Citation
Molecular Beacon Aptamers: Fundamentals for Fluorescence-Based Detection and for Protein Studies

Material Information

Title:
Molecular Beacon Aptamers: Fundamentals for Fluorescence-Based Detection and for Protein Studies
Creator:
VICENS-CONTRERAS, MARIE CARMEN ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Beacons ( jstor )
Cells ( jstor )
DNA ( jstor )
Enzymes ( jstor )
Excimers ( jstor )
Fluorescence ( jstor )
Molecules ( jstor )
Quenching ( jstor )
RNA ( jstor )
Signals ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Marie Carmen Vicens-Contreras. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/31/2010
Resource Identifier:
658210041 ( OCLC )

Downloads

This item has the following downloads:

vicenscontreras_m ( .pdf )

vicenscontreras_m_Page_050.txt

vicenscontreras_m_Page_074.txt

vicenscontreras_m_Page_002.txt

vicenscontreras_m_Page_013.txt

vicenscontreras_m_Page_056.txt

vicenscontreras_m_Page_095.txt

vicenscontreras_m_Page_128.txt

vicenscontreras_m_Page_030.txt

vicenscontreras_m_Page_137.txt

vicenscontreras_m_Page_114.txt

vicenscontreras_m_Page_126.txt

vicenscontreras_m_Page_043.txt

vicenscontreras_m_Page_011.txt

vicenscontreras_m_Page_031.txt

vicenscontreras_m_Page_124.txt

vicenscontreras_m_Page_094.txt

vicenscontreras_m_Page_028.txt

vicenscontreras_m_Page_069.txt

vicenscontreras_m_Page_089.txt

vicenscontreras_m_Page_037.txt

vicenscontreras_m_Page_047.txt

vicenscontreras_m_Page_082.txt

vicenscontreras_m_Page_131.txt

vicenscontreras_m_Page_106.txt

vicenscontreras_m_Page_121.txt

vicenscontreras_m_Page_006.txt

vicenscontreras_m_Page_134.txt

vicenscontreras_m_Page_100.txt

vicenscontreras_m_Page_048.txt

vicenscontreras_m_Page_083.txt

vicenscontreras_m_Page_041.txt

vicenscontreras_m_pdf.txt

vicenscontreras_m_Page_010.txt

vicenscontreras_m_Page_108.txt

vicenscontreras_m_Page_045.txt

vicenscontreras_m_Page_026.txt

vicenscontreras_m_Page_129.txt

vicenscontreras_m_Page_054.txt

vicenscontreras_m_Page_088.txt

vicenscontreras_m_Page_109.txt

vicenscontreras_m_Page_040.txt

vicenscontreras_m_Page_024.txt

vicenscontreras_m_Page_067.txt

vicenscontreras_m_Page_136.txt

vicenscontreras_m_Page_118.txt

vicenscontreras_m_Page_019.txt

vicenscontreras_m_Page_107.txt

vicenscontreras_m_Page_039.txt

vicenscontreras_m_Page_058.txt

vicenscontreras_m_Page_029.txt

vicenscontreras_m_Page_055.txt

vicenscontreras_m_Page_061.txt

vicenscontreras_m_Page_065.txt

vicenscontreras_m_Page_012.txt

vicenscontreras_m_Page_077.txt

vicenscontreras_m_Page_076.txt

vicenscontreras_m_Page_113.txt

vicenscontreras_m_Page_027.txt

vicenscontreras_m_Page_018.txt

vicenscontreras_m_Page_066.txt

vicenscontreras_m_Page_084.txt

vicenscontreras_m_Page_025.txt

vicenscontreras_m_Page_133.txt

vicenscontreras_m_Page_033.txt

vicenscontreras_m_Page_116.txt

vicenscontreras_m_Page_097.txt

vicenscontreras_m_Page_112.txt

vicenscontreras_m_Page_072.txt

vicenscontreras_m_Page_034.txt

vicenscontreras_m_Page_117.txt

vicenscontreras_m_Page_087.txt

vicenscontreras_m_Page_062.txt

vicenscontreras_m_Page_015.txt

vicenscontreras_m_Page_032.txt

vicenscontreras_m_Page_111.txt

vicenscontreras_m_Page_004.txt

vicenscontreras_m_Page_104.txt

vicenscontreras_m_Page_005.txt

vicenscontreras_m_Page_139.txt

vicenscontreras_m_Page_119.txt

vicenscontreras_m_Page_022.txt

vicenscontreras_m_Page_091.txt

vicenscontreras_m_Page_105.txt

vicenscontreras_m_Page_075.txt

vicenscontreras_m_Page_138.txt

vicenscontreras_m_Page_059.txt

vicenscontreras_m_Page_021.txt

vicenscontreras_m_Page_080.txt

vicenscontreras_m_Page_115.txt

vicenscontreras_m_Page_036.txt

vicenscontreras_m_Page_020.txt

vicenscontreras_m_Page_003.txt

vicenscontreras_m_Page_046.txt

vicenscontreras_m_Page_049.txt

vicenscontreras_m_Page_122.txt

vicenscontreras_m_Page_130.txt

vicenscontreras_m_Page_052.txt

vicenscontreras_m_Page_023.txt

vicenscontreras_m_Page_070.txt

vicenscontreras_m_Page_063.txt

vicenscontreras_m_Page_099.txt

vicenscontreras_m_Page_060.txt

vicenscontreras_m_Page_053.txt

vicenscontreras_m_Page_073.txt

vicenscontreras_m_Page_103.txt

vicenscontreras_m_Page_101.txt

vicenscontreras_m_Page_096.txt

vicenscontreras_m_Page_132.txt

vicenscontreras_m_Page_044.txt

vicenscontreras_m_Page_017.txt

vicenscontreras_m_Page_051.txt

vicenscontreras_m_Page_014.txt

vicenscontreras_m_Page_135.txt

vicenscontreras_m_Page_092.txt

vicenscontreras_m_Page_098.txt

vicenscontreras_m_Page_009.txt

vicenscontreras_m_Page_093.txt

vicenscontreras_m_Page_016.txt

vicenscontreras_m_Page_123.txt

vicenscontreras_m_Page_042.txt

vicenscontreras_m_Page_110.txt

vicenscontreras_m_Page_127.txt

vicenscontreras_m_Page_090.txt

vicenscontreras_m_Page_102.txt

vicenscontreras_m_Page_071.txt

vicenscontreras_m_Page_057.txt

vicenscontreras_m_Page_085.txt

vicenscontreras_m_Page_007.txt

vicenscontreras_m_Page_064.txt

vicenscontreras_m_Page_120.txt

vicenscontreras_m_Page_038.txt

vicenscontreras_m_Page_086.txt

vicenscontreras_m_Page_068.txt

vicenscontreras_m_Page_078.txt

vicenscontreras_m_Page_035.txt

vicenscontreras_m_Page_125.txt

vicenscontreras_m_Page_008.txt

vicenscontreras_m_Page_079.txt

vicenscontreras_m_Page_081.txt

vicenscontreras_m_Page_001.txt


Full Text











MOLECULAR BEACON APTAMERS: FUNDAMENTALS FOR
FLUORESCENCE-BASED DETECTION AND FOR PROTEIN STUDIES














By

MARIE CARMEN VICENS-CONTRERAS


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

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Marie Carmen Vicens-Contreras
















ACKNOWLEDGMENTS

I am grateful for all the efforts and kindness of all the people involved in this

proj ect. I especially thank my advisor Dr.Weihong Tan, for his guidance, encouragement,

and ideas throughout my graduate studies. I appreciate the opportunity of working in his

lab. He provided all of the resources that I needed for the past 5 years.

I appreciate the assistance of former group members of the Tan Lab, especially

Xiaohong Fang, for teaching me the basics to begin this proj ect; and Andrew Vanderlaan,

for his hard work and help during his time in the group.

Also, I thank Richard J. Rogers, Jon Stewart, James Winefordner, and Richard

Yost, for having agreed to be part of my graduate committee. I thank Arup Sen who

profoundly influenced me and helped me to get to where I am today, for sharing so many

ideas and offering me a new perspective on many things, and teaching me much about

learning and development.

My life as a graduate student was fun and cheerful thanks to my officemates. All of

them have taught me many things about life and their cultural heritage. For all the great

times we spent together celebrating life, I want to thank all of the members of the Tan

Group, especially Alina, Lisa, Tina, Prabodhika, Josh, Colin, James, Hong, and Shelly. I

thank Karen Martinez, for being a happy spirit and making me smile every day.

Intelligence and a deeply critical and analytical way of looking at ideas made Tim Drake

a fantastic colleague and great friend. I wish them all the best!










My adopted Puerto Rican family in Gainesville (which we like to call "el corillo"),

made my life j oyful. I especially thank Wilfredo Ortiz for one of the best friendships I

can ever imagine.

I am extremely thankful to "mi familiar especially my parents (Carmen and

Francisco) for their endless love, support, advice, and encouragement. I thank my

amazing brothers and sisters for taking care of me from a distance and calling me

whenever I forgot to call them.

At last, without the love, support, interest, sense of humor, assistance of all kind,

time, motivation and care from Jose I would be nowhere. It was his suggestion to come to

the University of Florida. His patience and his continuous encouragement enabled me to

succeed. He is one of the most wonderful human beings in this world. I am so indebted to

him that there is no way to repay him. I am overwhelmed with gratitude.


















TABLE OF CONTENTS


page

ACKNOWLEDGMENT S ............ ..... .__ .............. iii...


LI ST OF T ABLE S ............ ..... ._ .............. viii..

LI ST OF FIGURE S .............. .................... ix


AB STRAC T ................ .............. xii

CHAPTER


1 INTRODUCTION ................. ...............1.......... ......


Aptamers ................. ........... ...............2.......
W hat Is an Aptamer? ............. ......... ...... .... .. .. ... .... .......2
The Aptamer Selection Process: Systematic Evolution of Ligands by
Exponential Enrichment (SELEX) .............. ...............3.....
Aptamers versus Antibodies............... ...............
Aptamers in Bioanalytical Chemistry .............. ...............6.....
Fluorescence Methods for Signal Transduction .............. ...............10....
Fluorescence Quenching ................. .... ..._..._.. ............ ............. 1
Fluorescence Resonance Energy Transfer (FRET) ................. ............. .......12
M olecular Beacons ................ ................. ...............1
Molecular Beacons for Non-Specific Protein Detection ................ ................. 14
Aptamers and FRET for Protein Detection .............. ............. ........ .........1
Molecular Beacon Aptamer (MBA) Development for a Model Protein
System ................... .. ... .. ..... ...... .. ...... ...... ....................1
Aptamers for Platelet Derived Growth Factor-BB (PDGF-BB) .........................18

2 DEVELOPMENT AND CHARACTERIZATION OF A MOLECULAR
BEACON APTAMER FOR PLATELET DERIVED GROWTH FACTOR ............21

Introducti on .................. ...............21._____......
Materials and Methods ................ ......... ...............2
Synthesis and Purification of MBAs ........._ ....... ....._ .............23
Standard Fluorescence Quenching Assay .............. ...............25....
Results and Discussion .............. .... ........... ... ...........2
Choice of Aptamer Sequence for MBA Development............_. ............____...27
Evaluati on of Additi onal Fluorophore Quencher-P ai rs ................ ................. .2 9
Incubation Conditions for MBA Bioassay .............. ...............34....












Selectivity of the MBA-Based Fluorescence Quenching Assay .......................39
Fluorescence Quenching Assay Distinguishes Molecular Variants of PDGF ....42
Conclusions............... ..............4


3 IMPLICATIONS OF USING A MBA IN A FRET BASED AS SAY FOR PDGF
STUDIES IN BIOLOGICAL SAMPLES .............. ...............46....

Introducti on ................. ...............46.................
M materials and M ethods .............. ...............47....
In storm entati on.................. ....... ..... .. ...............47....

Preparation of a Simulated Biological Sample ................. ............... ...._...48
Cell Culture and Preparation of Conditioned Media ........._._......... .......... .....48
Results and Discussion ............... ...............49...
Use of MBA in Biological Sample ................. ................ ............... 49. ...
Implications of Total Protein Content on the MBA Assay ................ ...............53
Sample Pre-Treatment .....__ ................. ...............55.......
Conclusions............... ..............6


4 LIGHT SWITCHING APTAMER PROBE FOR PLATELET DERIVED
GROWTH FACTOR USING RATIOMETRIC AND TIME-RESOLVED
F LUORE SCENCE ME ASURE MENT S ........._ ....... ......__............6


Introducti on ............... ... ......__ ...............63...
Photophysics of Pyrene Excimers ................ ......... ....___ ............6
Time-Resolved Measurements .............. ...............66....
Experimental Section............... ...............68
Instrum ents .................. .................................6
Synthesis and Purification of Pyrene Labeled MBA............_._. ........._._.....69
Results and Discussion ................. ............. ... ........7
Light Switching Excimer Aptamer Probe Design ....._____ ..... .....___..........70
Light Switching Aptamer Prob e for PDGF -BB D etecti on .............. ..............7 1
Sensitivity and Selectivity of the Probe..................... ..............7
Direct Quantitative Detection of PDGF in Cell Medium .............. ..............77
Detection of PDGF-BB in Simulated Biological Sample .............. ..................82
Expression of PDGF-BB from Cells .........___....... ......__..........8
Conclusions............... ..............8


5 FLUORESCENCE RESONANCE ENERGY TRANSFER STUDIES FOR
MOLECULAR BEACON APTAMERS: UNDERSTANDING PROTEIN
INTERACTION AND VERSITILITY OF PROBES ................ ..................9


Introducti on ............ ..... ._ ...............90....
Results and Discussion ...................... ...............91
Backbone-Modification of DNA .............. ........... .. .... .... ._ ............9
MBAs Created from Homologous RNA and DNA Aptamers Display
Significant Differences in Eliciting FRET in Response to PDGF-BB ............94












Fluorescence Enhancement Assays with MBAs ............_... ..... ..__............99
Two-Step FRET Assays Using Restriction Enzymes Cleavage Sites ..............104
Conclusions............... ..............10


6 SUMMARY AND FUTURE WORK ................. ....___.....__ ............1


Summary ............ ..... .._ ...............112...
Future W ork ........... __.... .....__ ..... ....... ._ .. .. .. ..... ........ 1
New MBA Development for Other Proteins and Small Molecules ..................1 17
Enrichment and Simplification of Complex Samples ........._.._.. ...............1 17
Pyrene Application to Other Protein Recognition Systems .............. .............118
Signal Amplification of Aptamer Recognition through PCR. ........................1 19

LIST OF REFERENCES ............ .......__ ...............120.


BIOGRAPHICAL SKETCH ............ ...... ._ ...............126..

















LIST OF TABLES


Table pg

2-1 Percentage of quenching of different consensus sequences of 50 nM of MBA
for PDGF aptamer with 200 nM of PDGF ..........._ ..... .__ .......___.......2

3-1 Analysis for PDGF-BB spiked samples after fractionation on a G-100
Sephadex column using enzyme-linked immunosorbent assay (ELISA).................60

4-1 Pyrene-labeled oligonucleotide sequences used in this study. ................ ...............68

4-2 Amount of protein collected in 24 hours ................. ...............86..............

4-3 Amount of protein collected in 24 and 36 hours ................. .......... ...............87

5-1 Aptamers synthesized for PDGF-BB detection .............. ..... ............... 9


















LIST OF FIGURES


Finure pg

1-1 Overview of the systematic evolution of ligands by exponential enrichment
process ........... ....... __ ...............4...

1-2 Typical Jablonski diagram .............. ...............10....

1-3 Fluorescence resonance energy transfer diagram ................. ................ ....__.13

1-4 Working principle of molecular beacon ................. ...............15...............

1-5 Assumed biological foldings of Platelet Derived Growth Factor ................... .........19

1-6 Folding of 36t molecular beacon aptamer (MBA) ................ ........................20

2-1 Fluorescence intensity changes versus time of different aptamer sequences. ..........28

2-2 Proposed mechanism of PDGF-BB induced FRET responses for the 36t probe.....29

2-3 Evaluation of fluorescence quenching caused by PDGF-BB with MBAs.............31

2-4 Calibration curves for PDGF-BB .....__.....___ ..........__ ...........3

2-5 Effect of increasing temperature on fluorescence signal of free MBA in
solution ........... _.......__ ...............35....

2-6 Effect of increasing pH on fluorescence signal of free MBA in solution ...............37

2-7 Effect of divalent cations on fluorescence signal of free MBA in solution .............38

2-8 The effect of monovalent cations on fluorescence signal of free MBA in
solution ................. ...............39.................

2-9 Binding selectivity of the MBA ........._._.. .......... ...._.._ ...........4

2-10 Fluorescence quenching displayed by the MBA............... ...............41..

2-11 Cartoon representing PDGF-BB and its molecular variants .............. ..................43

2-12 Dose-response curves of PDGF variants: fluorescence signals of MBA for
PD GF-AA ........... .......__ ...............43..











3-1 Use of molecular beacon aptamer in biological samples ................ ................ ...51

3-2 Molecular beacon aptamer fluorescence quenching .............. ....................5

3-3 Fluorescence quenching caused by PDGF-BB in the presence of proteins in a
simulated biological specimen .............. ...............57....

3-4 Fluorescence intensity of fractions of simulated biological sample (SBS)
collected from G-100 column .............. ...............60....

3-5 Fluorescence intensity of dilutions of SBS .............. ...............61....

4-1 The chemical structure of pyrene ................ ...............65...............

4-2 Use of pyrene excimer to probe PDGF-BB ........._ ....... __ ........__........70

4-3 Fluorescence emission spectra of Pyr-MBA with and without PDGF-BB in
binding buffer. .............. ...............71....

4-4 Real-time response of excimer/monomer ratio .............. ...............73....

4-5 Dose-dependent excimer formation on additions of PDGF-BB .............. .............74

4-6 Visual detection of PDGF-BB............... ...............7

4-7 Response of excimer probe to different proteins .............. ...............75....

4-8 Response of 50 nM Pyr-MBA to 500 nM of different growth factors....................76

4-9 Increasing amounts of BSA incubated with 50 nM Pyr-MBA. ............. ................77

4-10 Fluorescence spectra of simulated biological sample .............. ....................7

4-11 Monomer time-resolved spectra collected at 398 nm .............. ....................8

4-12 Excimer time-resolved spectra collected at 480 nm............... ...................8

4-13 Fluorescence intensity as a function of wavelength at different times. .........._.......82

4-14 Steady-state and time-resolved fluorescence spectra of SBS samples.....................83

4-15 Fluorescence decays of 200 nM Pyr-MBA in simulated biological specimen ........84

5-1 Phosphorothioate DNA backbone modification .............. ...............92....

5-2 Fluorescence quenching produced by a phosphorothiate modified MBA. ..............93

5-3 Theoretical model of the 36tRNA-MBA ................. ...............96...............











5-4 Comparison of FRET- response of 36Gt-DNA-MBA and the homologous 3 6t-
RNA-MBA with PDGF-BB .............. ...............97....

5-5 Competition of FRET-response by unmodified 36Gt-DNA aptamer with the
standard 36t-DNA-MBA............... ..............9

5-6 Competition of FRET response by unmodified RNA aptamer with DNA-MBA
and by unmodified DNA aptamer with RNA-MBA ................ ......................100

5-7 Fluorescence enhancement FRET-assay ..........._ ..... ..__ .........__........10

5-8 Working principle of the enzyme-site modified MBA ...........__... .........__.....105

5-9 Molecular beacon aptamer variants with restriction enzyme cleavable stem
sequence response to magnesium ....__ ......_____ .......___ .............0

5-10 Molecular beacon aptamer variants with restriction enzyme cleavable stem
sequence response to PDGF-BB .............. ...............108....
















