles to bind to the cell surface. After the application of a magnetic field, the magnetic nanoparticles and anything bound to them are immobilized and the

PAGE 14

14 unbound materials can be washed away. This allows for the selective enrichment and detection of the target cells. The second assay uses gold nanoparticles instead of silica-based nanoparticles. The gold nanoparticles are in close proxim ity and their surface plasmons can interact. The interaction results in a red shift of the abso rption of the particles and an in crease in the extinction coefficient of the particles. Using these properties of the gold nanoparticles with the selectivity and affinity of the aptamers results in colorimetric assay wh ere a solution containing the target cells changes color. However, detection is only one important criterion for cancer treatment. A better and more complete understanding of the disease at a biomolecular level is critical to developing more effective treatments. By microinjecting multiple molecular beacons with different fluorophores inside of single breast carcinoma cells and m onitoring with advanced fluorescent microscopy, the expression of multiple genes can be simultane ously monitored inside of single living cells. The mRNA for B-actin, Manganese Superoxide Dismutase, and a control sequence were detected simultaneously using this method. Us ing ratiometric analysis as a basis for the measurements allows the different gene expression levels to be compared from cell to cell. Not only does this allow differentiation of indivi dual mRNA expression levels between multiple single cells, but it also allows fo r mRNA expression trend analysis at the single cell level. This can be further coupled with in vivo ion monitoring experiments to allow a more complete understanding of ce llular processes.

PAGE 15

15 CHAPTER 1 INTRODUCTION Challenges in Cancer Detection Cancer is the second leading cause of death in the United States. Half of all men and onethird of all women in the US will develop cancer during their lifetimes. Today, millions of people are living with cancer or have had cancer. It is well accepted that the sooner a cancer is found, correctly diagnosed, and treat ed then the better th e chances are for survival and remission of the disease. However except for skin cancer all other types of cancer manifest internally making detection and diagnosis difficult. There ar e currently several methods utilized for the diagnosis of cancer including magnetic reso nance imaging, X-rays, genetic analysis,1 biopsy,2 and immunophenotyping.3 Biopsy is typically used as a di agnostic method to histologically identify whether a previously detected tumo r, lesion, or other abnor mal tissue is cancerous. Magnetic resonance imaging is considered too co stly to be effective for cancer screening purposes however it has been shown to be useful for staging cancers a nd deciding on treatment avenues. Immunophenotypic analyses of leukemia cells use antibody probes to ex ploit the variation of specific surface antigens in order to different iate malignant cells from normal cell lines. The limitation to this method is that antigens used for cell recognition are no rmally not exclusively expressed on any single cell type, dramatically influencing sensitivity, and resulting in false positive results. Due to of this, immunophenotyp ic analyses often require multiple antibody probes for accurate cell detec tion, increasing both the complex ity and cost of the method. Genetic analysis and PCR based methods have proven to be highly sensitive diagnostic techniques for cellular recognition,1,4,5 but they are indirectly de tecting cells by monitoring RNA expression, and require prolonged R NA isolation steps before analysis. In addition, the variable

PAGE 16

16 sensitivity of PCR can limit its effectiveness as a diagnostic technique l eading to false-negative results, particularly with occult tumor cel ls where low-level si gnals are expected. Immunophenotypic analyses are also time consuming and costly, and therefore, there is still a need to develop new technologies fo r rapid, economical cell recognition. The work described herein will explore the use of nanomaterials and molecular probes to aid in the detection and analysis of cancer cells. In particular different types of nanoparticles with different characteristics will be utilized for the sensitive and selective detection of cancer cells. Two major directions will follow this c ourse including the development of a sensitive and selective screening method for cance r from even very complex samples. The other direction will focus on developing a colorimetric method for ca ncer cell detection for use in point of care diagnostics for circumstances where sending a samp le to a laboratory would not be practical or ideal. However detection of th e cancerous cells is only one as pect to improving the prognosis for cancer. More attention has to be paid to the underlying bi ology behind cancer and developing methods that would allow identifying the optimal treatment paths through the study of gene expression and other analytes on a cellular level. Nanomaterials and Nanotechnology To realize rapid, sensitive, a nd cost effective methods for the early detection of cancer many researchers have focused on the use of nanotechnology. Nanoparticles can be comprised of many different mate rials such as silica,6 gold,7 carbon,8 or silver.9 Silica nanoparticles in particular offer many advantages since they can be doped with a variety of materials such as fluorophores,10 semiconductor nanocrystals,11 or magnetic materials.12 Each material has its own characteristics and advantag es, yet by combining the materi als and exploiting the unique advantages of each material type sensitive, accu rate and cost effective methods for the early detection of cancer can be realized.

PAGE 17

17 Gold Nanoparticles Gold nanoparticles have several advantageous properties that have fu eled their increasing use for a wide variety of applications. Gold nanoparticles exhibit many aspects such as their assembly of multiple types involving materials science, the behavior of the individual particles, size-related electronic, magnetic a nd optical properties, and their applications to catalysis and biology. Their overall utility is mainly related to their different physical properties. In particular, the size dependant opti cal properties have made gold nanoparticles very attractive for biosensing and labeling schemes. Gold nanoparticles exhibit a strong plasmon absorption band absent in the bulk material due to a collective os cillation of the conduction electrons in response to optical excitation13 whereas its condition for optical ex citation of its plasmon band is also shifted compared to the bulk material.14,15 These properties are strongl y size dependant and also result in the size dependant color va riation exhibited by gold nanoparticles. The first reported synthesis of gold nanoparticles dates to 185716 although colloidal gold has seen use as a pigment17 and a curative agent18 for hundreds of years prior to that. While many methods of gold nanoparticle involve the re duction of gold complexes, the most widely used synthetic method is the citrate reducti on of HAuCl4 in water that was introduced by Turkevitch in 1951.19 It leads to gold nanoparticle appr oximately 20 nm in diameter with good uniformity and a negative surface charge due to the coverage of citrate. Since then many methods have described the synthesis of gol d nanoparticles is various shapes and sizes.20,21 Gold particle diameters can be tuned reliab ly and routinely from 1 nm to 200 nm,21 and have been modified with a vast array of protective a nd functional ligands. As such many different functional groups, ligands, and biomolecules have been incorporated wi th gold nanoparticles including carboxylic acids,22,23 phosphines,22 oligonucleotides,24-32 amines,33 proteins,34-36 enzymes,34 and small drug molecules.27,37

PAGE 18

18 Owing to their long established biostability and ease of bioconjugation gold nanoparticles have been used in a wide array of different applications. One of the most widely used applications is the use of gold nanoparticles to label biological molecules for detection. Gold particles adsorbed to antibodies,38 proteins,39 or peptides40-42 are widely used for the detection of molecular and macromolecular targets especially for localization studies. The most widely used example of this is a popular home pregnancy test kit that when positive develops a pink line. The pink line is actually th e congregation of 40nm gold nanopa rticles adsorbed to an antibody, antihuman chorionic gonadotropin (hCG).43 In terms of colorimetric labels, the extinction coefficient of 40nm gold nanoparticle s is significantly larger compar ed to single molecule labels such as fluorescein. This translates into an increas e in detection sensitivit y of several orders of magnitude if colloidal gold is used as a molecula r label rather than other single molecule labels allowing for the easy detection with the naked eye.44 It is well known that the optical signature of gold nanoparticles is strongly dependent on th e particle size and in terparticle distance45,46 because of the interaction of the surface plasmo ns between individual gold nanoparticles. When the interparticle distance is substantially gr eater than the average particle diameter, the nanoparticle suspension appears re d or pink, but as the interparti cle distance decreases to less than approximately the average particle diameter, the color shifts to blue or purple, depending on the size of the particles and the particle concentration. More impor tantly, the changes in optical properties have been widely used in colori metric detection sche mes for oligonucleotide conjugated gold nanoparticles, as developed by Mirkin.24,31,47,48 In these studies, the color changes from red to bluepurple accompanying th e aggregation of the gold nanoparticles after hybridization of the target DNA. This allows for the colorimetric detectio n of the DNA with the

PAGE 19

19 naked eye. A similar approach has also been developed for the detection of proteins using oligonucleotide based aptamers th at induce aggregation by hybridiz ing to their target protein. Magnetic Nanoparticles Magnetic nanoparticles have been studied fo r an array of differe nt fields including magnetic fluids,49 catalysis,50,51 biotechnology/biomedicine,52 magnetic resonance imaging,53,54 data storage,55 and environmental remediation.56,57 Magnetic nanoparticles function through exploiting the properties of magnetism for different functions The electrical basis for the magnetic properties of matter stems down to the at omic level. Electrons have both charge and spin and can therefore be though of as a charge in motion. The charge in motion can produce an atomic level magnetic field. In most atoms, th e electrons are paired within energy levels, according to the exclusion principle, so that the el ectrons in each pair have antiparallel spins thus canceling their atomic magnetic fields. However, in some atoms there are more electrons with spins in one direction resulting in a net ma gnetic field for the atom as a whole. All materials however are aff ected by magnetic fields to va rying degrees in one of two ways. Diamagnetic is used to describe materi als that line up at righ t angles to a nonuniform magnetic field and are slightly repelled by that field. Diamagnetism results from magnetic field's interference with the motion of el ectrons. When diamagnetic matter is placed in a magnetic field, the magnetic field causes the some electrons to speed up some electrons and others to slow down. This interferes with the motion of the magn etic field, so the atoms internally oppose the field causing the material to be slightly repell ed by the magnetic field. Paramagnetism generally occurs in materials possessing unpaired electrons. Many elements like iron, palladium, platinum, and the rare earth elements have single electr ons that generate a small magnetic field. These elements when placed in a magnetic field, result s in the field of the atom aligning with the applied magnetic field and causes the atom to be slightly attracted to that magnetic field.

PAGE 20

20 Superparamagnetism occurs when a naturally ma gnetic material is composed of nano-scaled particles from 1nm to 10nm. In this case, ambi ent thermal energy is sufficient to change the direction of electron spin in the entire crystallite. These vari ations in the dire ction of electron spin cause the magnetic field of the atom to can cel itself. However when an external magnetic field is applied, the materials act in paramagne tic nature exhibiting a strong attraction to the source of the magnetic field. Magnetic nanoparticles have been synthesized using a variety of elements and materials including Iron oxides,58 pure Iron,59 pure Cobalt,60 MgFe2O4,61 MnFe2O4,61 CoFe2O4,63 CoPt3,64 and FePt.65 Co-precipitation is common synthesi s method for Iron oxide nanoparticles from Fe3+/Fe2+ salts through the addition of a base at room or elevated temperatures.66 The size, shape, and composition of th e nanoparticles can then be tuned through the adjustment of the salts used, the Fe2+/Fe3+ ratio, temperature, pH, and the ionic strength of the solution.67 The addition of different organic molecules have also been shown to stabilize the co-precipitation method and results in a higher monodispersity of the nanoparticles synthesized through this route.68-70 Another method of magnetic nanoparticle s ynthesis is through thermal decomposition following similar procedures to semiconductor nanocrystal synthesis.71-73 Monodisperse magnetic nanoparticles can be formed through th e decomposition of organometallic compounds in boiling organic solvents containing certain stabilizing surfactants. The advantage of this approach is that elements such as Fe, Mn, Co, Ni, and Cr can be incorporated into the nanoparticle through using differe nt organometallic precursors.74-77 The size and morphology of the nanoparticles are controlled th rough the ratio of th e starting reagents like the organometallic

PAGE 21

21 compounds, solvents, and surf actants. In addition, the temperat ure, reaction time, and aging time can be varied to precisely control th e size and morphology of the nanoparticles.78-80 Magnetic nanoparticle can also be synthesized through microemulsion.81 In water-in-oil microemulsions, the aqueous phase is dispersed as microdroplets surrounded by a monolayer of surfactant molecules. In the microemuls ion methodology, two different microemulsions containing the desired reactants are mixed a nd through the interaction of the different microdroplets the nanoparticles are formed. By a dding an appropriate solven t to the mixture, the nanoparticle can be precipitated and then centrifuge d or filtered to obtain the nanoparticle. This method has been successful in producing metallic c obalt, cobalt/platinum alloys, and gold-coated cobalt/platinum nanopartic les in reverse micelles of cetyltr imethlyammonium bromide, using 1butanol as the cosurfactant and octane as the oil phase.82 MnFe2O4 nanoparticles with controllable sizes from about 4 nm are synthesi zed through the formation of water-in-toluene inverse micelles with sodium dodecylben zenesulfonate (NaDBS) as surfactant.83 Although a wide variety of magnetic nanoparticles have been synthesized via microemulsion, the particles have a higher variation in size in comparison to other technique s. In addition, the technique suffers from poor yields owing to the large amounts of solvents required making scale up of the synthesis difficult. The other common method for ma gnetic nanoparticle synthesi s is hydrothermal synthesis although it is less matured of the four synthetic r outes. The system consists of a solid metal linoleate, an ethyllinoleic aci d liquid phase, and a wateret hanol solution at different temperatures under hydrothermal conditions.84 This strategy is based on a general phase transfer and separation mechanism occurring at the interf aces of the liquid, solid, and solution phases present during the synthesis althoug h the precise mechanism of the formation of the particles has

PAGE 22

22 not been fully elucidated. Nonetheless, hi ghly Monodisperse 9nm to 12nm Fe3O4 and CoFe2O4 particles have been synthesized as well ferr ite spheres with tunabl e sizes from 200-800nm.84,85 After synthesis of the magnetic nanoparticle, the particle is typically post-coated to increase the stability of the nanoparticle a nd to also reduce aggreg ation. Many different materials have been used for this function incl uding polymers, precious metals like gold and silver, carbon, and silica.86-88 The post-coating step not only pr otects the nanopart icle, but it also allows the further functionalizati on of the particle for many diffe rent applications. The addition of different functional groups make magnetic nanopart icle ideal scaffold fo r catalyzing reactions. Magnetically driven separations make the recovery of catalysts in a liquid-phase reaction much easier than by filtration and centrifugation, especially when the catalysts are in the submicrometer size range allowing the efficient reuse of expensive catalysts. In addition, they offer a high potential for numerous biomedical applications, such as cell separation,56 automated DNA extraction,57 gene targeting,89 drug delivery,90 magnetic resonance imaging,91 and hyperthermia.92 When coated with, for example, an antibody, they can be applied for highly sensitive immunoassays93-95 or small substance recoveries.96 Furthermore, single-stranded DNA or oligonucleotide immobilized on magnetic pa rticles were successfully used for DNA hybridization analyses with the aim of identifying organisms97,98 and single-nucleotide polymorphism analyses for human blood.99,100 Thus, a wide array of applications for magnetic nanoparticles have been previous explored wi th particular emphasis on using their unique properties to facilitate the efficient separation of materials from even complex mixtures and samples. Fluorescent Nanoparticles Fluorescence is a highly usef ul phenomenon in which light used for the excitation of fluorescence in absorbed and converted to light with a different wavelength. This allows for the

PAGE 23

23 sensitive and efficient detection of a fluorescent species. As such fluorescence has been utilized in a tremendous number of applicat ions. Jablonski diagrams are ty pically used to illustrate the underlying mechanism behind fluorescence. As seen in Figure 1-1, the excitation light is absorbed by a fluorophore and the electrons are indu ced to an excited singlet state. The excited state typically undergoes certain non-photonic relaxation processes between vibrational levels called internal conversion before returning to the ground state. Wh en the electrons return to the ground state, a photon will typicall y be emitted. The emitted light is designated fluorescence. The emitted photon from this process is lower in energy, or longer in wavelength, than the absorbed excitation photon. The difference between the excitation and emission wavelengths of light is called the Stokes shift. Figure 1-1 Depiction of a typical Jablonksi diag ram illustrating the fluorescence processes wherein the photon is absorbed, the intern al conversion processes through which some energy is lost non-photonically, and the final photonic transition from the excited state to the ground state.

PAGE 24

24 While labeling with fluorescent molecules is a very sensitive t echnique it is often advantageous to increase the amount of fluorescen ce as much as possible. One very effective method for doing this is through th e use of fluorescent nanoparticles that can emit thousands of times as much fluorescence as a single fluor escent molecule. Fluor escent nanoparticles are generally either constructed from semiconducto r nanocrystals or thr ough the encapsulation of many fluorescent molecules in polymers or silica. Semiconductor Nanocrystals Fluorescent semiconductor nanocrystals are currently of wide interest fo r their potential for biological applications. As such they possess se veral advantageous prop erties including their small size, generally less than 10nm, very broad excitation spectra, and em ission spectra that are narrower than conventional fluorophores. The sp ectral properties of qua ntum dots are based on quantum confinement conditions un der which the crystal is on the order or smaller than the materials Exciton Bohr radius. This results in a discrete bandgap between the valence and conduction bands of electrons in the crystals that is not present in the bulk materials. After the electrons are induced to an excited state, they emit a photon when returni ng to the ground state. The emission bandwidth of the photons is much narrower as it correlates to the discrete and narrow bandgap between the valence and conduction bands. Both group II-VI quantum dots like CdSe, CdTe CdS, and ZnSe and group III-V quantum dots like InP and InAs have been synthesi zed and studied extensively in the past.101-104 Prior to 1993, quantum dots were mainly prepared in aqueous solution with different stabilizing agents such as thioglycerol or polyphosphate. This process produced low-quality QDs with poor fluorescence efficiencies and larg e size variations due to impe rfect crystal structures. In 1993, Bawendi105 reported the synthesis of highly lumine scent CdSe quantum dots by using a hightemperature organometallic procedure similar to the previous mention thermal decomposition

PAGE 25

25 methodology. The quantum dots possessed nearly perf ect crystal structures and narrow size variations, however the quantum yields were st ill low compared to conventional fluorophores. This limitation was later addressed through the de position of CdS or ZnS surface capping layers. The ZnS capping was particular successful as the Zn/S chemical bonds are similar in length to the Cd/Se bond lengths allowing the growth of thin ZnS films with the same crystalline orientation around the CdSe core. The ZnS capped CdSe crystals exhibite d quantum efficiencies around four times higher than CdSe nanocrystal s by themselves. More recently, quantum dots with excellent quantum efficiencies have been produced through th e use of nontraditional precursors to realize high quality quantum dots with no need for capping. The size of quantum dots can also be controlled through careful modi fication of the synthesi s temperature, reaction time, and reagents. One of critical aspects to any fluorescent moie ty however is the attachment to biological molecules for labeling or biosensor applications. As such many different molecules have been adapted for use as solubilization or crosslinking agents for quantum dots. These include ligand exchange with simple thiol-containing molecules105,106 or more sophisticated ones such as oligomeric phosphines,107 endrons,108 and peptides,109encapsulation by a layer of amphiphilic diblock110 or triblock copolymers,111 or in silica shells,112,113 phospholipid micelles,114 polymer beads,115 polymer shells,116 or amphiphilic polysaccharides.117 Also, combinations of different layers have also been utilized to instill nece ssary parameters like so lubility, stability, and attachment to molecular recognition elements. The development of so many surface chemistry has also led to the attachment of molecules th rough many different chem istries including thiol group,114,118,119 N-hydroxysuccinimyl ester moieties,109,112 or streptavidin/biotin conjugations.120

PAGE 26

26 Due to their excellent spectral properties, qu antum dots have been used for a variety of different applications. These including cell surface protein targeting through streptavidin,120,121 secondary122 or primary antibodies,123 receptor ligands such as ep idermal growth factor (EGF) 124,125 or serotonin,126 recognition peptides,127 and affinity pairs such as biotin-avidin after engineering of th e target protein.128 In addition, their stabilit y and brightness make them possible candidates for in vivo imag ing applications. As such they have been demonstrated in imaging blood vessels in live mice,129 as tissue specific vascular markers in mice,118 and targeted delivery through antibody conjugated quantum dots.111,130 The major issue preventing quantum dots for mo re widespread use in imaging studies is concerns to possible toxicity. While some studies have shown qua ntum dots are safe at lower concentrations many other studies have documen ted the toxicity of quantum dots including on embryo development,114 cytotoxicity in EL4 lymphoma cells,131 and DNA damage.132 It has also been demonstrated that the less protected the quant um dot is, the faster the onset of the adverse cell effects133 through the release of Cd2+ and Se2in core/shell and core only nanocrystal structures. Therefore, the work described here in focused on mainly the fluorescent nanoparticles that possess higher biostability a nd less cytotoxic side effects. Encapsulated Nanoparticles In lieu of using nanoparticles with very di fferent spectral properties from conventional fluorophores, another strategy to increase the fluorescence intensity is to simply add as many fluorophores as possible to a partic ular label or biorecognition element. Therefore, the label has the same spectral properties as a conventional fluorophore but has a much higher intensity. As it is impractical to covalently attach multiple fl uorophores to a label, the fluorophores are typically encapsulated inside of a materi al like silica or a polymer ma trix. The matrix imbues the fluorophores with many advantageous properties su ch as photostability, wider pH stability, and

PAGE 27

27 the use of simple well established conjugation schemes. As such, silica and polymer based fluorescent nanoparticles have been use for a wi de array of applications. While polymer nanoparticles possess many of the same advantages as silica nanoparticle s, the bioconjugation routes for silica are better establ ished and offer an additional degree of flexibility and versatility and thus will be the focus of the work described. One of the most important aspects of us ing nanoparticles is the synthesis of the nanoparticle and method for trapping the dye insi de its matrix. While other synthesis methods exist, the two most well established methods for silica nanoparticle s ynthesis are the Stober method134 and the microemulsion method.135 The microemulsion method incorporates the use of water, oil, and surfactant molecules in order to hydrolyze tetraethyl orthosilicate. In this method, the surfactant molecules form micelles around the water droplets containing a polar fluorophore. The size of the droplets can be control through th e water to oil ratio. Then, ammonium hydroxide and tetraethyl orthosil icate can be added to the mixture. Both hydrophilic species are brought together in the water drople ts and the ammonium hydroxide hydrolyzes the tetraethyl orthosilicate into a solid silica matrix. The hydrolysis of the tetraethyl orthosilicate forms produces Silicic acid which forms the silica matr ix through a subsequent condensation reaction. The matrix forms around the fluorophores in the water droplet entrapping them to yield highly fluorescent silica nanoparticles. Typically inorganic dyes are used for this method as they are more water soluble and they are typically posi tively charged and form stable electrostatic interaction with the silica matr ix preventing them from leak ing out of the nanoparticle. However, many strategies have been developed to use organic fluorophores such as introducing a hydrophobic silica precursor,136 using water-soluble dextranmolecule-conjugated dyes, and synthesizing in acidic conditions.137

PAGE 28

28 The other commonly used method for silica na noparticle synthesis is the Stober method. While microemulsion simply entraps the fluor ophore in the silica ma trix, the Stober method utilizes a covalent attachment to link th e fluorophore directly to the silica matrix.138,139 In the Stober method, a silicate precursor, generally tetraet hyl orthosilicate, is h ydrolyzed in a mixture of ethanol and ammonium hydroxide. To inco rporate a fluorophore, th e procedure requires a slight modification. First, the fluorophore, typi cally an isothiocyanate or succinimidil ester based fluorophore, is reacted with an amine containing silane form ing a covalent linkage. Then the fluorophore linked silane is adde d to the tetraethyl orthosilicate. The mixture is then allowed to hydrolyze in water, ethanol, and ammoni um hydroxide and undergo the subsequent condensation reaction to form the nanoparticle from both the te traethyl orthosilicate and the fluorophore linked silane. This method allows an y fluorophore with an amine reactive functional group to be combined into the silica matrix. One of the more advantageous aspects of employing silica based nanoparticles is the flexibility and versatility of the silica in terms of bioconjugation. The simplest method for conjugation involves the physical adsorption of a biomolecule like avidin followed by a crosslinking step to encase the nanoparticle in an avidin shell. Then any biotinylated ligand can easily be attached. There are also many methods that utilize a covalent attachment if other surface functionalities are wanted or required for the particular application. Ligands requiring a covalent attachment require first that the nanopa rticle surface is modifi ed with a particular functional group whether a thiol, amine, or ca rboxy group. This is achieved during a post coating step using a silane with the pr oper group and undergoing an additional round of hydrolysis and condensation that leaves the silane as part of the nanoparticle matrix and the functional group exposed on the surface.140-142 After the functionalization step the nanoparticles

PAGE 29

29 can be conjugated to oligonucleotides,143 antibodies,144 peptides,145 or organic molecules.146 Owing to their wide versatility, silica based nan oparticles have been used a wide variety of applications. These applications include biosensing,147 gene delivery,146 drug delivery,148 and labeling of biological molecules.147 Molecules for Bio-recognition While nanomaterials offer many benefits in terms of sensitivity, detection, and separation they possess no inherent selectiv ity. Therefore, any method usi ng nanomaterials is only as good as the species used for molecular recognition. The importance of molecular interactions has become increasingly clear in disease states. As such, many diseases have been shown to have characteristic biomarkers associated with th em. The identification of these biomarkers has greatly aided in the diagnosis a nd subsequent treatment of those diseases. However, the vast majority of diseases currently do not currently have specific biomarkers associated with them, greatly limiting the diagnosis and treatment capab ilities in the field of medicine. The problem lies in the great amount of time and effort that is necessary for biomarker discovery. It involves the systematic separation and identification of biological molecules from complex bodily fluids or tissues that requires a larg e effort with sufficient contro ls to avoid identifying the wrong molecule as a biomarker. Typically there ar e two classes of molecu les that are used for molecular recognition; antibodies or aptamers. Antibodies are generally large Y-shaped glycoproteins used to identify and neutralize foreign objects by the immune system. There ar e two major types of antibodies, polyclonal antibodies and monoclonal antibodies. Polyclonal antibodies are a combination of antibodies active against a particular antigen each recogni zing a particular epitope of the antigen. Polyclonal antibodies are normally obtained by inject ing the antigen into a small animal, such as a mouse or rabbit and then isolating the anti body from the animals blood although larger

PAGE 30

30 animals are used in cases were a large amount of the antibody is required. Monoclonal antibodies are instead produced in the laborat ory using a specially prepared an antibody producing cell line for a specific antigen. In research, purified antibodies widely used for the identi fication and localization of proteins or other biologically re levant molecules. Antibodies ar e used in flow cytometry to differentiate cell types,149 in immunoprecipitation to separate proteins in a cell lysate,150 in a Western blot to identify pr oteins after electrophoresis,151 or in immunohistochemistry to examine protein expression in tissues,152 and to examine the localizati on of proteins within cells by immunofluorescence.149 The major problems with antibodies is that since they are typically produced through biological pathways instead of th rough chemical synthesis they can be difficult to reproduce, difficult to chemica lly modify, are very sensitive to environmental factors, and have a finite shelf life. Aptamers are single-stranded DNA (ssDNA), RNA, or modified nucleic acids that have the ability to bind specifically to target mol ecules ranging in size from small organic molecules to large proteins.153-155 The dissociation constants of apta mers to targets can range from 10 M10 M making them comparable to antibodies in many instances. Aptamers recognize their targets with high specificity and are capable of discriminating between pr otein targets that are highly homologous or differ by only a few amino acids.156,157 The tertiary structures formed by the single-stranded oligonucle otide molecules are the basis for target protein recognition.158,159 These single-stranded oligonucle otide aptamers are selected by a process called SELEX (Systematic Evolution of Ligands by Exponential en richment), where the aptamers are selected from libraries of random sequences of synthe tic DNA or RNA by repetitive binding of these oligonucleotides to the target molecules.153,160-162 Through this in vitro selection process, single-

PAGE 31

31 stranded oligonucleotide aptamers with high specifi city and affinity to their targets can be obtained. Most of the aptamers repor ted so far have been selected using pure molecules, such as purified proteins as the targets. However, the ap tamer selection against complex targets (such as red blood cells and single protein on live trypanosomes) was also demonstrated and interesting aptamers have been generated.163-168 To produce probes for molecular profiling of tumor cells, researchers in our lab have developed a novel method, the cell-based aptame r selection process (c ell-SELEX) that has generated the aptamers for the work described here in. Instead of using a si ngle type of molecules as targets, the Cell-S ELEX process uses whole cells as ta rgets to select single stranded DNA aptamers that can distinguish target cells from control cells (see Figure1 ). In addition to low molecular weight, easy and reproduc ible synthesis, easy modification,169 fast tissue penetration, low toxicity or immunogenicity,170,171 easy storage, high binding affinity and specificity,153,172 the biggest advantage of the Cell-SELEX-based aptamer technology is the unique cell-based selection process illustrated in Figure 1-2. A gro up of cell-specific aptamers can be selected using a subtraction strategy in a relatively shor t period without knowing which target molecules are present on the cell surface. There is no eas y way to produce a similar panel of monoclonal antibodies in such a short time without purifie d antigens for any unknown an tigens. Compared to 2-D gel electrophoresis and mass spectrometry used for proteomic studies aiming at identifying proteins, the cell-SELEX will produce molecular pr obes in less time then use the specific probes to identify the target proteins. Thus, not only can the selected aptamers be used as molecular probes for molecular profiling, but also they can be used as tools for identifying new biomarkers expressed by tumor cells or other cells in disease status.

PAGE 32

32 Figure 1-2. Schematic of cell-based aptamer selecti on (cell-SELEX). Briefly, ssDNA pool is incubated with target cells. After wash ing, the bound DNAs are eluted by heating in binding buffer. The eluted DNAs are then incubated with control cells (Negative cells) for counter-selection. After ce ntrifugation, the unbounded ssDNAs in supernatant are collected, a nd then amplified by PCR. The amplified DNAs are used for next round selection. The selectio n process is monitored using fluorescent imaging by confocal microscope or fl uorescent analysis by flow cytometry. The aptamers can easily be c onjugated with fluorophores, radioi sotopes, or various other functional groups for in-vivo molecular imagi ng of cancer cells and tissue, for molecular profiling of cancers in flow cyto metry analysis, and for potential biomarker discovery in cancer samples. In the work described herein, aptamers will be used as a molecular recognition element for not only the detection of cancer ous cells but also the collection and enrichment of the cells.

PAGE 33

33 Using functional groups attached to the aptamers during the synthesis pro cess, the aptamers can be attached to the various types of nanomaterials that have been previously discussed to enable not only the easy colorimetric dete ction of cancer cells, but also the collecti on and enrichment of cancer cells. Once the cancer cells are collected and enriched they can be further studied to shed new light into the underlying diseas e processes and to help determ ine effective treatment options by exposing the cells to different cancer fighting agents. Thus the work described in this dissertation pertains not only to the development of new methods for cancer detection but also new methods on the study and characterization of cancer. Measuring and Analyzing Gene Expression Traditional Methods of Gene Expression Analysis The initial techniques used fo r gene expression analysis were developed to examine expression of individual known ge nes beginning in 1977 when the northern blot technique was introduced.173 In Northern Blot analysis, the total cell RNA is first pr epared and different size classes are separated electrophoretic ally. The targeted size is transf erred to nitrocellulose film and mixed with radiolabeled DNA probes for a particular mRNA. This is followed by radioautographic detection of DN A-RNA hybrids. This allows for the expression of particular gene expressions to be compared to other gene sequences to ga in a qualitative understanding into the relative expression levels in a group of cells. This technique is still widely used and is often performed to validate the results of other types of gene expression studies and newly developed approaches. Introduced in 1977, a method was de veloped that protect s a DNA labeled probe against degradation by singlestranded nuclease S1 by annealing the probe with RNA.174 This led to RNase protection assays that were devel oped to detect the expr ession of specific RNA sequence and to compare the levels of expre ssion through labeled cDNA. The cDNA forms a hybrid with its complementary mRNA and after exposure to a single-strand specific nuclease,

PAGE 34

34 unhybridized strands are degraded leaving only the targeted hybrids that are detected using gel electrophoresis. Another widely used technique, the serial analysis of gene expression (SAGE) allows for the quantitative measurement of entire transcriptomes. The method is based on averaging the individual cell responses through the use of seque ncing tags that are derived from different populations of cells.175 In 1993, subtractive hybridization techniques became available for constructing subtractive cDNA libraries. Th is methodology uses cDNA from one pool to hybridize to mRNA from the other allowing cD NA libraries to be constructed from the transcripts that are not hybridized and are used to identify specific mR NAs. A modification of this technique, representational difference analysis (RDA), also uses prefer ential amplification of non-subtracted fragments. In RDA, simplified vers ions or representations of the genomes being studied are created using restric tion enzyme digestion. While this tehcnique was first developed to study variations between different genomes, it ha s also proven useful for cloning differentially expressed genes. One more modern technique for gene expression analys is is fluorescence-based microarrays. Microarrays are simply ordered sets of DNA molecules of known sequences plated in an array format where different sequences ar e spatially separated. Usually rectangular, they can consist of a simple arrangement of a few se quences to hundreds of thousands of sequences for the massively multiplexed comparison of gene expression levels. Modern microarray technology employs sensitive and high resolution instrumentation to detect differences fluorescence intensities of thousands of different sequences as changes as small as 1.3 to 2-fold are detectable.176,177 This allows the detection of the upr egulated and downregulated expression levels of many genes compared to the basal level. However, the technique does suffer from a

PAGE 35

35 lack of sensitivity when analyzing changes in ge nes that have very high or very low expression, although there has been effort to wards nanoparticle based arrays in order to increase both the sensitivity and the overall limit of detection.178 Traditionally, mRNA subcellular localiza tion has been performed using in situ hybridization (ISH) in conjunction w ith light and fluorescent microscopy.179 ISH involves using labeled antisense DNA or RNA probes that hybridi ze to mRNA in pretreated fixed cells in culture or tissue sections. The unhybridized probes are then washed away from the cells and the hybridized probes are then detected using emulsion autoradiography, light microscopy, or fluorescence microscopy depending on the nature of the probe. ISH can provide very precise and sensitive localization information for the mR NAs. Unfortunately, ISH is very difficult to quantitate and is not practical fo r mRNAs that have a complex secondary and tertiary structure that sterically inhibits hybridization of the probe to the RNA of interest. Due to this phenomena, even highly abundant mRNAs can be undetectable by ISH if the mRNA has a strong secondary structure.180,181 Molecular Beacons Traditional RNA detection and localization methods such as in situ hybridization are generally used with fixed and pretreated cells or tissues.182 The fixed cells samples are dead cells which prevents monitoring dynamic cellular events such as synthesis and transport. With MBs on the other hand, living cells can be tested and explored allo wing for new applications and discoveries into the dynamic processes of the ce lls. Molecular beacons (MBs) are fluorescent nucleic acid probes that offer excellent specific ity and sensitivity for monitoring the expression of mRNA at the single cell level. Molecular beacons are synthetic DNA molecules in a hairpin structure with a fluorophore-que ncher pair that undergo a conformational change upon hybridization to a complementary nucleic acid ta rget. As hybridization occurs, the fluorophore-

PAGE 36

36 quencher pair is separated and the fluorescen ce of the fluorophore is restored. Fluorescence based detection methods can be highly sensitive due to the detection of emission photons that are spectrally separated from the excitation photons. While fluorescence has been previously discussed, an important consideration for mol ecular beacons is the concept of fluorescence quenching. The quenching of the fluorescence emission is accomplished primarily through two mechanisms. The first mechanism is referred to as dynamic or collisional quenching. In this case, a quencher molecule makes contact with the fluorophore during the lif etime of its excited state. The diffusion controlled collision betw een the two molecules results in the fluorophore returning to the ground state without emitting a phot on thus resulting in no fluorescence. Neither molecule is chemically altered during this pro cess and many different types of molecules can act as collisional quenchers. Howe ver, the collisional quenching is a random process and is not suitable for use in biosensor based applications as it would be very difficult to integrate a random diffusion based phenomena into a signal transduc tion scheme. The other type of quenching, static quenching, is far more suitable to biosensor applications. It involv es the formation of a complex between the fluorophore and quencher moie ties. While the fluorophore outside of the complex relaxes from the excited state thr ough a photonic emission, the complex relaxes through thermal and Vibrational modes due to the structur e of the quencher portion of the complex. This results in the prevention of the photonic emission and eliminates the fluorescence. While the two process can occur in the same, the molecular beacon is designed to rely on static quenching when the fluorophore and quencher are in close proximity in the stem region of the beacon. Thus upon undergoing the forced conformational change after target hybridiza tion, the quencher is

PAGE 37

37 separated from the fluorophore preventing form ation of the complex and allowing the fluorophore to emit fluorescence once again. Using molecular beacons in vivo for mRNA expression detecti on offers several advantages over traditional means of mRNA analysis. Traditional R NA analysis methods involve fixed pretreated cells or tissues that represent pooled averages from millions of cells, but may not represent individual cellular processes in living systems.182 Similarly, single cell RT-PCR exhibits variations in the amplif ication among different RNA sequences.183 Molecular beacons do not require any preor post-treatment of the cells, since, in theory, they will only open upon hybridization with their specific target. Thus, molecular beacon fluorescence occurs only when the complementary target is present, making separation of hybridized from unhybridized molecular beacons unnecessary. The molecular beacon detection of mRNA inside of the cell does not require amplification like RT-PCR so ther e is no chance of prefer ential amplification. Molecular beacons inherently possess the specif icity through Watson-Crick base pairing and the sensitivity through fluorescence detection that makes them ideal for intracellular mRNA detection. Successful use of MB s for intracellular mRNA detection depends on a valid sequence design, optimal delivery of the probe, and optimized fluorescence imaging conditions that will be explored in further detail in chapter 5. Intracellular Applications of Molecular Beacons To date MBs have been used for intracellular detection in a variety of cell systems. In 2001, the real-time hybridization of a MB to mRNA was visualiz ed inside of a single cell.184 In this study, MBs specific for -actin and -1 andrenergic mRNA were used to detect their respective mRNA targets in kangaroo rat kidney (P tK2) cells. The MBs were delivered to the cells using microinjection and then monitored for 18 minutes. In the PtK2 cells, an increase in fluorescence intensity was detected for both the -actin and -1 andrenergic MBs. In order to

PAGE 38

38 prove that increase of fluorescence was due to the intended interaction a negative control MB was also used in this experiment. The negative control MB, which has no complement inside of the cells, showed no increase in fluorescence insi de of the cell. This indicates that the fluorescence from the targeted MBs was due to h ybridization to its targ et as opposed to a nonspecific interaction since any non-sp ecific interaction or degradati on would have equally affected the negative control MB. This study demonstrated the real-time detection capabi lities of MBs for mRNA detection and showed conclusively th at MBs could be very valuable tools for detecting gene expression in side of single living cells. In 2003, Tyagi et al demonstrated that MBs could be used for the visualization of the distribution and transport of mRNA.185 In this study an MB for oskar mRNA was investigated in Drosophila melanogastar oocytes. Initially they demonstrated visualizing the distribution of oskar mRNA in the cell. Due to the fluorescen ce background exhibited from the MBs, they decided to use a binary MB approach that util ized two MBs that target ed adjacent positions on the mRNA. When both MBs were hybridized to the mRNA sequence the donor and acceptor fluorophore were brought within clos e proximity allowing FRET to occur. This generates a new signal that indicated hybr idization of both MBs with the mRNA. In addition to visualizing the mRNA distribution, they were al so able to track the migratio n of the mRNA throughout the cell and even into adjacent cells in the oocyte. Th is study demonstrated th e potential of MBs not only for studying the localization of mRNA inside of single cells bu t it also demonstrated that MBs could be used for tracking mRNA migration ev en into different cells. Other studies have built on this line of investigati on and imaged MBs on viral mRNA inside of host cells to study the behavior of the mRNA. This study investigat ed both the localization of the mRNA inside of cell and also utilized photoblea ching the fluorophore on the MB in order to study the diffusion of

PAGE 39

39 the MB mRNA hybrid. These studies demonstrate th e wealth of information that can be gained through the visualizatio n of MB hybridization inside a single cell. In addition to localization and distribution, expression levels of mRNA have also been studied inside of living cells using MBs. Using a two MB FRET approach the relative expression levels of K-ras and survivin mRNA were determined in human dermal fibroblasts.186 In this study, human dermal fibroblasts and panc reatic carcinoma cells were used to study the expression levels and the localiza tion of K-ras and survivin mRNA. In order to accomplish this, a two MB FRET approach was used. This enco mpassed designing two MBs for adjacent target regions. Once both MBs hybridized to the sa me mRNA, the fluorophores on the MB were brought within a close distance and FRET was allo wed to occur. Using this process they demonstrated the localization of the two mRNA sequences in different cell types and how the different mRNA was localized at di fferent regions inside of the cell. The use of the two MBs allowed for much greater specificity since the si gnal required two separa te hybridization events to be generated. One of the limitations of MBs fo r intracellular analysis has been the inherent variability in the fluorescent measurements obtained. However, it is difficult to attribute these variations to any one cause such as instrument or experimental variability, variations in gene expression, or simply different amounts of probe being delivered in to the cell. In an effort to eliminate the experimental and instrumental variations in orde r to study the stochasticit y of gene expression a method of ratiometric analysis was deve loped and applied to cancer cell genomics.183 In this approach a fluorophore labeled DNA reference probe was injected along with the MBs. The reference probe exhibited a consta nt level of fluorescence relative to the MB signal that resulted in a ratio value for the MB that was independent of instrumental varia tions like the amount of

PAGE 40

40 probe delivered to the cell and expe rimental variations like different cell volumes. This is due to the ratio of the reference probe to the MB being constant, therefore any in crease in the ratio was due to the increase of the molecular beacon fluorescence i ndicating target hybridization. While MBs have shown a great deal of prom ise in these applications, there are some limitations that currently exist. One of the most significant limitations is a susceptibility of the MB to endogenous nucleases and single strand bi nding proteins (SSBs). Both of these can induce a false positive response from the MB and great ly limit the viability of the MB inside of a cell. SSBs can bind to loop region of the MB causing the separation of the fluorophore and quencher while nucleases can cleav e apart the MB also separating the fluorophore and quencher. The opening of the MB from specific and non-specif ic interactions is im possible to distinguish. In the human breast cancer cells studied, MBs rega rdless of sequence are degraded inside of the cell after approximately 30 minutes. This greatly limits the types of applications and processes that can be studied. There have been many strategies developed to deal with these problems. These include the use of modified DNA backbones to imbue the MB with resistance to nucleases like 2-O-methyl bases,187 2-fluoro bases,188 peptide nucleic acids,189 phosphothiolates,190 and locked nucleic acid bases.191 However, the modifications have been shown to affect the hybridization rates of the MBs. Peptide nucleic acid bases have also been used to synthesize MBs however their neutral charge would create solubility issues insi de of living cells. phosphothiolates have also been show n to exhibit toxicity to cells.190 Therefore while different approaches have been tried the issue of intracellu lar stability still require s additional attention. Another issue confronting th e MBs is the problem of low expression genes. These are genes that while biological are very significant express very low copy numbers of mRNA. To overcome this biological issue, MBs require fu rther refinement in the areas of superior

PAGE 41

41 quenching and brighter fluorescence. The pr esence of background fluorescence even in the closed state can hinder the limit of detection of the MB inside of the cell. Therefore improved quenching efficiency is an area that has been explored with promising results. Both gold nanoparticles192,193 and novel superquenching moieties194 have been utilized to achieve lower fluorescence backgrounds. Additionally, brighter fl uorescent moieties have also been explored. These include the use of quantum dots195 and fluorescent polymers196/ to make MBs as bright as possible. The continuing improvement of MB de sign and synthesis furthe r extends its potential as an intracellular pr obe however many of these improvement s have yet to make the transition from buffer solutions to the complex realm of the cell. Intracellular Ion Levels Beyond mRNA expression, there are ot her types of molecules that are critically important to understanding the underlying biol ogical function inside of the ce ll. These are the proteins and the ions inside of the cell. Th e proteins are the functional struct ures that carry out the biological processes in the cells. While some methods fo r studying proteins expre ssion inside of the cell exist, they involve altering th e genome of cells through the use of fluorescent proteins, in particular they involve altering the gene sequence of the very pr oteins to be studied. Other methods involve the use of fluorophore labeled antibodies, however with no inherent signal transduction method it is difficult to diffe rentiate the bound antibodies from the unbound antibodies. Ions on the other ha nd also perform many of the critic al functions inside the cell as signaling molecules, stabilizing proteins structur es, and enabling many protein functions. Unlike proteins however, there many mol ecules that exist that can chel ate ions with a good degree of specificity. The careful manipula tion of there structure has produc ed molecules that will chelate selectively to certain ions to produce a fluores cent species. This al lows for the fluorescent

PAGE 42

42 detection of the ions even inside of cells. Ther efore if one can study gene and ion levels inside of single cells more in depth st udies of the fundamental biologi cal processes can be achieved. Therefore the work described herein not only focuses on the early detection of cancer cells in samples ranging simple to complex but also on developing methods to analyze the cells. This way once a type of cancer is detected, the actu al cancer cells themselves can be collected, cultured, and further studied to re veal a specific molecular diagnos is and what treatment options will be the most effective. This would allow a more personalized treatment of the illness leading to ultimately not only a more effective treatment but one with fewer side effects. Also, when the cancer is detected at an early stage it responds much more favorable to treatment and leads to ultimately a much more favorable prognosis for the patient.

PAGE 43

43 CHAPTER 2 DNA AND NANOPARTICLE SYNTHESIS DNA Synthesis Before using the various materials and probes that have been men tioned, the first step encompasses actually making and characterizing the various constituents. After their synthesis, the different parts can be combined together to create nanomaterials and probes that will allow the detection and study of cancer cells. Therefore, this chapte r will focus on the synthesis of the DNA portions of the materials and the synthesis and characterizati on of the nanomaterials. Once both types of materials have been synthesized they can either be used individually or conjugated together and used for various applications. One of the primary advantages of the aptamer is that it is based on DNA and therefore cannot only be chemically synthesized, it can be ch emically synthesized with different functional groups or other molecules integrated into its stru cture. This makes aptamers ideally suited for coupling to different nanoparticles using well established conjugation strategies. Similarly, molecular beacons also benefit from the chemi cal synthesis of DNA as the whole molecular beacon structure from the quencher to the fluoro phore can be synthesized on column. However, it is often advantageous to use off column coup ling of the fluorophores. While resulting in an overall lower yield due to reaction efficiencies and increased purification steps, the fluorophores are not subjected to the harsh deprotection schemes and higher qua lity fluorophores not available as a phosphoramidite can be used. The higher qual ity fluorophores, such as the Alexa Fluor line available from Invitrogen, have better environmental stabilitie s, resistance to photobleaching, higher quantum efficiencies, and higher molar abso rptivities yielding a more stable and sensitive molecular beacon. Therefore, probes based on DNA offer many advantages due to its the versatile and well established chemical synthesis process.

PAGE 44

44 Several different methods for the chemical s ynthesis of DNA have been reported including the use of phosphodiesters,197 phosphoramidites,198 and the H phosphonate method.199 Among these, the phosphoramidite method remains the mo st commonly used as it is useful in the production of long oligonucleotide strands with ar tificial modifications in high yields and purity which is much more difficult with other synthetic routes.200 Synthesizing oligonucleotides using phosphoramidite chemistry was first developed by Caruthers and later improved by Kosters and others. It involves the us e of a series of cycled controlled reactions utilizing either a liquid or solid phase. For each repetition of the phos phoramidite cycle, one base is added on the sequence until the desired DNA sequence has been completed. The aptamers and molecular Figure 2-1. In automated DNA chemical synthesi s three primary steps, Detritylation, Coupling, and Oxidation are repeated until the desi red DNA sequence has been synthesized.

PAGE 45

45 beacons covered herein were synthesized on an ABI 3400 DNA/RNA synthe sizer using standard phosphoramidite chemistry. As the solid phase synthesis using phosphoramidite chemistry was used in the production of the aptamers and molecu lar beacons, it will be covered in more detail. The reaction cycle of the solid phase synthesis can be broken down into three critical steps referred to as Detritylation, Coupling, a nd Oxidation as shown in Figure 2-1. For the solid phase synthesis, the starting mo lecule whether the DNA ba se, biotin, or other molecule is immobilized on a controlled porous gl ass support (CPG) at the 3 position. Thus, the synthesis proceeds from the 3 end to 5 e nd of the DNA sequence. The Detritylation step involves removing the dimethoxytrityl (DMT) prot ecting group from the nucleoside, leaving a hydroxyl group that forms the foundation for the ne xt reaction in the cycle. The DMT group is removed by adding a dilute acid so lution of either dichloroacetic acid or trichloroacetic acid in dichloromethane. The DMT group leaves the nucl eoside and the column is washed to remove any extra reagents or byproducts yielding the hydroxyl group. Next is the Coupling step in the reaction in which the next base is added onto the chain. This i nvolves the addition of the next nucleotide as a phosphoramidite deriva tive and tetrazole to the reaction mixture. As depicted in Figure 2-2, the tetrazole activat es the phosphoramidite cleaving o ff protecting group leaving the phosphoramidite vulnerable to the nucleophilic attack of the h ydroxyl group. This forms an unstable phosphate linage and the excess reactants and byproducts and washed away. After the coupling step, the uproduct is then st abilized in the Oxida tion step. The product is stabilized via the oxidation of the phosphite linkage to a phosphate linkage through the addition of dilute iodine, pyridin e, and tetrahydrofuran. The produc t is now about to repeat this series of steps until the full sequence of the strand is synthesized.

PAGE 46

46 In addition to these steps, a nother critical process referred to as capping occurs after the coupling step to ensure any failed sequences are prevented from further reaction. Failed sequences occur because the coupling step yield is not always 100%. This would leave some reactive hydroxyl groups that could react in subs equent cycles resulti ng in deletions in the sequence. The deletion of a ba se could drastically alter the properties of the probe being synthesized whether it is a molecular beacon or an aptamer. In addition, the deletion would be difficult to separate using standard purification techniques. Therefore, to remove the unreacted hydroxyl groups a Capping step is performed after the coupling step. The Capping step involves adding acetic anhydride and N-methylimidazole re sulting in the acetylation of any unreacted hydroxyl groups. The acetylation prevents the ch ain from ongoing further coupling reactions resulting in a much shorter fragment. The s horter fragments are easily purified from the complete sequence resulting in a much purer final product. Figure 2-2. The Coupling step in DNA synthesis de picting the activiation of the phosphoramidite derivative and the subsequent coup ling of the two nucleotides. After the synthesis is finish ed, the oligonucleotide must be deprotected and removed from the solid support. This is accomplished by incubating the DNA/CPG complex in ammonium hydroxide for 2 hours at 65C followed by centrifuga tion to remove the CPG. The DNA is then put in the SpeedVac to evaporate the ammonium hydroxide. This is followed by resuspension in

PAGE 47

47 0.1M TEAA. The sequence is then purified usin g reverse phase HPLC on a C18 column running a gradient of acetonitrile to 0.1M TEAA in HPLC-grade water. This protocol does require some optimization depending on the different constituents of the DNA sequence. The HPLC separates the full sequence from those sequen ces that failed during the synthesis procedures. This is an important st ep as any sequence that has not been properly synthesized can greatly affect future experiment al results. For example, if a molecular beacon sequence has been synthesized and some of th e sequences have been synthesized without a quencher, that probe will have a much highe r background fluorescence. This will limit the sensitivity and signal enhancement of the beaco n. Conversely, if a beacon sequence or an aptamer sequence has been synthesized without a fluorophore, its hybridization will not only be undetectable but it will block fluorophore labele d sequences creating a lower response to the target molecule. To illustrate this phenom enon, Figure 2-3 depicts a typical molecular beacon purification. The PDA detector of the HPLC can measure the absorbance spectra on column which allows for the detection of the individual constituents. The DNA po rtions absorb in the 250nm to 300nm range while the quen chers and fluorophores absorb in their characteristic range. Figure 2-3 shows that there are three major co mplexes that can be separated, a fluorophore labeled sequence, an unlabeled sequence, and the full probe sequence la beled with a fluorophore and quencher thus highlighting th e need for proper purification. Once the sequence has been pr operly prepared, it can then be used in experiments or undergo further modification through the conjugation to different nanomaterials, the next section details the preparation and char acterization of the nanomaterials used in the work described herein followed by the conjugation schemes us ed to attach the DNA based probes to the nanomaterials.

PAGE 48

48 Figure 2-3. Typical chromatograph of a molecular beacon purificat ion using reverse phase HPLC using absorbance for detection. Nanoparticle Synthesis In order to separate the target species fr om complex biological samples, it is often necessary to perform a separation procedure. However, when sepa rating targets as large as cells there are few techniques that can be applied to keep them int act. Centrifugation and magnetic separations are the two most co mmon procedures to cell separati ons. While centrifugation is a simple and effective method, it does not allow for the selective separation of cells. However, when selective biorecognition elements like apta mers or antibodies are conjugated to magnetic particles, cells can be separated from other ce lls with good selectivity. Therefore, magnetic nanoparticles were synthesized to enable the selective ex traction of the target cells from complex samples. The iron oxide core magnetic nanopart icles were prepared by coprecipitation of iron oxide salts. Solutions of amm onium hydroxide (2.5%) and iron ch lorides were added together under nitrogen, and continuously stirred at 350 RP M using a mechanical stirrer for 10 minutes with a final volume of around 255mL. The iron chlo ride solution was made from ferric chloride hexahydrate (0.5 M), ferrous chlo ride tetrahydrate (0.25 M), and HCl (0.33 M) at a final volume of approximately 100 mL. The ammonium hydroxide solution was diluted in a 500 mL beaker at

PAGE 49

49 a volume of about 155 mL. The bulk solution of th e formed iron oxide nanoparticles was stored as is at room temperature until needed for further experiments. The problem with iron oxide however is that it is very difficult to conjugate biomolecules to as it lacks any usable functional group and th ere are difficult to att ach to iron oxide. To compensate for this, the iron oxide nanoparticles were coated with silica following the Stober sol gel protocol. To begin, a 6 mL aliquot of th e magnetic particle solution was magnetically extracted and washed three times with water and once with ethanol, and the washed samples were suspended in an ethanol solution cont aining ~1.2 % ammonium hydroxide. The final concentration of the washed particles was ar ound ~7.5 mg/mL, which contained 60 mg of iron oxide in 6 mL of solvent. To this sample, 210 L of tetraethoxyorthosil icate (TEOS) was added and then sonicated for 90 minutes to complete th e hydrolysis process to form the silica coating. A postcoating layer of pure silica was created by adding an additional 10 L aliquot of TEOS, and the sample was sonicated for an additional 90 minutes. The resulting silica nanoparticles were magnetically extracted and washed three tim es with a 6mL aliquot of ethanol to remove excess reactants. The silica coating allows for a large variety of conjugation schemes to be used to prepare the aptamer conjugated nanoparticles. Next, the fluorophore doped nanoparticles were synthesized. The inorganic fluorophore based nanoparticles like the RuBpy nanoparticles were prepared by the reverse microemulsion method. First, 1.77mL Triton x-100 7.5mL cyclohexane, 1. mL nhexanol were added to a 20 mL glass vial with constant ma gnetic stirring. Then, 400 L of H2O and 80 L of 0.1M tris(2,2 bipyridyl) dichlororuthenium (II) hexahydrate (Rubpy) dye (MW=748.63) were added, followed by the addition of 100 L Tetraethyl Or thosilicate (TEOS). Afte r thirty minutes of stirring, 60L NH4OH was added to initiate silica polymerization.

PAGE 50

50 Tetramethylrhodamine (TMR SE) and Cy5 doped NPs were synthesized according to the following method: TMR SE and Cy5-NHS were each dissolved in DMSO at a concentration of 5 mg/mL, and (3-aminopropyl)triethoxysilane (APT S) was added at a molar ratio of 1.2:1 APTS:dye. The APTS was allowed to conjugate to the amine reactive dye for 24 h in the dark with shaking prior to synthesi s of the particles. Glass reac tion vessels and Teflon-coated magnetic stir rods were washed with 1 M NaOH solution for 30 min, rinsed with DI water and ethanol, and allowed to dry. This wa sh step was performed to clean the glass vessel and stir rods and smooth the inside surface of the glass vess el, which prevents unwanted seeding and NP formation. After conjugation, 4.19 mL of ethanol was mixed with 239 L of ammonium hydroxide solution in the reaction vessel. A 36 L volume of TMR-APTS conjugate or 54 L of Cy5-APTS conjugate was added to the reaction vessels, yielding 3.44 10-7 mol of dye/reaction (ratio of 2300 mol of silica/mole of dye). A 177 L volume of TEOS was added rapidly to the reaction mixture, and the vessels were sealed. The reaction was allowed to proceed for 48 h in the dark before the particles were recovered by centrifugation at 14 000 rpm. The particles were washed three times with phosphate buffer to re move any dye molecule s that are weakly bound. Nanoparticle Characterization. Following the synthesis of th e nanoparticle, they still needed characterization to ensure the proper sy nthesis of the iron oxide nanoparticles and the successful coating with silica. A Hitachi H-7000 transmission electron microscope (TEM) was used to obtain the size and shape of the formed nanoparticles from dried samples. The uncoated magnetite particle were found to be roughly 10 2 nm in diameter, however these particles had poor uniformity in shape. Based on the TEM im ages, the silica coated nanoparticles had an average diameter of 65 nm and were uniformly spherical in shape. However, the size of silica

PAGE 51

51 coated iron oxide particles was more variable ranging from 30nm to 90nm in diameter. Representative images for each are shown in Figure 2-4. Figure 2-4. TEM images of iron oxide nanoparticle (A) and silica coated iron oxide nanoparticles (B). Given the difficulty in analyzing samples on the TEM instrument, characterization of the particles was also performed usi ng light scattering for routine ch aracterizations. Light scattering and zeta potential measurements were perfor med on 1270 Brookhaven Zeta Plus instrument. The light scattering can effectivel y measure the size of the nanopart icles while the zeta potential is useful for probing the surface ch arge of the particles and was mainly used for verifying the surface modifications. After synthesis and characterization of th e nanoparticles, the nanoparticles need to be functiona lized for conjugation to the bi omolecules used for recognition. Fluorescent nanoparticles were ch aracterized in a similar fash ion although their fluorescence spectra were also analyzed to ensure that th e fluorophore was properly en capsulated inside tzhe silica matrix. Nanoparticle Bioconjugation to DNA Silica Nanoparticles Since both types of particles both magnetic a nd fluorescent were comprised of silica, the same conjugation schemes can be utilized for eith er type of particle. Tw o different strategies AB AB

PAGE 52

52 were utilized for attaching the biomolecules to the silica nanoparticles. The first involved the use of a biotin/avidin conjugation. Fo r avidin coating, a 0.1 mg/mL Fe3O4-SiO2 (silica coated magnetic nanoparticles) solution and a 5 mg/mL av idin solution were mixed and then sonicated for 5-10 minutes. The mixture was incubated at 4 C for 12-14 hours. The particles were then washed three times with 10 mM phosphate buffe red saline (PBS) pH 7.4 and dispersed at 1.2 mg/mL in 10 mM PBS, and the avidin coating was stabilized by cross-linking the coated nanoparticles with 1% glut araldehyde (1 hour at 25 C). After another separation, the particles were washed three times with 1M Tris-HCl bu ffer. Then, the particles were dispersed and incubated in the 1M Tris-HCl buffer (3 hours at 4 C), followed by three washes in 20 mM TrisHCl, 5 mM MgCl2, pH 8.0. In order to verify that the av idin had been coated on the surface, the Zeta potential before and after co ating was measured. Prior to th e coating step, the silica surface of the nanoparticle had a Zeta potential of -19.59 while after coating the Ze ta potential was -4.37. This change in surface charge can be explained by the addition of the positively charged avidin to the surface thus neutralizing the ne gative charge of the silica surface. Now that the avidin is on the surface of the nanoparticle, biotinylated DNA sequences can be attached. First the partic les are dispersed in 0.2 mg/mL in 20 mM Tris-HCl, 5 mM MgCl2, pH 8.0. Biotin labeled DNA was added to the solution at a concentration of 30 M. The reaction was incubated at 4 C for 12 hours, a nd three final washes of the particles were performed using 20 mM Tris-HCl, 5 mM MgCl2 at pH 8.0. With the DNA now attached to the nanoparticle the surface of the na noparticle is now more negatively charged due to the negative charge of the oligonucleotid es. This is reflected in the new Zeta potential for the particle that is now measured at -14.56. Now the DNA conjugated nanoparticle can be used for a variety of different applications.

PAGE 53

53 The other methodology used to conjugate the DNA to the silica nanopar ticles involves the functionalized silica nanoparticle s and exploiting those functional groups for the attachment of DNA also functionalized with different groups. Typically carboxy groups are added to the nanoparticle surface and amine modified DNA is us ed to conjugate the DNA to the nanoparticle. However, this basic methodology also works for amine modified nanoparticles with a carboxy or thiol functionalized oligonucleot ide or thiolated nanoparticles w ith thiol or amine modified DNA. To obtain carboxy modified nanopart icles, a post coating step is required. After the silica polymerization steps, the pos t coating process begi ns with adding 50 L TEOS, 40 L carboxylethylsilanetriol sodium sa lt, and 10 L 3-(Trihyd roxyl)propyl methyl phosphonate. At this point, polymerization proceeds for 18 hours, and particles were centrifuged, sonicated, and vortexe d four times with 95% ethanol followed by one wash with H-2O. The silica nanoparticles can then be modi fied with DNA by addi ng 1.2 mg EDC, 3.5 mg Sulfo-NHS, and 0.5 nmoles DNA with 2 mg of pa rticles dispersed in 1.5 mL of 10 mM MES buffer (pH= 5.5). The solution is then mixed for three hours. The nanopart icles are then washed by centrifugation at 14000 rpm three times w ith 0.1 M Phosphate Buffered Saline (PBS) (pH=7.2). The post coating and amine attachment can be verified for either method by again measuring the Zeta potential. After the addition of the carboxy groups on the surface of the nanoparticle, the Zeta potential drops to 43.27 due to the negatively charged carboxy groups. After the amine modification, the Zeta poten tial increases to -33.31 representing the neutralization of the carboxy groups through the reaction with the positively charged amines. After this process the DNA conjugated silica nanoparticles ar e ready for use. Gold Nanoparticles While the avidin and functionalized silica coatings are quite useful for silica based nanoparticles, a different strategy is required fo r gold nanoparticles. While either method in

PAGE 54

54 theory could be used, the silica and avidin co ating would insulate th e gold nanoparticle and prevent its surface plasmons from interacting with other gold nanoparticles. This would prevent the change in spectral propertie s that makes the gold nanopartic les so useful for detection. Therefore to attach oligonucleot ides to gold nanoparticles, the well established gold/thiol based chemistry is utilized to attach the aptamer directly to the na noparticle surface. In a 2mL microcentrifuge tube, 1mL of the 20nm gold colloid nanoparticles ( GNPs), containing 7.0x1011 particles/mL, taken directly from the manuf acturer (Ted Pella, Inc. Redding, CA) was centrifuged for 15 minutes at 14,000 RPM. Th e GNPs were washed three times with 1mL aliquots of 5mM phosphate buffer (PB) pH 7.5 by decanting the supernatant, adding fresh PB, dispersing by sonication, and centrifuging for 15 minutes at 14,000 RPM. From decantation to dispersion, the wash step was performed within 35 minutes. After the final wash step, the GNPs were dispersed in 1 mL of the PB. To each washed GNP sample, 150 L of a 1 M thiol labeled DNA sequence was incubated for 3-5 days at 4C. The samples were sonicated to disperse the GNPs every 12 hours. When the incubations were completed, the samples were centrifuged at 14,000 RPM for 5 minutes and the samples washed as described previously with the PB. After the final wash, each GNP sample was disper sed in 0.25mL PB with approximately 6.0x1011 particles, and the samples stored at 4C until used. Conclusion By following these various synthetic protoc ols, biofunctionalized nanoparticles can be constructed for a wide array of applications. Nanomaterials fo r magnetic separations, fluorescent detection, or colorimetric detection can then appl ied to the collection and detection of a variety of biomolecules such as proteins, peptides, DNA, RNA, or even entire cells. The next two chapters will focus on using the nanomaterials for the detection of can cer cells through both fluorescent and colorimetric detection. Chapter 3 will focus on the use of two different types of

PAGE 55

55 nanoparticles to accomplish both the collection of th e cells and their detec tion. This approach uses the magnetic nanoparticles to magnetically se parate the cells of interest while fluorescent nanoparticles enable their sensi tive detection. Then, Chapter 4 will demonstrate the use of gold nanoparticles for a colorimetric assay for cancer cell detection. After detection of the cells, further analysis of cancer cells will be de monstrated using other DNA based probes thus illustrating the investigatory power of nanoma terials combined with DNA based probes.