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

MOLECULAR BEACON APTAMERS: FUNDAMENTALS FOR
FLUORESCENCE-BASED DETECTION AND FOR PROTEIN STUDIES

By

Marie Carmen Vicens-Contreras

August 2005

Chair: Weihong Tan
Maj or Department: Chemistry

Molecular beacon aptamers (MBA) are created by combining synthetic

oligonucleotide ligands with high affinity for a protein target with a sensitive signal

transduction mechanism. In this study, an MBA was developed and examined for the

detection of platelet derived growth factor-BB (PDGF-BB) using it in a fluorescence-

based assay. The developed probe was capable of consistently detecting 10 nM PDGF-

BB in homogeneous solution. The influence of certain physical and chemical parameters

of incubation that would affect DNA conformation were examined and the results show

that this bioassay is compatible with pH, temperature, and monovalent cation levels

typically encountered in biological samples. After a complete investigation in

homogeneous solution, the MBA was examined for PDGF-BB detection in biological

samples. This is the first attempt to use single aptamer molecules with an inherent signal

transduction mechanism for the detection of PDGF-BB in biological samples. For this

particular aptamer and assay conditions, the presence of more than 100 ug/mL of serum









proteins resulted in non-specific fluorescence quenching of the probe. The results

indicated that a sample processing method prior to the assay is required to remove

potential biological interference.

Selective and quantitative protein analysis in complex biological fluids still

presents considerable challenges for future research, mainly due to the background

signals from both the fluorescent probe and the components of typical cellular samples.

To overcome the problem of background signals, the existing MBA was modified by

labeling each end with pyrene molecules. This produced a monomer-excimer light-

switching aptamer probe for protein monitoring by using both steady-state and time-

resolved fluorescence measurements. With a time-correlated single photon counting

technique and the long fluorescence lifetime of pyrene, the biological background signal

was separated from the excimer signal and 200 pM of PDGF-BB were detected in a

simulated biological sample.

We also examined FRET responses to PDGF-BB shown by MBAs with modified

aptamer sequences that affect DNA conformation. Our findings provide new insights into

the fundamental mechanisms underlying protein-induced FRET shown by MBAs and

have significant implications on the design strategy and use of MBAs in structure-

function analysis, and in bioassays for protein biomarkers and their molecular variants.














CHAPTER 1
INTTRODUCTION

Proteins are essential for the structure and function of all living cells. Protein have

different shapes and conformations that are responsible for many functions, such as

enzymatic reactions, regulation of cell growth, structure, movement, carriers, and

receptors. Thus, many scientists in the bioanalytical sciences are interested in developing

methods and probes to be used for protein studies and detection. Biological interactions

are used to provide either qualitative or quantitative information about protein content

and could yield important data about the molecular mechanism of action of biologically

active compounds. Methodologies are needed to study disease biomarkers that can help

diagnose a disease at the biochemical level or help distinguish the responses of different

patients to the same medical treatment.'

For protein biomarker detection and quantitation, several techniques are widely

used including: gel electrophoresis,2 isoelectric focusing, affinity chromatography,3

surface enhanced laser desorption time-of-flight mass spectrometry (SELDI),4 prOximity

ligation,5 cell-based bioactivity assayS6 and immunological assays, such as enzyme-

linked immunosorbent assays (ELISA).7 Many of the above methods are time-consuming

and labor intensive; and require multiple steps, such as immobilization, repeated

incubations and washings, and additional reagents to amplify the signal. Additional

methods, such as nucleic-acid molecules labeled with two signaling moieties, can be used

in different types of formats, for biomarker detection. Nucleic-acid sequences, known as

aptamers, can adopt complex three-dimensional structures capable of selectively









recognizing protein molecules.8 Thus, aptamers are good candidates for detecting diverse

protein targets, including those proteins that do not naturally recognize or bind to DNA or

RNA. Aptamer-based probes can be developed with the idea that when the target protein

binds to the detector molecules, a change in the optical properties of the reporter

molecules will be triggered, to indicate the recognition event. Unfortunately, little is

understood about the basic mechanism of action for aptamer-based signal transduction

mechanism or its applications. With careful design and systematic characterization,

aptamer-based probes could be useful in diagnostic applications and protein-biomarker

studies.

Aptamers

What Is an Aptamer?

Nucleic acid binding species generated by in vitro selection have been referred to as

aptamers.9 Aptamers can be RNA, modified RNA, single-stranded DNA or double-

stranded DNA, and have been selected to bind targets ranging from small organic

molecules to cells.10 In vitro selection has been used to identify aptamers against different

targets covering a wide range of sizes, including ions," small molecules,9 peptides,12

single proteins,13 Organelles,14 VITusCS,15 and even entire cells.16 In addition to carrying

genetic information, oligonucleotide sequences (aptamers) can adopt complex three-

dimensional structures, capable of specific binding to target molecules and, furthermore,

of catalyzing chemical reactions.l7 Aptamers have been generated for over 100 proteins,

including growth factors, transcription factors, enzymes, immunoglobulins, and

receptors.l0 At present, an aptamer databaselo that contains sequences drawn from 239

published in vitro selection experiments available for public use. A typical aptamer is 10

to 15 kDa in size (30 to 45 nucleotides), binds its target with subnanomolar affinity, and









discriminates against closely related targets (e.g., will typically not bind other proteins

from the same gene family). A series of structural studies show that aptamers can use the

same types of binding interactions (hydrogen bonding, electrostatic, hydrophobic

contacts, steric exclusion, etc.) that drive affinity and specificity in antibody-antigen

complexes and are exploited in systematic evolution of ligands by exponential

enrichment, the aptamer-selection method.

The Aptamer Selection Process: Systematic Evolution of Ligands by Exponential
Enrichment (SELEX)

In 1990, Gold 13,Ellington and Szostak9 Simultaneously reported the development

of an in vitro procedure for the selection of single-stranded nucleic acids to perform a

specific function. This in vitro procedure has been called systematic evolution of ligands

by exponential enrichment (SELEX). Since its introduction over 10 years ago, in vitro

selection has been widely adopted as a tool for the development of research reagents, and

shows promise generating diagnostic and therapeutic agents. The SELEX process consists

of exponentially selecting and amplifying aptamers from a large library (1015 to 10 ") of

oligonucleotide molecules.1-2 (Figure 1-1).

The starting point for the in vitro selection process is a combinatorial library

composed of single-stranded nucleic acids (RNA or DNA) typically containing 20 to 40

randomized positions. Randomization creates an enormous diversity of possible

sequences (e.g., four different nucleotides at 40 randomized positions gives a theoretical

possibility of 440 Or ~1024 different sequences). Since short, single-stranded nucleic acids

adopt somewhat rigid structures (imposed by their sequences), such a library contains

many molecular shapes or conformations. To identify high-affinity nucleic acid ligands to

a given target protein, the starting library of nucleic acids (in practice, ~1014 to 1015









different sequences) is incubated with the protein of interest. Nucleic acid molecules that

adopt conformations that allow them to bind to a specific protein are then partitioned

from other sequences in the library that are incapable of binding to the protein under the

conditions used.


1014 10 15

sequences nL2seifctre



Repeat process
with selected
sequences
Protein
target




Partition
Bound Unbound

AmO pD se


Figure 1-1.Overview of the systematic evolution of ligands by exponential enrichment
process. The initial library of nucleic acids is incubated with the protein
target of interest. Molecules that bind to the target protein are then partitioned
from other sequences in the library. The bound sequences are then amplified
to generate a library enriched in sequences that bind to the target protein.

The bound sequences are then removed from the protein and amplified by reverse

transcription polymerase chain reaction (RT-PCR) (for RNA-based libraries) or PCR (for

DNA-based libraries) to generate a reduced-complexity library enriched in sequences that

bind to the target protein. The DNA population is then used to generate a new library of

single-stranded molecules (transcription for RNA; strand separation for DNA) that is

again subj ected to the selection procedure. This process is repeated, until a group of










single-stranded molecules are selected whose binding affinities are sufficient for the

intended application.

Aptamers versus Antibodies

Most of the ligand-detection methods known in the bioanalytical sciences are based

on antibody binding. Like aptamer-based methods, antibody-based detection methods

require maintenance of the ligand-antibody complex, to generate a detectable signal. In

addition, antibody methods such as ELISA or competitive radioimmunoassay (RIA),

while robust and highly sensitive, have limited applicability because of the require

heterogeneous assay conditions:

* The detection must be done on a solid surface

* In most applications both a capture antibody and detection antibody are required

* For ELISA-based protein detection methods, the antibodies must recognize the
folded, native structure of the protein that is present in cell or tissue isolates.7;22;23

As an alternative (or complementary) molecule to antibodies, aptamers have

several inherent advantages that merit study. Since aptamers consist of a short, single

strand of DNA or RNA, they are much easier and economical to synthesize and have a

much longer shelf life. In terms of stability, aptamers are chemically robust and are

intrinsically able to regain activity after exposure to heat and denaturants, and can be

stored for extended periods (more than a year) at room temperature as lyophilized

powders. In contrast, antibodies must be stored refrigerated. Aptamers are chemically

synthesized and thus can be readily scaled as needed to meet production demand.

Whereas difficulties in scaling production currently limit the availability of some

biological products and the cost of a large-scale protein production plant is high, a single

large-scale synthesizer can produce more than 100 kg of oligonucleotide per year and









requires a relatively modest initial investment.l One of the most significant advantages

of aptamers is that their selection process mimics natural selection, so it is possible, in

theory, to develop a highly specific aptamer for virtually any target molecule with the

same high affinity similar offered by antibodies. The affinities of aptamers range from

dissociation constants of 0.3 to 500 nM, with most aptamers having binding affinities in

the range of 1 tol0 nM (comparable to or greater than the typical Fab fragments obtained

through immunization).

In comparison to engineering antibodies for particular applications, incorporating

site-specific labels or coupling sites into an aptamer is usually a minor procedure. Since

the aptamer itself lacks any means of signal transduction on binding a protein, this ease of

molecular labeling is crucial for their development into fluorescence probes for protein

studies.

Aptamers in Bioanalytical Chemistry

Aptamer molecules have been developed and used to detect specific analyte

molecules in different experimental settings. RNA or DNA molecules must create a

specific cavity to enclose and recognize a small molecule. Nonetheless, proteins possess

extensive surfaces (with ridges, grooves, projections, and depressions) with several

hydrogen-bond donors and acceptors; suggesting that proteins may be excellent targets

for aptamer-probe developmental shtdies.9 The first known nuclei-acid aptamer to a

protein that does not normally interact with RNA or DNA was a single-stranded DNA

aptamer against thrombin.24 The aptamer for thrombin has been used extensively for

multiple applications. In one case25, flUOrescently labeled anti-thrombin aptamers

attached to a glass surface were used to detect the presence of thrombin proteins in a

sample, by detecting changes in the optical properties of the aptamers. In this system,









binding of thrombin to the labeled aptamer is monitored by detecting fluorescent

emission of the dye-labeled aptamer on excitation by an evanescent field.

McGown and her group at Duke University26 USed the same aptamer for affinity

capture of thrombin with Matrix Assisted Laser Desorption and lonization-Time of

Flight-Mass Spectrometry (MALDI-TOF-MS). The aptamer was covalently attached to

the surface of a glass slide that served as the MALDI surface. Results showed that

thrombin is retained at the aptamer-modified surface while other proteins, such as

albumin, are removed by rinsing with buffer. Upon application of a low-pH MALDI

matrix, the G-quartet structure of the aptamer unfolded, releasing the captured thrombin.

After TOF-MS analysis, residual matrix and proteins were washed from the surface, and

buffer was applied to refold the aptamers, allowing the surface to be reused. The aptamer

demonstrated high selectivity against mixtures of thrombin and albumin and of thrombin

and mixtures of prothrombin from human plasma.

Aptamers can also be used to separate closely related compounds depending on

their dissociation constants.27 Deng et al 28 explored the use of a 42-mer DNA aptamer

for adenosine/ATP binding, as a weak affinity chromatography stationary phase using

biotin-avidin coupling. Their data demonstrated that affinity chromatography is useful for

characterizing the binding of aptamers to multiple targets. In addition, they evaluated the

aptamer stationary phase for separation of cyclic-AMP, NAD AMP, ADP, ATP, and

adenosine. Their data demonstrated that an aptamer immobilized by biotin-avidin

coupling can selectively retain and separate related target compounds by weak affinity

chromatography. The reported data suggested that it may be possible to develop

stationary phases that selectively retain and separate classes of compounds based on









aptamers and that affinity chromatography is a valuable and convenient technique for

characterizing the affinity and selectivity of aptamers.

Other nucleic acid-based detection schemes have exploited the ligand-sensitive

catalytic properties of some nucleic acids, such as ribozymes. For example, Robertson

and Ellington29demonstrated that a ribozyme that acquires a ligase activity on ligand

binding can be used to detect a ligand by monitoring the ligation of a small, labeled,

second oligonucleotide to the ribozyme. In a complementary approach, 30;31 labeled

allosteric ribozymes that undergo cleavage upon binding to a ligand have been used to

detect ligands by monitoring the release of the label from the ribozyme. However, the

maj ority of the detection techniques have the disadvantage that the ligand-activated

ribozyme is irreversibly modified in the course of generating a signal. Thus, ribozymes

can be used only once in an assay. Furthermore, signal generation is slow with ribozymes

and can take from 1 minute to 1 hour or more.

Manalis32 developed a method for label-free protein detection using a

microfabricated cantilever-based sensor that was functionalized with DNA aptamers to

act as receptor molecules. The sensor used two adj acent cantilevers that constituted a

sensor/reference pair and allowed direct detection of the differential bending between the

two cantilevers. In this case one cantilever was functionalized with aptamers selected for

Taq DNA polymerase while the other was blocked with single stranded DNA. They

reported that the polymerase aptamer binding induced a change in surface stress, which

caused a differential cantilever bending dependent on the ligand concentration. Protein

recognition on the sensor surface was specific and had a concentration dependence that is

similar to that in solution.









Another class of aptamer probe is based on monitoring the change in fluorescence

anisotropy on protein binding.21;25;33 Anisotropy probes are designed by labeling an

aptamer with a single fluorophore molecule. When the labeled aptamer is bound with its

target protein, the rotational motion of the fluorophore becomes much slower as a result

of the larger molecular weight of the aptamer-protein complex. As a consequence of the

binding event, a significant increase in fluorescence anisotropy can be monitored using

plane-polarized light. Anisotropy-based aptamer probes serves as an alternative

mechanism of signal transduction for aptamer-protein binding which results in 'on' and

'off' type signals. Some advantages of anisotropy-based aptamer probes versus analogous

antibody probes are a result of their respective sizes. Since most aptamers are

substantially smaller than antibodies, the relative increase in molecular weight and

fluorescence anisotropy will be much larger on binding with proteins for aptamer probes.

However, the sensitivity of aptamer anisotropy probes is generally not as good as

fluorescence resonance energy transfer-based probes.

For Platelet Derived Growth Factor (PDGF-BB) detection, an aptamer was

converted into an anisotropy probe through 5'-labeling with fluorescein.33 Addition of

PDGF-BB (nM concentration) to an equimolar solution of the labeled aptamer increases

the fluorescence anisotropy more than 2-fold. The binding of protein to the aptamer

clearly results in a large molecular-weight complex, thereby slowing down the rotational

diffusion of the fluorescein label. The binding reaction is quick, with equilibrium

achieved in less than 20 s, including mixing time. Controls using free fluorescein dye

with PDGF-BB show no effect on anisotropy, validating that the observed changes are a

direct result of the aptamer-PDGF-BB binding. All the reported applications of aptamers










demonstrate that with the use of aptamer-based molecules it is possible to detect analyte

binding in both solution (homogeneous) and on solid supports (heterogeneous).

Aptamers are required to work along with a signal transduction mechanism to

report the binding event to the target molecule. High affinity and selective aptamer

probes can be further exploited by incorporating a highly sensitive signal transduction

mechanism within the synthetic DNA sequence that would allow their use for single-step

detection schemes without the need of separating the aptamer-target complex from the

solution.

Fluorescence Methods for Signal Transduction

Fluorescence is a highly sensitive detection method. Figure 1-2 shows a typical

Jablonski diagram where So, S1 and S2 Stand for the singlet, first and second electronic

states, respectively, while T1 stands for triplet state. Following light absorption, a

fluorophore is excited to a higher vibrational level. By a process called internal

conversion, the molecules in condensed phases rapidly relax to the lowest vibrational

level of S1. Internal conversion is generally complete prior to emission; hence,

fluorescence emission generally results from the lowest-energy vibrational state of S1.



I INTERNAL CONVERSION
S, INTERSYSTEM
CROSSING







PHOSPHORESCENCE
h 0



Figure 1-2. Typical Jablonski diagram









Upon excitation into higher electronic and vibrational levels, the excess energy is

quickly dissipated leaving the fluorophore in the lowest vibrational level of S1. Since this

is a rapid relaxation, emission spectra are usually independent of the excitation

wavelength. One of the general characteristics of fluorescence is that it exhibits what is

known as the phenomenon of Stokes Shift. This phenomenon reveals that the energy of

the emission is typically less than that of absorption; hence, fluorescence typically occurs

at lower energies or longer wavelengths. The intensity of fluorescence can be decreased

by a wide variety of processes known as quenching.

Fluorescence Quenching

The process of decreasing fluorescence intensity is termed quenching and it usually

happens through two maj or mechanisms.34 The first mechanism, known as collisional

quenching, takes place when a fluorophore is excited by incident light and as a

consequence goes into an excited state and stays there for a short period of time

(nanoseconds). The time it stays in the excited state is called fluorescence lifetime. In this

case, the quencher must diffuse to the fluorophore during the lifetime of the excited state.

Upon contact, the fluorophore returns to the ground state, without emission of a photon.

Collisional quenching is described by the Stern-Volmer as shown in Equation 2-1; where

Fo and F are the fluorescence intensities in the absence and presence of quencher,

respectively, k, is the bimolecular quenching constant, to is the lifetime of the

fluorophore in the absence of the quencher, and [Q] is the concentration of the quencher.

Fo/F=1+kgro[Q]=1+KD [Q] (2-1)

The second mechanism of fluorescence quenching, named static quenching, is

characterized by the formation of a nonfluorescent complex between a fluorophore and a










quencher. When fluorophore-quencher complex absorbs light, it immediately returns to

ground state without emission of a photon.

Fluorescence Resonance Energy Transfer (FRET)

Fluorescence resonance energy transfer is an interesting fluorescence-related

phenomenon which relies on the radiationless transfer of energy from a donor

fluorophore to an acceptor fluorophore. The distance dependent transfer of energy is a

physical process that depends on the spectral overlap of the donor and acceptor spectra

and proper dipole alignment of the two fluorophores. FRET is considered to be one of the

few tools available for measuring nanometer scale distances and changes in distances,

both in vitro and in vivo.

In FRET, a donor fluorophore is excited by incident light, and if an acceptor is in

close proximity, the excited state energy from the donor can be transferred. This leads to

a reduction in the donor's fluorescence intensity and excited state lifetime, and an

increase in the acceptor' s emission intensity. One of the requirements for FRET is that

the two molecules need to be very close, typically less than 10 nm. In fact, Lakowics34

showed that the efficiency of this process (E) depends on the inverse of the distance,

between donor and acceptor, to the sixth power as shown in Equation 2-2, where E is the

energy transfer efficiency, r is the distance between the donor and the acceptor, and Ro is

the Forster distance at which half of the energy is transferred (typically 20 to 60 A+), and

depends on the spectral characteristics of the dyes and their relative orientation. When

non-fluorescent molecules are used as acceptors, the result of the energy transfer is

quenching of the donor fluorescence.