PAGE 56

56 CHAPTER 3 TWO PARTICLE ASSAY FOR THE COLLECT ION AND ENRICHMENT OF TARGETED CANCER CELLS Introduction and Scheme In the previous chapter, the synthesis and bioconjugation schemes were detailed for the preparation of aptamer conjugated nanoparticles for the detection of cancer cells. The majority of cancers including lung cancer arise from epithelial surfaces. Often as the tumors develop, they exfoliate cells spontaneously in to bodily fluids like blood an d sputum. These cancerous cells often contain genetic or molecu lar abnormalities, but most of th ese abnormalities are not clearly defined and there are essentially no biomarkers that can distinguish them from normal cells. The exfoliated cells are generally fa r outnumbered by normal cells in bodily fluids especially during the early development of the dis ease when it is most treatable. Therefore a tec hnology that is able to collect, enrich, and preserve abnormal ce lls from bodily fluids co uld greatly assist cancer diagnosis and the discovery of biomarkers whil e allowing earlier detect ion of the disease. To accomplish the collection and detection of cancer cells from bodily fluids, we have used a group of aptamers for specific cancer cells and then conjugated the aptamers to nanoparticles to allow not only th e collection, enrichment, and pres ervation of the cancer cells, but also the rapid detection of the cancer cells directly from bodily fluids. The method involves the use of two separate aptamer-conjugated na noparticles, one magnetic nanoparticle and one fluorescent nanoparticle. The magnetic nanoparticle allows for the collection and enrichment of the cancer cells while the fluorescent nanopartic le clearly marks the cell for fluorescence detection. The overall strategy is schematically shown below. While ma ny magnetic extraction techniques also collect non-targ et cells through non-specific inte ractions, the use of the second fluorescent nanoparticle offers a second level of di scrimination. This is because only the target cells will have high fluorescence intensity from being covered with the aptamer-conjugated

PAGE 57

57 fluorescent nanoparticles and the unbound fluorescent nanoparticles wi ll be washed away during the magnetic extraction. Thus, the use of the two different type s of nanoparticles gives the assay the ability for the efficient collection characte ristic of the magnetic nanoparticles while the fluorescent nanoparticles enable the sensitive detection of the cells. Magnet Magnet Figure 3-1. Schematic representation of the two pa rticle assay. In the assay, the magnetic (blue) and fluorescent (red) nanoparticles bind to the target cell. After a magnetic field is applied the magnetic nanoparticles immob ilize the target cells and any bound fluorescent nanoparticles wh ile the unbound cells and fl uorescent nanoparticles are washed away. Methods and Materials All materials were purchased from Sigma-Al drich (St. Louis, MO ) unless other noted. Whole blood samples were obtained from Res earch Blood Components, LLC (Brighton, MA). Fluo-4 was purchased from Molecu lar Probes (Eugene, OR), and car boxylethylsilanetriol sodium salt was purchased from Gelect, Inc. (Morrisv ille, PA). N-hydroxysulfosuccinimide (SulfoNHS) and 1-Ethyl-3-[3-dimet hylaminopropyl] carbodiimide Hydrochloride (EDC) were purchased from Pierce Biotechnology, Inc.( Rockfo rd, IL). Hydrochloric acid and Ammonium Hydroxide were obtained fr om Fisher Scientific. DNA Aptamer Synthesis. The following aptamers have been selected for the CCRF-CEM, Ramos, and Toledo cells respectively: 5'-TTT AAA ATA CCA GCT TAT TCA ATT AGT CAC ACT TAG AGT TCT

PAGE 58

58 AGC TGC TGC GCC GCC GGG AAA ATA CT G TAC GGA TAG ATA GTA AGT GCA ATC T-3'; 5'-AAC ACC GGG AGG ATA GTT CGG TGG CT G TTC AGG GTC TCC TCC CGG TG-3'; 5'-ATA CCA GCT TAT TCA ATT ATC GTG GGT CAC AGC AGC GGT TGT GAG GAA GAA AGG CGG ATA ACA GAT AAT AAG ATA GTA AGT GCA ATC T-3'. Both the amine and biotinylated versions of th e aptamer sequencers were synthesized in-house. An ABI3400 DNA/RNA synthesizer (Applied Biosystems, Foster City, CA) was used for the synthesis of all DNA sequences. A ProStar HPLC (Varian, Walnut Creek, CA) with a C18 column (Econosil, 5u, 250 4.6 mm) from Allt ech (Deerfield, IL) was used to purify all fabricated DNA. A Cary Bio-300 UV spectrometer (Varian, Walnut Creek, CA) was used to measure absorbencies to quantify the manuf actured sequences. All oligonucleotides were synthesized by solid-state phosphoramidite chemistry at a 1 mol scale. The completed sequences were then deprotected in concentrated ammonia hydroxide at 65 C overnight and further purified twice with reve rse phase high-pressure liquid chromatography (HPLC) on a C-18 column. Fluorescent Nanoparticle Synthesis Dye doped nanoparticles were synthesized by the reverse microemulsion method.21 First, 1.77 mL Triton x-100, 7.5 mL cyclohexane, 1.6 mL nhexanol were added to a 20 mL glass vial with constant magnetic stirring. Then, 400 L of H2O and 80 L of 0.1M tr is(2,2 bipyridyl) dichlororuthenium (II) hexahydrate (RuBpy) dye (MW=748.63) were added, followed by the addition of 100 L Tetraethyl Or thosilicate (TEOS). After thir ty minutes of stirring, 60L NH4OH was added to initiate silic a polymerization. After 18 hours, the carboxyl modified silica post-coating was initiated by adding 50 L TEOS, 40 L carboxylethylsilanetriol sodium salt, and 10 L 3-(Trihydroxyl)propyl methyl phosphona te. Polymerization proceeded for 18 hours, and particles were centrifuged, sonicated, and vorte xed four times with 95% ethanol, followed by

PAGE 59

59 one wash with H2O. Carboxyl functionalized RuBpy na noparticles were modified with DNA by adding 1.2 mg EDC, 3.5 mg Sulfo-NHS, and 0.5 nm oles DNA with 2 mg of particles dispersed in 1.5 mL of 10 mM MES buffer (pH= 5.5). The solution was then mixed for three hours. Particles were then washed by centrifugation at 14000 rpm three times with 0.1 M Phosphate Buffered Saline (PBS) (pH=7.2). RuBpy nanopart icles were stored at room temperature and were dispersed in cell media buffer at a final concentration of ~10 mg/mL. Tetramethylrhodamine (TMR SE) and Cy5 doped NPs were synthesized according to the following method: TMR SE and Cy5-NHS were each dissolved in DMSO at a concentration of 5 mg/mL, and (3-aminopropyl)triethoxysilane (APT S) was added at a molar ratio of 1.2:1 APTS:dye. The APTS was allowed to conjugate to the amine reactive dye for 24 h in the dark with shaking prior to synthesi s of the particles. Glass reac tion vessels and Teflon-coated magnetic stir rods were washed with 1 M NaOH solution for 30 min, rinsed with DI water and ethanol, and allowed to dry. This wa sh step was performed to clean the glass vessel and stir rods and smooth the inside surface of the glass vess el, which prevents unwanted seeding and NP formation. After conjugation, 4.19 mL of ethanol was mixed with 239 L of ammonium hydroxide solution in the reaction vessel. A 36 L volume of TMR-APTS conjugate or 54 L of Cy5-APTS conjugate was added to the reaction vessels, yielding 3.44 10-7 mol of dye/reaction (ratio of 2300 mol of silica/mole of dye). A 177 L volume of TEOS was added rapidly to the reaction mixture, and the vessels were sealed. The reaction was allowed to proceed for 48 h in the dark before the particles were recovered by centrifugation at 14 000 rpm. The particles were washed three times with phosphate buffer to re move any dye molecule s that are weakly bound. The synthesis method was found to reproducibly pr oduce a number average particle size of 50

PAGE 60

60 nm 5 nm with a monomodal distribution when measured w ith a Honeywell UPA 150 dynamic light scattering instrument. Magnetic Nanoparticle Synthesis The iron oxide core magnetic nanoparticles32 were prepared by means of precipitating iron oxide by mixing ammonia hydroxide (2.5%) and ir on chloride at 350 RPM using a mechanical stirrer (10 minutes). The iron chloride solution contains ferric chlori de hexahydrate (0.5 M), ferrous chloride tetrahydrate (0.25 M), and HCl (0.33 M). After three washes with water and once with ethanol, an ethanol solution containing ~1.2 % ammonium hydroxide was added to the iron oxide nanoparticles, yielding a fi nal concentration of ~7.5 mg/mL. To create the silica coating for the magnetite core particles, tetraethoxyorthosilicate (200 L) was added, and the mixture was sonicated for 90 minutes to complete the hydrolysis process. For post coating, an additi onal aliquot of TEOS (10 L) was added and additional sonication was performed for 90 minutes. The resulting nanopartic les were washed three times with ethanol to remove excess reactants. For avidin coating, a 0.1 mg/mL Fe3O4-SiO2 (silica coated magne tic nanoparticles) solution and a 5 mg/mL avidin solution were mi xed and then sonicated for 5-10 minutes. The mixture was incubated at 4 C for 12-14 hours. Th e particles were then washed three times with 10 mM phosphate buffered saline (PBS) pH 7.4 a nd dispersed at 1.2 mg/mL in 10 mM PBS, and the avidin coating was stabilized by crosslinking the coated nanoparticles with 1% glutaraldehyde (1 hour at 25 C). After another separation, the particles were washed three times with 1M Tris-HCl buffer. Then, the particles we re dispersed and incubated in the 1M Tris-HCl buffer (3 hours at 4 C), followed by three washes in 20 mM Tris-HCl, 5 mM MgCl2, pH 8.0.

PAGE 61

61 DNA was attached to the particles by dispersi ng the particles at 0.2 mg/mL in a buffer of 20 mM Tris-HCl and 5 mM MgCl2 at a pH 8.0. Biotin labeled DNA was added to the solution at a concentration of 31 M. The reaction was incubated at 4 C for 12 hours, and three final washes of the particles were perfor med using 20 mM Tris-HCl, 5 mM MgCl2 at pH 8.0. Magnetic nanoparticles were used at a final conc entration of ~0.2 mg/mL and stored at 4 C before use. Cell Culture CCRF-CEM cells (CCL-119 T-cell, human acu te lymphoblastic leukemia), Toledo cells (CRL-2631, non-Hodgkin's B cell lymphoma) a nd Ramos cells (CRL-1596, B-cell, human Burkitts lymphoma) were obtained from ATCC (American Type Culture Association). The cells were cultured in RPMI medium supplemen ted with 10% fetal bovine serum (FBS) and 100 IU/mL penicillin-Streptomycin. The cell density was determined using a hemocytometer prior to any experiments. After which, approximately one million cells dispersed in RPMI cell media buffer were centrifuged at 920 rpm for five minute s and redispersed in dye free cell media three times, and were then redispersed in 5 mL dye fr ee cell media. During all experiments, the cells were kept in an ice bath at 4oC. Two Particle Assay Methodology Approximately 105 of the cells were obtained in individual test tubes. To each cell sample, 5 L of the MNP solution was added, and the mixture was incubated for 15 min. After incubation, the cells were washed by magnetic extr action with 200 L of fresh cell media three times and resuspended in 200 L of the media buffer. The wash was performed by removal of the supernatant and addition of fresh buffer, and the sample was resuspended in the fresh buffer typically within 3-5 min. To complete the stepwise process, 2 L of FNPs was added and incubated for 5 min. The concentration of MNPs to FNPs in the samples was 2:1. Again, the

PAGE 62

62 sample was washed three times with 200 L of cell media as described previously and then dispersed in 20 L of media for imaging and microplate reader analysis. Fluorescence Imaging. All cellular fluorescent images were collected using the confocal microscope setup. The confocal consists of an Olym pus IX-81 automated fluorescence microscope with a Fluoview 500 confocal scanning unit. Ther e are three lasers providing la ser excitation at 458nm, 488nm, 514nm, 543nm, and 633nm. The TMR nanoparticles were excited at 543nm and collected at 570nm. Cy5 nanoparticles were excited at 633nm and collected at 660nm. RuBpy nanoparticles were excited at 488nm and the emission was collect ed at 610nm. Images were taken after a five to ten second period during which the instrument was focused to yield the highest intensity from the fluorescence channels. The images were assigned color representations for clarity and are not indicative of the actual emission wavelengths. Flow Cytometry Fluorescence measurements were also made using a FACScan cytometer (Becton Dickinson Immunocytometry Systems, San Jose, CA). To support imaging data, RuBpy fluorescence of pure sample s initially containing 105 cells were measured by counting 30000 events. Cell experiments were performed exactly as stated for imaging experiments, except all solutions were diluted to a final volume of 200 L Cell sorting allowed for accurate quantitative analysis of cell samples, as well as a plat form for collection efficiency determination. Plate Reader Measurements Fluorescence measurements were taken using a Tecan Safire Microplate Reader in a 384 well small volume plate. 20 L aliquots from each sample were deposited in the well and sample fluorescence intensity at defined wavelengths were measured at a constant gain at 5nm slit widths. TMR based nanoparticles were excited with 550nm light and its emission was measured

PAGE 63

63 at 575nm. RuBpy nanoparticles were excited at 458nm and its emission was measured at 610nm. Cy5 was excited at 640nm and its emission was detected at 670nm. Two Particle Assay Proof of Concept The first in evaluating the assay was simply trying to extract the ta rget cells and ensure that both the magnetic and fluorescent nanopar ticles were binding to the target cells. Additionally, it was similarly important that the pa rticles not bind to the control cells. If the assay was going to function well enough for any future clinical analysis it needed above all to be as selective as possible as the targeted cells in a clinical sample would be overwhelmed by nontarget cells and other species. Sensitivity is also a secondary concern however with the magnetic extraction larger initial sample vol umes could always to be used to detect a clinically relevant level of cells. To demonstrate the concept of our tw o particle based magnetic collection and detection methodology, individual CEM and Ramos cell samples were analyzed according to our two particle protocols, followed by fluorescent imaging and analysis using flow cytometry. Before nanoparticle incubation, cells were disp ersed in 500 L cell media buffer and centrifuged three times at 920 rpm for five minutes, and were then redispersed in 200 L media buffer. Fluorescent and magnetic nanopartic le solutions were then adde d to the cell samples using excess of the fluorescent nanopartic les. After a five minute inc ubation with the nanoparticles, the cells were magnetically extracted and wash ed with 500 L of buffer three times, and redispersed in 20 L buffer for imaging and 200 L buffer for flow cytometric analyses respectively. All pure sample experiments started with 1.0 x 105 5.0 x 105 cells before nanoparticle incubation. Each pure cell extracti on was repeated 10 times. Figure 3-2A shows representative confocal images of 2 L aliquots of target cell s after five minute incubation and three magnetic extractions and Figure 3-2B shows th e results for the contro l cells after the same treatment.

PAGE 64

64 Figure 3-2. Images of extracted samples from (A) target cells and (b) cont rol cells, and (C) flow cytometric comparison of target and cont rol signal after 5 minute incubation with magnetic and fluorescent nanoparticles, followed by three washes by magnetic separation. There is a noticeable change in both the am ount of cells present and fluorescent signal between the extracted cell solutions. Magnetic collection pulled out fe w control cells, while a significant number of target cells were extracted using the same procedures. In addition, the few control cells inadvertently collected by magne tic extractions were labeled with few RuBpy nanoparticles and had no significan t fluorescent signal. Conversel y, the target CEM cells that were subjected to the assay had very intens e fluorescent signals that made them easily distinguishable from the control cells. The flow cytometric analysis of the pure sample assay, Figure 3C confirms that fewer control cells were collected than target cells, and the control cells showed less fluorescent emissions than the extracte d target cells. Thus, the selectivity of the assay appears to be very good and if the assay can perform similarly on more complex samples the assay could prove quite useful for clinical samples. With the initial proof of concept demonstrated, the next series of experi ments will focus primarily on validation and characterization of the assay.

PAGE 65

65 Characterization of the Two Particle Assay Dye and Nanoparticle Fluores cent Intensity Comparison The rationale for using the fluorescent nanoparticles was to increase the amount of fluorescence signal thus making the assay more se nsitive than using si mply fluorophore labeled aptamers. Since the fluorescent nanoparticles in corporate thousands of individual fluorophores it should have a significantly higher intensity than a single fluorophore. However since fluorophores are known to self quench and the la rger nanoparticles may block some of the binding sites for other nanoparticles it was unclear the actual level of enhancement that would be obtained by using the nanoparticles. To demonstr ate the fluorescence enha ncement capabilities of RuBpy doped nanoparticles, individual RuB py probes were linked with our DNA aptamer and directly compared to RuBpy Nanoparticle-aptamer conjugates after immob ilization on our target cells. Equal concentrations of magnetic and RuBpy nanoparticle s (0.5 nM) were incubated with CEM cells, then washed by magnetic extraction with 500 L media buffer three times, and redispersed in 20 L buffer for imaging and 200 L buffer for flow cytometric analysis. Figure 3-3A and 3-3B compare cell extr actions labeled with fluorescen t nanoparticles to extractions labeled with RuBpy dye. There is a significant difference in th e amount of fluorescent signal seen in the two images. Flow cytometry was used to verify that the RuBpy nanoparticles provide enhanced fluorescence signal, and Figure 3-3C c onfirms over a 100-fold enhancement of RuBpy nanoparticle labeled cells to RuBpy dye labeled cel ls. This figure also shows the nanoparticle labeled cells in an apparent bimodal distributi on. While the exact cause of this pattern is unknown possible explanations include the formati on of nanoparticle aggr egates, the formation of cell/nanoparticle aggregates, different levels of receptors on cells, or simply an artifact of the experimental method used. Nonetheless the experi ment illustrates the significant advantage that the fluorescent nanoparticles po ssess over single fluorophores.

PAGE 66

66 Figure 3-3. Fluorescence images of extracted sa mples after five minute incubation with (A) 40 M RuBpy Dye-aptamer conjugates, and (B) 0.5 nM RuBpy nanoparticle-aptamer conjugates, followed by three magnetic washes. (C) Comparison of dye labeled cells to nanoparticle labeled cells by flow cytometric analysis. Collection Efficiency While magnetic nanoparticles have been successful in our lab in extracting DNA and proteins, no one had yet attempted using magnetic nanoparticle for detecti ng something as large as a cell. Therefore we wanted to verify that the nanoparticles c ould effectively extract the target cells from the samples. Values for the coll ection efficiency were obtained by incubating increasing amounts of magnetic nanopa rticles with the target CEM ce lls and Ramos control cells. The number of cells collected was determined by flow cytometry by the counting of signal events. In addition, the cell c ounting was performed on a control sample of both cell types that did not undergo the magnetic extraction and was ta ken as the total amount of the cells. The

PAGE 67

67 collection efficiency was calcula ted by dividing the number of ev ents for each sample by the total cell number. Figure 3-4. Flow cytometric determination of magnetic nanoparticle co llection and separation efficiencies between target and control cells. As seen in figure 3-4, the collection efficien cy of target cells from ranges from 30-80%, however the collection efficiency seems to plateau at around 80%. In addition, the Ramos control cells had collection effi ciencies ranging from 0.5-5% for the same magnetic nanoparticle concentrations. This indicates that the target ce lls can be preferentially extracted from a sample, while few of the Ramos cells are extracted using the same method. Since the use of 10 L of magnetic nanoparticles had the high est separation efficiency, this amount was used for sample assay experiments. Enrichment Validation In order to verify that the samples were indeed being enriched, several samples were extracted and then resuspended in varying amounts of buffer. Samples were extracted using the previously mentioned protocol s with only the final volume ch anging. After resuspension the samples were analyzed on the plate reader to co mpare their fluorescence in tensities. Figure 3-5 shows the results of these experiments normalized to the highest intensity. The results indicate

PAGE 68

68 that the volumetric based enrichment succeeded in enriching the samples. When the cells were resuspended in the original volume of the sample the cells had the lowest overall fluorescence while suspended in 50 times less buffer resulted in 48.7 fold increase in fluorescence intensity. Other suspension volumes had a similar linear en richment compared to the original sample volume. Thus the results clearly show the cel l samples can be enriched through the magnetic separation and resuspension process. Enrichment Effect0 0.2 0.4 0.6 0.8 1 1.2Normalized Intensity 20uL 50uL 100uL 200uL 500uL 1000uL Figure 3-5. In order to validate the enrichment effect of the magnetic extraction, samples were resuspended in different volumes and thei r fluorescence intensities were measured. Limit of Detection An important consideration for the assay is the limit of detection (LOD). While larger volumes of any clinical samples could be used to increase the total amount of cells, it is still necessary to determine the absolu te amount of cells that can be reliably detected so that the volume needed for clinical samples can be known. The LOD was determined using pure cell samples and the extractions were performed as described previously. The limit of detection threshold was taken to be three standard devia tions above the blank, and because of this any

PAGE 69

69 residual fluorescence in the bl ank was accounted for. The LOD was performed using CEM target cells. Each of the samples was then an alyzed with NPs using the previously mentioned protocols with the fluorescence intensity being determined on the microplate reader following completion of the ACNP technique. Figure 3-6. Limit of detecti on for the stepwise addition of the MNP and FNP using the microplate reader for detection A) the full calibration curve and B) an enlarged depiction of the lower concentration regime. The detection limit was computed by plotti ng the fluorescence intens ity versus the cell number present in the sample. Consequently, the plotted data pr oduced a linear response as seen in Figure 3-6A and Figure 3-6B di splays the data focused in on the lower sample concentrations. From this plot the detection limit was determin ed to be approximately 250 cells with a dynamic range covering more than two orde rs of magnitude. This indicates that the ACNP system has the ability to sensitively det ect low amounts of intact targets cells from a given sample, and that a wide range of cell concentrations can be anal yzed by this method with little to no sample

PAGE 70

70 preparation depending on the amount of ACNPs use d. Estimates of exfoliated cancer cells in bodily fluids such as blood put the number of cancer cells at around 20 cells per mL indicating that clinical samples of 15-20 mL would be requ ired for analysis. That is roughly the amount of blood collected in a single vial for normal diagnostic testing. Improving LOD with Multiple Aptamers One of major advantages of the Cell-SELEX methodology is that once the selection is complete there is generally a panel of different aptamer sequences that are specific for the cells of interest. Since the cells are a living biological system unto themselves, it is very possible that some of the aptamer targets on the surface will have a higher or lower expression depending on different circumstances like their cell cycle, ove rall health, or genetic stochasticity. Thus, in some cells the target for a single aptamer may be downregulated resul ting in a worse limit of detection. In addition, since the fluorescent and magnetic nanoparticles have the same aptamer, they must compete for binding sites which can also reduce the sensitivity of the assay. In order to address this possibility, magnetic nanoparticles were prepared with different aptamer sequences. In all four different magnetic na noparticles were compared corresponding to one, two, three, and four different aptamer sequences conjugated to the nanopa rticle. Figure 3-7A shows the intensity of the same amount of target cells with each type of magnetic nanoparticle. The results indicate that increases the numbe r of different sequen ces on the nanoparticle, increases the fluorescent signal from the sample. This increase in signal likely results from two different phenomena, more cells being collected and more fluo rescent nanopart icles being bound to the cells since there is less competition fo r the binding sites. To determine whether the increase in intensity results in an increase in the limit of detection the one and four aptamer nanoparticles were directly compared. Using va rying amounts of target cells, calibration curves were prepared for the one and four aptamer na noparticles. Based on the calibration curves the

PAGE 71

71 four aptamer magnetic nanoparticles had a limit of detection of 85 cells while the one aptamer nanoparticles had a limit of detection of 250 cells. While the increases in the intensity and limit of detection were not overwhelmi ng, it is important to consider that the samples used cultured cells. The cultured cells would likely have less va riation than the cells fro m different patients. Therefore, the multiple aptamer nanoparticles w ould likely be far more effective using actual patient samples than the results here indicate. Figure 3-7. Comparison of magnetic nanoparticle w ith multiple aptamer sequences conjugated to the surface. A) The fluorescent intensities of 30,000 cells with the different magnetic nanoparticles. B) Calibration curves of the one and four aptamer magnetic nanoparticles. Separations from Complex Samples Mixed Cell Sample Assays In order to evaluate the potential of the assay, complex samples needed to be tested to determine extraction and detection capabilities in complex matrices. Rather than start out with very complex biological samples, first sample s were prepared by simply mixing different cultured cell lines together In Figure 3-8, we show the results from our artificial complex sample where equal amounts of CEM and Ramos cells we re mixed and our two particle assay was

PAGE 72

72 applied. To differentiate CEM from Ramos cells Fluo-4, a fluorescent calcium ion indicator, was used to label Ramos cells pr ior to nanoparticle incubation. Figure 3-8. Fluorescence Images of A) 1:1 ratio of target cells mixed with Fluo-4 stained control cells. B) Fluo-4 signal, and C) RuBpy signal after 5 minut e, two particle incubation and three magnetic washes of the mixture in 3A D) 1:1 ratio of Fluo-4 stained target cells mixed with control cells. E) Fluo-4 signal, and F) RuBpy signal after 5 minute, two particle incubation and three magne tic washes of the mixture in 3D. Fluo-4 labeled control cells were mixed in a one to one ratio with unlabeled CEM cells shown in Figure 3-8A. Magnetic and fluorescen t nanoparticles were simultaneously added and incubated at 4C for five minutes with occasio nal gentle stirring. After incubation, a magnetic field was applied to remove cells which were not attached to the aptamer labeled iron oxide particles. A 2 L aliquot of the extracted samp le was then illuminated to monitor Fluo-4 and RuBpy fluorescence as shown in Figure 3-8B and 3-8C, respectiv ely. Based on the images, the assay was able to collect the CEM cells in the sample and bright fluorescence from the RuBpy nanoparticles made them easily distinguishable. Th e experiment was also performed by labeling CEM cells with Fluo-4 and mixing them with unlab eled control cells as in Figure 3-8D. The

PAGE 73

73 cells shown in Figure 3-8E were separated by th e two particle assay, and all exhibit a Fluo-4 signal. In Figure 3-8F, the same cells are shown with the RuBpy emission overlaid. The presence of the Fluo-4 fluorescen ce proves that only the CEM cells were collected and imaged. The lack of Fluo-4 signal in Figure 3-8B, along w ith the presence of the Fluo-4 signal in Figure 3-8E prove that only target cells are being collected using this method for extractions from 1:1 cell mixtures. These samples were repeated 5 times with similar results achieved for each experiment. Bone Marrow Samples Since leukemia cell samples were predominantly used for the verification of this methodology, we also sought to test the performa nce of the assay in bone marrow samples from actual patients. In these experiments bone ma rrow aspirates were obtained from the Pathology department of Shands Hospital. The bone marro w samples were from healthy patients and the target cells were then spiked in the bone marro w sample and analyzed followed the established procedures. The results were then analyzed us ing fluorescence microscopy and the plate reader with the results from the experiment shown in Fi gure 3-9. The imaging results clearly show that the target cells were extracted from the complex sample. In the unspiked sample, few cells were extracted although some cellular debris was pulled out In the plate reader results that measured the fluorescence intensity of the entire sample, th ere is a clear difference between the spiked and unspiked samples. This indicates that the two particle assay is i ndeed capable of being used even in bone marrow aspirates.

PAGE 74

74 Figure 3-9. A) Extraction of target cells from a spiked bone marrow sample. B) Extraction performed on an unspiked bone marrow sample. C) Fluorescence measurements from the spiked and unspiked bone marrow samples. Whole Blood Sample Assays Blood samples were also used to determ ine detection capabilities from complex biological solutions. Control expe riments indicated that the aptame r sequence used was stable in serum samples for up to 2 hours. Target cells were spiked into whole blood samples (500 L) and compared to unspiked samples after magnetic extraction to make certain that target cells could be detected in complex biological samp les. As shown in Figure 3-10, nonspecific interactions caused the unwanted collection of some red blood cells, but the lack of RuBpy fluorescent signal on the unwanted cells allows for target cells to still be accurately distinguished. For magnetic ex tractions from whole blood sample s, 40% of the spiked target cells were routinely recovered after three magnetic washes and after accounting for dilution. This is consistent with current extraction efficiency values repo rted by immunomagnetic separation.201,202 These experiments were repeated for to tal of five times with similar results being obtained in each sample. This experiment wa s meant to mimic a real clinical sample which normally would contain thousands of different specie s. By successfully ex tracting our target cell line from whole blood, we have shown that this method is applicable for biomolecular and cellular detection in real clinical applications. Ou r assay selectively removes our target cells from

PAGE 75

75 this complex mixture with collection efficienci es rivaling or surpassi ng current methods for cellular detection from clinical samples. Figure 3-10. Confocal images of extractions from whole blood. (A) Extracted sample from target cell spiked whole bl ood. (B) Extraction from unspi ked whole blood. (C) and (D) show magnified images of extracted cells from 4(A). Collection and Detection of Multiple Cancer Cells Instrumental Validation The first step detecting multiple cell types was first optimizing the confocal microscope and plate reader settings to ensure that the diffe rent cell samples could be detected reliably and without any crosstalk between channe ls. In order to accomplish this, 1 M samples of different fluorophores were analyzed on both instruments in order to determine which fluorophores could be best detected. The samples were first analyz ed on the plate reader as it was decided that demonstrating this more complex methodology on a simple instrument would illustrate the wide

PAGE 76

76 applicability of the technique. Figure 3-11 show s the plate reader results for six different fluorophores. Each fluorophore was measured at its excitation and emission maximum in addition, each fluorophore was measured at the ex citation and emission maximum of the five other fluorophores. This would determine the amount of crosstalk that existed. 430nm/480nm 4 8 8n m / 5 20 n m 5 4 0n m / 57 0 n m 590 n m / 615nm 633nm/650nm 45 8nm/610nm AF43 0 AF488 AF55 5 T XRD A F 647 R u B py 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000Multiplexing Potential of Plate Reade r Figure 3-11. Plate reader measurements of multip le dyes to test spectral compatibility. The excitation and emission wavelengths used in each channel are listed on the x-axis while the fluorophore measured is listed on the y-axis. The results indicate that five of the fl uorophores are still candidates for further consideration with only the Alexa Fluor 430 show ing an unacceptable amount of crosstalk. Next the fluorophores were evaluated on the confocal microscope in a similar format with 2 L aliquots of each fluorophore solution being placed onto a microscope slide. The fluorescence intensity of each drop was measured with five different optical channels optimized for each individual fluorophore. In thes e tests, only the Texas Red fluor ophore was eliminated due to the difficulty in its detection. The microscope c ould not efficiently excite the fluorophore so it was removed from further consideration. Since only three fluorophores were needed for the three

PAGE 77

77 available aptamer sequences, RuBpy, Cy5, and TMR were selected for use. FAM based nanoparticles had proved rather difficult to synthe size so the experiments went forward with the best fluorescent nanoparticles for the most sensitive detection. Single Cell Type Extractions As it is unlikely that a single aptamer would be able to recognize every patients cancer cells it is likely that multiple aptamer sequences will be require to obtain a reliable diagnosis using the two particle assay. In addition, obtaining the profiling information may prove advantageous in the future for making a more specific diagnosis and st aging the progression of the disease. To expand the concept of the two particle-based magnetic collection and detection technique three different cancer cell lines were an alyzed using three different aptamer sequences. Each aptamer was selective for a different cell type. CEM, Ramos, and Toledo cell samples were extracted using ACNPs followed by fluorescent imaging and analys is by the microplate reader. Each pure cell sample extraction was repeated 10 times. As was mentioned in the methods section, the control cells used for the CEM experiments were the Ramos and Toledo cell types, for Ramos were the CEM and Toledo, and for th e Toledo were the CEM and Ramos. Figure 312 shows representative confocal images of 2 L aliquots of the CEM target cells (top) and Ramos nontarget cells (bottom) using CEM ACNPs (red) (A), Toledo target cells (top) and CEM nontarget cells (bottom) using Toledo ACNPs (g reen) (B), and Ramos target cells (top) and CEM nontarget cells (bottom) using Ramos AC NPs (blue) (C) after NP incubations and magnetic washes. Ramos, CEM, and CEM cells we re used as the respective controls for those experiments. The other control cell types were al so performed and resulted in similar responses as the one presented in the representative im ages. In addition, some fluorescence spots were observed in the images. However when the samp les were analyzed with the microplate reader

PAGE 78

78 and compared to sample blanks treated with the MNPs and FNPs, the levels of fluorescence signal were the same (images not shown). Ba sed on the fluorescence images a significant difference is evident in both the amount of cells extracted and fluorescen t signal present between the target and control cells in all samples. Howe ver, some control cells that were inadvertently collected and even labeled with some FNPs, but no significant signal was indicated by the microplate reader data for those samples producing signals in the same realm as sample blanks (images not shown). Figure 3-12. Fluorescence images of pure cell sa mples in buffer after magnetic extraction and washes A) Image of CEM target cells (top) and Ramos nontarget cells (bottom) using CEM ACNPs B) Image of Toledo target cell s (top) and CEM nontarget cells (bottom) using Toledo ACNPs C) Image of Ramos ta rget cells (top) a nd CEM nontarget cells (bottom) using Ramos ACNPs. Table 1 provides the fluorescence data obtai ned from the microplate reader. The first column represents the cell sample that was an alyzed, the second column represents the signal

PAGE 79

79 produced by the CEM NPs, the third column re presents the signal produced by the Toledo NPs, and the fourth column represents the signal produ ced by the Ramos NPs. The rows in the table display the cell samples that were investigated using the ACNPs. Table 3-1: Single ce ll type extraction Sample cells CEM NP signal Toledo NP signal Ramos NP signal Ramos 945 965 7,574 CEM 48,967 1,056 314 Toledo 1,075 36,728 438 On the contrary, the target cells subjected to this procedure had ve ry intense fluorescent signals that made them easily discernible from the control cells. A closer look at the characterization of expanding the ACNP techniqu e to multiple cell types the microplate reader data (table 1) demonstrated that when using 100,0 00 cells in each of the pure cell samples at a collection efficiency of 85% (det ermined in a previous publication7), all target cell samples produced signals in upwards of 24 fold enhan cements above the background and as high as 50 fold. The target samples indicated in Table 1 for this experiment were th e CEM cells (row 2) for the CEM NPs (column 1), the Toledo cells (row 3) for the Toledo NPs (column 2), and the Ramos cells (column 3) for the Ramos NPs (c olumn 3). The control samples for these experiments were represented in the remainder of the table for each of the NPs and cell types. The signals for the control samples at the c onditions mentioned above and the collection efficiency for the MNP amounts used in these expe riments for the control cells determined to be no greater than 5% (determine d in a previous publication7), resulted in fluorescence signals at the same level as a buffer blank sample treated with the ACNPs. This data indicates that the MNPs were both selective for the target cells by discriminating against the control cells and reproducible in all samp le types investigated.