E = RO6 R6 r6 (2-2)










DONOR

ACCEPTOR


RI ENERGY TRANSFER







Figure 1-3. Fluorescence resonance energy transfer diagram. Whereas normally an
excited fluorophore returns back to the ground state with the emission of a
photon, FRET results in the excitation of the nearby acceptor fluorophore that
in turn emits a photon when it returns to the ground state. The occurrence of
FRET is characterized by a decrease in observed donor emission, and a
simultaneously sensitized (increased) acceptor emission.

Fluorescence resonance energy transfer is a distance-dependent phenomenon that

can be applied to the development of aptamer probes to study intermolecular and

intramolecular relationships in biophysical systems and cell biology. The development of

an efficient aptamer probe is not only dependent on the oligonucleotide sequence and

conformation but also on the selection of the fluorophore and the quencher pair. With the

high diversity of existing fluorophores and the growing number of the alternatives

quenchers, the aptamer probes can be modified to fulfill specific experimental needs.

Several reports have been published with updated information regarding the efficiencies

of FRET for commonly used fluorophore-quencher pairs and their effects on duplex

stability.35 Various approaches, including the well studied molecular beacons, have

exploited the ability of fluorescent compounds to absorb energy and transfer it to nearby

molecules for the study of molecular interactions.36









Molecular Beacons

Molecular beacons (MBs) 36 are synthetic oligonucleotide probes that possess a

stem-and-loop structure, designed for specific recognition of DNA or RNA targets. The

basis for target recognition is the hybridization of the nucleic acid to its complementary

target. The single-stranded loop portion of the MBs has a sequence complementary to the

target DNA and can report the presence of specific target nucleic acids. The stem

typically has five to seven base pairs which are complementary to each other but

unrelated to the target. Signal transduction in MBs is accomplished by resonance energy

transfer. A fluorophore (donor molecule) and a quencher (acceptor molecule) are

covalently linked to the two ends of the stem. The stem keeps the two moieties in close

proximity, causing the fluorescence of the fluorophore to be quenched by energy transfer.

When the probe encounters a target DNA molecule, the molecular beacon undergoes a

spontaneous conformational reorganization, leading to the formation of a hybrid that is

longer and more stable than that of the stem, that forces the stem apart and a consequent

fluorescence restoration. The conformational state of a molecular beacon is thus directly

reported by its fluorescence: in the closed state, the molecular beacon is not fluorescent;

in the open state, when the fluorophore and the quencher molecules are apart, it emits

intense fluorescence. Different molecular beacons can be designed by choosing loop

sequence and length. Also, the quencher and the fluorophores can be changed according

to the application.

Molecular Beacons for Non-Specific Protein Detection

Although molecular beacons were originally designed to bind and recognize

specific nucleic acids, the probes can also lead to increased fluorescence on binding to

certain proteins. Since the binding of proteins to DNA or RNA molecules can readily










i Target
DNA
Loop MB 'Y















Figure 1-4. Working principle of molecular beacon. Before hybridization with target
nucleic acid sequences the MB takes on a stem-loop structure which maintains
the close proximity of the fluorophore (F) and quencher (Q) moieties. The
resulting fluorescence is minimal due to static quenching of the fluorophore
molecule. After introducing target nucleic acid sequences, the loop sequence
of the molecular beacon hybridizes to the target and unstabilizes the stem
hybrid. Consequently, the two moieties spatially separate and result in the
restoration of the fluorescence.

disturb the conformation of the nucleic acid, it is expected that this binding would result

in spatial separation of the MB fluorophore and quencher. As an example, an E.coli

single stranded DNA binding protein (SSB) was used to demonstrate the protein

recognition capability of MBs. This DNA-binding protein is used in DNA

replication, recombination, and repair. Using a conventional fluorescence

spectrofluoremeter, the investigators detected as low as 20 nM of SSB using a MB

labeled with tetramethylrhodamine (TMR) and dimethylaminophenylazobenzoic acid

(DABCYL). The measured fluorescence intensity changes over time revealed that the

SSB-MB interaction is rapid, reaching equilibrium within 10 s. The MB-based

SSB assay is not, however, particularly specific. In fact, SSB leads to a fluorescence

enhancement nearly equal to that of the complementary DNA, but other proteins










(including histone and Rec A) can also bind with the MB and cause a fluorescence

intensity increase. The results demonstrated that while MBs were sensitive and somewhat

selective to DNA-binding proteins, they were not specific enough to be capable of

distinguishing a particular protein. In order to apply the principle of MBs for real-time

detection of protein biomarkers, a more selective protein recognition mechanism is

required.

Aptamers and FRET for Protein Detection

In addition to exploiting the inherent conformational changes that aptamers

undergo, aptamers have been engineered similar to MBs such that the addition of an

analyte results in a conformational change and simultaneous reduction or increase in a

fluorescent signal. Utilizing FRET between fluorophore and quencher moieties for signal

transduction of protein binding is one of the more popular designs. Depending on the

relative positions of the fluorophore and quencher before and after protein binding,

probes can result in either enhanced or reduced fluorescence on binding. It is important to

mention that the design of optimized probes is not always a trivial process. A maximum

change in fluorescence on binding is often achieved by gaining knowledge about the

conformations of the free aptamer and the aptamer-protein complex. Ideally, when this

structural information is available, one can strategically incorporate the fluorophore and

quencher moieties on the aptamer probe in such a way that the transition from bound to

unbound conformations causes a dramatic change in their relative positions.

An example of a FRET based quenching aptamer probe is the one developed using

a thrombin aptamer.37 The first reported aptamer for thrombin contains a 15-nucleotide

consensus sequence.38 When bound to thrombin, the aptamer exists primarily in its

quadruplex form containing two G-quartet structures, but in free solution, it can adopt









either conformation, dependent in part on the ionic strength and temperature. This

conformation shift provides the basis for an aptamer probe. By labeling the two ends of

the aptamer with a fluorophore and quencher pair and extending the aptamer by one base

on each end, aptamer binding of thrombin would force the quencher adj acent to the

fluorophore, resulting in a substantial decrease in fluorescence.

Another method of detecting binding of a target protein to an aptamer has also been

described which relies on the use of fluorescence-quenching pairs whose fluorescence is

sensitive to changes in secondary structure of the aptamer upon ligand binding.39

However, target protein-mediated changes in secondary structure were engineered into

the aptamer molecule via a laborious engineering process in which four to six nucleotides

were added to the 5' end of the aptamer that was complementary to the bases at the 3' end

of the thrombin binding region. In the absence of thrombin, this structure forms a stem

loop structure, while it forms a G-quartet structure in the presence of thrombin.

Fluorescent and quenching groups attached to the 5' and 3' end signal this change.

A related class of aptamer probes is the two fluorophore FRET probes.40 Binding of

target protein can be detected by monitoring the fluorescence of a second fluorophore

(F2) directly or preferably by the ratio of fluorescence of both fluorophores (F2/F l).

Ratiometric detection may provide enhanced sensitivity, with reported detection limits in

the low pM regime. All of the design considerations applicable to the quenching aptamer

probes are also crucial to this class of aptamer probes. Detailed structural information of

the aptamer and aptamer-protein complex allows optimized positioning of the

fluorescence donor and acceptor.









Molecular Beacon Aptamer (MBA) Development for a Model Protein System

Based on previous studies that showed that the high affinity and selectivity of

aptamers can be successfully combined to develop aptamer probes, it is of interest to

further investigate and develop aptamers for biomolecule studies. In this case, PDGF-BB

was chosen based on the significance of the protein in biological systems and the fact that

an aptamer had already been selected for this protein.

Platelet derived growth factor was discovered as a maj or mitogenic factor present

in serum but absent from plasma. The biological function of PDGF-BB is to stimulate the

division and proliferation of the cells through binding to receptors on the cell surface.

Studies showed41-44 that PDGF is not one molecule but three, each a dimeric combination

of two distinct but structurally related peptide chains designated A and B. Dimeric

isoforms PDGF-AA, AB and BB are differentially expressed in various cell types and

their effects are mediated through two distinct receptors, named oc and P. Platelet derived

growth factor is considered a cancer biomarker because it is expressed at low or

undetectable levels in normal cells, but it is found to overexpressed in many tumor cells

and the autocrine and paracrine effects of PDGF-BB increase with the degree of malignancy.

45;46 In addition to cancer, PDGF has been implicated in other diseases, including renal

disease. There is a hypothesis that PDGF is important to cell transformation processes,

tumor growth and progression and thus could be used as a cancer marker for diagnosis.

The assumed biological foldings for PDGF-BB are shown in the Figure 1-5.

Aptamers for Platelet Derived Growth Factor-BB (PDGF-BB)

The first step towards the development of a novel 1VBA is the selection of an

aptamer. In 1996, Louis Green and his group47 USed the SELEX procedure for the

development of an aptamer for high affinity and selectivity against the B chain of PDGF-









BB. The first selection was initiated by incubating approximately 3x1014 mOleCUleS Of

random ssDNA with PDGF-AB in binding buffer. The two best ligands from this group

were identified by their relative affinities for PDGF-BB and its molecular variants over a

range of protein concentrations.












Figure 1-5. Assumed biological foldings of Platelet Derived Growth Factor (PDGF) 44
The biological molecule is the macromolecule that has been shown to be or is
believed to be functional.

In both cases, the affinity of ligands for PDGF-BB was higher than the affinity for

the other PDGF isoforms. The selected aptamers, 41t and 36t, shared a secondary

structure motif: a three-way helix junction with a three-nucleotide loop at the branch

point.(Figure 1-6) Two of the helices end in highly variable loops at the distal end from

the junction, suggesting that the regions distal from the helix junction are not important

for high-affinity binding to PDGF. The highly conserved nucleotides are indeed found

near the helix junction. The 36t aptamer is composed of 39 DNA bases with a

dissociation constant of 0.093nM. The other selected aptamer, 41t, is a 44-mer sequence

with a higher dissociation constant of 0. 129 nM.

Based on theoretical determinations, both sequences appear to have several

possible conformations before and after binding of the protein. Using this information,

the possibility exists to develop a new aptamer based detection scheme for PDGF. Also,

given that the protein has multiple isoforms, this makes the PDGF system particularly










attractive to explore because we can evaluate the different interactions of a single

aptamer with a family of proteins that have, in this case, ~60 % identity. In the next few

chapters, a systematic investigation of the PDGF-BB protein-aptamer system will be

presented. A variety of conditions have been evaluated for the effective recognition of the

protein as well as several transduction mechanisms studied for potential use in biological

studies.


C C

GGCA AG AG ,CA -A /C



GC
AST
COG


Figure 1-6. Folding of 36t molecular beacon aptamer (1VBA)

The scope of the work presented here was to conduct a systematic investigation of

a novel design for an aptamer-based biomolecule, used to detect and study a target

biomarker PDGF-BB. The combination of a nucleic acid molecule with different

fluorescence phenomena as signal transduction mechanisms will be investigated.

Potential applications of this probe in multiple assay environments and assay formats will

be explored and evaluated for homogeneous solution and in cellular samples.















CHAPTER 2
DEVELOPMENT AND CHARACTERIZATION OF A MOLECULAR BEACON
APTAMER FOR PLATELET DERIVED GROWTH FACTOR

Introduction

Proteins play very important roles in almost all functions of life. Molecular probes

for specific and sensitive detection of proteins and their molecular variants are necessary

in many biotechnology applications and biomedical studies. Monoclonal antibody-based

immunoassays have been used for protein analysis and studies during the past decades.

Most immunological methods and many nucleic-acid-based methods involve multiple

steps to achieve amplification of the specific signal produced by a target protein. By

attaching a fluorophore and a quencher to an aptamer, the high degree of sensitivity

afforded by fluorescent signals40;48;4 can be combined with the selectivity of the DNA

binding to the target protein to form a single-step molecular beacon aptamer (MBA)

assay.

Fluorescence changes upon the interaction between a fluorophore and a quencher

have been studied using molecular beacons (MBs) for DNA and RNA targets. Both, MBs

and MBAs share a common characteristic: they have similar inherent signal transduction

mechanisms and allows for target detection and reporting without removing for probe-

target complex from the sample solution (detection without separation). Some differences

between the two probes include the fact that MBs are mostly used for DNA and RNA

detection and are based on the hybridization of the loop sequence to the target sequence.

MB probes interact with their complement primarily through hydrogen bonding at









various sites, depending on the length of the hybrid. In contrast, the use of MBAs is

mainly used for the detection of protein targets and relies on a mixture of forces:

hydrophobic, ionic and hydrogen bonding, at only a few specific sites. The basis for

differences in secondary structures between the free and bound forms of MBAs for

protein targets is not well understood, however interactions with different proteins are

likely to result in secondary structure changes that could be used to produce FRET

signa s.1 4

Platelet derived growth factor is a dimeric protein for which several natural

molecular variants are known, and at least two natural variants of cell-membrane

receptors have been described with different specifieity for the PDGF variants.41-43 The

expression of variants of PDGF and the PDGF receptors have been implicated in

malignancy and developmental abnormalities."o"s The use of MBAs for protein targets is

in its infancy and the basis for differences in secondary structures between the free and

bound forms of MBAs for protein targets is not well understood. Theoretical models

predicted that when the aptamers selected for PDGF-BB are free in solution, several

stable conformations are possible. Based on the previously described MBA for thrombin

which forms a G-quartet after binding to the protein, the possibility existed for the PDGF

aptamer to undergo a conformational change after binding to the protein as a result of

stabilizing one of the predicted conformations. Initially, it was unclear whether the stem

structure would be open or closed prior to protein binding. Therefore, the first PDGF-BB

probe was designed as an anisotropy probe in an effort to explore the application of an

aptamer anisotropy probe for protein analysis. The PDGF-BB aptamer was labeled with a

single fluorescein dye at the 5' end.33 PDGF-BB in the nM range, when added to the









aptamer solution, caused an anisotropy increase due to the larger molecular weight of the

complex. The results were our initial attempts toward developing a PDGF-BB probe.

Given the possibility of conformational changes that could occur due to binding of the

protein, a distance dependent signal transduction mechanism was incorporated into the

aptamer molecule. The initial hypothesis was that a FRET-based aptamer probe could be

designed for the PDGF aptamer. In this chapter, the result of a systematic study of the

chemical and physical properties of a fluorescence quenching DNA aptamer probe will

be presented.

Materials and Methods

Synthesis and Purification of MBAs

All DNA molecules were obtained from Geno-Mechanix, LLC (Gainesville, FL)

and the general synthesis protocols are described in this section. The DNA molecules

were synthesized using the standard phosphoramidite chemistry in an Expedite 8909

automated DNA synthesizer on controlled pore glass beads, deprotected in ammonium

hydroxide and purified by gel filtration. Unmodified DNA aptamers were purified by ion-

exchange HPLC followed by gel filtration to remove salts. The standard MBA was

synthesized using DABCYL immobilized on controlled pore glass beads (BioSearch

Technologies, Novato, CA) and fluorescein (6-FAM) was added at the 5'-end using 6-

carboxyfluorescein phosphoramidite (BioSearch Technologies, Novato, CA). The dually

modified DNA was deprotected at 550C for 6 hr with ammonium hydroxide (50%) in

methanol. The deprotected DNA was first subj ected to gel filtration followed by two

cycles of reverse phase HPLC using two different hydrophobic resins with

triethylammonium acetate (pH 6.0, 0.05 M) buffer and eluted with an acetonitrile









gradient. The Einal purified material was subj ected to gel filtration to remove solvents,

dried under vacuum and stored dry at -200C until use.

MBAs with tetramethylrhodamine (TMR)-6-FAM pair was synthesized and

purified as described above using TMR-immobilized on controlled pore glass (BioSearch

Technologies, Novato, CA), except that deprotection was carried out in potassium

carbonate (0.1M) in methanol (80%) for 4 hr at 550C before processing the deprotected

dual labeled DNA. For the synthesis of MBA with Black Hole Quencher 2 (BHQ2)-

Texas Red pair, synthesis was done using BHQ2 immobilized on controlled pore glass

(BioSearch Technologies, Novato, CA) and a 5'-modification with a functional amine

group using a C3 spacer. Following deprotection at 550C with ammonium

hydroxide/methanol and gel filtration, the DNA was reacted with the N-

hydroxysuccinimidyl ester of Texas Red (Molecular Probes, Eugene, OR) in sodium

borate buffer (pH 8.2, 0. 1 M) and the labeled material was again subj ected to gel

filtration prior to the two cycles of reverse phase gel filtration as described above.

For the synthesis of MBA labeled with Cy3 and Cy5, the DNA was synthesized

with a functional amine group immobilized on controlled pore glass beads and Cy5 was

added to the 5'-end using Cy5 phosphoramidite (Amersham Bioproducts, Piscataway,

NJ). Following deprotection with potassium carbonate (0.1M)/methanol and gel filtration,

the DNA was reacted with N-hydroxysuccinimidyl ester of Cy3 (Amersham Bioproducts,

Piscataway, NJ) and the dual labeled material was purified as described for Texas Red

labeling above. All the MBA stock solutions were prepared at 100 CIM concentration in

Tris/HCI buffer (pH 7.5, 20 mM) with NaCl (20 mM), and stored at -200C in aliquots.

The DNA aptamer sequence in the standard 36tMBA used in most of the studies










described here is the one designated 36t by Green et.al (5'-CACAGGCTACGGCAC

GTAGAGCATCACCATGATCCTGTG-3 ') and contains a 6 base complementary

sequence at the 5' and 3'- ends that can form a double stranded stem in a closed

conformation.47

Recombinant human PDGF-BB, PDGF-AB, and PDGF-AA were purchased from

R&D Systems (Minneapolis, MN). PDGF and its isoforms were reconstituted in 4 mM

hydrochloric acid with at least 0. 1% bovine serum albumin (BSA). Other recombinant

human growth factors used in the selectivity experiments, epidermal growth factor

(EGF), insulin-like growth factor-I (IGF l), were bought from Roche (Indianapolis, IN).

Human hemoglobin (HEM), porcine lactic dehydrogenase (LDH), myoglobin (MYO),

chicken lysozyme (LYS), and human gamma-thrombin (THR) were purchased from

Sigma (St. Louis, MO). Other biomolecules used include: BSA, (New Englands Biolabs,

MA), thrombin (Haematologic, Inc, VT), ovalbumin (US Biological, Sawmpscott, MA)

and, glycogen (Fisher Biotech, USA).The binding buffer used for all the characterization

experiments was 20 mM Tris-HCI (pH 7.5) with 20 mM sodium chloride.

Standard Fluorescence Quenching Assay

All the characterization experiments for the MBA measurements were carried out

in a 100 Cll cuvette (Starna Cells). Standard fluorescence quenching assays were

performed by incubating the desired concentration of protein target with a 50 nM solution

of the MBA in binding buffer at room temperature in a final volume of 100 Cll. All

fluorescence measurements were monitored using a Fluorolog-Tau-3 spectrofluorometer

(Jobin Yvon Inc., Edison, NY) equipped with a thermostat accurate to 0.10C. The sample

cell was a 100 Cll cuvette (Starna Cells,Atascadero, CA). The fluorescence emission of 6-









FAM was monitored at the excitation maximum, 480 nm, and the fluorescence intensity

was measured at the emission maximum, 520 nm. Both excitation and emission slits were

varied to yield the best signals. The results are reported as mean values of triplicates.

The effect of temperature on MBA fluorescence was monitored using a Fluorolog-

Tau-3 spectrofluorometer (Jobin Yvon Inc, NY) equipped with a thermostat accurate to

0.10C. The sample cell was a 100 Cll cuvette (Starna cells, Atascadero, CA). The

fluorescence intensity of a 50 nm MBA solution in the binding buffer was monitored

before and after the addition of 200 nM PDGF-BB.

In order to adapt the fluorescence quenching assay to high throughput format, a

Tecan Saphire (Durhan, NC) fluorescence microplate reader was used with 96-well flat

bottom microtiter plates (Nalge Nunc International, Rochester, NY). All experiments

were carried in a final volume of 100 CIL, and the excitation and emission maxima were

selected according to the fluorophores used.

Results and Discussion

Aptamer probes have significant appeal in the development of new methodologies

for purification, labeling, and inhibition and characterization of proteins; however, their

use as molecular probes has been under development and advances made only in the last

5 years.8;52 One of our maj or goals for this research was to develop a single-step assay

which could analyze protein concentrations of distinct biomarkers in biological samples.

The initial efforts focused on the protein, PDGF-BB, for which two potential DNA

aptamer sequences for probe development have been selected.47 To characterize the MBA

and determine if the probe could operate in a typical biological sample, a variety of

chemical and physical parameters were evaluated.









Choice of Aptamer Sequence for MBA Development

Based on the predicted secondary structures, the first step toward developing a

FRET-based aptamer probe for PDGF was to determine if a secondary structure change

could be observed on binding to PDGF. To accomplish this task, two MBAs which are

based on the aptamers previously characterized,47designated 36t and 41t, were evaluated

(Table 2-1). Each aptamer was converted into an MBA by attaching DABCYL at the 3'-

end and 6-FAM at the 5'-end. Using thermodynamic software calculations, the aptamers

were predicted to form a helix-loop conformation with a single stranded region at the 5'

and 3' end as a possible conformation with a reasonable free energy in the absence of

PDGF.53;54 As a result of labeling the two ends of the aptamer, the fluorescence intensity

of the free aptamer sequence is expected to be fairly high since the fluorophore and

quencher pair will be spatially separated. Upon the addition of PDGF-BB, if protein

binding induces a conformational change in the stem structure of the aptamer a change in

the fluorescence intensity should be observed. In a fluorescence versus time experiment,

both aptamers were first monitored for approximately 300 s and then PDGF-BB was

added at four-fold excess (Figure 2-1). Upon PDGF-BB addition the fluorescence signal

decreases and reaches equilibrium rapidly. The quenching efficiencies were considerably

different between the two aptamer sequences. The reported binding affinities

demonstrated that the 36t aptamer has a higher affinity for PDGF consistent with the fact

that this aptamer showed a higher fluorescence quenching. The labeling of the aptamer

molecule with a fluorophore and quencher at each end of the oligonucleotide sequences

should not affect the binding capabilities of the aptamers since it was reported47that the

regions distal from the trinucleotide loop of both sequences are not critical for binding.

Since FRET is dependent on the distance between the fluorophore and the quencher, it










Table 2-1. Percentage of quenching of different consensus sequences of 50 nM of MBA
for PDGF aptamer with 200 nM of PDGF
Name Sequence % quenching
36t CACAGGCTACGGCACGTAGAGCATCACCAT 94
GATCCTGTG
41t TACTCAGGGCACTGCAAGCAATTGTGGTCC 55
CAATGGGCTGAGTA


_______~___1___


800000


* 41T
* 36T


j'
~600000



400000

o-
o

LI.


O 100 200 300 400 500 600 700 800
Time (s)
Figure 2-1. Fluorescence intensity changes versus time of different aptamer sequences.

may be possible that preferential binding of PDGF stabilizes, or closes, the stem thus

resulting in a decreased FRET response.

Based on the observed decrease in the fluorescence signal, the 36t and 41t aptamer

sequences appear to have a conformation change on binding to the protein. The PDGF

FRET aptamer is proposed to have a "stem-open" conformation prior to binding, due to

the high fluorescence signal that was observed (See Figure 2-2). Prior to binding one or

several structural conformations could be possible, and PDGF most likely stabilizes the

"closed-stem" structure for the 36t probe. Since the 41t aptamer had similar fluorescence

prior to binding and quite different from after binding, most likely the structure of the 41It









aptamer after binding is not a "stem-closed" structure but a partially-closed structure.

Also, it could be possible that due to the decreased affinity of the 41It sequence for PDGF,

the equilibrium fluorescence would be less in comparison to the 36t control aptamer.

However since the experiments were done in 4-fold molar excess of PDGF, this is not

likely to be the case because most of the aptamer probes should be bound. Another

possibility for the observed difference could be the modifications themselves. If single

fluorophore labeled DNA impurities existed within the solution, one might expect that

the degree of quenching would be less. As a result other synthetic batches were tested and

still similar results were obtained.






PDGF



A B
Figure 2-2. Proposed mechanism of PDGF-BB induced FRET responses for the 36t
probe. When the MBA is free in homogeneous solution, multiple
conformations are possible. Upon binding of PDGF-BB, the probe
conformation shifts from the opened (A), or partially opened (B), to a closed
conformation. The fluorescence signal, therefore, is high prior to binding; and
on stem formation, the signal is quenched.

Evaluation of Additional Fluorophore Quencher-Pairs

Since the 36t aptamer probe was clearly superior to the 41t in producing a

quenching response, the 36t probe was selected to further develop and characterize the

assay. One of the advantages of the MBA assay is that the use of FRET as a signal

transduction allows for a single step detection of the target without having to isolate the

MBA-protein complex from the solution. With FRET, alterations in the emission of









fluorescent labels are measured. An important consideration in the design of aptamer

probes for protein binding assays is the efficiency of energy transfer between the

fluorophore and quencher used to label the probes. The versatility of the assay would be

significantly improved if various known fluorophore-quencher pairs were found to be

more effective for PDGF detection. Measurements of quenching efficiencies of different

fluorophore-quencher pairs can be used to aid in the design of different kinds of MBAs.

In the next section, several MBA molecules carrying known FRET pairs were studied and

compared to the standard MBA labeled with 6-FAM-DABCYL pair. The following

pairs: Texas Red-BHQ2, Cy5-BHQ2 and 6-FAM-TMR were selected for their high

quenching efficiency in other systems. The synthesis and purification of MBAs with

FRET pairs is described in the Experimental Section and tested in the presence of a four-

fold molar excess of PDGF-BB in the standard fluorescence quenching assay.

As shown in Figure 2-3, all the selected FRET pairs were quenched when PDGF

was added, albeit to different degrees. Two of the most commonly used quenchers,

DABCYL and TMR, were paired against 6-FAM. Both quenchers may affect the

sensitivity and flexibility of FRET assays. DABCYL is a dark quencher that absorbs

broadly without emitting light; consequently, when paired with 6-FAM it can reduce

more than 90% of the original fluorescence intensity of the MBA with high sensitivity

and low background fluorescence. On the other hand, TMR is not a dark quencher and

contributes to an overall increase in background because of its own native fluorescence.

When both quenchers are paired against the same fluorophore for the MBA, DABCYL is

superior in terms of overall fluorescence quenching against 6-FAM.










5 MBA 5 MBA + PDGF-BB


1 .0



S0.6

S0.4




0.0
FAM-DABCYL FAM-TMR TexasRed- CY5-BHQ2
BHQ2
Figure 2-3. Evaluation of fluorescence quenching caused by PDGF-BB with MBAs
containing different fluorophore-quencher combinations. PDGF MBA
molecules labeled with different fluorophore-quencher pairs were incubated in
the standard fluorescence quenching assay buffer (final volume 100 C1L) with
a four-fold molar excess of PDGF-BB (300 nM) to the concentration of the
MBA (75 nM).

Black Hole quenchers were designed to maximize the spectral overlap with many

fluorophores and it does not re-emit the absorbed fluorescence as light, thus decreasing

the background fluorescence. BHQ-2 was capable of quenching more than 50 % of the

initial fluorescence intensity when paired with two commonly used dyes, Texas Red (TR)

and Cy5 with different extents. In this particular case, the Cy-5 labeled MBA resulted in

74% quenching effciency versus the Texas Red analog with only 54 %, even though

previous reports35 predicted that both dyes, Texas-Red and Cy5 should have similar

quenching effciency mediated by contact quenching. Possible reason for the difference

in quenching effciency includes the possibility that the effective closed conformation of

the MBA on binding to PDGF-BB is favorable for the FRET effciency of the Cy5

labeled probe.34 In addition, when the BHQ-2-Cy5 pair is in close proximity, the dye-pair

interaction actually increases the stem stability.55;56 Since FRET is a distance dependent









phenomenon, it could be possible to improve the performance of other fluorophore-

quencher pairs by changing the location of the label molecules in the MBA.57 The

availability of choices in the selection of fluorophore and quencher is an important part of

MBA-based assay development because various biological samples might present

potential interference at certain excitation-emission spectra that may be effectively

addressed by selecting the proper FRET pair. In addition, the development of a multiplex

assays for the simultaneous detection of multiple proteins in a single homogeneous

incubation may be possible by using two or more MBAs that are selective for different

targets and that are labeled with different fluorophore-quencher pairs. The results above

support the notion that MBAs are compatible with various fluorophores; however, the

aptamer sequence, conformation, and binding principle will all potentially affect the

FRET process. Each MBA that is developed would need to be fully characterized in order

to obtain the most effective position and dye-pair for labeling.

Once the MBA was developed with an appropriate fluorophore-quencher pair, we then

investigated the applicability of the MBA for the in vitro detection of PDGF-BB in

homogeneous solution by incubating a fixed amount of the 36t MBA with increasing

amounts of PDGF-BB in binding buffer. As can be observed in Figure 2-4A, a dose-

dependent fluorescence quenching was observed on the addition of PDGF-BB and 1.5

nM of PDGF was detected in repeated experiments. In comparison, an ELISA-based

PDGF-BB assay was reported to have a detection limit of ~15 pg/mL (0.6 pM) PDGF-

BB (Figure 2-4B). The almost 100-fold difference in detection capabilities is

understandable in this case.










50000
-000


40000

200


30000


50000



30000

20000

10000

0


50
PDGF (nM)


A


2.0


1.5


e 1.0
O

0.5


0.0 *
0


200


800


1000


600


PDGF (pg/mL)


Figure 2-4. Calibration curves for PDGF-BB. (A)Dose-dependent fluorescence
quenching of the MBA 36Gt-MBA on addition of PDGF-BB. Increasing
concentrations of PDGF-BB were incubated in the standard fluorescence
quenching assay buffer (final volume 100 uL) with the 36tMBA labeled with
Dabcyl quencher at the 3'- end and 6-FAM at the 5'-end (fluorescence
measurements were made at the excitation and emission maxima for 6-FAM)
(B) Calibration curve obtained with PDGF-BB ELISA kit (RnDsystems)









In the MBA assay, the fluorescence response is produced as the result of a single (or

possibly two as will be shown in later chapters) fluorophore molecule being quenched in

the presence of a single PDGF-BB molecule. Since the ELISA assays utilizes an enzyme

to effectively amplify the signals for each bound protein, clearly the sensitivity and

detection limits of such assays will be better. However, ELISA does present the difficulty

of being extremely time consuming (~7 h for one set of analysis). The time factor is

essentially attributed to the washing and incubation steps which allow for binding of

target molecules and the removal of unbound species. Since the MBA bioassay involves

only the mixing of two solutions, the MBA solution and the analyte solution, and a short

binding time, the analysis time is much shorter (~30 min for the MBA assay). As a result,

in terms of sensitivity, if one can sacrifice analysis time and the detection limits and

sensitivity are not an issue, then the MBA assay will provide a simple single step

capability for screening purposes.

Incubation Conditions for MBA Bionssay

Effects of temperature. A critical factor in the MBA-based bioassay is the

difference in distance between the quencher and the fluorophore in the free MBA (open

conformation) and the target-bound MBA (closed conformation). Physical factors that

affect DNA conformation and the distance between the FRET pairs are expected to

influence the dynamic range of the assay. It is unclear as to the exact mechanism for

PDGF binding to the 36t control sequence; however, it is believed that the opened, or

partially open, conformation is bound by the protein and then the stem is consequently

stabilized. At higher temperatures, the free MBA would preferentially assume a more

open conformation and increase the fluorescence signal. Conversely, at lower









temperatures, the equilibrium would be shifted to a more closed conformation. Both

instances could potentially influence the binding of PDGF by either inhibiting it or

increasing its ability to bind. In order to assess the effect of temperature on the MBA

probe and binding, we measured the fluorescence intensity of the MBA at different

temperatures under conditions used in the standard fluorescence quenching assay and

then compared 250, 30o and 37oC for binding to the PDGF aptamer.


Sno PDGF-BB with 200 nM PDGF-BB
1800000

< 1500000

'5 1200000

3 900000

a 600000

S 300000

F4 0
15 25 30 37 45 52 60 67

Temperature (0 C)
Figure 2-5. Effect of increasing temperature on fluorescence signal of free MBA in
solution. A solution of MBA (75nM) in standard fluorescence quenching
assay buffer as described in Materials and Methods was subjected to a gradual
increase in temperature (50C per minute), held at the indicated temperatures
for 5 min and the fluorescence signal was measured.

The results presented in Figure 2-5 show that as the temperature increases, the

initial fluorescence intensity increases presumably resulting from a larger separation of

the arms of the predicted three-way helix junction where the six-base pair double

stranded region would be formed. The largest fluorescence quenching was observed for









the MBA incubated at 370C, which is the standard physiological temperature. The initial

fluorescence intensity does not significantly increase above 370C. Since the aptamer

probe can potentially exist in solution as multiple conformations (i.e., random coil,

partially open, and closed structures), this particular assay offers the unique opportunity

to observe not only the specific closing of the probe due to PDGF binding but nonspecific

interactions that could result in the opening of the probe. As a result, experiments were

performed at 300C to monitor events that induce conformational changes that may result

in higher fluorescence intensities. This temperature also allows to monitoring PDGF-BB

induced quenching.

Effects of pH. Since the 6-FAM/DABCYL dye pair was chosen for assay

development, the pH sensitivity of fluorescein could potentially affect the overall

performance of the FRET based assay. As a result, the effect of pH on the fluorescence

intensity of the probe without target protein was investigated. The fluorescence intensity

of a 75 nM solution of the MBA in binding buffer used in the standard fluorescence

quenching assay was also measured in the microplate reader except that the pH was

varied between 5.0 and 10. The scope of this experiment was limited by the pH

dependence of fluorescence intensity of fluorescein. The results (Figure 2-6) showed that

within a pH range of 6.5 to 9.5, the fluorescence intensity of the MBA is within + 10% of

that observed at pH 7.5. The pH-dependent intensity profile of the MBA correlated with

the pH-dependent spectra of fluorescein;5 consequently, the observed trend is believed to

be due to the fluorescence characteristics of the dye molecule rather than conformational

changes leading to quenching. Even though fluorescein is a highly pH dependent

fluorophore, the optical properties should not affect the measurements of the FRET based












50000


S40000 -


S30000


rl20000








4.5 5.5 6.5 7.5 8.5 9.5 10. 5

pH

Figure 2-6. Effect of increasing pH on fluorescence signal of free MBA in solution.
Fluorescence intensity measurements of a solution of MBA (75nM) in binding
buffer with varying pH conditions.

assay because the pH conditions during the experiments are similar to the optimal range

for fluorescein.47

Effects of ionic strength. Certain monovalent and divalent cations commonly

encountered in biological specimens are known to affect DNA conformation. A series of

experiments to study the effects of varying concentrations of two divalent cations

(magnesium and calcium) and two monovalent cations (sodium and potassium) on the

fluorescence intensity of the free MBA in solution were performed. The results are

presented in Figures 2-7 and 2-8. Each of the divalent cations caused a concentration-

dependent decrease in the fluorescence intensity (Figure2-7) and at more than 1 mM











1.2 1 Mg2+
u, H Ca2+







0.4

0.2


0.0 0.5 1.0 1.5 2.0 2.5 3.0
ion concentration (mMI)

Figure 2-7. Effect of divalent cations on fluorescence signal of free MBA in solution.
Increasing concentrations of divalent cations (CaCl2 and MgCl2) were added
to a solution of MBA (75nM) in the standard fluorescence quenching assay
buffer and the fluorescence signals were measured.

concentration, the extent of quenching was comparable to that obtained with an excess of

PDGF-BB. The results were not surprising since divalent cations stabilize secondary

structures in DNA molecules and would thus shift the equilibrium to a closed

conformation of the MBA.Monovalent cations sodium and potassium caused only a small

initial decrease in fluorescence intensity (Figure 2-8) and even at concentrations of more

than 50 mM the extent of quenching by the ions was less than 13% as compared to more

than 90% quenching obtained with a 4-fold molar excess (300 nM) of PDGF-BB in a

standard fluorescence quenching assay. Since typical combined concentrations ofNal

and K1+ in cell culture medium is between 150 mM to 160 mM, desalting columns can be

used to remove monovalent and divalent cations and prevent potential contributions

towards quenching of the probe.










1.2 1 K+ Na+





e 0.8



mo 0.6

-0.4






0 100 200 300 400 500 600

ion concentration (mM)

Figure 2-8. The effect of monovalent cations on fluorescence signal of free MBA in
solution. Increasing concentrations of monovalent cations (NaCl and KC1)
were added to a solution of MBA (75nM) in the standard fluorescence
quenching assay and the fluorescence signals were measured.

Selectivity of the MBA-Based Fluorescence Quenching Assay

Once the aptamer was selected and tested against PDGF for fluorescence

quenching on binding, the selectivity of the MBA was examined. We incubated 20 nM

MBA with 100 nM of either PDGF-BB or one of several extracellular proteins (bovine

serum albumin, hemoglobin, lactate dehydrogenase, lysozyme, myoglobin, and thrombin)

or unrelated growth factors (such as epidermal growth factor and insulin-like growth

factor-1). The results shown in Figure 2-9 indicated that only pure human PDGF-

BB causes a marked reduction in fluorescence signal. All other proteins tested failed to

cause any significant change in fluorescence, even at 10-fold higher concentrations than

that used for PDGF-BB.













3 0.8

cn 0.6-

.= 0.14




-0.2







Figure 2-9. Binding selectivity of the MBA. Relative fluorescence signals of incubation
mixtures containing MBA (20 nM) and one of the following proteins (200
nM): bovine serum albumin (BSA), hemoglobin (HEM), lactate
dehydrogenase (LDH), lysozyme (LYZ), myoglobin (MYO), thrombin
(THR); or one of the growth factors, epidermal growth factor (EGF), basic
fibroblast growth factor (bFGF), insulin-like growth factor-1 (iGFl), or
PDGF-BB.

In a subsequent experiment, some of the previously tested biomolecules including,

growth factors (EGF), common proteins that are abundant in biological specimens

(thrombin, ovalbumin and serum albumin) and glycogen (an abundant natural

polysaccharide in tissues) were tested individually in a dose dependent manner. Dose-

response curves were obtained by incubating increasing amounts of each biomolecule in

a standard fluorescence quenching assay. The results are presented in Figures 2-10A and

2-10B. The small dose-dependent decrease in fluorescence due to thrombin and EGF

indicated a slight interaction with the MBA probe. In comparison to similar






41




BSA Ovalbumin = Glycogen


60000



gj40000



~20000

L.


5 10
biomolecule concentration (pig/mL)

A


60000


+ EGF m Thrombin PDGF-BB


~40000


2000
O'
LI.

Ocnn


50
protein concentration (nM)


Figure 2-10. Fluorescence quenching displayed by the MBA in response to increasing
concentrations of PDGF and other biomolecules. In a typical fluorescence
quenching assay containing MBA (75 nM) in Tris (pH 7.5, 20 nM) and NaCl
(20nM) as described in Materials and Methods, fluorescence signal was
measured following the addition of increasing concentrations of various
biomolecules: [A] BSA, ovalbumin, glycogen; and [B] PDGF-BB, epidermal
growth factor, thrombin.