PAGE 80

80 Multiple Cell Type Extraction Method With the aptamers having demonstrated suffici ent selectivity to al low the extract each cell type effectively without extract the control cells, the extraction of multiple cell types could then be attempted. Determining the multiple extr action and detection capability of the ACNP was performed by creating artifici al complex samples of CEM, Ra mos, and Toledo cells. Figure 3-13 displays the schematic diagram of the multiple cell extraction procedure that was employed. A stepwise extraction protocol wa s used as it proved to be the most effective for the blood samples and initial experiments using a simulta neous extraction of multiple cell types proved unsuccessful. The samples with one, two, and thr ee cell types were analyzed using previously established ACNP protocols. The samples were prepared by obtai ning approximately 105 cells of each type for the respective sample type. Th e stepwise extraction protocol was performed by adding the specified amounts of MNPs for Ra mos cells, followed by CEM aptamer-conjugated MNPs, and finally with Toledo specific MNPs. E ach set of MNPs were incubated with the cell samples separately for 15 minutes. After th e Ramos MNPs were incubated with the cell samples, magnetic extraction was performed, and th e supernatant kept to be treated with the CEM specific MNPs. The remainder of the magnetic extractions was carried out as described in the magnetic extraction section. Th e sample was redispersed in 200 L cell media, followed by addition of the Ramos aptamer-conjugated FNPs with 5 minute incubation, and magnetic extraction procedure performed. Similarly, th e respective CEM and Toledo aptamer-conjugated FNPs were subsequently introduced to their samples. After the final wash, the cell sample was dispersed in 20 L media buffer. The samples were an alyzed using fluorescence imaging with the 2 L aliquots and with the plate reader spectrometer with 20 L aliquots.

PAGE 81

81 Figure 3-13. Schematic representation of the multiple extraction procedure with the MNPs being added and extracted stepwise and the co rresponding FNPs being added post magnetic extraction of cell samples. Mixed Cell Samples The power of the multiple extraction procedur e needed to be evaluated using complex sample mixtures. Figure 3-14 A, B, and C demo nstrates the results from artificial complex samples by mixing equal amounts of the appropriat e cell types for the th ree different multiple extraction samples diluted in cell media buffer, where CEM, Ramos, and Toledo cells were mixed and the ACNP process applied as describe d above. A total of 100,000 cells in all samples were used, cell and buffer volumes were adjusted accordingly. To exhibit that the MNPs indeed have the ability to selectively differentiate the cells from one another in a multiple cell mixture format, single, double, and triple cell mixed samples were evalua ted. Figure 3-14A illustrates the selective nature of the t echnique by performing the ACNP st eps with a single cell sample, Ramos cells. The single cell sample was firs t treated with CEM ACNPs followed by Toledo ACNPs, and finally Ramos ACNPs. The samples were incubated at 4 oC with the MNPs and FNPs as expressed in the previous section.

PAGE 82

82 Figure 3-14. Fluorescence images of buffer extracted mixed cell samples using the multiple extraction procedure A) cont ains only Ramos cells B) contains CEM and Toledo cells C) contains all three CEM, Toledo, and Ramos cells. Fluorescence images D, E, and F displays the mixed cell samples extracted from FBS using the multiple extraction procedure, where D) contai ns only Ramos cells E) c ontains CEM and Ramos cells and F) contains all three CEM, Toledo, and Ramos cells. Based on the fluorescence images, this method wa s able to selectively collect the Ramos cells (blue) only when the Ramos ACNP were in troduced to the cell sample Figure 3-14A. The Toledo and Ramos cells were used in single cell sa mple extractions as well (image not shown). This method was further tested by performing the ACNP steps with a mixture of two different cell types, CEM and Toledo. Figure 3-14B displays the selective natu re of this technique for the cells indicated. The fluorescence images again de monstrate selective isol ation of the CEM (red) and Toledo (green) cells. Other CEM, Toledo, and Ramos two cell type mixed samples were performed as well (image not shown). The fina l test was to perform this technique with a mixture of all three cell types in the same samp le using the CEM, Toledo, and Ramos cells at the same time. Figure 3-14C reveal s the selective nature of the method for each of the cells

PAGE 83

83 indicated. Fluorescence images again depict th e selective isolation of the CEM, Toledo, and Ramos cells, Figure 3-14C left (re d), middle (green), and right (b lue) images respectively. Table 3-2: Multiple cell type extraction Sample cells CEM NP signal Toledo NP signal Ramos NP signal Ramos 1,281 1,040 7,862 CEM 44,972 920 375 Toledo 1,025 34,972 320 CEM, Ramos 46,874 1,505 7,385 CEM, Toledo 43,890 37,896 414 CEM, Toledo, Ramos 42,145 32,945 7,524 Microplate reader data was in complete ag reement with the confocal image data and presented in Table 2. The first column represents the cell sample s that were analyzed, the second column represents the signal produced by the CEM NPs, the third column represents the signal produced by the Toledo NPs, and the fourth colu mn represents the signa l produced by the Ramos NPs. The rows in the table display the cell samp les that were investigated using the ACNPs. With 100,000 total cells present in all samples, samples containing target cells produced signals in upwards of 24 fold enhancements above the background and as high as 47 fold. The signals for the control samples at the conditions menti oned above resulted in fluorescence signals at the same level as a buffer blank sample treated w ith the ACNPs with the exception of the Toledo nontarget sample in sample 4. Standard devia tions determined for all these samples were determined to be 8-12%. This data indicates th at the MNPs were both selective for the target cells by discriminating against the control cells and repr oducible in all sample types investigated. Serum Samples To show applicability of the stepwise proce ss in real biological samples, fetal bovine serum (FBS) was used. FBS was spiked with each of the respective cell types for the corresponding one cell, tw o cell, and three cell ex traction experiments (500 L). The process

PAGE 84

84 was performed as described above in the buffer solution experiments. Confocal imaging and fluorescence microplate reader were used to characterize cell extr actions. Figure 3-14 D, E, and F illustrate the results of the FBS spiked complex samples by mixing equal amounts of the indicated cells at a total cell concentration of approximately 100,000 cells. Figure 3-14D, 3-14E, and 3-14F present the selective na ture of the technique for the single, double, and triple mixed cell type samples. The samples were treate d with CEM ACNPs followed by Toledo ACNPs, and finally Ramos ACNPs. For the single cell sample experiment, the sample was first treated with CEM ACNPs followed by Toledo ACNPs, and finally Ramos ACNPs. The samples were incubated at 4 oC with the MNPs and FNPs as expressed previously. Figure 3-14D fluorescence image shows the sample contains Ramos cells extr acted and labeled only af ter being treated with Ramos ACNP (blue). Extractions with the CE M and Toledo cell types were completed as well (images not shown). Figure 3-14E left image (re d) and right image (blu e) show the CEM cells and Ramos cells extracted when treated with CEM and Ramos ACNP for the two cell type extraction experiment. Other tw o cell type extractions with the CEM, Toledo, and Ramos cell types were performed as well (images not shown) Figure 3-14F left (re d), middle (green), and right (blue) images show the extraction of all three cells treated with all the ACNP. The fluorescence imaging data was confirme d by collecting fluores cence data using the microplate reader, Table 3. The table layout was the same as in the previous table: first column was cell samples, second column was the CEM NP signals, third column was Toledo NP signals, and fourth column was the Ramos NP signal. Th e rows in the table display the cells mixed to make the samples that were analyzed. The standa rd deviations determined to be 8-12% for all samples measured in the FBS. With 100,000 total cells present in ea ch sample dispersed in FBS, the signal enhancements determined above the background ranged from 10 to about 24. In all

PAGE 85

85 cases, the signals for all target samples were lo wer than those for the cell media buffer, and the background signals were all higher. The Toledo samples produced the lowest enhancement of all the extracted samples, which produced th e highest background signa l of the three ACNPs pairs. The Toledo aptamer is less selective then the other aptamers that were used, which would explain the higher background produ ced in this particular sample The fluorescence images and microplate reader data demonstrated that the MN Ps were both selective for the target cells by discriminating against the cont rol cells and reproducible even in spiked FBS samples. Table 3-3: Multiple cell type extraction in serum Sample cells CEM NP signal Toledo NP signal Ramos NP signal Ramos 1,845 3,241 6,776 CEM, Ramos 43,835 3,554 6,980 CEM, Toledo 40,767 31,240 452 CEM, Toledo, Ramos 42,973 33,112 7,078 Small Cell Lung Cancer Cell Extractions In order to demonstrate the effectiveness of this methodology on an exfoliated cancer cell type, the assay was used to detect small cell l ung cancer cells (SCLC). SC LC is a disease that can spread rapidly throughout th e body even early in its development making it a very lethal disease and difficult to detect at an early stag e. SCLC can even metastasize before being detectable using conventional means. Therefore, developing an effectiv e screening method for it could lead to earlier detecti on and diagnosis thereby improving the prognosis for the disease which currently has 5 year survival rate of only 15%. Therefore, an aptamer selected for SCLC using Cell-SELEX was conjugated to magnetic an d fluorescent nanoparticles. The cells were then extracted using the previous ly mentioned protocols with th e results shown in Figure 3-15. The results indicate that the two particle assa y works well with the SCLC cells as the image shows many target cells being extracted with fa r fewer of the non-target lung cancer cells being

PAGE 86

86 extracted. The plate reader measurements confirm these results as the target samples had a much higher fluorescence intensity compared to the nontarget sample. While these preliminary results are encouraging further optimization needs to be completed and more complex samples still need to be analyzed before any conclusions as to th e effectiveness of the me thodology can be made. Figure 3-15. A) Fluorescence image of target SCLC cells extracted. B) Fluorescence image of non-target lung cancer cells extracted. C) Plate reader results for each sample. Conclusion In this chapter, the two partic le assay for the collection and detection of cancer cells has been demonstrated. The assay has been shown to work well even in complex samples owing to the selectivity of the aptamers a nd unique properties of the nanomate rials used. Future directions for this project are mainly concerned with study ing clinical samples and applying the assay to other cancer types such as SCLC. In particular the application of the method to SCLC remains a priority as it can solve a great problem in th e medical field, an effective screening method for SCLC. In particular, SCLC is a good candidate for effective screening method as 99% of cases present in people who smoke or who have smoke d. Thus, the most susceptible population of people at risk for the disease has already b een identified. In add ition, over 100,000 people die every year from lung cancer thus confirming the need for more e ffective means of detection and

PAGE 87

87 diagnosis. These factors combine to form an ef fective rationale for a pplying the two particle assay for a screening method for SCLC.

PAGE 88

88 CHAPTER 4 A COLORIMETRIC ASSAY FOR THE DIRE CT DETECTION OF CANCER CELLS Introduction The key to the effective and ultimately successf ul treatment of diseases such as cancer is an early and accurate diagnosis. An early diagnosis is only possible with a sensitive method for the detection of the disease. Current methods are time consuming, expensive, and require advanced instrumentation. A more cost effectiv e method requiring simple or no instrumentation yet still providing great sensitivity and accuracy would be ideal for point of care diagnosis. To accomplish this, we have developed the first colori metric assay for the dir ect detection of cancer cells using aptamer conjugated gold nanoparticle s (ACGNPs). A colorimetric assay would enable diagnosis based on a simple color change enabling diagnostic assays for diseases where sophisticated instruments are unavailable. Due to the Cell-SELEX aptamers, colorimetric assays could be implemented for any disease that result s in the expression of different proteins on the cell surface including cancerous cells or cells exposed to viral infections. To make the assay colorimetric in nature gold nanoparticles were utilized for their biofunctionalization, biostability, and spectral properties. Due to the plasmon resonance of gold nanoparticles, they possess strong distance de pendant optical properties. Once the gold nanoparticles come into close proximity with one another their absorption spectra shift and their scattering profile changes resulting in a change in color and in the resulting absorption spectra of the sample.203,204 As a result of these prope rties, many techniques have been developed based on the aggregation of gold nanoparticle s to detect genes and proteins.205,206 However, instead of simple aggregation of the gold na noparticles using genes or prot eins, the ACGNPs are targeted to assemble on the surface of a specific type of cancer cell creating a virtual gold shell on the surface of the cell. The assembly of the gold nan oparticles around the cell surface causes a shift

PAGE 89

89 in the absorption spectra of the particles along with a significant incr ease in the extinction coefficient. This results in not only a change in the color of the solution but a large increase in the intensity of the measured absorbance of the sample. This allows for the visualization of the target cells with the naked eye or with a micr oplate reader for increased sensitivity and the potential for automated high throughput analysis The assembly of gold NPs on cell membrane surface due to cell receptor recognition by apta mers presents a novel approach for direct colorimetric detection of cancer cells. Experimental Methods DNA Aptamer Synthesis The following aptamers have been sel ected for the CCRF-CEM and Ramos cells respectively: 5'-TTT AAA ATA CCA GCT TAT TCA ATT AGT CAC ACT TAG AGT TCT AGC TGC TGC GCC GCC GGG AAA ATA CTG TAC GGA TAG ATA GTA AGT GCA ATC T-3'; 5'-AAC ACC GGG AGG ATA GT T CGG TGG CTG TTC AGG GTC TCC TCC CGG TG-3';. Both the thiol versions of the ap tamer sequencers were synthesized in-house. An ABI3400 DNA/RNA synthesizer (Applied Biosyste ms, Foster City, CA) was used for the synthesis of all DNA sequences. A ProStar HP LC (Varian, Walnut Creek, CA) with a C18 column (Econosil, 5u, 250 4.6 mm) from Alltech (Deerfield, IL) was used to purify all synthesized DNA. A Cary Bio-300 UV spectrometer (Varian, Wal nut Creek, CA) was used to measure absorbances to quantify the manufact ured sequences. All oligonucleotides were synthesized by solid-s tate phosphoramidite chemistry at a 1 mol scale. The completed sequences were then depr otected in concentrated ammonia hydroxide at 65 C overnight and further purified twice with reverse phase highpressure liquid chromatography (HPLC) on a C-18 column.

PAGE 90

90 Aptamer Conjugated Gold Nanoparticle Synthesis In a 2mL microcentrifuge tube, 1mL of the 20nm gold co lloid nanoparticles (GNPs), containing 7.0x1011 particles/mL, taken directly from the manufacturer (Ted Pella, Inc. Redding, CA) was centrifuged for 15 minutes at 14,000 RPM. The GNPs were washed three times with 1mL aliquots of 5mM phosphate buffer (PB) pH 7.5 by decanting the supe rnatant, adding fresh PB, dispersing by sonication, and centrifuging fo r 15 minutes at 14,000 RPM. From decantation to dispersion, the wash step was performed with in 3-5 minutes. After the final wash step, the GNPs were dispersed in 1 mL of the PB. To each washed GNP sample, 150 L of a 1 M thiol labeled DNA sequence was incubated for 3-5 days at 4C. The samples were sonicated to disperse the GNPs every 12 hours. When the in cubations were completed, the samples were centrifuged at 14,000 RPM for 5 minutes and the samp les washed as described previously with the PB. After the final wash, each GNP sample was dispersed in 0.25mL PB with approximately 6.0x1011 particles, and the samples stored at 4C until used. Cells CCRF-CEM cells (CCL-119 T-cell, human acute lymphoblastic leukemia) and Ramos cells (CRL-1596, B-cell, human Burkitts lym phoma) were obtained from ATCC (American Type Culture Association). The cells were cu ltured in RPMI medium supplemented with 10% fetal bovine serum (FBS) and 100 IU/mL penicillin-Streptomycin. The cell density was determined using a hemocytometer prior to any experiments. After which, approximately one million cells dispersed in RPMI cell media buffe r were centrifuged at 920 rpm for five minutes and redispersed in dye free cell media three times and were then redispersed in 5 mL dye free cell media. During all experiments, the cells were kept in an ice bath at 4oC.

PAGE 91

91 Assay Protocol For each assay, 1.0x1010 ACGNPs were utilized in a total volume of 300 L. The assay were incubated for 30 minutes at 4oC in a 500 L microcentrifuge tube and then transferred to a BD Falcon 96-well transparent microplate (Fisher Scientific, Pittsburgh, PA ). After incubation, the assays in the microplates were imaged w ith an Epson Stylus CX3200 flatbed scanner while the absorbance spectra from 400nm to 900nm were measured in a Tecan Safire Microplate reader (Mannedorf, Switzerland) for each sample All data analysis was performed using Microsoft Excel. Results and Discussion To demonstrate the principle behind the a ssay, it was first determined whether the ACGNPs could differentiate between target cells and control cells. Fo r these experiments, CCRF-CEM acute leukemia cells were used as targ et cells while Ramos, a Burkitts lymphoma cell line, were used as control cel ls. An aptamer sequence with high selectivity and affinity to the CCRF-CEM cells was conjugated to 20nm gold nanoparticles th rough a thiol functional group on the aptamer sequence. The five samples are: 1.0x1010 ACGNPs, 1.0x1010 ACGNPs with 10,000 target cells, the ACGNPs w ith 10,000 control cells, 10,000 targ et cells with no ACGNPs, and 1.0x1012 ACGNPs in cell media. The absorption spectrum for each sample is shown in Figure 4-1. The results establish that the abso rbance of the ACGNPs with 10,000 target cells is significantly higher than the same amount of ACGNPs with 10,000 control cells, the same amount of target cells without ACGNPs, or with only the ACGNPs. These results indicate that the ACGNPs are binding selectively to the target cells and that the assembly of the ACGNPs around the target cells causes an increase in th e extinction coefficient of the solution. The increase of the extinction coefficient, predominan tly in the scatter coefficient, was also observed

PAGE 92

92 colorimetrically as the target cell sample was deep purple in coloration compared to the other samples that were mainly clear and colorless. However, the signals from the non-binding sa mples were too low to determine whether there was a shift in the spectra accompanying the binding of the ACGNPs to the target cells. To determine whether a shift occurred, a spectru m from a larger amount of ACGNPs was also measured and plotted in Figure 4-1. This spectr um indicates that a significant shift in the absorption occurs when the ACGNPs bind to the ta rget cells. These results are consistent with the spectral shifts observed by others usi ng the aggregation of gold nanoparticles for detection.203-206 Figure 4-1. Plots depicting the absorption spectr a obtained for various samples analyzed using ACGNPs. The spectra illustrate the differences in spectral characteristics observed after the ACGNPs bind to the target cells.

PAGE 93

93 Nanoparticle Size Effect The next step in the development of the assa y was to determine the best gold nanoparticle size to use. Four batches of aptamer conjuga ted nanoparticles were synthesized with four different gold nanoparticle sizes. The sizes us ed were 5nm, 20nm, 50nm, and 100nm. Based on published results,207 the larger nanoparticles should have the longest re d shift which would make detection easiest but would also have a highe r background. Therefore, the nanoparticles would be evaluated on the basis of its red shift and the difference between target and control cell samples. Figure 4-2 depicts the plots of each na noparticle type spectra with its target cells. Based on these results, the 20nm and 50nm nanopart icles had the most significant red shift upon target hybridization each having absorbance spectra reaching past 700nm. The 5nm nanoparticles exhibited very little change in spectra with the target cells pe rhaps indicating the 0 0.2 0.4 0.6 0.8 1 1.2 400500600700800900WavelengthNormalized Absorbance 5nm 20nm 50nm 100nm Figure 4-2. Spectra of the different sizes of Apta mer Conjugated Gold Nanoparticles with target cells.

PAGE 94

94 nanoparticles were too far apart to effectivel y interact or that an unknown problem occurred during the conjugation step. While the 100nm e xhibited an effective enhancement upon target cell addition, they did no t exhibit as significant of a red shift as the 20nm or 50nm nanoparticles. Based on the spectra the 20nm and 50nm nanoparticle s performed the best in terms of red shift while the 20nm particles had the greatest enhancem ent of the target cells relative to the control cells. Based on the results obtained from these limited tests, the 20nm gold nanoparticles were used in subsequent experiments due to their performance. However, in the future the nanoparticle sizes will be evaluated more rigorou sly to obtain the optimized performance for each size. Assay Response With the best size of nanopartic les identified, the next step was to verify that the assay is indeed colorimetric and that the change in colo r was proportional to the am ount of cells present. In order to accomplish this, 1.0x1010 ACGNPs were incubated with increasing amounts of target cells. This was repeated with the same amounts of control cells for comparison. The image of both cell types is shown in Figure 4-3A. The resu lts clearly show that the samples with more target cells have a darker color while with the control cells, the samples remain almost colorless and there is no significant difference between th e samples regardless of the amount of cells present. Thus the assay allows for the detection of target cells with the naked eye. In order for a more sensitive detection than allowed by the human eye, the samples were also analyzed using a microplate reader. The sp ectra for the different amounts of control and target cells are shown in Figures 4-3B and 4-3C, respectively. The absorption spectra correlate well with the colorimetric result s in that the samples with an increasing amount of target cells absorb light more intensely. There is little ch ange in the ACGNP absorption spectra of the

PAGE 95

95 control cell samples regardless of the amount of cells present. Th is is most likely a consequence of the selectivity of the aptamer itself and also the nature of the gold nanoparticles. Realistically, absolute selectivity is difficult to achieve regardless of the molecular recognition element employed and it is highly likely that a few of the ACGNPs bind non-selec tively to the control cells. However, in order to generate a si gnal from ACGNPs there needs to be many gold nanoparticles in close proximity, therefore even if there is a limited amount of non-selective binding it is very unlikely it will be to the extent to cause a positive response to the assay further increasing the apparent selectivity of the assay to the target cells. Figure 4-3. A) Images of ACGNPS with increasi ng amounts of target (top) and control cells (bottom). The amount of cells used in each sample is given in the legend on the bottom right. B) Absorption spectra of th e control cell samples with ACGNPs in Figure 2A. C) Absorption spectra of the target samples with ACGNPs in Figure 2A.

PAGE 96

96 In order to determine the limit of detection for the assay, the prev ious experiment was repeated four more times. The absorbance at 650 nm was then recorded for each amount of cells and plotted in Figure 4-4A. The assay showed an excellent dynamic range with standard deviations ranging from 6-10%. Based on three times the standard deviation of the blank measurement, the limit of detection of the target cells was calculated to be 90 cells. In addition, this experiment was repeated with control cells to measure their response to the assay. The ACGNPs had no response to the cont rol cells at the lower cell con centrations as these samples Figure 4-4. A) Calibration curve i llustrating the relationship between the amount of cells and the absorbance intensity at 650nm for both targ et cells (black) and control cells (gray). The assay shows a very good dynamic range in addition to excellent sensitivity. B) Bar graph showing the change in intensity be tween the target cells and control cells at 650nm in both cell media (CM) and fetal bovine serum (FBS) for both cell types.

PAGE 97

97 had signals comparable to the blank. At the hi gher cell concentrations, the ACGNPs had a small response to the control cells a lthough it was still signi ficantly lower than even the smallest concentration of target cells th at were evaluated. Based on these results, the assay for direct cell detection has demonstrated excelle nt sensitivity and selectivity. In order to truly evaluate the assay, more comp licated samples also needed to be analyzed to determine whether the assay could be useful fo r actual samples. In order to accomplish this, the assay was also used on several samples of fetal bovine serum (FBS). In these samples, the 50,000 cells were spiked into FBS and then the 1.0x1010 ACGNPs were added. Target and control cells were spiked into different samples and then after incubation with the ACGNPS their spectra were measured using the microplate read er. The absorbance at 650 nm from the three samples was averaged and plotted in Figure 44B, for comparison, the signals of the same amount of cells in cell media were also measur ed. The target cells in FBS clearly show a significantly higher signal than the control cells in the FBS, indicat ing that the assay functions as expected in even complex environments. Colori metric determination, however, proved difficult due to the color of the FBS. In future experiments with more complex samples, further sample preparation to remove any colored species may be necessary for any colorimetric detection. Regardless, spectroscopic detection of the target cells in FBS was achieved without any further sample preparation steps. In an effort to show the assay is applicable to other cell types, this experiment was repeated along with the same expe riment in cell media using an aptamer that is selective for the Ramos cell line while using th e CCRF-CEM cells as a negative control. Again, the ACGNPs showed excellent sensitivity and selectiv ity to the target cells regardless of the line of cells targeted in both the cell media and in the FBS.

PAGE 98

98 In conclusion, ACGNPs have been demonstrated for the sensitive and selective detection of cancer cells through utilizing the unique spec tral properties of gol d nanoparticles and the excellent selectivity of aptamers. The assay is the first colorimetric assay for the direct detection of cancer cells using molecular aptamers dir ectly selected from whole cells. In addition, spectroscopic detection using ACGNPs proved su ccessful for even complex samples like FBS and demonstrated excellent sens itivity and selectivity. Little n on-selective binding was observed, most likely due to the characte ristics of the gold nanoparticles in that any binding of a few nanoparticles to the cont rol cells would not have been enoug h to drastically alter the spectral properties of the particles. We have proven th at gold NPs can be assembled on cell membrane surface for spectral change, providing a direct vi sualization of cancer cells. Future work will include the further development of the assay in cluding optimizing the particle size, incubation times, and sample preparation methods for comp lex samples. After optimization, the assay will be tested on more complex samples including blood samples using both colorimetric and spectroscopic detection. This assay has the po tential to provide a rapid, high throughput, sensitive, and cost effective approach for the early and accurate detection of cancer through the utilization of aptamers and nanotechnology. Once the assay has been sufficiently optimized, it may prove more useful for other diseases than cancer. Given its rapid, co lorimetric based detection, the a ssay may prove more suitable for the rapid screening of multiple samples in a point of care or triage situation of a viral or bacterial infection outbreak. Infected cells often have different or variable cellular surface marker expression, thus if an aptamer could be select ed using Cell-SELEX to exploit the differences between the normal and infected cells, the aptamer could easily be adapted to the assay allowing the colorimetric detectio n of the infection. This would enable the rapid screening of patients in

PAGE 99

99 hospitals if a particular infec tion strikes a hospital such as Legionnaires disease or after a bioterrorism incident.

PAGE 100

100 CHAPTER 5 MOLECULAR BEACON DESIGN, DELIVERY, AND EVALUATION Introduction Since their development in 1996,208 molecular beacons (MBs) ha ve seen ever growing use in a wide variety of fields from chem istry to biology to medical sciences. 209-211 One emerging application of MBs has been their use for intracellular studies of gene expression and localization. Cellular analysis using MBs offe r many advantages to traditional forms of gene expression analysis. One distinct advantage is that MBs allow the probing of gene expression without destroying the cell unlike Northern Bl ot analysis and RT-PCR based methods. To understand why MBs are ideal for in tracellular analysis one has to first understand a MB. MBs are oligonucleotide based probes that use fluorescence for detection. MBs are composed of two discrete oligonuc leotide regions; the loop region and the stem region. Due to their unique hairpin structure, MBs undergo a conformational change upo n hybridizing to their targets. The conformational change causes the st em portion of the MB to melt and separate. The melting of the stem causes the fluorophore and qu encher on each end of the stem to become spatially separated. When unhybridized the fluoro phore and quencher are in close proximity, the fluorescent signal is suppressed. After the fluorophore and quench er pair is separated after hybridization, the fluorescence signal is restored. This method of signal transduction gives the MB allows the bound and unbound MBs to be easily distinguished allowing for the detection of bound MBs without separating out the unbound MBs. This feature of MBs is very significant for detection of mRNA inside of living cells since previous met hods of mRNA detection inside of cells required fixed or pretreated ce lls so that the unbound probes could be removed.212 Such techniques do not allow analysis on living ce lls thus the MB seems to possess the ideal characteristics for analysis on living cells. In addition, MBs inhere ntly possess excellent

PAGE 101

101 specificity through Watson-Crick base pairing, thus specific gene sequences can be probed inside of cell. The sensitivity of fluorescence detection also makes them ideal for intracellular mRNA detection. The ability to monito r mRNA expression inside of cells selectively and in real time gives MBs the potential to improve the understandi ng of many biological processes. It can shed light on many fundamental processes like m echanisms and kinetics of mRNA production, transportation and localization of mRNA inside of a living cell, and single cell responses to external stimuli. In addition, MBs have other im portant characteristics for intr acellular experiments. Since MBs are comprised of DNA, they are readily adapted for intrace llular experiments since DNA is non-toxic to cells. Another very important characteristic of MBs is that they can be designed for virtually any mRNA sequence, in principle it is possible to design a MB for any known gene sequence. Combined with their inherent signa l transduction method that allows for detection without separation, and the ability to design a MB for any gene, MBs have a wide applicability for various intracellular studies of biochemical, biological, and medical significance. However when designing and using molecular beacons fo r intracellular experiments there are several important criteria that cannot be ove rlooked that will be explored in more detail in this chapter. In addition, some critical experimental parameters such as probe delivery and in vitro testing of probes will also be covered. Methods and Materials Equipment Fluorescence imaging was conducted with a conf ocal microscope setup consisting of an Olympus IX-81 inverted microscope with an Olympus Fluoview 500 confocal scanning system and three lasers, a tunable Argon Ion laser (458nm, 488nm, 514nm), a green HeNe laser (543nm), and a red HeNe laser (633nm) with three separate photomu ltiplier tubes (PMT) for

PAGE 102

102 detection. The cellular images were taken with a 40x 1.4 NA oil immersion objective. A Leiden microincubator with a TC-202A temperature contro ller (Harvard Apparatus, Holliston, MA) was used to keep the cells at 37oC during injection and monitori ng. An EXFO Burleigh PCS-6000150 micromanipulator was used for positioning the injector tip. An Eppendorf Femtojet microinjector with 0.5 m Femtotips was used to inject the molecular beacons and reference probe into the cells. All analysis was conducted on the Fluoview 500 software, followed by processing of the data using Microsoft Excel. Fluorescence Imaging All cellular fluorescent images were collected using the confocal microscope setup. The confocal consists of an Olym pus IX-81 automated fluorescence microscope with a Fluoview 500 confocal scanning unit using either a 20X ai r objective or a 40X oil objective based on experiment. There are three lasers providing laser excitation at 458nm 488nm, 514nm, 543nm, and 633nm. The Alexa Fluor 488 were excited at 488nm and collected at 520nm. The Alexa Fluor 555 and TMR based MBs were excited at 543n m and collected at 560nm. The Alexa Fluor 647 and Cy5 based MBs were excited at 633nm a nd collected at 660nm. RuBpy was excited at 488nm and the emission was collected at 610nm. Imag es were taken after a five to ten second period during which the instrument was focuse d to yield the highest intensity from the fluorescence channels. The images were assigned color representations for clarity and are not indicative of the actual emission wavelengths. Plate Reader Experiments Fluorescence measurements were taken using a Tecan Safire Microplate Reader in a 384 well small volume plate. 20 L aliquots from each sample were deposited in the well and sample fluorescence intensity at defined wavelengths were measured at a constant gain at 5nm slit widths. Cy3 and Alexa Fluor 555 were excited at 550nm and their emission was detected at

PAGE 103

103 575nm. Alexa Fluor 488 and FAM were excited at 488nm and their emission was measured at 520nm. Alexa Fluor 647 and Cy5 were excited at 640nm while their emission was detected at 670nm. The data for the experiments utilizing the plate reader was analyzed using Microsoft Excel. Cell Culture MDA-MB-231 breast carcinoma cells (American Type Culture Collection, Manassas, VA) were maintained in Dulbeccos Modification of Eagles Medium (DMEM, Fisher Scientific) with 10% fetal bovine serum (Invitrogen, Carles bad, CA) and 0.5 mg/ml Gentamycin (Sigma, St. Louis, MO) at 37oC in 5% CO2/air. Cells were plated in 35 mm glass bottom culture dishes and grown to 80% confluency (MatTek Corp., Ashland, MA) for 48 hours prior to use. Fluorescent Probes All MBs were designed in-house based on pub lished mRNA sequences. The Alexa Fluor based MBs were synthesized by Ge nomechanix (Gainesville, FL) while the rest of the MBs were synthesized in-house. An AB I3400 DNA/RNA synthesizer (Applie d Biosystems, Foster City, CA) was used for the synthesis of all in-house MB sequences. A ProStar HPLC (Varian, Walnut Creek, CA) with a C18 column (Econosil, 5u, 250 4.6 mm) from Alltech (Deerfield, IL) was used to purify all fabricated DNA. A Cary Bio-300 UV spectrometer (Varian, Walnut Creek, CA) was used to measure absorbances to quantify the manufactured sequences. All oligonucleotides were synthe sized by solid-state phosphor amidite chemistry at a 1 mol scale. The completed sequences were then deprotec ted in concentrated ammonia hydroxide at 65 C overnight and further purified twice with reverse phase highpressure liquid chromatography (HPLC) on a C-18 column. The fluorophores Alexa Fluor 555 (AF555), Alexa Fluor 488 (AF488) and Alexa Fluor 647 (AF647) were purchas ed from Invitrogen (Carlsbad, California). All other reagents were purchased from Glen Re search (Sterling, VA). The quenchers used for

PAGE 104

104 the MBs were Dabcyl, Blackhole Quencher 2 (B HQ2), and Blackhole Quencher 3 (BHQ3). The sample solutions used in the experiments contained 1 M of each MB and in 20 mM Tris, 50mM NaCl and 5mM Mg2Cl2 buffer. Sequences and fluorophore que ncher combinations are given in Table 5-1. Table 5-1: Molecular beacon sequences MB name Sequence with fluorophore/quencher pair Control MB1 5 -Cy5-CCT AGC TCT AAA TCG CTA TGG TCG CGC TAG G-BHQ3-3 Control MB2 5 -AF555-CCT AGC TCT AAA TCG CTA TGG TCG CGC TAG G-BHQ2-3 -actin MB1 5'-TMR-CCG TCG AGG AAG GAA GGC TGG AAG AGC GAC GG-BHQ2-3 -actin MB2 5'-AF488-CCG TCG AGG AAG GAA GGC TGG AAG AGC GAC GG-BHQ1-3 Cyclin D1 MB1 5 -Cy3-ACG ACG GCC ACC ACG CTC CCC GCT GCC ACC GTC GT-BHQ2-3 Cyclin D1 MB2 5 -Cy3-GCA GCA TCC AGG TGG CGA CGA TCT TGC TGC-BHQ2-3 MnSOD MB 5 -AF647-CCG AGC CAG TTA CAT TCT CCC AGT TGA TTG CTC GG-BHQ3-3 Molecular Beacon Design The use of MBs for intracellular RNA detect ion and localization requires the design of the MB for mRNA inside of a cell. Designing a MB involves three majo r areas; the loop, the stem, and the fluorophore/quencher pair. Th e primary concern in designing MBs for intracellular use is designing the loop region. The first step of wh ich is selecting an appropriate target region on the mRNA sequence for the MB. This is critical due to the complex secondary structure exhibited by the large mRNA sequences. The MB must be able to reach to its complementary sequence through the secondary structure or the MB will not hybridize and produce the fluorescent signal for detection. Unfortunately current mRNA folding programs do

PAGE 105

105 not give a completely reliable secondary structure, thus a series of MBs is generally selected for each mRNA sequence. Once possible target se quences are identified through the RNA folding programs, MBs are designed for them. The stems are created after the l oop region is devised. This is to make certain that the stem region is not complementary to any region in the loop. Once the DNA portion of the MB is designed the fluorophore and quencher pair can be decided. This is generally based on the instrumentation av ailable for the experiment to ensure efficient excitation and detection of the fluorophore in the open state and hi gh quenching efficiency in the closed state. A series of probes is then synthesized and tested in vitro to make sure that the MB is functional. Further testing is then done in vivo until a probe is found that can hybridize with mRNA inside of a cell. Another more practical concern when desi gning and synthesizing molecular beacons for any application is the final nucleotide base positioned before the fluorophore. Several investigators have observed that nucleotid es can quench the fluorescence emission of fluorophores.213-215 While the quenching efficiency of the bases is smaller than those of commonly used quenchers, it can still significantly effect the performance of the molecular beacon. Guanine (G) is the best quencher among the bases, followed closely by adenine (A), while cytosine (C) and thymine (T) exhibit a much lower quen ching efficiency. The quenching properties of guanosine are centered on its elec tron donating ability that allows the energy transfer between the base and the fluorophore.45 As a result, molecular beacons designed with a C connecting the fluorophore and a G linking to the quencher moiety display better signal enhancements than other arrangements. When designing MBs for any application, an important factor to consider is the fluorophore and quencher pair. Recently, a syst ematic study of various static quenching

PAGE 106

106 efficiencies for different fluorophore and quencher pairs has been documented.213 Properly matching the fluorophore with an effective que ncher molecule can le ad to substantial improvements in detection cap abilities. This is achieved by reducing the background fluorescence from the molecular beacon in th e absence of target DNA. Similarly, using quenching moieties that are better able to absorb or transfer en ergy from the fluorophore can also lead to better fluorescence enha ncements. It has also been shown that spectral overlap is less important for effective quenching as the primar y mechanism of quenching is static quenching.213 One observation that further suppor ts the static quenching mechan ism of molecular beacons is the alteration that is observed in the spect ral properties of the fluorophore and quencher.216 The hairpin-loop structure brings the fluorophore and quencher into su ch close proximity that it disturbs their electr onic structure. As such, while many probes are initially te sted with FAM/Dabcyl fluorophore/quencher pairs these are often replaced for the actual applications to achieve better sensitivities. Figure 51 shows the fluorescence enhancements of severa l molecular beacons using FAM/Dabcyl pairs and then the higher quality pairs used in subseque nt applications. In th ese experiments only the fluorophore quencher pair was changed, each mol ecular beacon was at a concentration of 1 M with a target concentration of 10 M. The optimal fluorophore/que ncher pairs are Alexa Fluor 647 and Blackhole Quencher 3 for the MnS OD molecular beacon, Alexa Fluor 555 and Blackhole Quencher 2 for the control mol ecular beacon, Alexa Fluor 488 and Blackhole Quencher 1 for the B-actin molecular beacon, an d Cy3 and Blackhole Quencher 2 for both of the Cyclin D1 molecular beacons. The complete sequences and fluorophore quencher pairs are listed in Table 5-1.

PAGE 107

107 MnSOD MB Control MB B-actin MB Cyclin D1 MB1 Cyclin D1 MB2 FAM/DABCYL Optimal 0 10 20 30 40 50 60 70 80 90Enhancement Figure 5-1. Comparison of the enhancements of molecular beacons with FAM/Dabcyl and higher quality fluorophore quencher pairs. Molecular Beacon Delivery One of the more important decisions to be made concerning intracellular analysis with MBs is how to get the MBs delivered into the ce lls. Delivery of the MBs inside of the cell has been an area where a lot of effort has been a pplied and it has resulted in many very effective options for intracellular delivery. The most common delivery methods include microinjection6, electroporation7, peptide assisted delivery8, and reversible permeabilization9. Although any method that has been used to deliver oligonucleotid es inside of a cell has the potential to be an effective delivery mechanism of MBs inside of a cell. As each method has its own strengths and weaknesses, the proper means of delivery generally depe nds on the application. Microinjection involves the use of special equipment in order to physically pierce the cell and deliver the MBs. This is accomplished throu gh the use of a micromanipulator that allows for the precise positioning of an injector tip. The tip s are generally pulled fused silica capillary or

PAGE 108

108 commercially purchased tips. The MB solution is loaded into the tips and connected to a microinjector. Once the tip is positioned inside of the cell, the microinjector sends a pulse of pressure through the tip that forces the MB so lution into the cell. The amount of solution delivered can be finely tuned by changing the pre ssure and duration of the pulse. Microinjection has several advantages, most im portantly, it allows for the immedi ate monitoring of the cell for the response of the probe. Secondly, it delivers relatively reproducible am ount of probe to the cell of choice of the experimenter. The disadvant ages of using microinjec tion are related to the technique itself in that it requires additional in struments and expertise while being a very low throughput technique. Electroporation and reversible permeabilizati on are analogous techniqu es that deliver MBs through passive diffusion through pores created in the cell membrane. The two techniques differ through the process in which they create the pores. Electroporati on can create pores in the cell membrane through the application of an electrical pulse to the cells. The pulse causes the cell membrane to develop pores which allows any probes surrounding the ce lls to pass through the cell membrane. Reverse Permeabilization also delivers the MBs through pores created in the cell membrane. This involves the use of the chem ical, Streptolysin O to create pores in the cell membrane. Activated Streptolysin O binds to ch olesterol molecules in the cell membrane to form channels approximately 30nm in diameter in serum free media. The channels allow the passive diffusion of materials in to the cell. However the pores in both methods can also allow the loss of materials from inside of the cell a nd there are variations in the amount of probes delivered into each cell. In addition, electroporation also requ ires additional equipment to accomplish the delivery. Reverse Permeabilization also requires time to form the pores and then additional time to allow th e pores to reseal.

PAGE 109

109 Peptide-assisted delivery and liposome based transfection methods allow the probes to pass through the cell membrane without disturbing the cell. It allows any materi als conjugated to the peptide or inside the liposome to be endocytoti cally passed through the cell membrane and into the cell. This allows delivery of the MB into the cell without fo rming artificial pores or physical injection. However peptide assisted delivery requ ires the peptide to be conjugated to the probe which can increase the cost and complexity of th e probe synthesis. Both methods also require time for the probe to be delivered inside of the cell. The one encompassing theme is that one must select the proper delivery mechanism based on the application and also on th e properties of the molecular pr obes to be delivered. Probes based on non-standard bases can survive longer inside the cell allowi ng for longer incubation times to be used. In order to evaluate which method would work best for the study of cancer cells using MBs, several different delivery methodologies were tested based on normal DNA molecular beacons. Electroporation Electroporation is common method for gene transf ection and can refer to either single cell electroporation or the mass deliv ery of materials to a whole population of cells. The major advantage of electroporation is th at it can achieve a mass delivery of the probes with a very short (less than one minute) incubation time, thus allowing real time monitoring of the probe like in microinjection without the limitations of microinj ection. For this experiment, the focus is on electroporation as a mass delivery method as single cell electroporation only differs slightly from microinjection. Electroporation was performed using a BTX ECM 830 pulse generator (Harvard Apparatus) with a PP35-2P Petri Pulser electrode. The electrode is designed to fit a 35mm cell culture dish and be able to deliv er a sufficient electrical pulse to deliver the DNA into the cell.

PAGE 110

110 The initial settings were base d on a protocol for MDA-MB-231 cells provided by BTX using a pulse of 300V for 50 s with a 1 M concentration of fluorophore labeled DNA. Figure 5-2. A) Fluorescence images of cells after electroporation in a 5 M DNA spiked medium. B) Average cell intensities of ten cells after electroporation. This protocol produced no noticeable change in the fluorescence of th e cells. The voltage, pulse length, concentration of DNA and number of pulses were th en increased until reaching the instrument maximums. The results of a repres entative experiment performed with five 50 s 1,000V pulses in Figure 5-2 with the fluorescence image (A) and th e average cell intensities (B) indicating little to no uptake of the DNA transpired. These experiments were repeated using trypsinized cells in BTX electropor ation cuvettes and the cell were imaged after replating in a 35mm culture dish however no fluorescence signal reached levels suitable for use as a MB delivery method. These results indicate that while electroporation functions well for gene transfection were only low copy nu mbers of the genes need to be delivered, the method does not translate well for DNA probe delivery. This is due to the need for a large amount of the probes to be delivered to the cells in order for the gene expression to be reliably detected. The failure of the electroporation delivery method means that obtaining high throughput delivery of the probes for real time measurements may not be feasible at this juncture as other methods for mass

PAGE 111

111 cellular delivery require incubation times of at least 30 minutes and there is no practical method for the high throughput use of microinjection. Reversible Permeabilization In order to evaluate the use of reversib le permeabilization as a delivery method for molecular beacons, two primary criteria were esta blished. The first and most important was the efficiency of the delivery of the probe material. If an insufficient amount of probe is delivered to the cell, then the overall sensitivity of the meas urement is affected. Also, the amount delivered to the cell needs to be relatively reproducible, there will likely be some cell to cel l variations in delivery however many variations can be compensated for by using ratiometric analysis which will be discussed at a later point in the chapter. The second criterion is the incubation time required for the probe to be de livered in effectiv e quantities. Since unmodified DNA based molecular beacons will be used for subsequent experiments, the incubation time for the delivery must take less than 30 minutes. This is due to the degradation of the molecular beacon inside the cell that occurs after 30 minutes. The MB degrada tion also causes the rest oration of fluorescence thereby making it indistinguishable from target hybridization. Th erefore, in measurements taken after 30 minutes it will be difficult to prove what signal it due to target hybridization and what signal is due to probe degradation. Fluorophore labeled were delivered into liv ing cells using a previously published reversible permeabilization protocol.211 The protocol utilizes stre ptolysin O (SLO), which was shown to be a rapid, efficient, less damagi ng and more versatile compared with many conventional transfection methods.212 Specifically, SLO was first activated by adding 5 mM of TCEP to 2 U/ml of SLO for 30 min at 37 C. Cells grown in 35mm dishes were incubated for 10 min in 200 m l of serum free medium containing 0.2 U/ml of activated SLO (0.5 U SLO per 106 cells) and 1 M of the AF488 labeled DNA. The cells were then incubated for different

PAGE 112

112 incubation periods of 15 minutes, 30 minutes, 45 minutes, and 60 minutes. The cells showed uptake of the DNA after 15 minutes however the uptake did not plateau until 45 minutes (data not shown). Figure 5-3A shows a representati ve fluorescence image, Figure 5-3B shows the average fluorescence intensity of ten random cells. After 45 minutes, the cells showed a relatively reproducible fluoresce nce intensity that was quite suitable for imaging or other applications. While the incubation time wa s too long for conventional DNA MBs reversible permeabilization will likely be an effective delive ry method for probe enhanced with nuclease or protein resistance. Figure 5-3. A) Fluorescence image of fluorophor e labeled DNA delivered through reversible permeabilization. B) Average intensities of ten random cells after delivery of probe using reversible permeabilization. Liposome Delivery Another common used method for the delivery of DNA into the cells is the use of lipid based transfection reagents. These reagents form a bilayer similar to the cell membrane when placed in aqueous environments. Any material stab le in the aqueous environment is then trapped inside the membrane. When mixed with the cells, the liposome can deliver their contents inside of the cell either through inco rporation of the lipid bilayer into the cell membrane or the breakdown of the bilayer by lysosome inside of the cell both of which cau se the release of the

PAGE 113

113 materials into the cell. In order to determin e whether the lipid based transfection would be suitable for probe delivery a commercially av ailable transfection r eagent Lipofectamine 2000 (Invitrogen) was used to deliver the AF488 labeled DNA. Prior to any experiments, the cells were incubated in antibiotic free media to prevent any uptake of the antibiot ic into the cell. The uptake of the antibiotic into the cell through the liposome complexes is highly toxic to the cells. The stock DNA was then diluted to 1 M concentration while 10 L of Lipofectamine was diluted in a separate 50 L solution of cell media. Then the diluted Lipofectamine was gently mixed with 50 L of the DNA solution and incubated for 30 mi nutes. The cell media in the cell culture dish was then replaced with the transfection mixture and incu bated for one hour. After the incubation period, the transfection reagents were removed and the cells were washed three times with cell media. The cells were then imaged w ith the fluorescence image of the cells shown in Figure 5-4A. The average intensities of ten rand om cells are plotted in Figure 5-4B. Incubation times of less than 1 hour were found to have sign ificantly lower fluorescence intensities. The liposome based transfection results however indicate that this me thodology can deliver sufficient amounts of material inside of th e cell for analysis using MBs. However the incubations times required are unsuitable for DNA based MBs. As an interesting side note, the liposomes appear to have an affinity for the culture dish surface as well the cell membrane of the cultured cells. In each experiment conducted with the liposomes th e bare cell culture dish surface generally had several highly fluorescent spots on the surface even after several washing steps. While this affinity did not appear to eff ect the overall results, the presence of the spots on the dish can impair the determination of the fluorescence bac kground and lead to some uncertainty in those determinations. In addition, the liposome based methods had the greatest variability in the delivery and further use would likely re quire some form of normalization.

PAGE 114

114 Figure 5-4. A) Fluorescence image of cells af ter delivery of fluorophor e labeled DNA using a lipid based transfection reagent. B) Aver age fluorescence inte nsities of ten random cells after delivery of fluorophore labeled DNA. Viral Vector Delivery Following the establishment of the first infe ctious clone of AAV serotype 2 (AAV2) in 1982,223 viral vectors have rapidly gained popularity in gene therapy applic ations, due to their lack of pathogenicity, wide rang e of infectivity, and ability to establish long-term transgene expression.224 Recombinant AAV2 vectors have been tested in preclinical studi es for a variety of diseases such as hemophilia, 1 anti-trypsin deficiency, cystic fibrosis, Duchenne muscular dystrophy, and rheumatoid arthritis. At least 20 clinical trials have been completed or initiated with 15 different AAV2-based vectors being admini stered in several hundred patients thus far.224 Also, AAV vectors have been used for gene deli very in a variety of cell types including liver,225227 muscle,226 brain,228 retina,229 and cancer cells.230 Therefore, we have sought to determine whether they have the potential as a deliver y vehicle for DNA based molecular probes. AAV viral vectors that were modified to express several biotin molecule s on their surface were obtained from the labs of Kenne th Warrington. Prior to incuba tion with the ce lls, the virus vectors were incubated with excess FITC-labeled streptavidin for 12 hours. In theory, the

PAGE 115

115 streptavidin would then be used to attach seve ral biotinylated sequences onto the surface of the vector, however for this proof of concept the labele d streptavidin alone was used for the quantification of the amount deli vered. After incubation with the streptavidin, th e viral vectors were incubated with the cells for 1 hour that was based on published reports at a concentration of 1,000 viral particles per cell. Figure 5-5. A) Fluorescence image of cells after incubation with a FITC-S treptavidin conjugated viral vector. B)The average intensity of ten random cells after incubation with FITCstreptavidin linked viral vectors. Figure 5-5A depicts a represen tative fluorescence image of the cells while Figure 5-4B shows the average intensities of 10 random cells. Overall the viral vectors delivered a fair amount of material into the cell although it was less than many of the other methods tested. It should be noted however, that once any biotinylated probe was adde d to the vector, 2-3 times as much probe material could be delivered due to the multiple biotin binding sites situated on the streptavidin. While further optimization of the delivery protocol would have proved beneficial, there was insufficient virus for s ubsequent experiments. This indi cates that anothe r limitation of the viral vector based delivery is the lack of co mmercially available vector suitable for use with synthetic DNA probes.

PAGE 116

116 Microinjection Microinjection is a well establis hed technique for the delivery of materials into the cell. The primary criteria for the delivery of the AF488 labeled DNA using microinjection was the amount of DNA delivered. Since no incubation tim e is necessary for microinjection, it allows the real time monitoring of hybrid ization events inside of the cell. However, the low throughput nature and high technical expertise required make systematic studies us ing microinjection much more difficult and time consuming. A Leiden microincubator with a TC-202A temperature controller (Harvard Apparatus, Holliston, MA) was used to keep the cells at 37oC during injection and monitoring. An EXFO Burleigh PCS-6000-150 micr omanipulator was used for positioning the injector tip. An Eppe ndorf Femtojet microinjector with 0.5 m Femtotips was used to inject the 1 M AF488 labeled DNA sequence into the cell. The cells were injected utilizing a 40x 1.35 NA oil immersi on objective. The cells were injected with the DNA at 8psi for a duration of 10 s. The cells to be injected were selected base d on morphology in an effort to inject healthy cells. Cells displaying a small round morphol ogy or a large blob-like morphology were not injected. Figure 5-6A shows a re presentative injected cell while Figure 5-6B shows the average intensity from 10 different injected cells. Ov erall the injected cells had the highest signal intensity of the delivery methods tested and showed good reproducibility. The use of microinjection produces several adva ntages in that the lack of an incubation time allows real time monitoring, an effective amount of probe can be delivered, and the approach is valid for most adherent cell types. That being said the previously ment ioned disadvantages cannot be overcome as they are inherent to the microinjection based delivery.

PAGE 117

117 Figure 5-6. A) A representative image of a cell injected with the fluor ophore labeled DNA. B) the average cell intensities of ten different cells inject with the fluorophore labeled DNA sequence. Incubation The delivery efficiency of a simple inc ubation of the AF488 labeled DNA sequence was also performed. This experiment was conducted more as a control for the previous techniques that required an incubation time than as a true delivery method. For these experiments, a 1 M solution of AF488 labeled DNA was prepared using cell media. The cells were incubated for 1 hour in the DNA spiked cell media after which th e DNA spiked media was replaced with normal cell media. The fluorescence images and analysis were performed at that time. Figure 5-7A shows the cells after the incubation while Figure 5-7B illustrates the average cell intensities of the live cells obtained after in cubation. While many of the d ead cells (based on morphology) exhibited a high fluorescence inte nsity indicating an uptake of the probe, living cells showed little to no uptake of the fluor ophore labeled DNA. This indicat es that incubation alone is insufficient as a delivery method and that the re sults from the previous experiments where uptake of the DNA occurred were not due to simple incubation of the probe.

PAGE 118

118 Figure 5-7. A) Fluorescence image of cells incubated with the fluorophore labeled DNA sequence for 1 hour. B) Plot of the aver age cell intensities of ten random cells incubated with the fluorophore labeled DNA. Delivery Method Conclusions In order to more effectively compare th e different delivery methods, the average fluorescent intensities of the cells analyzed for each deliver y method were calculated and plotted in Figure 5-8. Microinjection, reversible permeabilization, and the liposome based transfection all performed well with each method deliveri ng experimentally significant amount of the fluorophore labeled DNA. The viral vector shows the potential for being a solid method for MB delivery although a great deal of work is still requ ired to achieve that goal. Electroporation and incubation were both unsuccessful in delivering the fluorophore la beled DNA inside of the cells. However, in selecting a delivery method for futu re studies with DNA MBs, there were two main criteria established, one being that sufficient amounts of mate rial can be delivered. The other criterion that is equally importa nt is that any incubation must be less than 30 minutes. The two methods that achieved delivery in less than 30 mi nutes were microinjecti on and electroporation. The electroporation did not deliver sufficient materi al for use in the intracellular studies leaving microinjection as the only method left standing. However, when effective probes that are more

PAGE 119

119 stable in the intracellular environment are use d, delivery methods like the liposome transfection and reversible permeabilization should be very effective optio ns for MB delivery assuming the similar delivery efficiencies w ith the non-standard MBs. The us e of microinjection does create some additional problems. Certain problem s like being low thr oughput and technically demanding are inherent to the method and cannot be effectively addressed. The variability that can be seen in the amount of probes delivered and other experimental variations can be addressed through different normaliza tions procedures. In particular ratiometric analysis can be an effective tool in obtaining more reliable results from even technically demanding intracellular studies. 0 500 1000 1500 2000 2500 3000 3500Average Intensity Microinjection Viral Vector Electroporation Incubation Reversible Perm. Liposome Figure 5-8. Average intensities from the cells for each DNA delivery method tested. Ratiometric Analysis The potential of using MBs to detect DNA and mRNA inside living cells has previously been demonstrated. However, such intracellula r measurements have been based on measuring

PAGE 120

120 the intensity of a single fluores cence signal. This fluorescence si gnal can be affected by signal changes from light scattering by the sample, excitation source fluctuations, variability in the intracellular delivery, and variability in the intracellular environment.231,232 As a result, the use of MBs has been limited to visual ization and not quantitation. Ra tiometric analysis is one methodology that can address the problems posed by taking intensity measurement231,233,234 since they normalize these effects by taking the ratio of the signal intensit y caused by the specific probe for a target over the signal intensity produ ced by an unrelated reference probe that works as an internal standard. The ratio value is ca lculated by dividing the MB signal intensity by the signal intensity of the reference probe. Since th e ratio between the two molecules stays constant change in the ratio only occurs when there is a change in the fluorescen ce intensity in one probe and not the other, like when the MB hybridizes to its target. The ratiometric analysis allows for the normalization of the fluorescence intensities fr om a myriad of experi mental factors thus providing more reliable data. To demonstrate the advantage of the ratiomet ric analysis used in future experiments a series of injections were made into the M DA-MB231 cells. For each injection the time duration of the injection was varied so that different amounts of material would be delivered to each cell. The probe solution consisted of 1 M concentrations of DNA labeled with either AF488 or AF647. The ratio of the DNA mixture was held c onstant. The cells were then monitored for nine minutes after injection and the fluorescent intensities of both probes were measured. Figure 5-9A shows the AF647 intensities of five cells in jected with different amounts of fluorophore. This represents how the fluorescence intensity from experiment to experiment can altered in a controlled format as the cells showed dramatically different fluorescence in tensities. However, when the same intensities are normalized through the ratiometric analysis, the result is a fairly

PAGE 121

121 constant ratio as illustrated in Figure 5-9B. This allows for more reproducible results to be obtained in the intracellular experiments and for more reliable gene expression data. Figure 5-9. A) Fluorescent intensit ies over time for five cells inje cted with different amounts of fluorophore labeled DNA. B) The same cells af ter ratiometric analysis is used to normalize the fluorescent intensities. In Vitro Testing of Molecular Beacons In order to use the MBs for multiplexed st udies, their individual performance for both in vitro and in vivo had to first be evaluated. The MBs were first tested in simple buffer experiments to confirm that they could hybridize selectively to th eir targets and that their signal enhancements were sufficient for intracellular an alysis. Each MB that will be used in the subsequent intracellular experime nts was first tested in buffer solution to verify its performance prior to intracellular demonstration. Each MB was diluted to a 1 M solution, each MB sample was then incubated with either buffer, 10 M target DNA, or 10 M random DNA. The signal intensities from the samples with target and ra ndom DNA were then divided by the closed MB intensity determined from the sample incubated with buffer. This ratio is referred to as the signal enhancement of the MB and the enhancements fr om the target and random DNA were plotted in Figure 5-10 for each MB sequence and fluorophore quencher pair.

PAGE 122

122 Figure 5-10. Signal enhancements in buffer soluti on for each MB utilized in the intracellular experiments. The sequences and fluorophore quencher pairs fo r the MBs are listed in Table 5-1. The figure illustrates that each MB has excellent se lectivity and a good signal enhancement of at least 15. The establishment of the 15 fold criteria simply indicates that the beacon undergoes a significant signal transduction upon bi nding to its target however the beacons were also assessed on their overall intensity as intracellular measur ements require probes with very high signal intensities. Therefore, the beacons used in the subsequent two chapte rs were thoroughly and effectively evaluated prior to use. Conclusions Based on the results in this chapter, the opt imal parameters for future intracellular experiments were determined. The sequences for the MBs were tested and the fluorophore and

PAGE 123

123 quencher pairs were tested and optimized. Th e delivery methods for oligonucleotides were thoroughly evaluated leaving micr oinjection as the preferred me thod for DNA MB studies. The future application of this work is for the effec tive detection and monitoring of the MBs inside of single cells. The right selecti on of fluorophores and sequences w ith the most effective delivery method should allow for the sensitive detection of multiple mRNA sequences inside of a single cell. This would allow biological studies into how the expression levels of different genes are related and allow for new trends and correlations to be observed.