]'5 *









concentrations of PDGF, the final extent of quenching was substantially lower than the

quenching observed for PDGF. Conversely, minor changes in fluorescence intensity were

observed for unrelated, commonly occurring biomolecules (bovine serum albumin,

ovalbumin and glycogen, up to 12 ug/mL). Since only PDGF was capable of producing a

large change in fluorescence signal in comparison to other proteins tested individually at

the respective concentrations, the assay should be selective and sensitive for PDGF

detection in the presence of other proteins commonly found in biological specimens.

Fluorescence Quenching Assay Distinguishes Molecular Variants of PDGF

Green et al. reported that the PDGF aptamers selected for binding to PDGF-BB

by using isotropic labeling bound to three molecular variants of PDGF (Figure 2-11i),

namely BB, AB, and AA, albeit with different affinities, presumably because of the

amino acid sequence homology (60%) between the A and the B chains. We tested the

three common PDGF variants for their effects on the fluorescence quenching assay.

Serial dilutions of each protein were incubated with 10 nM MBA in binding buffer and

the dose-response curves for fluorescence quenching were compared. Parallel sets of

experiments were conducted with varying amounts of the MBA. The calibration curves

obtained with 10 nM MBA are shown in Figure 2-12.

The results presented in Figure 2-12 indicated that the slopes and the concentration

of protein required to attain 50% of the maximum quenching are distinct for the three

molecular variants of PDGF. Although this data indicates that the MBA response shown

for PDGF-AB is more similar to that for PDGF-BB (in comparison to the response for

PDGF-AA), the difference between PDGF-AB and PDGF-BB is much more easily

distinguishable than that reported method.47










AA AS


Figure 2-11. Cartoon representing PDGF-BB and its molecular variants (PDGF-AA and
PDGF-AB) and its receptors alpha (a) and beta (P).


350000
Denatured PDGF

~v280000 *I

S210000

140000 A
9 a ". .PDGF-AA
70000
E PDG-BB DGF-AB

0 5 10 15 20 25 30
PDGF (nM)

Figure 2-12. Dose-response curves of PDGF variants: fluorescence signals of MBA for
PDGF-AA (A), PDGF-AB (m), PDGF-BB (*), and denatured PDGF-BB (+).
The concentration of the MBA was 10 nM.

When PDGF-BB is reduced by dithiothreitol and denatured with sodium dodecyl

sulfate, the resulting protein fails to cause any fluorescence quenching. Also, based on the

slopes of the linear portion of the serial dilution curves it is clear that the binding affinity

of each isoform is different for the aptamer probe. The results collectively indicated that

the fluorescence quenching assay is not merely dependent on the primary sequence of the










protein chains, but is capable of distinguishing conformational characteristics or the

folding of highly related protein molecules.

Conclusions

For the development of a molecular beacon aptamer probe that will successfully

report the binding event to a target protein in a single step, a signal transduction

mechanism must be incorporated into the recognition molecule. Theoretical calculations

using commercially available software showed that when the selected aptamers for PDGF

are free in homogeneous solution, one possible conformation was the spatial separation

of the 5' and 3' end.53;54 Taking advantage of the inherent structural properties of the

aptamer, a distance dependent signal transduction mechanism, such as FRET, was

incorporated to create a probe for the detection and study of PDGF-BB. Once the aptamer

binds PDGF-BB, the distance between the ends of the aptamer molecule was reduced,

resulting in the formation of a 6-base long stem.

Also, the results obtained with varying incubation conditions described, in this

chapter, support the notion that MBA-based FRET may be used in assays for protein

detection in biological specimens since the FRET assay is effective at physiological pH

and temperature. Divalent cations show interference with the assay by causing quenching

in the absence of PDGF target probably due to a stabilizing effect on the stem of the

probe. The results offer us new insights into, and justify the need for, future development

and optimization of this one step MBA-based FRET assay. It was also demonstrated that

three highly related molecular variants of PDGF (AA, AB, and BB dimers) can be

distinguished from one another. In comparison to ELISA, the quenching assay developed

is less sensitive, detection limits are higher, and has a higher throughput. Also, based on

product literature and experimental controls for the ELISA assay, cross reactivity of the









PDGF-BB antibody is prevalent for the AA isoform, which is often the case for antibody

assays. As a result, depending on the information and degree of scrutiny needed from a

given experiment one method may be preferred over the other.

The use of fluorescence quenching as a measure of binding between the aptamer

probe and the target protein eliminates the potential false signals that may arise in

traditional fluorescence enhancement assays as a result of degradation of the aptamer by

contaminating nucleases. Utilizing the conformational change of the aptamer to signal

protein binding may be applied to essentially all proteins, especially those that do not

naturally bind to DNA. Finally, other DNA aptamers with desirable selectivities for

chosen target proteins can be synthesized and coupled to different fluorophore-quencher

combinations to allow simultaneous detection of multiple target proteins in the same

solution at different excitation/emission wavelengths or, conversely, a multiple-well

microtiter plate can be used for many distinct samples for monitoring the same protein.















CHAPTER 3
IMPLICATIONS OF USING A MBA INT A FRET BASED AS SAY FOR PDGF
STUDIES INT BIOLOGICAL SAMPLES

Introduction

The development of a bioanalytical method includes the characterization of all

procedures and parameters that demonstrate that a particular method can be used for

quantitative measurements of analytes (in this case PDGF). The performance

characteristics of the method should be suitable, reproducible and reliable for the

intended analytical applications. In the previous chapter, detailed characterization

experiments revealed that the developed MBA can detect nanomolar amounts of PDGF in

a homogeneous solution, and that the optimal conditions for the MBA-PDGF complex

formation are compatible with the conditions typically found in a biological sample. The

next step in the development of the MBA-based bioassay will be focused on the

implications of using this aptamer probe to study and detect PDGF in cellular samples.

The aptamer selected for PDGF, was reported to have a 700 fold higher affinity for

PDGF versus other random oligonucleotide sequences.47 This characteristic is especially

important for the detection of PDGF in the presence of a complex biological sample.

Nonetheless, in a biological sample such as cell culture medium, many other species may

be present that could potentially interfere with the binding and the subsequent monitoring

of the fluorescence change on binding. It has been reported, that PDGF and PDGF-like

growth factors can be over-expressed and secreted into the media of various types of

cancer cells; including MCF-7,46MDA-MB-23 1,5o and, PC3 cancer cell lines.59;60 In










addition to PDGF, high levels of other proteins typically encountered in cell cultures may

non-specifieally bind to the probe, induce conformational changes, and as result produce

false positive quenching. Also, many species in biological samples may present optical

interference for fluorescence measurements, due to absorption and auto-fluorescence.

The above issues and others need to be carefully considered and explored before

attributing the total signal change solely to PDGF binding. As a result, two directions

were concurrently taken. One specifically designed to address non-specific interactions of

the aptamer and the other addresses signal differentiation of PDGF responses and cellular

interference.

This chapter will be aimed towards the determination of the robustness of the

fluorescence quenching assay against other species present in the biological samples.

Specifically, the incorporation of sample processing prior to carrying out the assay to

remove potential biological interference was investigated to minimize the effect of the

factors mentioned above. The first part will include studies with simulated biological

samples containing PDGF. For the second part, cancer cell conditioned media will be

used in the first attempt to detect PDGF in biological samples using a MBA.

Materials and Methods

Instrumentation

Fluorescence was monitored using a Fluorolog-Tau-3 spectrofluorometer (Jobin

Yvon Inc., Edison, NY). The sample cell was a 100 Cll cuvette (Starna Cells, Atascadero,

CA). The fluorescence emission of 6-FAM was monitored at the excitation maximum,

480 nm, and the fluorescence intensity was measured at the emission maximum, 520 nm.

Both excitation and emission slits were varied to yield the best signals. For high

throughput format assay, a Tecan Saphire (Durhan, NC) fluorescence microplate reader









was used with 96-well flat bottom microtiter plates (Nalge Nunc International, Rochester,

NY). All absorbance measurements were carried out in a Cary 300 UV-Vis

spectrophotometer (Varian, Palo Alto, CA).

Preparation of a Simulated Biological Sample

A simulated biological sample (SBS) was prepared by combining Dulbeccos

Modified Eagle's Medium (DMEM, Mediatech, Hendon,VA) with serum proteins from

fetal bovine serum(Invitrogen, Carlsbad, CA). SBS will be used as an intermediate

between a homogeneous solution and a complex biological specimen, such as cell culture

media, to investigate the applicability of using MBAs for protein detection. Total protein

content was determined using the Bradford Protein Assay Kit61 (Biorad, Hercules, CA).

Cell Culture and Preparation of Conditioned Media

The cell lines used were human breast carcinoma MDA-MB-231, and murine

BALB/c-3T3 fibroblasts (American Type Culture Collection,Manassas, VA), were grown

in DMEM supplemented with 1 % Gentamicin (Sigma, Saint Louis,MI) and either 10 %

fetal bovine serum for MDA-MB-231 or 10 % calf serum (Invitrogen,Carlsbad,CA) for

BALB/c-3T3. Prostate cancer cells, PC-3(American Type Culture Collection,

Manassas,VA), were grown in R12K medium (American Type Culture Collection,

Manassas,VA) supplemented with 10 % fetal bovine serum and 1 % Gentamicin. The

cells were incubated in humidified air containing 5 % CO2 at 370C. For the collection of

serum-free conditioned media, the culture medium of confluent cultures was replaced

with serum-free DMEM when the cell monolayers reached 80-90 % confluence. After

approximately 60 minutes, the medium was replaced with fresh serum-free DMEM and

the cells were incubated for an additional 24 hours. The serum-free CM was collected

and clarified by centrifugation. Acetic acid was added to the supernatant to a final









concentration of 0. 1 M and the material was lyophilized. The lyophilized powder was

resuspended in 0.1 M acetic acid, clarified by centrifugation at high speed in a microfuge

for 5 minutes, and the supernatant containing the soluble proteins was subj ected to size

exclusion chromatography utilizing a NAP-5 (Amersham Biosciences, Piscataway, NJ)

column packed with Sephadex G-25 in 0.1 M acetic acid. The excluded volume fractions

with the highest absorbance at 280 nM were collected and lyophilized again. The protein

preparation was resuspended in binding buffer (Chapter 2) to maintain appropriate pH

conditions and used as the stock solution from which dilutions were made for incubation

in the fluorescence assay.

Results and Discussion

Use of MBA in Biological Sample

A SB S sample was used to determine the potential application of this one-step

fluorescence-quenching assay for PDGF-BB detection in the presence of low levels of

serum proteins. In the first case, the sample was lyophilized and split in two equal

portions, one of which (Sa) was additionally supplemented with 100 nM recombinant

human PDGF-BB (the other was labeled Sb and did not contained PDGF). Each portion

(Sa and Sb) was resuspended in 0.1M acetic acid and the protein components were

collected using a gel filtration column packed with Sephadex G-25 to remove salts and

small molecules. As was mentioned on Chapter 2, high levels of monovalent and divalent

cations resulted in background quenching and need to be removed prior analysis.

Fluorescence quenching activity of the two preparations was compared by adding 50 nM

MBA. As shown in Figure 3-la, Sa caused a marked reduction in fluorescence,

comparable to that observed with pure human PDGF-BB (see Figure 2-4A), and Sb

caused a small fluorescence decrease.33 Analytical denaturing and reducing










polyacrylamide gel electrophoresis of Sa and Sb showed that the low protein

compositions of the two samples were indistinguishable from each other, which may

indicate that the differences in fluorescence quenching presented by both samples was

due to the PDGF-induced conformational change. Nonetheless, 20 % of quenching in the

nonspiked sample may indicate nonspecific binding of other proteins in the sample. It

is conceivable that as a result of binding to the MBA a reduced fluorescence is observed

from the sample, as seen in Chapter 2 for thrombin and EGF. The results do, however,

demonstrate that, when in the presence of similar total protein concentrations, the MBA

has the potential to qualitatively distinguish the presence of nanomolar quantities of

PDGF-BB in SBS.

The next sample to be examined for PDGF detection was conditioned media from

cultured cells. Conditioned media (CM) is a general term to describe media in which cells

have been cultivated for a period of time. It contains many mediator substances that were

secreted by the cells which were produced in this medium. The mediator substances

contain growth factors, such as PDGF, and may promote the growth of new cells. A

human breast carcinoma cell line, MDA-MB-231, has been reported to secrete PDGF-BB

in culture medium.'" Serum-free CM was collected from human HTB cells and normal

murine BALB/3 T3 fibroblast cells in culture. The samples were collected and processed

as described in Materials and Methods and serial dilutions of protein preparations from

each cell line were incubated with a fixed amount of the MBA and fluorescence

measurements were obtained to construct a dose-response curve. Parallel sets of dose-

response curves were obtained by using different amounts of the MBA for each such set.
































HTB cells



BALB cells






PDGF serial dilutionl
i M~~ralilegn


300000




S200000


o 1000


O C~


MB~A


MBA+Sa


M/BA+Sb


350000

300000

250000 -

200000

150000

100000


50000

0


Log Vd ------


Figure 3-1. Use of molecular beacon aptamer in biological samples.(A) Fluorescence
signals of incubation mixtures of MBA (50 nM) with serum proteins without
(Sb) or supplemented with (Sa) PDGF-BB, as compared to the MBA alone.
(B) Dose-response curves/calibration curves for conditioned media and PDGF
standard samples: fluorescence signals of mixtures of MBA (10 nM) with
serial dilutions of protein preparations of conditioned cell media from HTB
cells (triangles) or BALB cells (squares). Serial dilutions of a 500 nM solution
of PDGF-BB (circles) were used as a control. Vd is the dilution factor.33









For each set, a control dose-response curve for fluorescence quenching was obtained with

serial dilutions of a solution of pure human PDGF-BB. The results obtained with 10 nM

MBA are presented in Figure 3-1B.33 Typically, several factors may indicate the presence

of PDGF or PDGF-like molecules when the dose-response curves from the CM samples

are compared with the one obtained using standard PDGF-BB. The factors include a

concentration-dependent fluorescence intensity decrease, the similarity of the slope, and

the final extent of quenching. For HTB conditioned media, the final extent of quenching

as well as the dose-dependent decrease in fluorescence is comparable to the standard

PDGF-BB .

In the case of BALB conditioned media, the final extent of quenching is lower than the

one obtained with HTB cells and standard PDGF-BB. It is important to mention here that

initially BALB cells were selected as a negative control since it has not been reported that

this particular cell line over-expresses PDGF. The observation that serial dilutions of

BALB conditioned media produced a dose dependent quenching in some way similar to

the standard PDGF-BB reveals that this cell line may have measurable levels of PDGF

(or PDGF-like molecules) or high levels of interfering proteins that bind non-specifically

to PDGF and produce a conformational change of the MBA. Although the final extent of

quenching for both samples was different indicating that the total amount of PDGF is

different in each case, the use of MBA for PDGF quantitation and detection in cancer cell

media samples may not be reliable due to the presence of interfering species in the

biological environment that may lead to false positive results. An additional contribution

to the change in fluorescence intensity may be due to filtering effects in the conditioned

media that masked the true fluorescence signal.









Implications of Total Protein Content on the MBA Assay

The MBA appear to have the ability to qualitatively discriminate between cellular

samples that contain PDGF; however, in order to verify the reliability of the results, the

potential interfering compounds that could exist within cellular samples needs to be

explored. For protein assay development, it is critical to determine the tolerable amount

of protein that would allow a reliable fluorescence measurement. Protein aggregation and

non-specific protein binding may have significant implications in the formation of the

MBA-PDGF complex.

Two samples of protein mixtures containing a total of 10 Clg/mL and 100 Clg/mL

were prepared in binding buffer to study solely the protein effect on the MBA binding by

combining the following representative biomolecules: bovine serum albumin, albumin,

ovalbumin, lyzozime, glycogen, myoglobin, thrombin and a few growth factors

including: insulin-like growth factor (IGF), epidermal growth factor (EGF) and vascular

endothelial growth factor (VEGF).

In the first case, a solution containing 10 Clg/mL of total protein was incubated with

50 nM of MBA in binding buffer and compared with a solution of 10 Clg/mL of pure

PDGF-BB incubated with 50 nM MBA. The results presented in Figure 3-2 revealed that

10 Clg/ml of a protein mixture was capable of causing a 17 % quenching of the MBA.

PDGF was added to the sample the fluorescence intensity drastically decreased, up to

96 %, indicating that PDGF is necessary for a complete quenching of the MBA at this

protein concentration. In a separate experiment, the total protein content of the sample

was increased by an order of magnitude. With 100 Clg/mL in the sample, 76 % of the










700

600

500

g 400

S 300

S200

0 100


MBA 10 ug/mL 10 ug/mL 100 ug/mL 100 ug/mL
PDGF-BB serum serum serum
proteins proteins proteins and
PDGF-BB

Figure 3-2. Molecular beacon aptamer fluorescence quenching caused by a protein
mixtures containing 10 Clg/mL.

fluorescence of the MBA is quenched indicating that non-specific binding may be

responsible for the decrease in fluorescence intensity. Nonetheless, the addition of 10

Clg/mL of pure PDGF-BB in the solution further reduces the fluorescence intensity of the

aptamer probe.

Based on the above results, a plausible explanation for the observed fluorescence

quenching in cellular samples is non-specific binding. Thus, the dose dependent

fluorescence decrease observed in cellular samples can not only be attributed to the

presence of PDGF but to a combination of specific and nonspecific binding of proteins to

the aptamer probe. Another alternative includes that the the MBA may be recognizing

similar binding sites in other growth factors that may have a degree of identity with

PDGF. In order to be able to use the PDGF aptamer in an assay for protein detection,

alternative strategies for sample clean-up before analysis needed to be developed.










Sample Pre-Treatment

The development of recognition molecules for any type of assay should be

experimentally convenient, sensitive and easy to perform in a short period of time. The

developed MBA meets the characteristics for homogeneous solution analysis.

Unfortunately, the response of the aptamer may be compromised by the presence of

excess proteins in solution. Having the protein of interest in a biological environment is

somewhat difficult for reliable detection by DNA-based methods. Based on the data

presented in the last section, a mixture with total protein content higher than 100 Clg/mL

in a 50 nM MBA solution may lead to false positive results due to non-specific binding.

For this reason, the protein of interest needed to be isolated or the sample itself needs to

be partially cleaned-up in order to obtain more reliable results. The ability to study

proteins in mixtures with complete control of their environment (salts, temperature, pH,

total protein content, etc.) is critical for the development of an in vitro assay using the

PDGF-MBA.

To simplify the sample before analysis, one must begin with the starting material

(in this case cellular CM) and fractionate it using any one of a large number of physical

or biochemical approaches: centrifugation, solid precipitation, binding to affinity

columns or separation by sizing columns (gel filtration chromatography). In this case,

size exclusion chromatography was used. After the cell medium was collected, clarified

by centrifugation and lyophilized, it was run through a commercial NAP-5 or in a G-100

column to desalt the samples and remove low molecular weight protein components. In

gel filtration chromatography, the stationary phase consists of porous beads with a well-

defined range of pore sizes. The stationary phase for gel filtration is said to have a









fractionation range, meaning that molecules within that molecular weight range can be

separated. Proteins that are small enough can fit inside all the pores in the beads and are

said to be included. The small proteins have access to the mobile phase inside the beads

as well as the mobile phase between beads and elute last in a gel filtration separation.

Proteins that are too large to fit inside any of the pores are said to be excluded. They have

access only to the mobile phase between the beads and, therefore, elute first. Proteins of

intermediate size are partially included meaning they can fit inside some but not all of

the pores in the beads. The proteins will then elute between the large ("excluded") and

small ("totally included") proteins. For our experiments we tested two different packing

materials: Sephadex G-25 (PDGF will be included in the pores) and Sephadex G-100

(PDGF will be excluded from the pores).

A control SBS was prepared to mimic conditioned media collected from tissue

cultured cells that are commonly used to study the expression of proteins by normal and

deceased cells. The sample contained DMEM supplemented with serum proteins and no

PDGF-BB. It was processed by lyophilization to reduce the volume, diluted to 1 mL with

0.1 M acetic acid, loaded into a NAP-5 column equilibrated with acetic acid, and

collected at 0.5 mL fractions with 0.1 M acetic acid running buffer. After lyophilization

and reconstitution in the binding buffer, each fraction was analyzed using the standard

fluorescence quenching assay. Each fraction contained ~16.5 Clg/ml on average of serum

proteins, as calculated with the Bradford assay. The results for the control SBM (Figure

3-3 A) showed that Fractions 6 to 16 caused different extents of quenching even though

PDGF-BB had not been added.










50000

40000 -


~30000

~20000

10000



2 4 6 10 12 14 16 MBA
Fraction number

A


60000




S40000




S 20000






MBA Fraction 12 50 ng PDGF 250 ng PDGF


B
Figure 3-3. Fluorescence quenching caused by PDGF in the presence of proteins in a
simulated biological specimen. Protein fractions were obtained from a
Sephadex G-25 gel filtration of acetic acid soluble materials in a simulated
biological specimen as described in Materials and Methods and each fraction
was resuspended in 130 ul of a stock buffer Tris (pH 7.5, 10mM) with NaCl
(20 mM). (A) 65 ul of each fraction indicated on the X-axis was incubated in a
standard fluorescence quenching assay (final volume 100 ul) with the MBA
(75nM) and the resulting fluorescence was measured. (B) Indicated amounts
of PDGF-BB were added to a dilution of Fraction 12 with serum-derived
proteins and incubated in a standard fluorescence quenching assay.62









Based on repeated experiments conducted with PDGF-BB in buffer solution, it was

determined that fractions 11 to 13 should contain PDGF-BB if added to the control SBS.

Given that in fraction 6 there is ~50% quenching without the presence of PDGF-BB, gel

filtration does appear to minimize the interference by removing them from the PDGF-

BB containing factions. To investigate this possibility and establish the limit of detection

for the PDGF assay in the presence of serum proteins, varying amounts of PDGF-BB

were added to fraction 12 of the control SBS. The results (Figure 3-3 B) showed a

detectable decrease in fluorescence on the addition of as low as 50 ng of PDGF-BB.62

This data shows that the current assay can detect as little as 50ng/16.5 Clg of serum

proteins. This leads us to believe that for 1VBA applications to be effective in biological

samples the model system chosen for any protein of interest, should adequately express

the protein at high enough levels to allow for the detection of the target in the presence of

typical amounts of serum proteins.

We further tested the reliability and effective range of operation for the

fluorescence quenching assay by trying to improve the sample clean-up procedure using a

Sephadex G-100 gel filtration column. For the G-100 packing material, the solid phase

has larger pores and, as a result, retains PDGF in the column longer. This property should

improve the exclusion of interfering impurities from PDGF containing fractions. Also, a

direct comparison between an ELISA analysis and IVBA assay was conducted. The

ELISA not only allowed for the identification of the distinct fractions for which PDGF

was eluted, but also, enabled us to compare the quenching results with the PDGF-BB

responses in the presence of serum proteins. For the next experiment, a buffer sample

containing only 50 Clg/mL PDGF was used. A Sephadex G-100 gel filtration column was









equilibrated with 0.1 M acetic acid and the sample loaded. Then, 0.5 mL fractions were

collected as the sample was eluted. To determine the fraction(s) that contained PDGF-

BB, each one of the collected 0.5 mL fractions was tested using a commercial ELISA kit

and PDGF was found in fractions 19 to 22. Using this information, we ran a SBS spiked

with PDGF-BB in the Sephadex G-100 gel-fi1tration column and tested fractions 11 to 25

with the fluorescence quenching assay. Similar to the previous results obtained with the

NAP-5 column (Figure 3-3A), several fractions exhibited different degrees of quenching

when analyzed with the fluorescence quenching assay (Figure 3-4). The quenching assay

indicated a significant amount of quenching in the fractions 17 to 22 after a 2-fold

dilution of the original sample which was required for assay analysis. To determine the

amount of PDGF in each fraction of the spiked SBS sample, fractions 7 to 18, 19 to 20,

and 21 to 22 were pooled and analyzed using ELISA. PDGF-BB was found in samples 19

to 22 as expected based on the spiked ELISA experiments (Table 3-1). Even though

fractions 17 to 18 did not contain PDGF-BB based on the ELISA experiment, a large

degree of quenching was still observed for the IVBA assay. The results further suggest

the need for sample pre-treatment in order to minimize the interfering proteins such as

those found in fractions 17 to 18.

Unfortunately, it is still not understood as to whether the amount of quenching is

due to PDGF versus non-specifie binding to the probe. Interestingly, it was also observed

that when the total protein concentration in individual fractions that did not contain

PDGF from the two gel-fi1tration columns studied here was higher, the extent of

quenching was higher. This finding points out the possibility that dilutions could be










performed to minimize the extent of perceived interference. A control sample that did

not contain PDGF was made and processed as before.

No PDGF was found in fractions 19-22 using ELISA. Since fractions 21-22

contained the highest amount of PDGF-BB, a side-by-side comparison of the PDGF and

non-PDGF containing fractions was conducted with the MBA assay (Figure 3-5).

Table 3-1. Analysis for PDGF-BB spiked samples after fractionation on a G-100
Sephadex column using enzyme-linked immunosorbent assay (ELISA)
Eluted Fraction # I g/mL PDGF
7-18 0.8
19-20 2.4
21-22 3.0


50000-

S40000

S30000

S20000



S10000





fraction number

Figure 3-4. Fluorescence intensity of fractions of simulated biological sample (SB S)
collected from G-100 column. Each fraction was 0.5 mL. After lyophilization
samples were resuspended in binding buffer for the fluorescecence quenching
assay. MBA concentration was 50 nM.

Both fractions were adjusted to the same total protein concentration and the

fluorescence quenching was monitored with the MBA assay. As shown in figure 3-5, the

measured florescence intensity of the combined fractions spiked with PDGF-BB was

more than the un-spiked control. As the dilution factor increases, the corresponding










fluorescence intensities increased for both samples. This indicated that the cause of

quenching in the control sample was capable of being diluted. To determine if PDGF

could interact with the aptamer to cause complete quenching of the probe in the samples,

an excess of PDGF-BB (250 ng) was spiked into the control samples after processing.

The aptamer appeared to respond and bind to this addition of PDGF even in the presence

of the serum proteins regardless of the dilution factor. The results were consistent with

the above results and told us that with dilutions and gel-filtration PDGF-BB could be

monitored in the presence of serum proteins. The amount required to obtain full

quenching of the probe would, however, be significant considering the sample would

need to be diluted. Upon a 4-fold dilution of the total proteins, the aptamer appeared to be

unaffected by the serum proteins and capable of distinguishing between the control

sample with and without PDGF.

Spiked with pdgf-bb H no pdgf-bb added O pdgf-bb added to spl 21
50000


S40000


S30000


S20000


~10000


O -
2 fold 4 fold 6 fold 8 fold 10 fold buffer MBA MBA +
PDGF

Figure 3-5. Fluorescence intensity of dilutions of SBS. Green columns represent fraction
spiked with an additional 200 nM PDGF-BB post processing. MBA
concentration is 50 nM.









Conclusions

The results presented in this chapter were used to explore the applicability of the

PDGF-MBA probe to monitor PDGF secretion from cancer cells in cell media. Initially,

the assay showed the ability to detect PDGF in spiked biological samples containing low

levels of serum proteins. The conditioned media from two cell lines, which express

different levels of PDGF, was analyzed with the assay. For HTB conditioned media, the

final extent of quenching as well as the dose-dependent decrease in fluorescence

illustrated a noticeable increase in the PDGF concentration in comparison to the BALB

conditioned media. The MBA appeared to have the ability to qualitatively discriminate

between cellular samples that contain PDGF; however, higher levels of serum proteins

may result in false positive quenching. Simple size exclusion chromatography allowed

for more reliable results to be obtained and showed that 50 ng of PDGF could be detected

in the presence of ~16.5 Clg/mL of serum proteins. Further improvements in the sample

processing could be explored; however, signal differential methodologies could provide

more reliable results.















CHAPTER 4
LIGHT SWITCHING APTAMER PROBE FOR PLATELET DERIVED GROWTH
FACTOR USINTG RATIOMETRIC AND TIME-RESOLVED FLUORESCENCE
MEASUREMENTS

Introduction

Fluorescent techniques offer excellent choices as signal transduction mechanisms

to report binding events because most samples remain intact under normal excitation

conditions and the sensitivity afforded by fluorescence detection. Fluorescence resonance

energy transfer (FRET),40;62-64 anisotropy 25;33and lifetime-based measurements are

examples of some fluorescent methods that, combined with a selected aptamer sequence,

can be used for bioassay development. All signal transduction techniques have their

individual strengths for different experimental needs. Nonetheless, FRET and anisotropy

suffer from some limitations that could hamper their effective applications in complex

biological samples. For instance, although fluorescence anisotropy only requires singly

labeling of one dye molecule on each aptamer sequence, it entails non-standard

instrumentation and data interpretation. FRET or fluorescence quenching based probes

quantify target concentrations with changes in fluorescence intensity, but the two

methods are sensitive to the solution environment and protein content as was showed in

the last chapter. More importantly, depending on the selection of fluorophores, directly

applying them to biological samples in their native environments may be difficult due to

fluorescence interference.

The maj or goal of the experiments discussed in this chapter was to explore the

possibility of designing a DNA probe that has the ability to differentiate between PDGF










responses and the biological background responses. Typically, protein analysis in their

native environments has two significant contributors to background signals. The first one

is the probe itself. For example, in a quenching based FRET molecular probe for protein

studies, the probe may suffer from incomplete quenching, resulting in significant

background. Moreover, in a native biological environment, there are many potential

sources for non-specific quenching of the probe as discussed in the previous chapter. This

greatly limits exploring the full potential of FRET-based MBAs for protein detection.

Concurrent with the experiments presented in this chapter, the effect of extracellular

proteins in cellular samples was also addressed in Chapter 3 and a sample processing

procedure to aid in the removal of most interfering proteins. Nonetheless, the effect of

background signal from auto-fluorescing components of the cell media still needed to be

investigated. Several molecular species exist in a biological environment, some of which

will yield a strong fluorescence background signal when light is applied to the dye

molecules. If the auto-fluorescence intensity of some of the cell media components is

comparable to or larger than the fluorescence intensity of the fluorescence of the MBA,

the fluorescence quenching becomes masked and the analysis of the fluorescence signals

becomes difficult. Great efforts have been made to solve this problem 65; however,

effective solutions are limited. The interesting spectroscopic properties of pyrene will be

employed and combined with time-resolved measurements as an alternative approach to

FRET. 34;66-68

Photophysics of Pyrene Excimers

Pyrene is a four-ringed polycyclic aromatic hydrocarbon (PAH) that gives

monomers and excimer emissions depending on its concentration. An excimer is a









dimmer which is associative in an electronic excited state and dissociative in its ground

state. The pyrene excimer was the first species of this type to be discovered.69;70








Figure 4-1. The chemical structure of pyrene

The formation of a pyrene excimer requires an electronically excited pyrene

encountering with a second pyrene in its ground electronic state. The excimer

fluorescence of pyrene can be described by the following equilibrium:

M* +M ++D*

Molecular absorption occurs to produce an excited monomer (M*), which then

interacts with its neighbor (M) to form the excimer (D*). When excimer formation takes

place in any given pyrene-containing system, it can be easily monitored by a steady-state

fluorescence spectrum: the broad, featureless emission centered at 480 to 500 nm is

extremely easy to recognize, even when extensive monomer emission occurs; since the

monomer fluorescence takes place in the 380 to 400 nm wavelength range.67

Another good example of a dye that exhibits the excimer phenomena is BODIPY,

whose monomer emits at 520 nm, and excimer has emission of 620 nm.71;72 The

formation of the excimer is useful to probe spatial arrangement of some molecules.

Similar to FRET, the distance dependent property of excimer can be used as a unique

signal transduction in the development of aptamer-based probes that change their

secondary structures on binding of its target, such as the aptamers selected against,

PDGF,33;62;63 COcaine,73thrombin,39;74 and HIV1 Tat protein. 5The emission wavelength










switching property solves the background signal problem that occurs with FRET

molecular probes. This finding encouraged us to apply the use of pyrene molecules in a

new aptamer probe design to develop oligonucleotide probes.

Time-Resolved Measurements

Time-resolved fluorescence spectroscopy is a well-established technique for

studying the emission dynamics of fluorescent molecules, i.e. the distribution of times

between the electronic excitation of a fluorophore and the radiative decay of the electron

from the excited stated producing an emitted photon. The temporal extent of this

distribution is referred to as the fluorescence lifetime of the molecule. This technique is

widely used in fluorescence spectroscopy because it often contains more information than

is available from steady-state data. With steady-state measurements some of the

molecular information is lost during the averaging process. Many macromolecules can

exist in more than a single conformation and the decay time of a bound probe may

depend on the conformation. The intensity decay may reveal two different decay times

indicating the presence of more than one conformational state. The steady state intensity

will only reveal an average intensity dependent on a weighted average of two decay

times. There are additional reasons for using time-resolved experiments. For example, in

the presence of energy transfer, the intensity decays may reveal how acceptors are

distributed in space around the donors. Also, time resolved measurements reveal whether

quenching is due to diffusion or to complex formation with the ground state fluorophores.

In fluorescence, much of the molecular information content is available only from the

time-resolved measurements.34

Fluorescence decay kinetics gives a complete picture of the fluorophore and its

interactions within the microenvironment. Complex intermolecular interactions can be









revealed by lifetime measurements made across an emission spectrum which has little

structure. Time correlated single photon counting (TCSPC) is a commonly used method

for the detection of the fluorescence lifetimes. TCSPC is a digital counting technique that

relies on the concept that the probability distribution for emission of a single photon after

an excitation event yields the actual intensity against time distribution of all the photons

emitted as a result of excitation. By sampling the single photon emission following a

large number of excitation flashes, the probability distribution can be constructed.

As a result of the above spectroscopic properties afforded by lifetime

measurements together with the use of a pyrene-based aptamer probe, it was of interest to

evaluate its potential for use with biological samples. Both steady-state and time-resolved

fluorescence measurements were conducted to provide in vitro monitoring of biological

samples with MBAs. In this approach, the MBA was labeled with pyrene molecules,

similar to what has been reported in using pyrene for molecular beacons.66 The optical

properties of pyrene were exploited, including that with the same excitation, the excimer

emitted at a longer wavelength than that for the monomer.

Although excimer formation is a very attractive characteristic of pyrene for protein

detection in solution, the excimer phenomena alone can not solve the problem of signal

from the multiple species in the biological environment. One feature of the pyrene

excimer is that it has a long lifetime67 COmpared with other potential fluorescent species.

The lifetime of the pyrene excimer could be as long as 100 ns, while that for most of the

species that contribute to the biological background signal is shorter than 5 ns. This

property should allow the separation of the biological background signal from that of the

excimer signal using time-resolved fluorescent measurements.










Experimental Section

All the aptamer-based probes listed in Table 4-1 were synthesized in-house and

purified with RP-HPLC. Reagents for the synthesis of DNA were purchased from Glen

Research (Sterling,VA). The pyrene labeled aptamer used for this set of experiments,

Pyr-MBA, is a shorter version of the original 36t aptamer designed specifically to reduce

pyrene background emission due to the close proximity of the dye molecules at each end

of the probe. Since the stem sequence is not important for the high affinity binding to

PDGF-BB47, in Order to reduce the background emission, the stem was shortened

gradually to identify an aptamer sequence that was in fully open conformation in the

absence of target while reserving good binding affinity. All aptamers were labeled with

pyrene at both termini. SGL-MBA is single labeled with pyrene at its 5' end. SCR-MBA

is a 39-mer scramble oligonucleotide sequence with pyrene molecules at both ends.

Table 4-1. Pyrene-labeled oligonucleotide sequences used in this study
Sequence Name Seauence
Pyr-MBA Pyrene -AGGC TAC GGCAC GTAGAGCAT CAC CAT GATC CT -
Pyrene
SGL-MBA Pyrene-AGGC TAC GGCAC GTAGAGCAT CAC CAT GAT CC T
SCR-MBA Pyrene- GGAAC GTAAT CAAC TGGGAGAAT GTAAC TGAC TGC -
Prene

Pyrene butylic acid was purchased from Sigma (St Louis, MO). Recombinant

human PDGF-BB, PDGF-AB, PDGF-AA, and, TNF-oc were purchased from R&D

Systems (Minneapolis, MN) and dissolved in 4 mM HCI with 0.1% BSA and then diluted

in 20 mM Tris buffer (pH 7.5) before use. Other recombinant human growth factors,

including recombinant human epidermal growth (EGF) and insulin-like growth factor 1

(IGF-1), were from Roche (Indianapolis, IN). Bovine serum albumin (BSA), human

hemoglobin (HEM), horse myoglobin (MYO), chicken lysozyme (LYS), human ot-









thrombin (THR) and other chemicals were from Sigma. The binding buffer used for all

the characterization experiments was 20 mM Tris-HCI (pH 7.5) with 20 mM sodium

chloride.

To prepare a simulated biological sample (SBS), Dulbecco's Mofication of Eagle' s

Medium (DMEM) (Mediatech, Inc, Hendon, VA) was supplemented with fetal bovine

serum (Invitrogen, Carlsbad, CA) and used for the detection of PDGF in biological

samples.

Instruments

An ABI3400 DNA/RNA synthesizer (Applied Biosystems, Foster City, CA) was

used for DNA synthesis. Probe purification was performed with a ProStar HPLC (Varian,

Walnut Creek, CA) where a C18 column (Econosil, Su, 250x4.6 mm) from Alltech

(Deerfield, 1L) was used. UV-Vis measurements were performed with a Cary 300 UV-

Vis spectrophotometer (Varian, Palo Alto, CA).

Steady-state fluorescence measurements were performed on a Fluorolog-Tau-3

spectrofluorometer (Jobin Yvon, Inc., Edison, NJ). For emission spectra, 349 nm was

used for excitation. Time-resolved measurements were made with a single photon

counting instrument (OB900, Edinburgh Analytical Instrument), where a nitrogen flash

lamp was used as the excitation source (h=337 nm).

Synthesis and Purification of Pyrene Labeled MBA

Several methods have been reported for oligonucleotide labeling.66;76

Unfortunately, for dually labeling pyrene on both nucleic acid termini, the reported

procedures proceeded in low yields partially due to solubility incompatibility of pyrene

derivatives and oligonucleic acids. A new solid phase coupling method that allow multi-









labeling organic dyes to DNA sequences with coupling yields as high as 80 % was used

to synthesize all the pyrene labeled aptamers.n7

Results and Discussion

Light Switching Excimer Aptamer Probe Design

As a proof of principle, the excimer signaling approach was used to develop a

probe for PDGF-BB. An excimer switching aptamer probe was developed by labeling

both ends with pyrene molecules that can form excimers. When the dual-pyrene-labeled

aptamer probe is free in solution without the target protein, both pyrene molecules will be

spatially separated, based on the FRET experiments in Chapter 2, and only the monomer

emission peaks (at 375 nm and 398 nm) will be observed. The binding of the aptamer

probe to protein brings the pyrene molecules at 3' and 5' ends close together, allowing the

excimer to form. Thus, the emission peak around 485 nm will be observed. The change in

emission color serves as a rapid process for qualitation and the excimer fluorescence

intensity can be used for highly sensitive real-time quantitation of PDGF in homogeneous

solutions.