PAGE 124

124 CHAPTER 6 STUDY OF CANCER CELLS THR OUGH MULTIPLE GENE MONITORING Introduction A method in which the mRNA expression prof ile of a single cell can be collected and compared against another single cell would provide a powerful tool for many biological or medical applications that invol ve gene expression. In part icular, the ability to monitor cytoplasmic mRNA expression of multiple genes woul d facilitate the study of cells to shed more light on disease processes and drug responses at th e single cell level. In our research, we have sought to demonstrate the abil ity to monitor the mRNA expr ession of multiple genes and differentiate between different le vels of mRNA expression. We have also sought to show that the data collected can be further analyzed to expose trends in multiple indivi dual cells that would go unnoticed in single gene monitoring experiments. In these experiments, we used three separate molecular beacons with three un ique fluorophores to measure the expression levels of different genes in the cytoplasm of a singl e cell. Along with the molecular beacons, a reference probe is also injected in order to conduct the ratiometric analysis. The molecular beacons were designed to be complementary to -actin mRNA and manganese super oxide dismutase (MnSOD) mRNA, while a control molecular beacon was designed th at has no complement inside of the cells. MDA-MB-231 breast carcinoma cells were used bo th at basal expression levels and after exposure to lipopolysaccharide (LPS), which is an inflammatory mediator involved in E. coli bacterial sepsis and is known to stimulat e MnSOD mRNA expression in multiple mammalian cells. One of the limitations of molecular beac ons use for intracellular analysis has been the variability of the fluorescent si gnal inside of cells from an array of causes including light scattering in the sample, excitation source fluctuat ions, and variations in microinjection delivery of the molecular beacons into the cell. In an effo rt to achieve more reliable results for molecular

PAGE 125

125 beacons, in vivo ratiometric analysis was conducted to compensate for the instrumental and experimental sources of these fluctuations. Rati ometric analysis involves injecting a reference probe along with the molecular beacons inside of the cell. The reference probe emits a stable constant fluorescence signal that acts as an intern al standard for analyzing the signals from the molecular beacons. This technique thus allows the molecular beacon signals from different cells to be directly compared. Methods and Materials Cell Culture MDA-MB-231 breast carcinoma cells (American Type Culture Collection, Manassas, VA) were maintained in Dulbeccos Modification of Eagles Medium (DMEM, Fisher Scientific) with 10% fetal bovine serum (Invitrogen, Carles bad, CA) and 0.5 mg/ml Gentamycin (Sigma, St. Louis, MO) at 37oC in 5% CO2/air. Cells were plated in 35 mm glass bottom culture dishes and grown to 80% confluency (MatTek Corp., Ashl and, MA) for 48 hours prior to injection. To stimulate MnSOD mRNA expressio n, cells were incubated in 1 g/ml LPS from E. coli serotype 055:B5 (Sigma, St. Louis, MO) fo r 4 hours prior to injection. RNA Isolation and Northern Blot Analysis MDA-MB-231 cells were grown as describe d on 100 mm tissue culture plates until 7090% confluent prior to treatment with L PS concentrations of 0.1microgram/mL or 1 microgram/mL for 1, 2, and 4 hours. Total R NA was isolated by th e acid guanidinium thiocyanate-phenol-chloroform extraction me thod described by Chomczynski and Sacchi235 with modifications.236 Twenty micrograms of total RNA wa s size-fractionated on a 1% agarose formaldehyde gel and electrotransferred to a charged nylon membrane (Zetabind, Cuno Laboratory Products, Cuno Inc., Meriden, CT) and UV covalently cross-linked. The membrane

PAGE 126

126 was hybridized with 32P-labeled human MnSOD or glyceral dehye-3-phosphate dehydrogenase (GAPDH) (as an RNA loading control) cDNAs and subjected to autoradiography. Equipment Fluorescence imaging was conducted with a conf ocal microscope setup consisting of an Olympus IX-81 inverted microscope with an Olympus Fluoview 500 confocal scanning system and three lasers, a tunable Argon Ion laser (458nm, 488nm, 514nm), a green HeNe laser (543nm), and a red HeNe laser (633nm) with three separate photomu ltiplier tubes (PMT) for detection. The cellular images were taken with a 60x 1.4 NA oil immersion objective. A Leiden microincubator with a TC-202A temperature contro ller (Harvard Apparatus, Holliston, MA) was used to keep the cells at 37oC during injection and monitori ng. An EXFO Burleigh PCS-6000150 micromanipulator was used for positioning the injector tip. An Eppendorf Femtojet microinjector with 0.5 m Femtotips was used to inject the molecular beacons and reference probe into the cells. All analysis was conducted on the Fluoview 500 software, followed by processing of the data using Microsoft Excel. Fluorescence Imaging All cellular fluorescent images were collected using the confocal microscope setup. The confocal consists of an Olym pus IX-81 automated fluorescence microscope with a Fluoview 500 confocal scanning unit. Ther e are three lasers providing la ser excitation at 458nm, 488nm, 514nm, 543nm, and 633nm. Alexa Fluor 488 based probes were excited at 488nm and collected at 520nm. The Alexa Fluor 555 based MBs were exci ted at 543nm and collected at 560nm. The Alexa Fluor 647 based MBs were ex cited at 633nm and collected at 660nm. RuBpy was excited at 458nm and the emission was collected at 610nm. Images were taken after a five to ten second period during which the instrument was focuse d to yield the highest intensity from the

PAGE 127

127 fluorescence channels. The images were assigned color representations for clarity and are not indicative of the actual emission wavelengths. Molecular Beacon Design for Multiple Gene Monitoring One of the first issues confronting these experiments was selec ting the appropriate fluorophores to use on the three molecular beacons and the reference probe to ensure that no crosstalk or fluorescence resonance energy transfer (FRET) was taking place. Due to their high quantum yields and separation of their emi ssion and excitation wavelengths AF488, AF555, and AF647 were chosen for the fluorophores of the molecular beacons while RuBpy was chosen for its excitation wavelength which proved not to ex cite the molecular beacon fluorophores (data not shown). In addition, experiments were conducted to test the stability of each dye inside of the cell. Each dye including RuBpy was conjuga ted to a random oligonucleotide sequence and injected inside of a single cell. Those experiments showed that the fluorophores demonstrated excellent stability in a cellular environment for the length of the measurement (data not shown). The injections of dye labeled ol igonucleotide also allowed the calibration of the channels since the dye labeled oligonucleotide would mimic th e response of fully open molecular beacons inside of the cell and allowed for the optimiza tion of the optics to use for each channel to eliminate cross-talk between the different channe ls in the confocal system. Experiments were also conducted to verify that each molecular be acon could function properly inside of the cell by itself before conducting any experiments with multip le molecular beacons (data not shown). Part of these experiments included monitoring the stabili ty of the control molecu lar beacon inside of the cell to determine the lifetime of the probes inside of the cell. The control molecular beacon has no target inside of the cell and therefore any increase in signal from the molecular beacon is due to non-specific binding or degradation of the probe. We found that the molecular beacon was stable inside of the cell for 35 minutes afte r which there was an increase in the fluorescence.

PAGE 128

128 This indicates the molecular beacons have a finite lifespan inside of the cell of approximately 30 minutes, after that the molecular beacon can no longer be considered functional inside of a cell. Multiple Gene Monitoring With the suitable control experiments comple ted, the next experiments involved the actual detection of mRNA inside of a single cell and monitoring its e xpression over a set amount of time. Figure 6-1 shows an image of each molecu lar beacon detection channel and the reference channel to demonstrate the stability of the refere nce probe over time and the basal expression of -actin. In this experiment femtoliters of 1 M of each of the three molecular beacons and 20 M of the RuBpy reference probe were injected into a single human breast carcinoma cell. Using the signal from the reference probe, the instrument was focused to yield the highest fluorescence intensity in the reference channel. After an initi al image is collected, the cell is monitored for 14 minutes with the fluorescent signal from each channel being measured every minute. This allows ample time for the molecular beacon to hy bridize with its mRNA complement without the increase on fluorescence that may accompany the degradation of the mo lecular beacon after 30 minutes. The PMT for each channel was set to be able to measure the full enhancement of each molecular beacon as determined in the control e xperiments. The nucleus of the cell in each channel was discarded during analysis of the sign al since nuclear uptake mechanisms exist that cause nucleic acids to be transferred to and non-sp ecifically opened or degraded in the nucleus. This process can cause a false-positive signal, although it is uncl ear whether the signal is from degraded probes or simply the concentration of the fluorescent probes in the nucleus. Thus the analysis of the fluorescence signals from the cell cytoplasm is used to determine the mRNA expression levels.

PAGE 129

129 Figure 6-1. Time elapsed fluorescent images of each MB inside of a single MDA-MB-231 cell. A is the image of the -actin MB (green), B is the imag e of the control MB (red), C is the image of the MnSOD MB (blue) and D is the image of the RuBpy reference probe (orange). In the channel B, the channel monitoring th e control molecular b eacon, there is a very small amount of background fluorescence from the molecular beacon, most likely from incomplete quenching of the fluorophore or some form of degradation. The signal however remains at a very low level throughout the m onitoring period, even as the other molecular beacons show an increase in their fluorescence signal. Due to the overall low level of fluorescence and the lack of any increase of fluorescence, it appears the control molecular beacon remains closed inside of the cell ove r the monitoring period, which shows that the

PAGE 130

130 molecular beacon is mostly stable and also re presents the absence of its complement. The significance of the negative response of the control molecular beacon is that it indicates that the signal produced from the other molecular beac ons inside of the cel l is a result of the hybridization with its mRNA complement and not due to degradation of th e probe inside of the cell. If degradation were taking place during the monitoring period, then the control molecular beacon as well as the MnSOD and -actin molecular beacons should have been affected equally producing fluorescence. In channel A, the channel monitoring the -actin molecular b eacon, the fluorescence signal starts at a barely detectable level a nd increases over time to a level in which the fluorescence saturates the PMT. The increase of fluorescence represents the hybridization of the molecular beacon and its target, the -actin mRNA. This is consiste nt with the high expression anticipated for -actin mRNA. The MnSOD molecular be acon monitored in channel C shows a slightly higher fluoresce nt signal than the control molecular b eacon but not to the extent seen in the -actin molecular beacon. This is consistent with basal levels of MnSOD mRNA, which are usually at a low expression level. From this e xperiment we have seen that it is possible to monitor more than one gene simultaneously inside a single living cell. Monitoring Stimulated Gene Expression The next goal was to demonstr ate the methods ability to m onitor gene expression changes in response to cellular treatment. For these experiments two groups of cells were used, one control group of human breast car cinoma cells and a second group where the gene expression of MnSOD had been induced by LPS stimulation. The inducible expressi on of MnSOD mRNA by LPS has been previously described236 and was verified by Northern an alysis prior to the cellular monitoring experiments. Northern analysis allo ws for the accurate determination of steady-state mRNA expression from a large number of cells. MDA-MB-231 cells were incubated with LPS

PAGE 131

131 for time intervals of 1, 2, or 4 hours prior to total RNA isolation and northern analysis. The results shown in Figure 2 confir m that MnSOD mRNA expression is significantly increased after incubation with 0.1 g/ml or 1.0 g/ml LPS. During this short LPS-exposure experiment, the greatest increase in MnSOD mRNA expression was observed after a four hour incubation with 1.0 g/ml LPS. Figure 6-2. Northern analysis of LPS-i nducible MnSOD mRNA expr ession in MDA-MB-231 cells. Based on the ability of the cells to produ ce a significant increase in MnSOD mRNA production after a four hour inc ubation with LPS, the three molecular beacons and the reference probe were injected into a contro l and an induced human breast can cer cell in order to monitor a change in the MnSOD mRNA expression. The expe rimental parameters were similar to the

PAGE 132

132 initial mRNA expression monitoring experiment with each molecular b eacon at a concentration of 1 M and the RuBpy reference probe at a concentration of 25 M. Each cell was monitored for 14 minutes and the same signal coll ection and analysis protocols we re observed as used in Figure 6-1. The signals from the cytoplasm were then measured along with those of the background. The background in each channel was subtracted from both the molecular beacon signal and the reference signal, after which the ratio values we re calculated. The experiment was repeated multiple times and the results of a repres entative cell are shown in Figure 6-3. Figure 6-3. Ratiometric analysis of the real time monitoring of gene expression (A) Real-time monitoring of gene expression in a single control MDA-MB-231 cell. (B) Real-time monitoring of gene expression in a single LPS induced MDA-MB-231 cell. In both cells the control mol ecular beacon exhibits the same constant low level of fluorescence. Since this molecular beacon shows no increase in signal and has no target inside of the cell it is believed that it remains clos ed during the monitoring ti me. Unlike the control molecular beacon, the -actin molecular beacon shows an incr ease in fluorescence in both cells, using the same rationale as befo re it is believed that the -actin molecular beacon is hybridizing with the -actin mRNA. The level of -actin is similar in both cells which is expected since LPS has not been shown to impact the expression of -actin mRNA. The signal from the MnSOD molecular beacon however change s significantly in th e induced cell when compared to the noninduced cell. Whereas no significant increase in the MnSOD molecular beacon signal is

PAGE 133

133 observed in the control cell, there is a large increase in signal from the MnSOD molecular beacon in the induced cell. Based on th e lack of change in the control and -actin molecular beacon it is believed that this large increase in signal is due to the increased expression of MnSOD mRNA. This is consistent with the No rthern analysis, which also showed a large increase in MnSOD mRNA expression in cells e xposed to LPS. Thus the results from the Northern analysis correlate with the result s obtained from the multiple gene monitoring by molecular beacon. The gene expression profile was determined twenty times for both the control and LPSinduced cells, the average peak ratio for each molecular beacon was calculated as the average ratio value after the signal plateaued inside of the cell. This value is used as the final signal from each molecular beacon. Consistently in each of the cells measured, high expression levels of actin mRNA were measured along with constant low signals from the control molecular beacon. There was a degree of variation in both the LPS induced and non-i nduced cells that showed some variation in how highly -actin mRNA was expressed. This is consistent with the findings of other investigators that there are cell to cell variations in gene expression.237-239 In each induced cell there was a high expression le vel of MnSOD, which was far different from the control cells, which generally showed a very low expression level with the exception a few that showed a moderate level of MnSOD expression. The induced cells were consistent with our expectations based on the results of the Northern Blot. In agreement with the -actin mRNA expression, the induced and non-induced MnSOD mRNA expression showed some variability which is again consistent with the belief that there is no aver age cell and that mRNA ex pression will vary from cell to cell.

PAGE 134

134 In order to compare the results from all of th e cells, the average peak signals from each cell were collected and plotted in a histogram in Figur e 6-4. The histograms cl early show a shift in the MnSOD mRNA expression in the induced cell s when compared to the non-induced cells. The average ratio of signals for each molecular beacon was found to be 1.8 for -actin, 0.2 for the control MB, and 0.6 for MnSOD in th e untreated control cells; and 2.1 for -actin, 0.3 for the control MB, and 2.0 for MnSOD in the induced gr oup of cells. Thus in each of the twenty induced cells there was a significant increase in the ratio of the MnSOD molecular beacon when compared to the control cells. The -actin and control molecular beacons had relatively similar ratios in both the induced and control cell line s, throughout the entire series of experiments, although the -actin expression showed a small increase in expression in the LPS-induced cells. Figure 6-4. Histograms showing distribution of ratiometric re sponses for control and LPS induced MDA-MB-231 cells with control MB (red), MnSOD MB (blue) and -actin MB (green) measured simultaneously in each cell. Gene expression pattern comparison Since we are able to detect and monitor th e mRNA levels of different genes inside the same single cell, comparisons can be made at the single cell level and trends can be elucidated. In determining the gene expression profiles of bot h groups of cells, there were variations in the

PAGE 135

135 levels of gene expression in both -actin and MnSOD mRNA. Base d on results we sought to determine if the variations formed a pattern of gene expression or were merely random fluctuations. It was unknown if any such patt ern existed since no prev ious literature was found that compared the mRNA expression of multiple genes inside the same cell to expression levels inside of other single cells. In this comparison the MnSOD and -actin mRNA expression levels were set relative to the control molecular beacon to compensate for any va riances in the cellular environment that would cause the expression levels to appear ar tificially high or low. The relative ratios from each cell were then plotted against each other in Figure 6-5 to determine whether any pattern existed. Figure 6-5. Plots showing the re lationship between the relative -actin mRNA expression and the relative MnSOD mRNA expression in both the LPS-induced and non-induced cells. This analysis yielded a relationship betw een the expression levels of MnSOD and -actin. In the plot this relationship based on R2 values yielded a linear re lationship but it is unknown whether this is truly representa tive. The line mainly serves as a guideline for comparison. The analysis does show that when the -actin mRNA expression is high in comparison to other cells,

PAGE 136

136 the basal and induced levels of MnSOD mRNA is also high. This correlation is possibly explained by effects of MnSOD expression on the cell cycle. Previous work on different cell lines indicates that high MnSOD expre ssion retards the transition from the G1 phase to the S phase.240 Further research has shown that -actin expression is upregulated in the G1 phase.241 This indicated that a likely expl anation for the trend is that th e higher MnSOD expression causes the cell to remain in the G1 cell cycle phase which also causes the upregulation of -actin mRNA. On the other hand it is also possible that the trend is due to an artifact in the experimental method caused by biology phenomena such as differential transport into the nucleus or localization of the probe s to different parts of the cell given their different sequences. Future studies will examine the uptake and trans portation kinetics of the probes that may provide information to better distinguish between tr ends and artifacts. However, the observed relationship demonstrates the potential of the t echnique to explore the in ner workings of cell by giving a more complete picture of processe s involving mRNA expres sion by detecting the expression of multiple genes in single cells instead of focusing on merely a single mRNA expression. Conclusions We have demonstrated a novel method to monitor multiple molecular beacons simultaneously using confocal microscopy allowing determination of multiple genes expression profiles within single human breas t carcinoma cells. Once the suitable control experiments were conducted, the mRNA expression was measured in i ndividual cells that expr essed both basal and stimulated levels of mRNA. After satisfactor y results were obtained, the method was reproduced on twenty control and LPS induced cells. The -actin molecular beacon and the control molecular beacon demonstrated similar expression levels in the control and LPS induced cells however the MnSOD molecular beacon showed a significantly higher response when the

PAGE 137

137 expression of MnSOD was induced with LPS. This trend agreed with the Northern analysis of LPS induced MDA-MB-231 cells (Figure 2). The i ndividual cells in both the control and induced groups displayed variation in gene expression wh ich is consistent with the varied expression profiles seen by other inve stigators using fixed cells (17,18,19). We have demonstrated the ability to differentiate between basal and s timulated levels of mRNA expre ssion. Furthermore we showed the ability draw correlations between the levels of mRNA expression on a single cell level. This methodology should also have a wide applicabil ity since in theory any mRNA for which a molecular beacon can be designed can be studied along with controls or to simply monitor the expression of multiple genes inside of any single cell. Future research in this area includes determin ing the effect of the cell cycle on the levels of mRNA measured, evaluating the toxicity of the probes on the ce ll, and measuring the kinetic parameters of the probes inside of the cell. Th ese experiments will allow us to limit the cell to cell variability and allow for better distinctions to be made between cellula r trends and artifacts of analysis. Future applications of this me thod involve applying it to disease states to study effects on gene expression. These include issues of medical relevance such as investigating the effects on gene expression in human breas t carcinoma cells after treatment with chemotherapeutic drugs. Other issues of biol ogical importance involve investigating multiple mRNA expression patterns in sing le neuron cells as a means of investigating the processes involved in memory and learning. Furthermore, plans to incor porate nuclease resistance into molecular beacons would allow extended measurem ents of molecular beacon fluorescence inside living cells to the order of hours or days and incorporating different types of analytes like ions and proteins, thus truly broadening the scope of future experiments. Ultimately, utilizing molecular beacons linked to cell-penetrating peptides may allow the delivery of multiple

PAGE 138

138 molecular beacons to thousands of cells simultane ously. Development of such techniques might eventually permit clinical laboratories to util ize MBs in fluorescent plat e readers for diagnostic and drug sensitivity assays in cells obtained from human biopsy specimens for a variety of disease states.

PAGE 139

139 CHAPTER 7 STUDY OF CANCER CELLS BY MONITORING MULTIPLE ANALYTES In the previous chapter, fluorescence micr oscopy was used to monitor multiple gene sequences inside of a single hum an breast cancer cell. However, only looking at a single analyte type only gives a partial look at the complex processes inside the cell as genes are only one important class of molecule insi de of the cells. In most studies the detection and monitoring of analytes is restricted to a single type of analyte. By monitoring di fferent types of analytes inside of the cell, more complex phenomena can be studi ed including different signaling pathways. By studying the different analytes at the single cell level, trends and correlations can be elucidated by comparing the signal levels in side of different groups of signa l cells. In this way, the groups can be compared on a cell to cell instead of rely ing on the average responses of millions of cells. In addition to genes, molecules like ions have many vital functions insi de of the cell. In particular Ca2+ is an important signaling molecule to help regulate muscle contraction, neurotransmitter release, cell migration, gene expression, and cytoskeleton manangement. In order to advance the utility of single cell studies, we propos e using MBs and fluorescent Ca2+ indicators simultaneously in the same single cel l. By combining thes e different types of fluorescent probes for intr acellular analysis we can devel op a more complete understanding of biological processes. The ability to detect and monitor these two ve ry different types of analytes simultaneously will have far reaching applications from basic biological investigations to the study of diseases like cancer. Methods and Materials Fluorescent Probes All MBs were designed in-house based on pub lished mRNA sequences. The Alexa Fluor based MBs were synthesized by Ge nomechanix (Gainesville, FL) while the rest of the MBs were

PAGE 140

140 synthesized in-house. The fluorophores Alexa Fluor 555 (AF555), and Alexa Fluor 647 (AF647) were purchased from Invitrogen (Carlsbad, Ca lifornia). All othe r reagents including Tetramethylrhodamine (TMR) and Cy5 fluorophores were purchased from Glen Research (Sterling, VA). The quenchers used for the MBs were Blackhole Quencher 2 (BHQ2) and Blackhole Quencher 3 (BHQ3). The sequences and fluorophore/quencher pair combinations used are listed in Table 1. The pr obe solution used for injections in the experiments contained 1 M of each MB and in 20 mM Tris, 50mM NaCl and 5mM Mg2Cl2 buffer. All MB sequences were tested in vitro prior to use and all exhi bited enhancements greater than 15 fold upon target hybridization. Cell-permeant Fluo-4 Calcium i on indicator was purchased from Invitrogen (Carlsbad, California) and dilute d in DMSO to a concentration of 5mM. The Fluo-4 was then further diluted to 1 M in cell media. The cells were inc ubated in the Fluo-4 media for one hour prior to injection of the MBs. A ll probes were first evaluated separa tely inside of the cells prior to any multiplexing experiments to verify and evaluate their performance. Cells MDA-MB-231 breast carcinoma cells (American Type Culture Collection, Manassas, VA) were maintained in Leibovitz's L-15 Medium (American Type Culture Collection, Manassas, VA) with 10% fetal bovine serum (Invitroge n, Carlesbad, CA) and 0.5 mg/ml Gentamycin (Sigma, St. Louis, MO) at 37oC. Cells were plated in 35mm glass bottom culture dishes and grown to 80% confluency (MatTek Corp., Ashla nd, MA) for approximately 48 hours prior to any experiments. TSA exposed cells were incubate d in 300ng/ml Trichostatin A from Streptomyces sp. (Sigma, St. Louis, MO) for 24 hours prior to injection. Protein isolation and Western analysis MDA-MB-231 cells were grown as describe d on 100 mm tissue culture plates until 7090% confluent prior to treat ment with a TSA concentrations 50, 100, and 300ng/mL for 24

PAGE 141

141 hours. Total cellular protein wa s harvested utilizing CellLytic -M Mammalian Lysis Extraction Reagent (Sigma-Aldrich Co.) and Complete prot ease inhibitors (Roche Biochemicals). Protein concentrations were measured by modified Brad ford assay (Pierce Biot echnology) and proteins (20 micrograms of total protein for each sample) were size fractionated on a 10% Tris-Glycine polyacrylamide gel (Invitrogen) and electro-tran sferred to a cellulose acetate membrane for western analysis. Rabbit anti-MnSOD antibody and mouse monoclonal anti--tubulin antibody (Stressgen Bioreagents Corp.) were incubate d as primary antibodies followed by secondary horseradish peroxidase-linked IgG antibodies (Amersham Biosciences). Chemiluminescence detection of protein was performed using Supe rSignal chemiluminescence substrate (Pierce Biotechnology) followed by autoradiography. Equipment Fluorescence imaging was conducted with a conf ocal microscope setup consisting of an Olympus IX-81 inverted microscope with an Ol ympus Fluoview 500 confocal scanning system. The cellular images were taken with a 40x 1.35 NA oil immersion objective. A Leiden microincubator with a TC-202A temperature contro ller (Harvard Apparatus, Holliston, MA) was used to keep the cells at 37oC during injection and monitori ng. An EXFO Burleigh PCS-6000150 micromanipulator was used for positioning the injector tip. An Eppendorf Femtojet microinjector with 0.5 m Femtotips was used to inject the MBs into the cells. Fluorescence Imaging All cellular fluorescent images were collected using the confocal microscope setup. The Fluo-4 was excited at 488nm and collected at 520nm. The Alexa Fluor 555 and TMR based MBs were excited at 543nm and collected at 560nm. The Alexa Fluor 647 and Cy5 based MBs were excited at 633nm and collected at 660nm. Im ages were taken after a five to ten second period during which the instrument was focuse d to yield the highest intensity from the

PAGE 142

142 fluorescence channels. The images were assigned color representations for clarity and are not indicative of the actual emission wavelengths. Data Analysis All data collected were analyzed using the Fl uoview analysis software. Only the signal in the cytoplasm was used due to the nuclear uptake of the MBs. The same area of cytoplasm was analyzed in each channel for each measurement. The equilibrium state fluorescence for each probe was determined as the level where the probe fluorescence remained constant. Prior to further analysis, the background was subtract ed from all of the probe intensities. Simultaneous Detection of Ca2+ and Gene Expression inside of Single Cells The first step to developing this methodology was determining the best loading methodologies for both the MB and the Fluo-4 Ca2+ indicator. From a conceptual standpoint, methods that used membrane pore formation such as electroporation or reversible permeabilization for the delivery of MBs were not considered. The pores allow passive diffusion in and out of the cells and therefore the ion c oncentrations may reach equilibrium with the cell media preventing their accurate measurement. Methods requiri ng incubation times of greater than 30 minutes were also discounted. It has been shown that inside of cells DNA based MBs remain viable for thirty minutes191, therefore incubations needing longer than thirty minutes makes it difficult to determine whether the MB re sponse is from specific hybridization or merely due to degradation from the cellular environment. Finally, microinjectio n was selected since it had neither of these limitations and it also is reproducible from injection to injection and applicable to all adhe rent cell types. Next, the delivery of the indicator was investig ated. Our first effort was to deliver the indicator using microinjection al ong with the MBs. However, the indicator suffered from very poor retention inside the cell even when conj ugated to larger Dextran molecules (data not

PAGE 143

143 shown). The poor retention time resulted in the Fluo-4 leaking from the cell prior to the MB signal reaching equilibrium. This would have prevented taking the measurements from the cells at the same time point possibly in troducing a source of variation or error into the data. Instead, a two-step delivery mechanism, wherein cell perm eable Fluo-4 indicator was incubated with the cells prior to injection of the MBs was utilized. Using this protocol th e Fluo-4 exhibited very reproducible performance inside of the cell and ex cellent retention. The advantage of this twostep technique is that it can be applied to virt ually any type of adherent cells without any prior alteration to the cells. With a strategy in place, the first step was to validate the protocols using positive and negative controls. -actin mRNA was chosen as a positive control due to its high intracellular expression at all points in the life cycle of the cell. The second MB chosen was a negative control MB that was designed to have no comple mentary target inside of human cells. The actin MB and the Control MB have spectrally distinct fluorophores that enable their simultaneous detection along with the Fluo-4 in dicator. The similar fluorophores have been previously demonstrated to be spectrally compatible with our imaging equipment.242 The MBs were injected into MD-MB-231 human breast car cinoma cells and all th e fluorescent channels were monitored for 15 minutes with one image take n every minute. Figure 7-1a shows the time elapsed images from a representative cell. As can be seen in the image, the Fluo-4 signa l remains at a high level of fluorescence throughout the monitoring period. The high constant signal from the Fluo-4 indicates that it has reached equilibrium with the free Ca2+ prior to injection. Based on this constant signal, it was also concluded that the addition of the MB solu tion volume had a negligible effect on the concentration levels of the indicator and Ca2+. If there had been an appreciable effect, a decrease

PAGE 144

144 in the signal intensity of the indicator woul d have been observed, however no decrease was observed during the measurement or upon comparison of the signal obtained pr ior to injection. Figure 7-1. a) Time resolved fluorescence images from the Fluo-4 channel (green), -actin MB (red), and the negative control MB (blue). b) Plots of the averag e intensities versus time for each probe for the ten cells being monitored. The -actin MB signal is at a low level of in tensity during the first images but then increases steadily until plateauing after 10-12 minutes indicating an equilibrium occurring between the MB and the -actin mRNA. The Control MB has no target inside of the MDA-MB-

PAGE 145

145 231 cells and therefore remains at a low intensity level throughout the monitoring period. This is an important observation as it signifies that no nonspecific interactions are taking place during the monitoring period, which validates the increase of signal seen for the -actin MB. If that increase had been due to a nonspecific intera ction then the Control MB would have been similarly affected. Studying the Effects of Trichostatin A on Ca2+ levels and Gene Expression Histone deacetylase inhibitors such as Tric hostatin A (TSA) have been shown to inhibit the proliferation of tumo r cells in culture and in vivo by causing cell cycle arrest and/or apoptosis.243 One of genes affected by TSA is Ma nganese Superoxide Dismutase (MnSOD). Superoxide dismutases like MnS OD scavenge superoxide anions and catalyze their conversion to H2O2. Decreased levels of superoxide anions and other reactive oxygen species (ROS) have been shown to be related to tumorigenesis in several experimental models and human cancers.244,245 It has been reported that MnSOD ove r-expression in seve ral cancer cell types restores a relatively normal phenotype,246-248 and it has also been theo rized that MnSOD could be considered an anti-oncogene.249 The effect of TSA on MnS OD expression was verified using Western Blot analysis as illustrated in Figure 7-2e. MnSOD protein levels are dramatically increased after TSA exposure. Thus, TSA ha s been shown to affect both the MnSOD mRNA and protein expression in the MDA-MB-231 cells. There have also been studies showing that Ca2+ concentrations are affected by ROS250 and that intracellular ROS generation leads to an increase in the Ca2+ inside of cells.251 These two phenomena then might be connected in that the in crease in MnSOD would result in a decrease in ROS. The decrease in ROS may then result in the Ca2+ levels returning to normal. Therefore, we will use the methodology we developed to explore whether these intracellu lar events are linked as suggested or merely coincidental.