PDG F



Figure 4-2. Use of pyrene excimer to probe PDGF-BB. PDGF-BB aptamer was labeled
with pyrene molecule at both ends. The two pyrenes will be away from each
other because of the open structure of the aptamer. After binding to PDGF-
BB, the aptamer adapts a close conformation, bringing two pyrenes close to
each other. As a consequence, pyrene excimer forms and green light emits.







71


Light Switching Aptamer Probe for PDGF-BB Detection

A light switching aptamer probe, named Pyr-MBA (sequence shown in Table 4-1),

was prepared by labeling a 33-nucleotide sequence, a shorter version of the 36t MBA

with pyrene molecules at both 3' and 5' ends.47 Figure 4-3 shows the fluorescence

emission spectrum from a solution containing 100 nM of Pyr-MBA. Two emission peaks

at 375 and 398 nm corresponds to pyrene monomer emission. No significant excimer

emission was observed from this solution (Figure 4-3). Upon addition of 50 nM PDGF-

BB into the Pyr-MBA solution, an aptamer-protein target complex was formed leading to

a secondary structure where the 3' end sequence hybridizes to the 5' end sequence,

forming a stable stem. This stem brings both pyrene molecules together, resulting in

excimer emission at 485 nm (Figure 4-3).


450000~ Pyr-MBA
Pyr-MBA + PDGFBB
S400000-

S350000-
S300000-
250000-
4 200000-
S150000-
100000-
50000-

400 450 500 550 600
Wavelength (nm)


Figure 4-3. Fluorescence emission spectra of Pyr-MBA with and without PDGF-BB in
binding buffer.

As shown in Figure 4-3, protein-bound probe gives emission peaks, two monomer

peaks at 375 nm and 398 nm respectively, and excimer peak at 485 nm. This allows for

ratiometric measurements. By taking the intensity ratio of the excimer peak to either one









of the monomer peaks, one could effectively eliminate signal fluctuation and minimize

impact of environmental quenching on the accuracy of measurement. We further

explored excimer formation by designing two control DNA sequences, SGL-MBA and

SCR-MBA. The sequences were prepared to confirm that the observed excimer emission

was a result of the aptamer-protein binding. The first sequence SGL-MBA was the same

sequences of the Pyr-MBA aptamer labeled with pyrene at the 5' end. Green et al.

reported that PDGF-BB has two binding sites; the phenyalanine-84 residue in the B-

chain of PDGF forms a point of contact with a specific nucleotide residue in loop at the

helix junction of the selected aptamer However, the addition of PDGF-BB into the SGL-

MBA solution did not change the emission spectrum of the solution, indicating that two

pyrene molecules need to be in close proximity. This suggests that the distance between

adj acent molecules was not enough for the excimer formation and emission. This data

also showed that PDGF-BB itself has no measurable effect on the optical properties of

pyrene molecules.

The second sequence, SCR-MBA, is a random DNA sequence with its 3' and 5'

ends labeled with pyrene. This sequence was used as a control aptamer with no or low

affinity for PDGF-BB to demonstrate that the excimer emission is a direct result of the

formation of the Pyr-MBA/PDGF-BB complex. This dually labeled scramble sequence,

did not give any excimer emission after the addition of PDGF-BB. Both results indicated

that the excimer emission showed in Figure 4-3 was indeed from the aptamer

conformation change on specific binding to PDGF-BB.







73




2.4 -( .......................
2.2-g
2.0 --m- pyr-MBA
1.8- -*- SGL-MBA

E1.6- SCR-MBA

LL1.2-


S0.8- addition
-- ~ 6 0.61- ( D F
0.4-

0.2-
e 50 100 150 200
t (s)


Figure 4-4. Real-time response of excimer/monomer ratio. It showed that the binding of
PDGF-BB aptamer to PDGF-BB takes place within seconds. It affords a rapid
protein assay.

Sensitivity and Selectivity of the Probe

The dose-dependent excimer formation was determined to evaluate the sensitivity


of the pyrene-labeled aptamer probe (Figure 4-5). A linear response was observed with

the addition of increasing amounts of PDGF-BB in concentrations ranging from 0 to 40


nM. This result confirmed the formation of the excimer due to the spatial proximity of the


two pyrene molecules (and the appropriate geometrical arrangement required for excimer

emission) at the ends of the MBA. According to the measurements for blank samples, the

limit of detection was calculated an about 200 pM were consistently detected in


homogeneous solution. Moreover, the high sensitivity together with the emission

wavelength switching and detection without separation nature of this probe, enabled


detection of the presence of 40 nM PDGF-BB just with the naked eye (Figure 4-6). A

clear green color was observed when PDGF-BB was added to 100 nM of excimer probe

solution.





































360 380 400 420 440 460 480 500 520 540 560
Wavelength (nM)


- OnM
- 2nM
4nM
- 6nM
8 nM
-10nM
12nM
- 14nM
- 16nM
- 20 nM
- 24nM
- 30nM
- 35nM


400000

350000 1

300000 -

250000 1

200000-

150000 1

100000 -

50000 1

0-I


15 20 25 30 35 40


PDGF-BB (nM)




Figure 4-5. Dose-dependent excimer formation on additions of PDGF-BB to a Pyr-MBA.
A)in binding buffer, B)calibration curve.

















Figure 4-6. Visual detection of PDGF-BB. Solution of the 100 nM pyr-MBA under UV
light (left), and 100 ul of 100 nM pyr-MBA with 40 nM of PDGF-BB (right).

To detect PDGF-BB in the presence of other biomolecules the selectivity of the

pyr-MBA probe is critical. Similar to the FRET-based MBA, the pyrene labeled MBA

was mixed with excess extra-cellular proteins such as albumin, hemoglobin, myoglobin,

and lysozyme. Even at 10 times the PDGF-BB concentration, the proteins did not

produce significant changes in signal (Figure 4-7).




1 .0


.c 0.8-





S0.2





S0.0-
SLSM HMG MYG THR BSA PDGF BB

Figure 4-7. Response of excimer probe to different proteins. Concentration of PDGF-BB
was 50 nM, while the concentration of other proteins was ten times higher
(500 nM).

Furthermore, we tested the selectivity of the excimer probe for proteins and

peptides potentially coexisting with PDGF-BB in biological samples. The results showed

this shorter version of the original MBA probe retain its high selectivity for PDGF-BB in










homogeneous solution. The responses to epidermal growth factor (EGF), vascular

endothelial growth factor (VEGF), or insulin-like growth factor 1 (IGF-1) were minimal.

PDGF-AA and PDGF-AB, both of which have shown lower response to the aptamer

sequence in the FRET system, gave lower fluorescence-signal response in this case as well

(Figure 4-8).47



1.0-





v,0.6-





0- 0.2-


0.0-
EGF IGF VEGF PDGFAA PDGFAB PDGFBB


Figure 4-8. Response of 50 nM Pyr-MBA to 500 nM of different growth factors.

Bovine serum albumin (BSA) belongs to the class of serum proteins called

albumins, which make up about half of the protein in plasma and are the most stable and

soluble proteins in plasma. It is very common for laboratories developing immunoassays,

mostly due to its availability, solubility and the numerous functional groups present for

coupling to haptens. For PDGF-BB reconstitution, a buffer containing BSA as a carrier

protein was used and as result BSA will be present throughout. To rule out any possible

contribution of B SA to the measured fluorescence emission, increasing amounts of BSA

were incubated with the pyr-MBA. In this case, high concentrations of BSA did not alter










the optical properties of pyr-MBA; the measured excimer emission is negligible as can be

observed from the data presented in Figure 4-9.


20000
0.1 mg/mL
ar: 0.15 mg/mL
j 0.2 mg/mL
< 15000 -I;i 8 0.25 mg/mL
rz *Omg/mL


-10000 -



5000




375 425 475 525 575
Wavelength (nM)

Figure 4-9. Increasing amounts of BSA incubated with 50 nM Pyr-MBA.

Direct Quantitative Detection of PDGF-BB in Cell Medium

To be applicable for a potential bioassay development, the MBA probe must

tolerate interference typically present in a biological sample. In the previous chapter we

showed that when the MBA is incubated with a protein mixture containing high levels of

cellular proteins, false-positive quenching signals were obtained. On the other hand,

when the protein levels are monitored and adjusted to acceptable levels, the MBA binds

to PDGF-BB with high affinity and responds by a decrease in fluorescence intensity

attributable to the formation of the PDGF-BB-MBA complex. In this chapter, the

interference presented by autofluorescencing components in the media will be

investigated. For the purposes of simplicity and to investigate solely the effect of auto-

fluorescence cell media components, a simulated biological specimen (SBS) was made










by combining cell medium with low levels of serum proteins. Figure 4-10 shows the

spectra of 200 nM Pyr-MBA in a Tris-HCI buffer solution and cell medium (DMEM).


900000- 200 nM Pyr-MBA in buffer
80000-200 nM Pyr-MBA + 50 nM PDGFBB in buffer
200 nM Pyr-MBA in DMEM
S700000- DMEM
1 I \200 nM Pyr-MBA + 50 nM PDGF-BB in DMEM
S600000-

S500000-
S400000-
E 300000-
u- 200000-
100000-

350 400 450 500 550 600 650
Wavelength (nm)


Figure 4-10. Fluorescence spectra of simulated biological sample. (DMEM, 200 nM Pyr-
MBA in DMEM, 200 nM Pyr-MBA and 50 nM PDGF-BB in DMEM, 200
nM Pyr-MBA in Tris-HCI buffer, and 200 nM Pyr-MBA and 50 nM PDGF-
BB in Tris-HCI buffer)

In binding buffer, the MBA showed an intense excimer emission which was

observed when the target protein was added to the probe solution. Unfortunately, intense

background fluorescence, contributing from some species typically encountered in the

cell medium such as riboflavin, nicotinamide, pyridoxine, tryptophan, tyrosine as well as

phenol red, was observed, which masked the signal response from the probe and made it

indistinguishable. This result indicates that simple steady-state fluorescence

measurements are questionable for direct detection of PDGF-BB in a biological sample.

As mentioned before, the MBA labeled with pyrene molecules not only has the

excimer formation capability but also the advantage of a long fluorescence lifetime. Most

of the background fluorescence has a lifetime of less than a few nanoseconds(~10 ns),

while the monomer and excimer emission of pyrene have much longer lifetimes. Since









we were interested in studying the possibility the pyrene molecules for signal

differentiation based on lifetime, the total amount of protein was maintained at acceptable

levels (~10 Clg/mL) for the PDGF-BB aptamer. Also, based on the steady state

fluorescence spectra (Figure 4-10) there is considerable background signals at this

concentration produced by SBS at the wavelength of interest which impede the ability for

pyrene to be useful in standard fluorescence measurements. Therefore, there is a need to

study this format by alternative fluorescence measurements and explore the potential of

using the lifetime characteristics of pyrene.

Time correlated single photon counting measurement of the Pyr-MBA and Pyr-

MBA-protein complex solutions were investigated to characterize their lifetimes and

identify the nature of excimer formation. The results obtained by TCSPC suggested that

the lifetimes of both pyrene monomer and excimer were around 40 ns. This is one

magnitude longer than lifetimes of most organic fluorophores and fluorescent

components in cell medium and cells, which should allow temporal resolution of the

excimer signal from intense background fluorescence from cell medium. Three cell

media samples were then subjected to SPC measurements: simulated biological sample

(SB S), 200 nM Pyr-MBA in SBS, and 200 nM MBA-Pyr-MBA and 50 nM PDGF-BB in

SBS. The data reported on Figure 4-11 and 4-12 showed that the fluorescence emission

from both pyrene monomer and excimer can be differentiated from background

fluorescence. Although pyrene monomer has a longer lifetime, the difference in intensity

between Pyr-MBA in SBS and SBS itself is small (~3 times). This may be because many

components of the SB S fluoresce at this region and their concentrations are higher than

the MBA concentration.












100000~
10 0


10000


1000


100


10 -


I


* SBS
* SBS +Pyr-MBA
SBS +Pur-MBA + PDGFBB


0~ 25 50b 7'5 1~00 125 150O 175 200
Time (ns)
Figure 4-11. Monomer time-resolved spectra collected at 398 nm (monomer emission)
for a sample of simulated biological medium (SBS), 200 nM Pyr-MBA in
SBA and 50 nM Pyr-MBA and PDGF-BB in SBS.


10000- SBS
*SBS +Pyr-MBA
~SBS + Pyr-MBA + PDGFBB

S1000-



-c 100-







0 25 50 75 100 125 150 175 200
Time (ns)
Figure 4-12. Excimer time-resolved spectra collected at 480 nm (excimer emission) for a
sample of simulated biological medium (SB S), 200 nM Pyr-MBA in SBA and
Pyr-MBA and 50 nM PDGF-BB in SBS.

Time resolved emission spectra revealed changes in emission spectra on a

nanosecond time scale (Figure 4-13). Due to its short fluorescence lifetime, the

fluorescence from the SBS decayed rapidly, getting to 0. 1% of its original signal 40

nanoseconds after the excitation pulse reached maximum intensity. In contrast, the

excimer emission decayed slower and retained reasonably high emission intensity even









after 40 ns of decay. The highest signal to noise ratio at 480 nm was observed from 40 to

about 100 ns after the excitation pulse reached maximum intensity. Before 40 ns, there

was still significant background fluorescence from the SBS while the signal to noise ratio

dropped below 3 after 100 ns of decay. In contrast, the excimer emission from a mixture

of protein and probe in SBS decayed much slower (Figure 4-13). The emission remained

after 40 ns of decay.

The temporal separation of intense background from weak signals is evident in

Figure 4-14, where time-resolved fluorescence emission spectra of SB S samples are

compared with steady-state fluorescence emission spectra. For steady-state measurements

no resolved peak around 485 nm was observed when PDGF-BB was added to Pyr-1VBA

in the SBS sample due to a significant amount of background signal. However, in time-

resolved emission spectra the long lifetime emission peak at 480 nm was well resolved.

This peak corresponded to excimer emission from protein-bound probe, which was

supported by two observations: its intensity varied with changes of protein concentration,

and no such peak was observed in either spectrum of cell medium or cell medium

solution containing PDGF-BB and a single-pyrene-labeled aptamer sequence

(SGL-1VBA). Thus, this characteristic emission peak could be used to examine

the presence of PDGF-BB in biological fluids.







82



1A 19 A I~20-3Bos 40-50m











110D-139s 1 4-5ns 1210-10




Waeent (nm)
Fiue4-3 lurseceitnst sa ucio fwveegh tdfern ims B
(gee) 200n Py-B nSS(e) n 0n y-B ith5n

PDG-BBin cel SBS (bue at2 .Ectton37m ie0n
corspn to th iea hc h xiato us ece aiu
intensity.
Dtctino D FB in Siuate ilgclSml
Wihtesnl htncutn ehiue h y-B rb a ulttvl
detct aret rotinin omgenoussoutin ad n asiulaed ioogialsample000
cnainn o eeso oeta nefrn rtis hntecnetaino D F
BBwsicesd h loesec nest fteecie eki iersle
emisin spcrmicesdacringyFiue1-5sostedc sof20n










aptamer probe Pyr-MBA in the cell medium with increasing concentrations of PDGF-BB

at 480 nm emission.



-4 Steady State: Pyr-MBA in SBS
-0 Steady State: Pyr-MBA + PDGFBB in SBS
Time-resolved: Pyr-MBA in SBS
1.2-
-o- Time-resolved: Pyr-MBA + PDGFBB in SBS





0.8-


S0.6-

0.







0.0-

360 380 400 420 440 460 480 500 520 540 560 580 600

Wavelength (nm)

Figure 4-14. Steady-state and time-resolved fluorescence spectra of SBS samples. Steady
state fluorescence emission spectra of 200 nM Pyr-MBA in SBS and 200 nM
Pyr-MBA with 50 nM PDGF-BB. Time-resolved fluorescence spectra of 200
nM Pyr-MBA in SB S, and 200 nM Pyr-MBA with 50 nM PDGF-BB in SBS.
Time resolved spectra were taken over 60 ns to 79 ns after the excitation pulse
reaches the maximum intensity.

Fluorescence intensity from the response of each solution could be calculated by

integrating photons emitted over an optimized time window. Photons emitted between 40

and 100 ns were counted and integrated for each concentration to construct a

calibration curve. The resulting fluorescence intensity is proportional to the PDGF-BB

concentration (Figure 14-14). This linear response of fluorescence intensity to PDGF-BB







84


in cell medium demonstrates the feasibility of detection of the target proteins in simple

biological samples containing low levels of proteins.


Time (ns)
A


200000-


150000


O 100000-
-


S50000
0-


0 20 40 60
PDGF-BB (nM)


80 100


Figure 4-15. Fluorescence decays of 200 nM Pyr-MBA in simulated biological specimen.
(A) Simulated biological sample with various concentrations of PDGF-BB
and response of fluorescence intensity to the change of protein concentration.
Excitation= 337 nm. The collection time for each decay was 3000 seconds.
(B)Photons emitted between 40 and 100 ns after excitation pulse reached
maximum were counted to construct the calibration curve.









Expression of PDGF-BB from Cells

Several reports have indicated that PDGF-BB can be over-expressed by various

types of cancer cells, including MCF-7, MDA-MB-231 and PC3 cultured cell lines.46

Unfortunately, the literature is ambiguous as to the specific amounts of PDGF-BB that

can be expressed under any given condition. This is largely limited to the fact that most

of methodologies currently used provide only qualitative information and often not just

specific to PDGF-BB expression.78;7 As a result, ELISA was employed to determine

specific amounts of expression that could be expected from PC-3 tissue cultured cells.

PC-3 cells are adherent prostate cancer cells and on incubation with TNF-a over-express

PDGF-BB. TNF-a (also called cachectin) is a maj or immune response modifying

cytokine produced primarily by activated macrophages. TNF-a also induces the

expression of other autocrine growth factors, increases cellular responsiveness to growth

factors and induces signaling pathways that lead to proliferation. TNF-a acts

synergistically with EGF and PDGF-BB in some cell types.

In order to determine the levels of PDGF-BB present in conditioned media, ELISA

tests were performed on different samples. The levels of PDGF-BB were measured from

samples which were cultured to confluency and spiked with Recombinant Human TNF-a

to induce PDGF-BB production (Table 4-2). The total protein concentration was

determined by the Bradford Assay. Since the addition of TNF-a to PC-3 cells in culture

should increase the levels of PDGF-BB in the conditioned media, it was first important to

Eind out if we could monitor PDGF-BB induction. PC-3 cells were cultured in 3 5 mm cell

culture dishes in a Einal volume of 2 mL. Twenty mg of TNF-a was added to each cell









dish and the conditioned media was collected at 0, 3, 6, 9, 12 and 24 hours after

induction.

Table 4-2. Amount of protein collected in 24 hours
PDGF- PDGF- Total protein
Incubation
BB BB concentration
time (hrs)
(pg/mL) (pM) (Clg/mL)
0 < 15 <0.6 5893
3 < 15 <0.6 6915
9 16.3 0.7 6787
12 42.9 1.7 6934
24 197 7.9 7340

In this case it can be observed that TNF-a induced the expression of PDGF-BB in

PC-3 cells. Non induced samples have less than 15 pg/mL of PDGF-BB. On the other

hand, after 9 hours of incubation, an increase in PDGF-BB levels can be observed, up to

197 pg/mL after 24 hours. The total protein concentration of all the samples was

measured and it ranged from approximately 5,900 Clg/mL to more than 7000 Clg/mL.