PAGE 146

146 Figure 7-2. a) Fluorescence image of Fluo-4 in the basal cell group. b) Fluorescence image of Fluo-4 in the TSA exposed cell group. Th e fluorescence is significantly decreased upon exposure to TSA c) Histogram of the range of intensities of Fluo-4 in the basal cell group. d) Histogram of the range of inte nsities of Fluo-4 in the TSA exposed cell group. Each histogram contains from the sign al intensities from thirty cells in each group. e) e) Western Analysis of the upregulation of MnS OD after exposure to TSA. The MnSOD protein is most upregulated after exposure to 300ng/ml of TSA. This concentration was then used for the further experiments involving TSA. -Tubulin is used as the loading control. As the effect of TSA on the Ca2+ levels inside the MDA-MB -231 cells has not yet been studied, the first step was to dete rmine whether TSA can affect the Ca2+ concentration inside the cell. For this study two groups of cells were utili zed. In the first group no changes were made to the cell media while in the second group 300ng/ml of TSA was added into the normal cell media. Figure 7-2a and 7-2b show fluorescence images of representative cells from the basal and TSA exposed group respectively. The images clearly show that the fluorescence intensity of the Fluo4 indicator decreases after exposur e to TSA. The fluorescent intens ities of thirty cells from each group were then analyzed and plotted in a histog ram in Figure 7-2c and 7-2d to illustrate the overall trend and to show the range of fluorescen t intensities that are observed. These results

PAGE 147

147 indicate that TSA exposure decreases the free Ca2+ concentration in the MDA-MB-231 cells. Thus, TSA has been shown to stimulate the MnSOD mRNA expression while decreasing the Ca2+ levels in the MDA-MB-231 cells. However, it is interesting to note th at there is a wide variation in the observed Ca2+ levels even between neighboring cells in the TSA exposed group. If it could be shown that the lower Ca2+ levels correlate with higher MnSOD expression levels and vice versa, then there would be a strong argu ment that these two analytes would be closely linked inside of the cell. To determine whether the increase in Mn SOD is related to the decrease in Ca2+, two groups of cells were utilized, one group at basal levels and another exposed to TSA. Each group of cells was then incubated with Fluo-4 fo r sixty minutes. Ten cells from each group were injected with MBs complementary for MnSOD a nd a negative control MB. Figure 7-3a shows a representative cell from each gr oup 15 minutes after injection. In the basal cell, the Fluo-4 fluorescence is at a very high intensity indicating a high Ca2+ level, while the fluorescence from the Control MB and the MnSOD MB are at a much lower intensity. This indicates that the Control MB is functioning as e xpected, since no target is presen t, the MB should not open. The MnSOD MB also remains at a constant low fluorescence level indicating the MnSOD mRNA is below our detection limits. This is not unexpe cted as the MnSOD expression is significantly downregulated in the MDA-MB231 cell line. In the TSA expos ed cells the Fluo-4 signal is significantly decrease d, indicating a decrease in the free Ca2+ concentration. The Control MB remains at the same low intensity as the basal ce lls as the MB has no targ et in either group of cells. The MnSOD MB however now shows an incr ease in its intensity as the TSA upregulates the MnSOD expression inside of the cells. Again the negative control MB validates this result since any nonspecific event that would cause the MnSOD MB to open would also affect the

PAGE 148

148 Control MB resulting in a simila r increase in fluorescence. Sin ce no increase in the Control MB is detected then the increase in the MnSOD MB must be due to a specific hybridization with its mRNA target. Figure 7-3: a) Fluorescence Imag es of Fluo-4 (green), the C ontrol MB (red), and the MnSOD MB (blue) for both the basal group and TS A exposed group of cells. The images indicate that the Fluo-4 intensity decreas es after exposure to TSA while the MnSOD MB intensity increases after exposure to TSA. The Control MB intensity does change from one group to the other. b) Th e average intensities of each probe in the two groups of cells. The av erages show a significant d ecrease in the Fluo-4 signal after TSA incubation while the MnSOD expre ssion increases. The control MB is not affected by the TSA exposure. The average signals from ten cells of each group were then determined and plotted in Figure 7-3b. This is consistent with the overa ll trends observed initially in the Western Blot analysis and the Fluo-4 averaged results wit hout injection of MBs. However no further information can be effectively determined from looking at the average values. To determine whether there were any patterns in the expression levels of the probes in the two groups, the equilibrium signals from each probe were plotte d in histograms. The histograms also allow

PAGE 149

149 observing the overall trends while utilizing the indi vidual intensities so th e distribution patterns can be studied. The histograms are shown in Figure 7-4. Figure 7-4. Histograms showing the distribution of intensities from each probe in the basal cell group (solid) and the TSA exposed cell gr oup (striped). In addition to the distributions, the overall trends in each group of cells can be seen from the histograms such as the increase in the MnSOD levels and the decreases in the Ca2+ levels inside of the cell. While the histograms show the same overall trends as observed in Figures 7-2 and 7-3, they also show the distribu tion patterns of the TSA exposed Fluo-4 and the MnSOD MB. The cells in this study exhibited a wide range of intensity levels in their Fluo-4 signal and cells with very high and low signals were chosen for further examination. If the Ca2+ and MnSOD levels were related one would expect them to exhibit a similar distribution pattern of intensities, such that cells with a high Fluo-4 intensity would ha ve low MnSOD levels and those with low Fluo-4 signals would have a higher MnSOD expression. However, the Fluo-4 signals show a varying yet level range of intensities while the MnSOD distribution appears to be Gaussian. This indicates that perhaps the two analytes are not closely linked inside of the cell. To further explore the differences between the distribution patterns, co rrelation plots of the signal levels were prepared. In the correlation plots, the signal in tensities of two different probes in the same cell are plotted against each other. If the intensity leve ls form a pattern such as a line or an exponential curve, the anal yte levels should share a similar relationship. For these plots the

PAGE 150

150 R2 value of the line indicates how well the anal yte signals correspond to that relationship. Correlation plots for Fluo-4 and the Control MB against MnSOD expre ssion are displayed in Figure 7-5. Figure 7-5. Correlation plots compar ing the levels of MnSOD to th e control and Fluo-4 levels. The degree of correlation is determined by fitting a trend line and using the R2 value. a) The plot of the MnSOD MB signal vers us the Control MB for the TSA exposed cells (blue) and the unexposed cells (red). b) The plot of the MnSOD MB intensity versus the Fluo-4 intensity for each group of cells. In Figure 7-5a, when basal intensities of the Control and MnSOD MBs are compared against each other a strong linear relationship is observed between those probes with an R2 value of higher than 0.900. Since the MnSOD levels are repressed in this can cer cell line, the basal expression of MnSOD is below the detection li mit of the MB. Therefore the fluorescence measured from the channel corresponds to the background from the MB, just like the negative control MB. Since the intensitie s are based solely on the amount of the MBs injected there is a strong correlation between th e intensities as the ra tio of the injected probes remains constant. This relationship is no longer visible in the TSA exposed cells however as now the MnSOD MB fluorescence is based primarily on the amount of MnSOD mRNA inside of the cell, not simply on the amount injected into the cell. In Figur e 7-5b, no patterns are observed for the basal and TSA exposed cells when trying to correlate the MnSOD and Ca2+ concentrations. R2 values of

PAGE 151

151 less than 0.100 are obtained for these plots indicating there is no correlation between these points. Based on the histograms and the correlation plot there does not appear to be any connection between the leve ls of MnSOD mRNA and Ca2+ inside of the TSA exposed cells. Indeed further literature re search into the known effects of TSA on the MDA-MB-231 cells reveals that TSA also upre gulates the mRNA and protei n expression of Gelsolin.252 Gelsolin is an actin binding protein that has several Ca2+ binding motifs that are utilized when the protein binds to actin. Therefore, the decrease of the Ca2+ concentration inside of the TSA exposed cells could be due to the upregulation of Gelsolin. Ho wever further work must be completed before it is known whether this is the co rrect explanation or if the phe nomena we observed is due to another process or protein altogether. Regardle ss, the technique has demonstrated its potential for aiding the understanding of biolog ical phenomena and disease states. Conclusion Advancing the understanding of biological processes and disease states requires the development of more complex forms of analysis In order to addre ss this need we have developed a methodology that allows for the dete ction and monitoring of different types of analytes simultaneously inside of the same single cell. Our technique was first illustrated using positive and negative controls along w ith a basal concentration of Ca2+. The demonstration convincingly showed that the detection of i on concentrations and mRNA expression were possible simultaneously. Also, it was done without any pretreatment of the cells and is readily applicable to almost any type of adherent cell. In addition, any gene expression system in theory could be studied by simply changing the MB sequences to different targeted mRNA. To demonstrate the utility and the potential of this technique, it wa s applied to the study of the effects of an anti-cancer agent in human breast cancer cells. In this study we sought to

PAGE 152

152 determine whether the levels of Ca2+ where related to the expr ession of MnSOD mRNA since the level of Ca2+ has been previously shown to be related to the levels of ROS in side of the cell. Despite the fact that no rela tionship was found between those levels, the study still provided valuable information into those biological processe s. No other method exists for this level of biological analysis into single cel ls with the ability to compare and investigate such different biologically important analytes. This methodology will allow further inves tigation into a wide range of medically and biologically relevant sy stems that simply is not possible with other techniques. The future of intracellular analysis in terms of single cell experiments is still largely unexplored. There are countless si gnaling pathways and other biol ogical functions that can be studied by detecting multiple analytes on the single cell level. In particular studies using other ions such as ROS and gene expression could be particularly useful to study the effects of anticancer agents on different cell types. Howeve r, one area of intracellular analysis still needs more improvement. While the MBs can function ve ry well inside of the cells for a short time, the MBs do not remain stable inside of the cel ls for longer than 30 minutes. This obviously limits the ability of the MBs fo r intracellular monitoring and requi res that different groups of cells be used to study cellular processes through chemical stimulati on. If more stable MBs were developed, this would enable monitoring for hours thus allowing the effects of chemical stimulation to be monitored in the same single cell. Therefore to address this limitation, the next chapter will focus on the integrati on of locked nucleic acid bases in to the MB in an effort to improve the stability of the MBs for intracellular analysis.

PAGE 153

153 CHAPTER 8 INTRACELLULAR APPLICATIONS OF LOCKED NUCLEIC ACID MOLECULAR BEACONS Introduction In the previous chapters, the potential for intracellular analysis using MBs was demonstrated. However, one of the problems with MBs is that they generate false-positive signals due to nuclease degr adation and protein binding253,254 leading a half life in complex biological systems like the inside of a cell of only 15 to 20 minutes.255 Also, its not only degradation that leads to the generation of fa lse positive signals as even simple binding to entities like single strand binding proteins can disr upt the stem structure of the beacon leading to the restoration of fluorescence.256 In order to minimize the effect of the in tracellular environment, nuclease-resistant backbones such as phosphorothioate and 2'-O-methyl RNA bases have been incorporated into molecular beacons. In addition, va rious groups have demonstrated neutral peptide nucleic acids (PNAs) for use in the MBs. While the backbone a nd sugar modifications have their advantages, the non-standard bases also possess different drawbacks depending on the base used. For example, those that retain the repeating ch arge continue to beha ve like natural nucleic acids in their hybridizing properties, while in troducing new problems such as the toxicity demonstrated by phosphorothioa te-containing oligonucleotides.257 Non-toxic modifications that more closely resemble natural nucleic acids can still be opened by intracellular DNAand RNAbinding proteins, many of which recognize a repe ating backbone negative charge to initiate binding to the oligonucleotide sequence. MBs with 2'-O-methyl RNA bases possess better nuclease resistance, higher affinity to DNA a nd RNA, increased specificity, and a superior ability to bind to struct ured targets compared to their DNA counterparts. However, 2'-O-methyl modified MBs still produce a fl uorescence signal non-specifically in cells, most likely as a result

PAGE 154

154 of protein binding.258 With no repeating backbone charges, PNA is resistant to nucleases and binding proteins. In addition, thei r hybridization products with R NA are thermally more stable compared with DNA-RNA and RNA-RNA hybrids. It has also been reported that using PNAMBs instead of DNA-MBs as in situ hybridization probes would be nefit cell detection under a wide range of conditions.259 However, the neutral backbone creates several new problems for using molecular inside of the cell. For instan ce, PNA has a tendency to aggregate in aqueous environments due to its neutral charge and even fo ld in ways that interfer e with the hybridization of the probe to its target.260 Also, the physical properties of PNA like solubility can change dramatically with small changes to the sequence of the strand.261 Since the environmental conditions inside of a living cell can not be optim ized for the solubility of PNA, intracellular applications of PNA are limited. Locked nucleic acids (LNA) offer one possi ble solution to this quandary by possessing a negatively charged backbone with enough m odification to disrupt protein binding.262-265 LNA is a conformationally restricted nucle ic acid analogue, that possesses a simple 2'-O, 4'-C methylene bridge locking the ribos e ring into a rigid C3'end conformation as illustrated in Figure 8-1A. LNA has many advantageous properties such as high binding affinity, ex cellent base mismatch discrimination capability, and decreased susc eptibility to nuclease digestion. When LNA hybridizes to DNA or RNA, there is a large in crease in melting temperature ranging from +3.0 to +9.6oC per LNA modification compared to normal DNA duplexes. Furthermore, LNA oligonucleotides can be synthesized using c onventional phosphoramidite chemistry allowing the automated synthesis of both fully modified L NA and mixed oligonucleotide sequences using combinations of LNA, DNA, and RNA. Other advantages of LNA stemming from its close structural resemblance to nativ e nucleic acids, include good so lubility and no toxicity in

PAGE 155

155 physiological conditions. This allows the LNA based probes to be used for intracellular applications utilizing any of the previous established delivery mechanisms. The combination of these properties give LNA a great deal of potential for intracellular applications. Figure 8-1. A) Illustration of the structure of the Locked Nucleic Acid bases. B)Illustration of a molecular beacon. Methods and Materials Equipment An ABI3400 DNA/RNA synthesi zer (Applied Biosystems, Fost er City, CA) was used for target DNA synthesis and MB probe preparati on. Probe purification was performed with a

PAGE 156

156 ProStar HPLC (Varian, Walnut Creek, CA) wh ere a C18 column (Econosil, 5u, 250.6 mm) from Alltech (Deerfield, IL) was used. UV-Vis measurements were performed with a Cary Bio300 UV spectrometer (Varian, Walnut Creek, CA ) for probe quantitation. Fluorescence measurements were performed on a Fluorolog-Ta u-3 spectrofluorometer (Jobin Yvon, Inc., Edison, NJ). Fluorescence imaging was conducted w ith a confocal microscope setup consisting of an Olympus IX-81 inverted microscope w ith an Olympus Fluoview 500 confocal scanning system and three lasers, a tunable Argon Ion la ser (458nm, 488nm, 514nm), a green HeNe laser (543nm), and a red HeNe laser (633nm) with three separate photomu ltiplier tubes (PMT) for detection. The cellular images were taken with a 60x 1.4 NA oil immersion objective. A Leiden microincubator with a TC-202A temperature contro ller (Harvard Apparatus, Holliston, MA) was used to keep the cells at 37oC during injection and monitori ng. An EXFO Burleigh PCS-6000150 micromanipulator was used for positioning the injector tip. An Eppendorf Femtojet microinjector with 0.5 m Femtotips was used to inject the molecular beacons and reference probe into the cells. All image analysis was conducted on the Fluoview 500 software, followed by processing of the data using Microsoft Excel. Molecular Beacon Synthesis Molecular Beacons possessing locked nucleic acid bases were synthesized on an Applied Biosystem 3400 DNA/RNA synthesi zer by using locked nucleic acid phorsphoramidites. The controlled-pore glass columns used for these syntheses introduced a DABCYL (4-(4(dimethylamino) phenylazo) benzoi c acid) molecule at the 3' e nds of the oligonucleotides. FAM (6-carboxyfluorescein) phosphoramidite was used to couple FAM to th e 5' ends of the sequence. The complete MB sequences were then deprotec ted in concentrated ammonia overnight at 65C and purified by high-pressure liquid chromat ography. The collection from the first HPLC

PAGE 157

157 separation was then vacuum dried, incubated in 200 l 80% acetic acid for 15 min, incubated with 200 l ethanol and vacuum dried before the second round of HPLC. Hybridization Study Unless otherwise indicated, hybridization experiments were conducted with 100 nM MBs, 500 nM complimentary target sequences in a total volume of 200 L. All experiments were conducted in 20 mM Tris-HCl (p H 7.5) buffer containing 5 mM MgCl2 and 50 mM NaCl. DNase I Sensitivity To test the nuclease sensitivity of MBs, the fluorescence of a 200 l solution containing 20 mM Tris-HCl (pH 7.5), 5 mM MgCl2, 50 mM NaCl and 100 nM MBs was measured as a function of time at room temperature. One unit of ribonuclease-free DNase I was added, and the subsequent change in fluorescence was recorded. Cell Culture MDA-MB-231 breast carcinoma cells (American Type Culture Collection, Manassas, VA) were maintained in Dulbeccos Modification of Eagles Medium (DMEM, Fisher Scientific) with 10% fetal bovine serum (Invitrogen, Carles bad, CA) and 0.5 mg/ml Gentamycin (Sigma, St. Louis, MO) at 37oC in 5% CO2/air. Cells were plated in 35 mm glass bottom culture dishes and grown to 80% confluency (MatTek Corp., Ashl and, MA) for 48 hours prior to injection. To stimulate MnSOD mRNA expressio n, cells were incubated in 1 g/ml LPS from E. coli serotype 055:B5 (Sigma, St. Louis, MO) fo r 4 hours prior to injection. Fluorescence Imaging All cellular fluorescent images were collected using the confocal microscope setup. The confocal consists of an Olym pus IX-81 automated fluorescence microscope with a Fluoview 500 confocal scanning unit. The TMR based MBs were excited at 543nm and collected at 560nm. The Alexa Fluor 647 based MBs were excited at 633nm and collected at 660nm. The Alexa

PAGE 158

158 Fluor 488 reference probe was excited at 488nm and its emission was detected at 520nm. Images were taken after a five to ten second pe riod during which the inst rument was focused to yield the highest intensity from the fluorescen ce channels. The images were assigned color representations for clarity a nd are not indicative of the actual emission wavelengths. Results and Discussion While the LNA MB has demonstrated impressive performances in buffer solutions the true measure of the intracellular potential of LNA MB can only be evaluated in side of living cells. As a first test of the LNA MB its intracellula r background and signal enhancement were tested inside living MDA-MB-231 cells. In these expe riments, the LNA MB was compared directly against a normal DNA MB counterpart. Both be acons possessed the Control MB sequence with a TMR/BHQ2 fluorophore/quencher pair. The beacons were in jected into the cell at 1 M concentrations with and wit hout a pre-incubation with 10 M target DNA. The microscope settings were kept constant to allow a fa ir comparison between the two probes. A 1 M AF488 reference probe was injected along with the b eacons to allow ratiometric analysis to be performed. This minimizes any experimental fluctu ations that might have influenced the results of the experiment. A representative fluorescen ce image for each sample is shown in Figure 8-2. When comparing the two beacons, the DNA MB has a higher fluorescence intensity after the incubation with target DNA although the signal from the LNA counterpart is still high. The most pronounced difference between the two probes however is in the unhybridized state. The LNA MB has a much lower fluorescence background than the DNA MB. This is mostly likely due to decreased thermodynamic fluctuations in stem since the LNA has a much higher binding affinity. Degradation of the probe is unlikely as a suspect as the images were taken less than a minute after injection of the probe mixtures.

PAGE 159

159 Figure 8-2. A) Fluorescence image of cell inject ed with and LNA MB (red) and the reference probe (orange) without target. B) Fluor escence image of cell injected with and DNA MB (red) and the reference probe (orange) without target. C) Fluorescence image of cell injected with and LNA MB (red) and th e reference probe (orange) with target. D) Fluorescence image of cell injected with and DNA MB (red) and the reference probe (orange) with target. The probe mixtures were injected in a total of three cells for each sample. The signals obtained for each MB were then divided by the reference probe intensity, averaged, and then plotted in Figure 8-3. The plot illustrates that the low background and also the superior signal enhancement of the LNA MB. Thus the LNA MB seems to have outperformed the DNA MB based on the criteria of the lower fluorescen ce background and the better signal enhancement. Figure 8-3. Comparison of the LNA and DNA MBs with and without target inside of living cells.

PAGE 160

160 The next test for the LNA MB was to determin e whether it had better stability inside of the cell compared to the DNA MB. In these experi ments, one of the major disadvantages of the MB design actually becomes beneficial. A probl em with the MBs is that when the MB is degraded, the fluorescence signal is restored just like when the MB hybridizes to its target. Normally, this is a problem as there is no way to distinguish between target hybridization and degradation. In these experiments however, MBs w ith no target inside of the cells can be used and the fluorescence intensity can be measured to monitor the degradati on of the MB. As no target is present, any increase in the fluorescen ce signal must be due to the degradation of the probe. For these experiments, 1 M of the LNA or DNA MB was injected into the cell along with 1 M of the reference probe. The signals from each cell were monitored for one hour with the results from three cells from each probe being plotted in Figure 8-4. The results show that after 30 minutes the fluorescence fr om the DNA MBs increases despit e a lack of target. This indicates that the probe is being de graded inside of the cell. This is in stark contrast to the LNA MB which in addition to exhibiting a lower back ground signal, remains at the same constant low level throughout the monitoring period. Ther efore, the LNA MB seems to possess better stability in the cellular environment than normal DNA MBs. Figure 8-4. Plots of the monitori ng of the ratiometric values of the probes inside of cells depicting the degrada tion of the probes.

PAGE 161

161 While the LNA MB has many advantages, thro ugh the course of these experiments one glaring weakness of the LNA MB was made clear. This is that slow hybrid ization kinetics of the MB prevent it from being used for most applic ations. In order for the LNA MB to produce a comparable hybridized signal to the DNA MB it requi red an incubation time of 24 hours. That length of time makes the LNA MB unsuitable for most applications especially intracellular experiments where real time measurements and the actual lifetime of the cell are critical parameters. LNA MB Optimization In order to render the LNA MB useful for widespread applicati ons, its hybridization kinetics need to be drastically improved. Howeve r, the beneficial aspects of the LNA MB also need to be conserved like its biostability. Wh ile its possible many strategies could accomplish this aim the following research focused on lowe ring the energy barrier of the opening of the stem. In theory, in order for the MB to open and generate a signal, two processes must occur; the hybridization of the loop to the target sequence and the dehybridization or melting of the stem. As the LNA has a higher affinity for DNA than the DNA has for itself the problem is not likely to be the hybridization of the loop. The higher binding affinity of the LNA however could cause the dehybridization or melting of the stem region to be much slower as the thermodynamic energy barrier of the process may be higher. In pursuing this line of reasoning, the major modification to the LNA MB were tested is th e incorporation DNA bases along with the LNA in the MB sequence. This should partially destabilize the stem allowing for faster hybridization kinetics without sacrificing the im proved biostability of the probes. Solution Experiments The first step in determining the optimal strategy for the beacon composition was using different ratios of DNA to LNA and determini ng in solution whether th e change in composition

PAGE 162

162 had a sufficient effect on the hybr idization rate. In this expe riment, the different molecular beacons were monitored in their closed state brie fly to target addition. After target was added the increase in fluorescence was monitored for each beacon. All signals were normalized to the closed molecular beacon signal intensity to dir ectly compare the hybridization rates for each in order to show the data on the same scale so th ey can be accurately compared. These are plotted in Figure 8-5. 02468101214 0 2 4 6 8 10 12 14 16 18 Fluorescence intensity changeTime (min) DNA-MB LNA-MB-E0 LNA-MB-E1 LNA-MB-E3 Figure 8-5. Hybridization rates of the different molecular beacon compositions. The data shows the full DNA MB has the fastest hybridizat ion rate, while each hybridization rate slows as more LNA is substituted into the sequence. The full DNA molecular beacon clearly has the fastest hybridizat ion rate reaching equilibrium after 5 to 6 minutes. The LNA-MB -E3 that possesses a DNA to LNA ratio of 3:1

PAGE 163

163 had the next fastest hybridization rate reaching e quilibrium after 12 to 13 minutes. The full LNA sequence designated LNA-MB-E1 and the L NA-MB-E1 sequence containing 50% LNA had slower hybridization rates both taking longer than the 15 minutes in th e plot to reach equilibrium. The LNA-MB-E1 required approximately one hour to reach equilibrium while the LNA-MB-E0 required over 24 hours to reach equilibrium. While this MB appears to reach a decent level of fluorescence quite quickly based on the figure, it is important to unders tand that this is the signal enhancement being monitored which is a ra tio between the hybridized and unhybridized fluorescence intensities. Therefore despite reac hing a signal enhancement of roughly half of the DNA MB, the actual intensity of th e beacon at this point is at a very in tensity due to the extremely low background of the LNA MB. Ho wever, both the LNA-MBE1 and LNA-MBE3 had higher intensities making them adequate for intracellular analysis based on their hybridization rates. The next set of experiments was to determ ine whether the LNA/DNA beacons could still withstand nuclease degradation. This is an importa nt determination as th e primary rationale for the modified MBs is to make them more stable in a biological environment. If the optimized probe can no longer resist DNAse cleavage then further experiments with that sequence would prove fruitless as it would have the same li mitations as DNA based probes along with slower hybridization rates. In these experiments each probe was in cubated with DNase I and the fluorescence intensity was monitored. It is important to note that the degradation of the MB by a nuclease causes the cleaving of the bases that make up the probe. This results in the spatial separation of the fluorophore and quencher a nd restoring the fluores cence emission of the fluorophore. Thus degradation of the probes can be monitored just like the hybridization of the probes.

PAGE 164

164 -1001020304050100110120130 0.0 0.2 0.4 0.6 0.8 1.0 Relative Fluorescence Intensity ChangeTime (min) LNA-MB-E0 LNA-MB-E1 LNA-MB-E2 LNA-MB-E3 LNA-MB-E4 LNA-MB-E5 DNA-MB Figure 8-6. Plot showing the respon se of the different MB compos itions to the addition of DNase I that causes the degradation of the probes. The response for each probe to DNase I is illustrated in Figure 8-6. Of all the compositions tested only two were sufficiently stab le in the presence of DNase I, the probe with all LNA and the probe with 50% LNA. None of the other compositions exhibited any resistance to the LNA indicating that the only the LNA-MB-E 1 had the potential for intracellular analysis based on biostability and hybridization kineti cs. Therefore, the LNA MB-E1 sequence was further evaluated inside of the cell. Neuron Imaging To thoroughly evaluate the performance of the LNA-MB-E1 sequence, it was tested inside of neurons from Aplysia Californica Aplysia Californica is a type of marine sea slug with a comparatively simple nervous system that is often used as a model for studies trying to elucidate the foundations of learni ng and memory. In order to trul y evaluate the po tential of the LNA-MB-E1 composition however, different sequen ces were required. For these experiments

PAGE 165

165 the same LNA percentage and composition as th e LNA-MB-E1 were applied to sequences that were more suitable for use in the Aplysia Californica neurons. One sequence was complementary for B-tubulin mRNA, a highly expr essed gene in all of the neurons while the other sequence did not have a complement in the known Aplysia genome. Each sequence was injected inside a single Aplysia neuron and mon itored for six hours. Representative images for one cell with each probe is shown in Figure 8-7. Figure 8-7. Representative images for neurons injected with the ne gative control sequence (left) and the B-tubulin sequence (ri ght) directly after the injection (top) and after six hours (bottom). The images indicate that the negative cont rol sequence remained closed throughout the monitoring period with no increase in the fluores cence intensity of the probe. The cell with the B-tubulin sequence however displa yed a significant increase in in tensity. This is due to high expression of the B-tubulin mRNA in side of the Aplysia neuron. This experiment was repeated once more for each probe. Additional neurons were not directly available which prevented further experiments from being conducted. The fluorescence intensities versus time for each neuron injected is plotted in Figure 8-8. The plot reinforces the images in that each negative

PAGE 166

166 control cell remains at a level low intensity fo r the duration of the monitoring. However, the signal from the B-tubulin probe increase s gradually over the monitoring period. Figure 8-8. Plots showing the fluorescence intensity versus time for each neuron injected with the alternating LNA/DNA molecular beacon. While successful the experiments indicate a prob lem with sequences used. The problem is that the hybridization kinetics are still not fast enough as the probe requires approximately four hours to reach equilibrium. This indicates that additional steps will be required before the LNA MBs will be suitable for routine intr acellular analysis. Most likely this will involve further steps to destabilize the stem such as using fewer bases or more DNA bases in the stem while conserving the alternating or full LNA modi fication of the loop region of the probe.

PAGE 167

167 LIST OF REFERENCES 1. Faderl, S.; Kantarjian, H. M.; Talpaz, M.; Estrov, Z. Blood 1998 91 (11) 3995-4019. 2. Grabstald, H. CA Cancer J Clin 1965 15 134-138. 3. Paredes-Aguilera, R.; Romero-Guzman, L.; Lopez-Santiago, N.; BurbanoCeron, L.; Camacho-Del Monte, O.; Nieto-Martinez S. Amer. J. Hematol. 2001 68 69-74. 4. Ghossein, R.A. Bhattacharya, S. European Journal of Cancer 2000 36 1681-1694. 5. Iinuma, H.; Okimaga, K.; Adachi, M.; Suda, K.; Sekine, T.; Sakagawa, K.; Baba, Y.; Tamura, J.; Kumagai, H.; and Ida, A. Int. J. Cancer 2000 89 337-344. 6. Zhao, X.; Tapec-Dytioco, R.; Tan, W. J. Am. Chem. Soc. 2003 125 11474-11475. 7. Simard, J.; Briggs, C.; Boal A. K.; Rotello, V. M. Chem. Commun. 2000 1943-1944. 8. Schvedova, A.; Kisin, E.; Mercer, R.; Murray, A.; Johnson, V.; Potapovich, A.; Tyurina, Y.; Gorelik, O.; Arepalli, S.; Schwegler-Berry, D. Am. J. Physiol. Lung. Cell Mol. Physiol. 2005 289 L698-L708. 9. Stepanov, A.L.; Krenn, J.R; Difbacher, H.; H ohenau, A.; Steinberger, B.; Leitner, A.; Aussenegg, F.R. Optics Letters 2005 30 (12) 1524-1526. 10. Tan, M.Q.; Wang, G.L.; Hai, X.D.; Ye, Z.Q.; Yuan, J.L.; J. Mater. Chem. 2004 14 2896-2901. 11. Zhu, M.Q.; Chang, E.; Sun, J.; Drezek, R.A.; J. Mater. Chem. 2007 17 800-805. 12. Takafuji, M.; Ide, S.; Ihara, H.; Xu, Z.; Chem. Mater. 2004 16 1977-1981. 13. Bohren, C.F.; Huffman, D.R. Absorption and Scattering of Light by Small Particles ; Wiley: New York. 1983. 14. Kerker, M.; J. Colloid Interface Sci. 1985 105 297-314. 15. Moskovits, M.; Suh, J.S.; J. Phys. Chem. 1984 88 5526-5530. 16. Faraday, M. Philos. Trans. 1857 147 145-181. 17. Savage, G. Glass and Glassware ; Octopus Book: London, 1975. 18. Kunckels, J. Nuetliche Observationes oder Anmerkungen von Auro und Argento Potabili; Schutzens : Hamburg, 1676. 19. Turkevitch, J.; Stevenson, P. C.; Hillier, J. Discuss. Faraday Soc. 1951 11 55-75.

PAGE 168

168 20. Grabar, K.C.; Freeman, R.G. ; Hommer, M.B.; Natan, M..J. Anal. Chem. 1995 67 735 743. 21. Grabar, K.C.; Brown, K.R.; Keating, C.D.; Stranick, S.J.; Tang, S.L.; Natan, M.J. Anal. Chem. 1997 69 471. 22. Franzen, S.; Folmer, J.C.W. ; Glomm, W.R.; ONeal, R. J. Phys. Chem. A 2002 106 6533. 23. Simard, J.; Briggs, C.; Boal, A.K.; Rotello, V.M. Chem. Commun. 2000 1943. 24. Storhoff, J.J.; Mirkin, C.A. Chem. Rev. 1999 99 1849. 25. Storhoff, J.J.; Elghanian, R.; Mucic, R.C.; Mirkin, C.A.; Letsinger, R.L. J. Am. Chem. Soc. 1998 120 1959. 26. Storhoff, J.J.; Lazarides, A.A.; Mucic, R.C.; Mirkin, C.A.; Letsinger, R.L.; Schatz, G.C. J. Am. Chem. Soc. 2000 122 4640-4650. 27. Wang, G.L.; Zhang, J.; Murray, R.W. Anal. Chem. 2002 74 4320-4327. 28. Park, S.J.; Taton, T.A.; Mirkin, C.A. Science 2002 295 1503-1506. 29. Taton, T.A.; Mirkin, C.A.; Letsinger, R.L. Science 2000 289 1757-1760. 30. Demers, L.M.; Mirkin, C.A.; Mucic, R.C., Reynolds, R.A.; Letsinger, R.L.; Elghanian, R.; Viswanadham, G. Anal. Chem. 2000 72 5535. 31. Elghanian, R.; Storhoff, J.J.; Mucic, R.C.; Letsinger, R.L.; Mirkin, C.A. Science 1997 277 1078. 32. Mirkin, C.A. Inorg. Chem. 2000 39 2258-2272. 33. Gomez, S.; Philippot, K.; Colliere, V.; Chaudret, B.; Senocq, F.; Lecante, P. Chem. Commun. 2000 1945. 34. Hayat, M.A. Colloidal Gold: Principles Methods and Applications ; New York: Academic Press; 1989 Vol. 1. 35. Tkachenko, A.G.; Xie, H., Coleman, D.; Glo mm, W.; Ryan, J.; Anderson, M.F.; Franzen, S.; Feldheim, D.L. J. Am. Chem. Soc. 2003 125 4700. 36. Xie, H.; Tkachenko, A.G.; Glomm, W.R.; Ryan, J.A.; Brennaman, M.K.; Papanikolas, J.M.; Franzen, S.; Feldheim, D.L. Anal. Chem. 2003 75 5797.

PAGE 169

169 37. Templeton, A.C.; Chen, S.W.; Gross, S.M.; Murray, R.W. Langmuir 1999 15 66. 38. Faulk, W.P.; Taylor, G.M. Immunochemistry 1971 8 1081. 39. Wang, B.L., Scopsi, L., Nielsen, M.H., and Larsson, L.I. Histochemistry 1985 83 109 115. 40. Mentlein, R.; Buchholz, C.; Krisch, B. Cell Tissue Res. 1990 262 431-443. 41. Krisch, B.; Buchholz, C.; Mentlein, R.; Turzynski, A. Cell Tissue Res. 1993 272 523 531. 42. Krisch, B.; Feindt, J.; Mentlein, R. J. Histochem. Cytochem. 1998 46 1233. 43. Niemeyer, C.M. and Mirkin, C.A. Nanobiotechnology: Concep ts, Applications and Perspectives Weinheim: Wiley-VCH, Verlag, 2004. 44. Gribnau, T.C.J., Leuvering, J.H.W., and Vanhell, H. J. Chromatogr. 1986 376 175. 45. Brust, M.; Kiely, C.J. Colloid Surf. A Physicochem. Eng Asp. 2002 202 175. 46. Kreibig, U.; Genzel, L. Surf. Sci. 1985 156 678. 47. Mirkin, C.A.; Letsinger, R.L.; Mucic, R.C.; Storhoff, J.J. Nature 1996 382 607. 48. Reynolds, R.A.; Mirkin, C.A.; Letsinger, R.L. J. Am. Chem. Soc 2000 122 3795. 49. Chikazumi, S.; Taketomi, S.; Ukita, M.; Mizukami, M.; Miyajima, H.; Setogawa, M.; Kurihara, Y. J. Magn. Magn. Mater. 1987 65 245. 50. Lu, A.H.; Schmidt, W.; Matoussevitch, N.; Bonnermann, H.; Spliethoff, B.; Tesche, B.; Bill, E.; Kiefer, W.; Schuth F. Angew. Chem. 2004 116 4403. 51. Tsang, S.C.; Caps, V.; Paraskevas, I.; Chadwick, D.; Thompsett, D. Angew. Chem. 2004 116 5763. 52. Gupta, A.K.; Gupta, M. Biomaterials 2005 26 3995. 53. Mornet, S.; Vasseur, S.; Grasset, F.; Verveka, P.; Goglio, G.; Demourgues, A.;Portier, J.; Pollert, E.; Duguet, E. Prog. Solid State Chem. 2006 34 237. 54. Li, Z.; Wei, L.; Gao, M.Y.; Lei, H. Adv. Mater. 2005 17 1001. 55. Hyeon, T. Chem. Commun. 2003 927. 56. Elliott, D.W.; Zhang, W.X. Environ. Sci. Technol. 2001 35 4922.

PAGE 170

170 57. Takafuji, M.; Ide, S.; Ihara, H.; Xu, Z. Chem. Mater. 2004 16 1977. 58. Neveu, S.; Bee, A.; Robineau, M.; Talbot, D. J. Colloid Interface Sci. 2002 255 293. 59. Park, S.J.; Kim, S.; Lee, S.; Khim, Z.; Char, K.; Hyeon, T.; J. Am. Chem. Soc 2000 122 8581. 60. Puntes, V. F. ; Krishan, K. M. ; Alivisatos, A. P. Science 2001 291 2115. 61. Chen, Q.; Rondinone, A.J.; Chakoumakos, B.C.; Zhang, Z.J. Magn. Magn. Mater. 1999 194 1. 62. Park, J.; An, K.; Hwang, Y.; Park, J.G.; N oh, H.J.; Kim, J.Y.; Park, J.H., Hwang, N.M.; Hyeon, T. Nat. Mater. 2004 3 891. 63. Bonnemann, H.; Brijoux, W.; Brinkmann, R.; Ma toussevitch, N.; Waldoefner, N.; Palina, N.; Modrow, H. Inorg. Chim. Acta. 2003 350 617. 64. Sun, S.; Murray, C.B.; Weller, D.; Folks, L.; Moser, A. Science 2000 287 1989. 65. Shevchenko, E. V.; Talapin, D.V.; Rogach, A.L.; Kornowski, A.; Haase, M.; Weller, H. J. Am. Chem. Soc. 2002 124 11480. 66. Grasset, F.; Labhsetwar, N.; Li, D.; Park, D. C.; Saito, N.; Haneda, H.; Cador, O.; Roisnel, T.; Mornet, S.; Duguet, E.; Portier, J.; Etourneau, J. Langmuir 2002, 18 8209. 67. Bee, A.; Massart, R.; Neveu, S. J. Magn. Magn. Mater. 1995 149 6. 68. Lee, J.; Isobe, T.; Senna, M. Colloids Surf. A 1996 109 121. 69. Cornell, R.M.; Schindler, P.W. Colloid Polym. Sci. 1980 258 1171. 70. Willis, A.L.; Turro, N.J.; OBrien, S. Chem. Mater. 2005 17 5970. 71. Murray, C.B.; Norris, D.J.; Bawendi, M.G. J. Am. Chem. Soc. 1993 115 8706. 72. Peng, X.; Wickman, J.; Alivisatos, P. J. Am. Chem. Soc. 1998 120 5343. 73. OBrien, S.; Brus, L.; Murray, C.B. J. Am. Chem. Soc. 2001 123 12085. 74. Farrell, D.; Majetich, S.A.; Wilcoxon, J.P. J. Phys. Chem. B 2003 107 11022. 75. Jana, R.J.; Chen, Y.; Peng, X.; Chem. Mater. 2004 16 3931. 76. Samia, A.C.S.; Hyzer, K.; Schlueter, J.A.; Qin, C.J.; Jiang, J.S.; Bader, S.D.; Lin, X.M. J. Am. Chem. Soc. 2005 127 4126.

PAGE 171

171 77. Li, Y.; Afzaal, P.; OBrien, J. J. Mater. Chem. 2006 16 2175. 78. Yamada, Y.; Suzuki, T.; Abarra, E.N. IEEE Trans. Magn. 1998 34 343. 79. Tegus, O.; Bruck, E.; Buschow, K.H.J.; de Boer, F.R. Nature 2002 415 150. 80. Song, Q.; Zhang, Z.J. J. Am. Chem. Soc. 2004 126 6164. 81. Langevin, D. Annu. Rev. Phys. Chem. 1992 43 341. 82. Carpenter, E.E.; Seip, C.T.; OConnor, C.J. J. Appl. Phys. 1999 85 5184. 83. Liu, C.; Zou, B.; Rondinone, A.J.; Zhang, Z.J. J. Phys. Chem. B 2000 104 1141. 84. Wang, X.; Zhuang, J.; Peng, Q.; Li, Y. Nature 2005 437 121. 85. Deng, H.; Li, X.; Peng, Q.; Wang, X.; Chen, J.; Li, Y. Angew. Chem. Int. Ed. 2005 44 2782. 86. Euliss, L.E.; Grancharov, S.G.; OBrien, S.; Deming, T.J.; Stucky G.D.; Murray, C.B.; Held, G.A. Nano. Lett. 2003 3 1489. 87. Kobayashi, Y.; Horie, M.; Konno, M.; R odriguez-Gonzalez, B.; Liz-Marzan, L.M. J. Phys. Chem. B 2003 107 7420. 88. Liu, Q.; Xu, Z.; Finch, J.A.; Egerton, R. Chem. Mater. 1998 10 3936. 89. Jiles, D.C. Acta Mater. 2003 51 5907. 90. Salata, O.V. J. Nanobiotechnol. 2004 2 3. 91. Barratt, G. Cell. Mol. Life Sci. 2003 60 21. 92. Neuberger, T.; Schopf, B.; Hofmann, H. ; Hofmann, M.; von Rechenberg, B. J. Magn. Magn. Mater. 2005 293 483. 93. Berry, C.C. J. Mater. Chem. 2005 15 543. 94. Tartaj, P.; del Puerto Morales, M.; Veinte millas-Verdaguer, S.; Gonzalez-Carreno, T.; Serna, C.J. J. Phys. D 2003 36 R182. 95. Gu, H.; Xu, K.; XU, C.; Xu, B. Chem. Commun. 2006 941. 96. Kalambar, V.S.; Han, B.; Hammer, B.E.; Shield, T.W.; Bischof, J.C. Nanotechnology 2005 16 1221.

PAGE 172

172 97. Huber, D.L. Small 2005 1 482. 98. Batlle, X.; Labarta, A. J. Phys. D 2002 35 R15. 99. C. M. Sorensen, Magnetism in Nanoscale Materials in Chemistry Wiley-Interscience Publication, New York, 2001. 100. Iwaki T.; Kakihara, Y.; Toda ,T.; Abdullah, M.; Okuyama,K. J Appl. Phys. 2003 94 6807. 101. Henglein, A. Chem. Rev. 1989 89 1861-1873. 102. Schmid, G. Chem. Rev 1992 92 1709-1727. 103. Alivisatos, A.P. Science 1996 271 933-937. 104. Nirmal, M.; Brus, L.E. Acc Chem Res 1999 32 407-414. 105. Murray, C.B.; Norris, D.J.; Bawendi, M.G. J. Am. Chem. Soc. 1993 115 8706-8715. 106. Han, M.Y.; Gao, X.; Su, J.Z.; Nie, S.M. Nat. Biotechnol. 2001 19 631-635. 107. Dabbousi, B.O.; Rodriguez-Viejo, J.; Mikulec, F. V.; Heine, J.R.; Mattoussi, H.; Ober, R.; Jensen, K.F.; Bawendi, M.G. J. Phys. Chem. B 1997 101 9463-9475. 108. Peng, X.G.; Schlamp, M.C.; Kadavanich, A.V.; Alivisatos, A.P. J. Am. Chem. Soc. 1997 119 7019-7029. 109. Hines, M.A.; Guyot-Sionnest, P. J. Phys. Chem. B 1996 100 468-471. 110. Peng, X.G.; Wickham, J.; Alivisatos, A.P.; J. Am. Chem. Soc. 1998 120 5343-5344. 111. Desmet, Y.; Deriemaeker, L.; Parloo, E.; Finsy, R. Langmuir 1999 15 2327-2332. 112. Dahan, M.; Laurence, T.; Schumacher, A.; Chem la, D.S.; Alivisatos, A.P.; Sauer, M.; Weiss S. Biophys J. 2000 78 2270. 113. Dahan, M.; Laurence, T.; Pinaud, F.; Chemla, D.S.; Alivisatos, A.P.; Sauer, M.; Weiss, S. Opt. Lett. 2001 26 825-827. 114. Hines, M.A.; Guyot-Sionnest, P. J. Phys. Chem. B 1998 102 3655-3657. 115. Guzelian, A.A.; Katari, J.E.B.; Kadavanic h, A.V.; Banin, U.; Hamad, K.; Juban, E.; Alivisatos, A.P.; Wolters, R.H.; Arnold, C.C.; Heath, J.R. J. Phys. Chem. 1996 100 7212-7219.

PAGE 173

173 116. Prieto, J.A.; Armelles, G.; Groenen, J.; Cales, R. Appl. Phys. Lett. 1999 74 99-101. 117. Micic, O.I.; Cheong, H.M.; Fu, H.; Zunger, A.; Sprague, J.R.; Mascarenhas, A.; Nozik, A.J. J. Phys. Chem. B 1997 101 4904-4912. 118. Katari, J.E.B.; Colvin, V.L.; Alivisatos, A.P. J. Phys. Chem. 1994 98 4109-4117. 119. ONeil, M.; Marohn, J.; McLendon, G. J. Phys. Chem. 1990 94 4356-4363. 120. Alivisatos, A.P.; Johnsson, K.P.; Peng, X.; Wils on, T.E.; Loweth, C.J.; Bruchez, M.P. Jr.; Schultz, P.G.; Nature 1996 382 609-611. 121. Peng, X.G.; Wickham, J.; Alivisatos, A.P. J. Am. Chem. Soc. 1998 120 5343-5344. 122. Winter, J.O.; Liu, T.Y.; Korgel, B.A.; Schmidt, C.E. Adv. Mater. 2001 13 1673. 123. Jaiswal, J.K.; Mattoussi, H.; Mauro, J.M.; Simon, S.M. Nature Biotechnol 2003 21 47. 124. Lidke D.S.; Nagy, P.; Heintzmann, R.; Arndt-Jovin, D.J.; Post, J.N.; Grecco, H.E.; JaresErijman, E.A.; Jovin, T.M. Nature Biotechnol 2004 22 198. 125. Derfus, A.M.; Chan, W.C.W.; Bhatia, S.N. Nano Lett. 2004 4 11. 126. Rosenthal, S.J.; Tomlinson, I.; Adkins, E.M.; Schroeter, S.; Adams, S.; Swafford, L.; McBride, J.; Wang, Y.; DeFe lice, L.J.; Blakely, R.D. J. Am. Chem. Soc. 2002 124 45864594. 127. Winter, J.O.; Liu, T.Y.; Korgel, B.A.; Schmidt, C.E. Adv. Mater. 2001 13 1673. 128. Pinaud, F.; King, D.; Moore, H.P.; Weiss, S. J. Am. Chem. Soc. 2004 126 6115. 129. Larson D.; Zipfel, W.R.; Williams, R.M.; Cl ark, S.W.; Bruchez, M.P.; Wise, F.W.; Webb, W.W. Science 2003 300 1434-1436. 130. Kim, S.; Lim, Y.T.; Soltesz, E.G.; De Grand, A.M.; Lee, J.; Nakayama, A.; Parker, J.A.; Mihaljevic, T.; Laurence, R.G.; Dor, D.M. ; Cohn, L.H.; Bawendi, M.G.; Frangioni, J.V. Nature Biotechnol. 2004 22 93-97. 131. Xiao, Y.; Barker, P.E. Nucleic Acids Res. 2004 32 28E. 132. Hoshino, A.; Hanaki, K.; Suzuki, K.; Yamamoto, K. Biochem. Biophys. Res. Commun. 2004 314 46. 133. Derfus, A. M.; Chan, W. C. W.; Bhatia, S. N. Nano Lett. 2004 4 (1), 11. 134. van Blaaderen, A.; Imhof, A.; Hage, W.; Vrij, A. Langmuir 1992 8 1514.

PAGE 174

174 135. Yamauchi, H.; Ishikawa, T.; Kondo, S. J. Colloids Surf. 1989 37 71. 136. Schmidt, J.; Guesdon C.; Schomcker R. J. Nanoparticle Res. 1999 1 267. 137. Zhao, X. J.; Bagwe, R. P.; Tan, W. H. Adv. Mater. 2004 16 173. 138. van Blaaderen, A.; Vrij, A. Langmuir 1992 8 2921. 139. Verhaegh, A. M. N.; van Blaaderen, A. Langmuir 1994 10 1427. 140. Deng, G.; Markowitz, M. A.; Kust, P. R.; Gaber B. P. Mater. Sci. Eng. 2000 11 165 172. 141. Santra, S.; Yang, H.; Stanley, J. T.; Holloway, P. H.; Tan, W. H.; Moudgil, B. M.; Mericle, R. A. Chem. Commun. 2004 24 2810. 142. Santra, S.; Xu, J. S.; Wang, K. M.; Tan, W. H. J. Nanosci. Nanotechnol. 2004 4 590 599. 143. Ohno, T.; Takeshi, M.; Hisao, S.; Masayuki, F. Adv. Powder Tech. 2006 17 (2) 167-179. 144. Roy, I.; Ohulchanskyy, T.Y.; Bharali, D.J.; P udavar, H.E.; Mistretta, R.A.; Kaur, N.; Prasad, P.N. Proc. Nat. Aca. Sci. 2005 102 (2), 279-284. 145. Hermanson, G. T. Bioconjugate Techniques Academic Press: San Diego, CA, 1996. 146. Gerion, D.; Pinard, F.; William, S. C.; Parak, W. J.; Zanchet, D.; Weiss, S.; Alivisatos, A P. J. Phys. Chem. B 2001 105 8861. 147. Tapec, R.; Zhao, X.J.; Tan, W. J Nanosci Nanotechnol. 2002 2 (4), 405-409. 148. Giri, S.; Trewyn, B. G.; Stellmaker, M. P.; Lin, V. S. Y.; Angew. Chem. Int. 2005 44 5038. 149. Brehm-Stecher, B.; Johnson, E.; Microbiol Mol Biol Rev 2004 68 (3), 538-59. 150. Williams, N. Methods Cell Biol 2000 62 449-53. 151. Kurien, B.; Scofield, R. Methods 2006 38 (4), 283-93. 152. Scanziani, E. Methods Mol Biol 2001 104 133-40. 153. Gold, L. J. Biol. Chem. 1995 270 13581-13584. 154. Famulok, M.; Mayer, G. Curr. Top. in Microbiol. and Immun. 1999 243 ,123-136.

PAGE 175

175 155. Brody, E. N.; Gold, L. Rev. Mol. Biotechnol. 2000 74 5-13. 156. Conrad, R.; Keranen, L.M.; Ellington, A.D.; Newton, A.C. J Biol Chem. 1994 269 32051-32054. 157. Hirao, I.; Spingola, M.; Peabody, D.; Ellington, A.D. Mol Divers 1998 4 75-89. 158. Macaya, R. F.; Schultze, P.; Smith, F. W.; Roe, J. A.; Feigon, J. Proc. Nat. Acad. Sci. 1993 90 3745-3749. 159. Kelly, J. A.; Feigon, J.; Yeates, T.O. J. Mol. Biol. 1996 256 417-422. 160. Gold, L.; Polisky, B.; Uhlenbeck, O.; Yarus, M. Annu. Rev. Biochem. 1995 64 763797. 161. White, R.R.; Sullenger, B.A.; Rusconi, C.P. J. Clin. Invest. 2000 106 929-934. 162. Bock, L.C.; Griffin, L.C.; Latham, J.A.; Vermaas, E.H.; Toole, J.J. Nature 1992 355 564-566. 163. Hicke, B.J.; Marion, C.; Chang, Y.F.; Gould, T. ; Lynott, C.K.; Parma, D.; Schmidt, P.G.; Warren, S. J. Biol. Chem. 2001 276 48644-48654. 164. Daniels, D.A.; Chen, H.; Hicke, B.J.; Seiderek, K.M.; Gold, L. Proc. Nat. Acad. Sci. 2003 100 (26), 15416-15421. 165. Wang, C.; Zhang, M.; Yang, G.; Zhang, D.; Ding, H.; Wang, H.; Fan, M.; Shen, B.; Shao, N. J. Biotechnol. 2003 102 15-22. 166. Morris, K.N.; Jensen, K.B.; Julin, C.M.; Weil, M.; Gold, L. Proc. Nat. Acad. Sci. 1998 95 2902-2907. 167. Homann, M.; Goringer, H.U. Nucleic Acids Res. 1999 27 2006-2014. 168. Blank, M.; Weinschenk, T.; Priemer, M.; Schluesener, H. J. Biol. Chem. 2001 276 16464-16468. 169. Li, J.; Fang,, X. H.; Tan, W. Biochem. Biophys. Res. Commun. 2002 292 31-40. 170. Knox, K.K.; Brewer, J.H.; Henry, J.M.; Harrington D.J.; Carrigan, D.R. Clin. Infectious Diseases 2000 31 (4), 894-903. 171. Cerchia, L.; Hamm, J.; Libri, D.; Tavitian, B.; Franciscis, V. FEBS Letters 2002 528 1216. 172. Hermann, T.; Patel, D.J. Science 2000 287 820-825.

PAGE 176

176 173. Harding, J. D.; MacDonald, R. J.; Przybyla, A. E.; Chirgwin, J. M.; Pictet, R. L.; Rutter, W. J. J. Biol. Chem. 1977 252 7391-7397. 174. Hallick, R. B.; Chelm, B. K.; Gray, P. W.; Orozco, E. M., Jr. Nucl. Acids Res. 1977 4 3055-3064. 175. Velculescu, V. E.; Zhang, L.; Vogelstein, B.; Kinzler, K. W. Science 1995 270 484-487 176. Lockhart, D. J.; Dong, H.; Byrne, M. C.; Follettie, M. T.; Gallo, M. V.; Chee, M. S.; Mittmann, M.; Wang, C.; Kobayashi, M.; Norton, H.; Brown, E. L. Nature Biotechnology 1996 14 1675-1680. 177. Schena, M.; Shalon, D.; Davis, R. W.; Brown, P. O. Science 1995 270 467-470. 178. Drake, T. J.; Zhao, X. J.; Tan, W. Nanobiotechnology 2004 444-457. 179. Eberwine, J.; Belt, B.; Kacharmina, J. E.; Miyashiro, K. Neurochemical Research 2002 27 1065-1077. 180. Eberwine, J. H.; Spencer, C.; Newell, D.; Hoffman, A. R. Microscopy Research and Technique 1993 25 19-28. 181. Spencer, C. M.; Eberwine, J. DNA and Cell Biology 1999 18 39-49. 182. Dirks, R. W.; Molenaar, C.; Tanke, H. J. Histochemistry and Cell Biology 2001 115 311 183. Drake, T.J.; Medley, C.D.; Rogers R.J.; Tan W. ChemBioChem 2005 6 2041-2047. 184. Perlette, J.; Tan,W. Anal.Chem. 2001 73 5544-5550. 185. Bratu,D.P.; Cha,B.J.; Mhlanga,M.M.; Kramer,F.R.; Tyagi,S. Proc. Natl. Acad. Sci. U.S.A. 2003 100 13308-13313. 186. Santangelo, P.J.; Nix, B.; Tsourkas, A.; Bao, G. Nucleic Acids Res. 2004 32 (6), e57/1e57/9. 187. Tsourkas, A.; Behlke, M.A.; Bao, G. Nucleic Acids Res. 2002 30 (23), 5168-5174. 188. Ono, T.; Scalf, M.; Smith, L.M. Nucleic Acids Res 1997 25 (22), 4581-4588. 189. Ortiz, E.; Estrada, G.; Lizardi, P.M. Mol. Cell. Probes 1998 12 (4), 219-26. 190. Braasch, D. A.; Corey, D. R. Chem. Biol. 2001 8 1.

PAGE 177

177 191. Wang, L.; Yang, C.J.; Medley, C.D.; Benner, S.A.; Tan, W. J. Am. Chem. Soc. 2005 127 (45), 15664-15665. 192. Dubertret, B.; Calame, M.; Libchaber, A.J. Nature Biotechnology 2001 19 (4), 365-370. 193. Maxwell, D.J.; Taylor, J.R.; Nie, S. J. Am. Chem. Soc. 2002 124 (32), 9606-9612. 194. Yang, C.J.; Lin, H.; Tan, W. J. Am. Chem. Soc. 2005 127 (37), 12772-12773. 195. Yang, C.J.; Pinto, M.; Schanze, K.; Tan, W. Angew.Chemie Int. Ed. 2005 44 (17), 2572-2576. 196. Kim, J.H.; Morikis, D.; Ozkan, M. Sensors and Actuators, B: Chemical 2004 B102 (2), 315-319. 197. Tomita, N.; Morishita, R.; Kaneda, Y.; Higaki, J.; Ogihara, T. Drug News Perspect. 2000 13 206. 198. Gryaznov, S.M.; Banait, N.S. Expert Opin. Ther. Pat. 2002 12 543. 199. Hayakawa, Y. Bull. Chem. Soc. Jpn. 2001 74 1547. 200. Verma, S.; Eckstein, F. Annu. Rev. Biochem ., 1998 67 99. 201. Stanciu, L.A.; Shute, J.; Holgate, S.T.; Djukanovic, R. J. Immun. Methods 1996 189 107-115. 202. Benez, A.; Geiselhart, A.; Handgretinger, R.; Schiebel, U.; Fierlbeck, G. J. Clin. Lab. Analysis 1999 13 229-233. 203. Sandrock, M. L.; Foss, C. A., Jr. J. Phys. Chem. B 1999 103 11398-11406. 204. Storhoff, J. J.; Lazarides, A. A.; Mucic, R. C. ; Mirkin, C. A.; Letsinger, R. L.; Schatz, G. C. J. Am. Chem. Soc. 2000 122 4640-4650. 205. Park, S.-J.; Lazarides, A. A.; Mirkin, C. A.; Letsinger, R. L. Angew. Chem. Int. Ed. 2001 40 2909-2912. 206. Norsten, T. B.; Frankamp, B. L.; Rotello, V. M. Nano Lett. 2002 2 1345-1348. 207. Jain, P.K.; Lee, K.S.; El-Sayed, I.H.; El-Sayed, M.A. J. Phys. Chem. B 2006 110 72387248. 208. Tyagi, S.; Kramer, F.R.. Nature Biotechnology 1996 14 (3), 303.

PAGE 178

178 209. Tan, W., K. Wang and T. Drake. Current Opinion in Chemical Biology 2004 8 (5), 547553. 210. Piatek, A.S.; Tyagi, S.; Pol, S.C.; Telenti, A.;Miller, L.P.; Kramer, F.R.; Alland, D. Nature Biotechnology 1998 16 (4), 359-363. 211. Sokol, D.L.; Zhang, X.; Lu, P.; Gewirtz, A.M. Proc. Nat. Acad. Sci. 1998 95 (20), 11538. 212. Marras, S. A. E.; Kramer, F. R.; Tyagi, S. Nucl. Acids Res. 2002 30 art-e122. 213. Crockett, A. O.; Wittwer, C. T. Analytical Biochemistry 2001 290 89-97. 214. Seidel, C. A. M.; Schulz, A.; Sauer, M. H. M. Journal of Physical Chemistry 1996 100 5541-5553. 215. Tyagi, S.; Bratu, D. P.; Kramer, F. R. Nature Biotechnology 1998 16 49-53. 216. Dirks, R.W.; Molenaar, C.; Tanke, H.J. Methods 2003 29 51-57. 217. Golzio, M.; Rols, M.P.; Teissie, J. Methods 2004 33 126-135. 218. Nitin, N.; Santangelo, P.J.; Kim, G.; Nie, S.; Bao, G. Nucleic Acids Res. 2004 32 (6), e58/1-e58/8. 219. Santangelo, P.J.; Nitin, N.; Bao, G. J. Biomedical Optics 2005 10 (4), 044025 1-6. 220. Santangelo, P.J.; Nix, B.; Tsourkas, A.; Bao, G. Nucleic Acids Res. 2004 32 (6), e57/1e57/9. 221. Faria, M. et al. Nat. Biotechnol. 2001 19 40-44. 222. Samulski, R. J.; Berns, K. I.; Tan, M.; Muzyczka, N. Proc. Natl. Acad. Sci. 1982 79 2077. 223. Carter, B. J. Hum. Gene Ther. 2005 16 541. 224. McCarty, D. M.; Monahan, P. E.; Samulski, R. J. Gene Ther. 2001 8 1248. 225. Wang, Z.; Ma, H.I.; Li, J.; Sun, L.; Zhang, J.; Xiao, X. Gene Ther 2003 10 2105. 226. McCarty, D.M.; Fu, H.; Monahan, P.E.; Toulson, C.E.; Naik, P.; Samulski, R. J. Gene Ther. 2003 10 2112. 227. Fu, H.; Muenzer, J.; Samulski, R.J.; Breese, G.; Sifford, J.; Zeng, X.; McCarty, D.M. Mol. Ther. 2003 8 911.

PAGE 179

179 228. Yang, G. S.; Schmidt, M.; Yan, Z.; Lindbloom, J. D.; Harding, T. C.; Donahue, B. A.; Engelhardt, J. F.; Kotin, R.; Davidson, B. L. J. Virol. 2002 76 7651. 229. Xu, D.; McCarty, D.; Fernandes, A.; Fish er, M.; Samulski, R.J.; Juliano, R.L. Mol. Ther. 2005 11 523. 230. Lakowicz, J. R.; Editor. Topics in Fluorescence Spectroscopy ; Plenum Press: New York, 1994 231. Perlette, J.; Tan, W. H. Anal. Chem. 2001 73 5544-5550. 232. Barker, S. L.; Clark, H. A.; Swallen, S. F. ; Kopelman, R.; Tsang, A. W.; Swanson, J. A. Anal. Chem. 1999 71 1767-1772. 233. Tan, W.; Shi, Z. Y.; Smith, S.; Birnbaum, D.; Kopelman, R. Science 1992 258 778-781. 234. Chomczynski, P.; Sacchi, N. Anal. Biochem. 1987 162 156-159. 235. Visner, G.A., Dougall, W.C.; Wilson, J.M.; Burr, I.A.; Nick, H.S. J. Biol. Chem. 1990 265 2856-2864. 236. Elowitz, M.B., Levine, A.J.; Siggia, E.D.; Swain, P.S. Science 2002 297 1183-1186. 237. Levsky, J.M.; Singer, R.H. Trends in Cell Biology 2002 13 4-6. 238. Zak, D.E., Gonye, G.E.; Schwaber, J.S.; Doyle, F.J. Genome Research 2003 13, 23962405. 239. Venkataraman, S.; Jiang, X.; Weydert, C.; Zhang, Y.; Zhang, H.J.; Goswami, P.C.; Ritchie, J.M.; Oberley, L.W.; Buettner, G.R. Oncogene 2005 24 77-89. 240. Spellman, P.T.; Sherlock, G.; Zhang, M.Q.; Iyer, V.R.; Anders, K.; Eisen, M.B.; Brown, P.O.; Botstein, D.; Futcher, B. Molecular Biology of the Cell 1998 9 3273-3297. 241. Medley, C.D.; Drake, T.J.; Tomasini, J.M.; Rogers, R.J.; Tan, W. Anal. Chem. 2005 77 (15), 4713-4718. 242. Yoshida, M.; Horinouchi, S.; Beppu, T. Bioessays 1995, 17 423. 243. Galeotti, T.; Wohlrab, H.; Borrello, S.; Deleo, M.E. Biochem. Biophys. Res. Commun. 1989, 165 581-589. 244. Sun, Y. Free Radical Biol. Med. 1990, 8, 583-599. 245. Bravard, A.; Hoffschir, F.; Sabatier, L.; Ri coul, M.; Pinton, A.; Cassingena, R.; Estrade, S.; Luccioni, C.; Dutrillaux, B. Int. J. Cancer 1992, 52 797-801.

PAGE 180

180 246. Bravard, A.; Petridis, F.; Luccioni, C. Free Radical Biology and Medicine 1999, 26 1027-1033. 247. Li, N.; Oberley, T.D.; Oberley, L.W.; Zhong, W.X. Prostate 1998, 35 221-233. 248. Bravard A.; Sabatier L.; Hoffschir F.; Ri coul M.; Luccioni C.; Dutrillaux B. Int. J. Cancer 1992, 51 476-480. 249. Goldman, R.; Moshonov, S.; Zor, U. Arch. Biochem. Biophys. 1998, 350 ,10-18. 250. Alexandratou, E.; Yova, D.; Loukas, S. Free Radical Biol. Med. 2005, 39 1119-1127. 251. Mielnicki, L.M.; Ying A. M.; Head, K.L.; Asch, H.L.; Asch, B.B. Exp.Cell Res. 1999, 249 161. 252. Li, J.; Fang, X.H.; Schuster, S. M.; Tan, W. H. Angewandte Chemie-International Edition 2000 39 1049. 253. Tan, W. H.; Fang, X. H.; Li, J.; Liu, X. J. Chemistry-A European Journal 2000 6 11071111. 254. Uchiyama, H.; Hirano, K.; KashiwasakeJibu, M.; Taira, K. J. Biol. Chem. 1996 271 380-384. 255. Fang, X. H.; Li, J. J.; Tan, W. H. Anal. Chem. 2000 72 3280-3285. 256. Braasch, D. A.; Corey, D. R. Chemistry & Biology 2001 8 1-7. 257. Tsourkas, A.; Behlke, M. A.; Bao, G. Nucleic Acids Res. 2002 30 5168-5174. 258. Xi, C. W.; Balberg, M.; Boppart, S. A.; Raskin, L. Appl. Environ. Microbiol. 2003 69 5673-5678. 259. Dueholm, K. L.; Egholm, M.; Behrens, C.; Christensen, L.; Hansen, H. F.; Vulpius, T.; Petersen, K. H.; Berg, R. H. ; Nielsen, P. E.; Buchardt, O. J. Org. Chem. 1994 59 57675773. 260. Gildea, B. D.; Casey, S.; MacNeill, J.; Perry -O'Keefe; H., Sorensen, D.; Coull, J. M. Tetrahedron Lett. 1998 39 7255-7258. 261. Koshkin, A. A.; Rajwanshi, V. K.; Wengel, J. Tetrahedron Lett. 1998 39 4381-4384. 262. Koshkin, A. A.; Nielsen, P.; Meldgaard, M.; Rajwanshi, V. K.; Singh, S. K.; Wengel, J. J. Am. Chem. Soc. 1998 120 13252-13253. 263. Wengel, J. Acc. Chem. Res. 1999 32 301-310.

PAGE 181

181 264. Vester, B.; Wengel, J. Biochemistry (Mosc). 2004 43 13233-13241. 265. Faulds, K.; Smith, E.W.; Graham, D. Anal. Chem. 2004 76, 412-417.

PAGE 182

182 BIOGRAPHICAL SKETCH Colin D. Medley was born in Toledo, OH in 1979. He completed his undergraduate carrer at the University of Tennessee in Knoxville under the research di rection of S. Douglass Gilman graduating Magna Cum Laude in 2003. He continue d his education at the University of Florida under Weihong Tan graduating in 2007.