A second set of samples was prepared with the idea of increasing the amount of

PDGF-BB present per sample. Changes included: incubation in a larger flask, with a

small volume (enough to cover the surface of the flask), and with a higher amount of

TNF-u. A large surface area of adherent cells in a small volume should allow for a higher

amount of PDGF-BB collection. The second change in the sample handling includes a

concentration step by lyophilization (25 ml were reduced to 5 mL). In this case, PC-3

cells were cultured in five 75 cm2 flasks with 5 mL of cell media in each. Cells were

grown to confluency and 25 mg of TNF-a was added to each flask and the conditioned

media was collected after 24 hours for one set and after 36 hours for the following. The

total volume of conditioned media collected was 25 mL. After lyophilization the volume

was reduced to 5 mL. The results from the ELISA and Bradford assays are shown in









Table 4-3. It is illustrated that the amount of PDGF-BB is indeed higher with the

modified sample collection method, doubling in a period of 12 hours.

Table 4-3. Amount of protein collected in 24 and 36 hours
Incubation PDGF-BB PDGF-BB Total protein
time (hrs) (pg/mL) (pM) concentration
(Clg/mL)
24 865 35 9334
36 1775 71 11934

The two sample collections methods showed an increase in PDGF-BB production

after induction with TNF-ot, but the levels of PDGF-BB are lower than the current limits

of detection of the pyrene and the FRET assay. In addition, even if the levels of PDGF-

BB were detectable, the total amount of protein per sample is about 2 orders of

magnitude higher than the allowed protein concentration determined for the quenching

assay. Since the interfering non-specific interactions would be similar for the Pyr-MBA

and FRET MBA (essentially, non-specific binding to the aptamer), it is reasonable to

assume that the total protein concentration would limit both assays. If the total amount of

protein commonly found in serum exceeds 100 Clg/mL before sample processing, non-

specific interactions will result in false positive signals for this particular aptamer as

shown in Figure 3-2. A longer incubation time may be a possibility for increasing PDGF-

BB production, but after 36 hours of incubation, cells are 100% confluent and as result

they begin to die. Also, the total amount of total protein concentration increased as well.

Conclusions

The method of excimer light switching is an excellent signal transduction for

aptamer probe development. Pyr-MBA is able to detect protein in homogeneous solution

and simple biological samples containing low, or acceptable levels of proteins for the

aptamer. The generation of the excimer emission requires the conformation change of the




Full Text

PAGE 1

MOLECULAR BEACON APTA MERS: FUNDAMENTALS FOR FLUORESCENCE-BASED DETECTION AND FOR PROTEIN STUDIES By MARIE CARMEN VICNS-CONTRERAS 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 2005

PAGE 2

Copyright 2005 by Marie Carmen Vicns-Contreras

PAGE 3

ACKNOWLEDGMENTS I am grateful for all the efforts and kindness of all the people involved in this project. I especially thank my advisor Dr.Weihong Tan, for his guidance, encouragement, and ideas throughout my graduate studies. I appreciate the opportunity of working in his lab. He provided all of the resources that I needed for the past 5 years. I appreciate the assistance of former group members of the Tan Lab, especially Xiaohong Fang, for teaching me the basics to begin this project; and Andrew Vanderlaan, for his hard work and help during his time in the group. Also, I thank Richard J. Rogers, Jon Stewart, James Winefordner, and Richard Yost, for having agreed to be part of my graduate committee. I thank Arup Sen who profoundly influenced me and helped me to get to where I am today, for sharing so many ideas and offering me a new perspective on many things, and teaching me much about learning and development. My life as a graduate student was fun and cheerful thanks to my officemates. All of them have taught me many things about life and their cultural heritage. For all the great times we spent together celebrating life, I want to thank all of the members of the Tan Group, especially Alina, Lisa, Tina, Prabodhika, Josh, Colin, James, Hong, and Shelly. I thank Karen Martinez, for being a happy spirit and making me smile every day. Intelligence and a deeply critical and analytical way of looking at ideas made Tim Drake a fantastic colleague and great friend. I wish them all the best! iii

PAGE 4

My adopted Puerto Rican family in Gainesville (which we like to call el corillo), made my life joyful. I especially thank Wilfredo Ortiz for one of the best friendships I can ever imagine. I am extremely thankful to mi familia, especially my parents (Carmen and Francisco) for their endless love, support, advice, and encouragement. I thank my amazing brothers and sisters for taking care of me from a distance and calling me whenever I forgot to call them. At last, without the love, support, interest, sense of humor, assistance of all kind, time, motivation and care from Jos I would be nowhere. It was his suggestion to come to the University of Florida. His patience and his continuous encouragement enabled me to succeed. He is one of the most wonderful human beings in this world. I am so indebted to him that there is no way to repay him. I am overwhelmed with gratitude. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT......................................................................................................................xii CHAPTER 1 INTRODUCTION........................................................................................................1 Aptamers.......................................................................................................................2 What Is an Aptamer?.............................................................................................2 The Aptamer Selection Process: Systematic Evolution of Ligands by Exponential Enrichment (SELEX)....................................................................3 Aptamers versus Antibodies..................................................................................5 Aptamers in Bioanalytical Chemistry...................................................................6 Fluorescence Methods for Signal Transduction.........................................................10 Fluorescence Quenching.....................................................................................11 Fluorescence Resonance Energy Transfer (FRET).............................................12 Molecular Beacons.....................................................................................................14 Molecular Beacons for Non-Specific Protein Detection.....................................14 Aptamers and FRET for Protein Detection.........................................................16 Molecular Beacon Aptamer (MBA) Development for a Model Protein System..............................................................................................................18 Aptamers for Platelet Derived Growth Factor-BB (PDGF-BB).........................18 2 DEVELOPMENT AND CHARACTERIZATION OF A MOLECULAR BEACON APTAMER FOR PLATELET DERIVED GROWTH FACTOR............21 Introduction.................................................................................................................21 Materials and Methods...............................................................................................23 Synthesis and Purification of MBAs...................................................................23 Standard Fluorescence Quenching Assay...........................................................25 Results and Discussion...............................................................................................26 Choice of Aptamer Sequence for MBA Development........................................27 Evaluation of Additional Fluorophore Quencher-Pairs.......................................29 Incubation Conditions for MBA Bioassay..........................................................34 v

PAGE 6

Selectivity of the MBA-Based Fluorescence Quenching Assay.........................39 Fluorescence Quenching Assay Distinguishes Molecular Variants of PDGF....42 Conclusions.................................................................................................................44 3 IMPLICATIONS OF USING A MBA IN A FRET BASED ASSAY FOR PDGF STUDIES IN BIOLOGICAL SAMPLES..................................................................46 Introduction.................................................................................................................46 Materials and Methods...............................................................................................47 Instrumentation....................................................................................................47 Preparation of a Simulated Biological Sample....................................................48 Cell Culture and Preparation of Conditioned Media...........................................48 Results and Discussion...............................................................................................49 Use of MBA in Biological Sample......................................................................49 Implications of Total Protein Content on the MBA Assay.................................53 Sample Pre-Treatment.........................................................................................55 Conclusions.................................................................................................................62 4 LIGHT SWITCHING APTAMER PROBE FOR PLATELET DERIVED GROWTH FACTOR USING RATIOMETRIC AND TIME-RESOLVED FLUORESCENCE MEASUREMENTS....................................................................63 Introduction.................................................................................................................63 Photophysics of Pyrene Excimers.......................................................................64 Time-Resolved Measurements............................................................................66 Experimental Section..................................................................................................68 Instruments..........................................................................................................69 Synthesis and Purification of Pyrene Labeled MBA...........................................69 Results and Discussion...............................................................................................70 Light Switching Excimer Aptamer Probe Design...............................................70 Light Switching Aptamer Probe for PDGF-BB Detection..................................71 Sensitivity and Selectivity of the Probe...............................................................73 Direct Quantitative Detection of PDGF in Cell Medium....................................77 Detection of PDGF-BB in Simulated Biological Sample...................................82 Expression of PDGF-BB from Cells...................................................................85 Conclusions.................................................................................................................87 5 FLUORESCENCE RESONANCE ENRGY TRANSFER STUDIES FOR MOLECULAR BEACON APTAMERS: UNDERSTANDING PROTEIN INTERACTION AND VERSITILITY OF PROBES................................................90 Introduction.................................................................................................................90 Results and Discussion...............................................................................................91 Backbone-Modification of DNA.........................................................................91 MBAs Created from Homologous RNA and DNA Aptamers Display Significant Differences in Eliciting FRET in Response to PDGF-BB............94 vi

PAGE 7

Fluorescence Enhancement Assays with MBAs.................................................99 Two-Step FRET Assays Using Restriction Enzymes Cleavage Sites...............104 Conclusions...............................................................................................................109 6 SUMMARY AND FUTURE WORK......................................................................112 Summary...................................................................................................................112 Future Work..............................................................................................................117 New MBA Development for Other Proteins and Small Molecules..................117 Enrichment and Simplification of Complex Samples.......................................117 Pyrene Application to Other Protein Recognition Systems..............................118 Signal Amplification of Aptamer Recognition through PCR............................119 LIST OF REFERENCES.................................................................................................120 BIOGRAPHICAL SKETCH...........................................................................................126 vii

PAGE 8

LIST OF TABLES Table page 2-1 Percentage of quenching of different consensus sequences of 50 nM of MBA for PDGF aptamer with 200 nM of PDGF...............................................................28 3-1 Analysis for PDGF-BB spiked samples after fractionation on a G-100 Sephadex column using enzyme-linked immunosorbent assay (ELISA).................60 4-1 Pyrene-labeled oligonucleotide sequences used in this study..................................68 4-2 Amount of protein collected in 24 hours..................................................................86 4-3 Amount of protein collected in 24 and 36 hours......................................................87 5-1 Aptamers synthesized for PDGF-BB detection.......................................................95 viii

PAGE 9

LIST OF FIGURES Figure page 1-1 Overview of the systematic evolution of ligands by exponential enrichment process........................................................................................................................4 1-2 Typical Jablonski diagram.......................................................................................10 1-3 Fluorescence resonance energy transfer diagram.....................................................13 1-4 Working principle of molecular beacon...................................................................15 1-5 Assumed biological foldings of Platelet Derived Growth Factor............................19 1-6 Folding of 36t molecular beacon aptamer (MBA)...................................................20 2-1 Fluorescence intensity changes versus time of different aptamer sequences...........28 2-2 Proposed mechanism of PDGF-BB induced FRET responses for the 36t probe.....29 2-3 Evaluation of fluorescence quenching caused by PDGF-BB with MBAs...............31 2-4 Calibration curves for PDGF-BB.............................................................................33 2-5 Effect of increasing temperature on fluorescence signal of free MBA in solution.....................................................................................................................35 2-6 Effect of increasing pH on fluorescence signal of free MBA in solution................37 2-7 Effect of divalent cations on fluorescence signal of free MBA in solution.............38 2-8 The effect of monovalent cations on fluorescence signal of free MBA in solution.....................................................................................................................39 2-9 Binding selectivity of the MBA...............................................................................40 2-10 Fluorescence quenching displayed by the MBA......................................................41 2-11 Cartoon representing PDGF-BB and its molecular variants....................................43 2-12 Dose-response curves of PDGF variants: fluorescence signals of MBA for PDGF-AA.................................................................................................................43 ix

PAGE 10

3-1 Use of molecular beacon aptamer in biological samples.........................................51 3-2 Molecular beacon aptamer fluorescence quenching................................................54 3-3 Fluorescence quenching caused by PDGF-BB in the presence of proteins in a simulated biological specimen.................................................................................57 3-4 Fluorescence intensity of fractions of simulated biological sample (SBS) collected from G-100 column..................................................................................60 3-5 Fluorescence intensity of dilutions of SBS..............................................................61 4-1 The chemical structure of pyrene.............................................................................65 4-2 Use of pyrene excimer to probe PDGF-BB.............................................................70 4-3 Fluorescence emission spectra of Pyr-MBA with and without PDGF-BB in binding buffer...........................................................................................................71 4-4 Real-time response of excimer/monomer ratio........................................................73 4-5 Dose-dependent excimer formation on additions of PDGF-BB..............................74 4-6 Visual detection of PDGF-BB..................................................................................75 4-7 Response of excimer probe to different proteins.....................................................75 4-8 Response of 50 nM Pyr-MBA to 500 nM of different growth factors.....................76 4-9 Increasing amounts of BSA incubated with 50 nM Pyr-MBA................................77 4-10 Fluorescence spectra of simulated biological sample..............................................78 4-11 Monomer time-resolved spectra collected at 398 nm..............................................80 4-12 Excimer time-resolved spectra collected at 480 nm.................................................80 4-13 Fluorescence intensity as a function of wavelength at different times.....................82 4-14 Steady-state and time-resolved fluorescence spectra of SBS samples.....................83 4-15 Fluorescence decays of 200 nM Pyr-MBA in simulated biological specimen........84 5-1 Phosphorothioate DNA backbone modification......................................................92 5-2 Fluorescence quenching produced by a phosphorothiate modified MBA...............93 5-3 Theoretical model of the 36tRNA-MBA.................................................................96 x

PAGE 11

5-4 Comparison of FRETresponse of 36t-DNA-MBA and the homologous 36t-RNA-MBA with PDGF-BB.....................................................................................97 5-5 Competition of FRET-response by unmodified 36t-DNA aptamer with the standard 36t-DNA-MBA..........................................................................................98 5-6 Competition of FRET response by unmodified RNA aptamer with DNA-MBA and by unmodified DNA aptamer with RNA-MBA..............................................100 5-7 Fluorescence enhancement FRET-assay................................................................103 5-8 Working principle of the enzyme-site modified MBA..........................................105 5-9 Molecular beacon aptamer variants with restriction enzyme cleavable stem sequence response to magnesium...........................................................................107 5-10 Molecular beacon aptamer variants with restriction enzyme cleavable stem sequence response to PDGF-BB............................................................................108 xi

PAGE 12

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MOLECULAR BEACON APTAMERS: FUNDAMENTALS FOR FLUORESCENCE-BASED DETECTION AND FOR PROTEIN STUDIES By Marie Carmen Vicns-Contreras August 2005 Chair: Weihong Tan Major Department: Chemistry Molecular beacon aptamers (MBA) are created by combining synthetic oligonucleotide ligands with high affinity for a protein target with a sensitive signal transduction mechanism. In this study, an MBA was developed and examined for the detection of platelet derived growth factor-BB (PDGF-BB) using it in a fluorescence-based assay. The developed probe was capable of consistently detecting 10 nM PDGF-BB in homogeneous solution. The influence of certain physical and chemical parameters of incubation that would affect DNA conformation were examined and the results show that this bioassay is compatible with pH, temperature, and monovalent cation levels typically encountered in biological samples. After a complete investigation in homogeneous solution, the MBA was examined for PDGF-BB detection in biological samples. This is the first attempt to use single aptamer molecules with an inherent signal transduction mechanism for the detection of PDGF-BB in biological samples. For this particular aptamer and assay conditions, the presence of more than 100 ug/mL of serum xii

PAGE 13

proteins resulted in non-specific fluorescence quenching of the probe. The results indicated that a sample processing method prior to the assay is required to remove potential biological interferences. Selective and quantitative protein analysis in complex biological fluids still presents considerable challenges for future research, mainly due to the background signals from both the fluorescent probe and the components of typical cellular samples. To overcome the problem of background signals, the existing MBA was modified by labeling each end with pyrene molecules. This produced a monomer-excimer light-switching aptamer probe for protein monitoring by using both steady-state and time-resolved fluorescence measurements. With a time-correlated single photon counting technique and the long fluorescence lifetime of pyrene, the biological background signal was separated from the excimer signal and 200 pM of PDGF-BB were detected in a simulated biological sample. We also examined FRET responses to PDGF-BB shown by MBAs with modified aptamer sequences that affect DNA conformation. Our findings provide new insights into the fundamental mechanisms underlying protein-induced FRET shown by MBAs and have significant implications on the design strategy and use of MBAs in structure-function analysis, and in bioassays for protein biomarkers and their molecular variants. xiii

PAGE 14

CHAPTER 1 INTRODUCTION Proteins are essential for the structure and function of all living cells. Protein have different shapes and conformations that are responsible for many functions, such as enzymatic reactions, regulation of cell growth, structure, movement, carriers, and receptors. Thus, many scientists in the bioanalytical sciences are interested in developing methods and probes to be used for protein studies and detection. Biological interactions are used to provide either qualitative or quantitative information about protein content and could yield important data about the molecular mechanism of action of biologically active compounds. Methodologies are needed to study disease biomarkers that can help diagnose a disease at the biochemical level or help distinguish the responses of different patients to the same medical treatment.1 For protein biomarker detection and quantitation, several techniques are widely used including: gel electrophoresis,2 isoelectric focusing, affinity chromatography,3 surface enhanced laser desorption time-of-flight mass spectrometry (SELDI),4 proximity ligation,5 cell-based bioactivity assays6 and immunological assays, such as enzyme-linked immunosorbent assays (ELISA).7 Many of the above methods are time-consuming and labor intensive; and require multiple steps, such as immobilization, repeated incubations and washings, and additional reagents to amplify the signal. Additional methods, such as nucleic-acid molecules labeled with two signaling moieties, can be used in different types of formats, for biomarker detection. Nucleic-acid sequences, known as aptamers, can adopt complex three-dimensional structures capable of selectively 1

PAGE 15

2 recognizing protein molecules.8 Thus, aptamers are good candidates for detecting diverse protein targets, including those proteins that do not naturally recognize or bind to DNA or RNA. Aptamer-based probes can be developed with the idea that when the target protein binds to the detector molecules, a change in the optical properties of the reporter molecules will be triggered, to indicate the recognition event. Unfortunately, little is understood about the basic mechanism of action for aptamer-based signal transduction mechanism or its applications. With careful design and systematic characterization, aptamer-based probes could be useful in diagnostic applications and protein-biomarker studies. Aptamers What Is an Aptamer? Nucleic acid binding species generated by in vitro selection have been referred to as aptamers.9 Aptamers can be RNA, modified RNA, single-stranded DNA or double-stranded DNA, and have been selected to bind targets ranging from small organic molecules to cells.10 In vitro selection has been used to identify aptamers against different targets covering a wide range of sizes, including ions,11 small molecules,9 peptides,12 single proteins,13 organelles,14 viruses,15 and even entire cells.16 In addition to carrying genetic information, oligonucleotide sequences (aptamers) can adopt complex three-dimensional structures, capable of specific binding to target molecules and, furthermore, of catalyzing chemical reactions.17 Aptamers have been generated for over 100 proteins, including growth factors, transcription factors, enzymes, immunoglobulins, and receptors.10 At present, an aptamer database10 that contains sequences drawn from 239 published in vitro selection experiments available for public use. A typical aptamer is 10 to 15 kDa in size (30 to 45 nucleotides), binds its target with subnanomolar affinity, and

PAGE 16

3 discriminates against closely related targets (e.g., will typically not bind other proteins from the same gene family). A series of structural studies show that aptamers can use the same types of binding interactions (hydrogen bonding, electrostatic, hydrophobic contacts, steric exclusion, etc.) that drive affinity and specificity in antibody-antigen complexes and are exploited in systematic evolution of ligands by exponential enrichment, the aptamer-selection method. The Aptamer Selection Process: Systematic Evolution of Ligands by Exponential Enrichment (SELEX) In 1990, Gold 13,Ellington and Szostak9 simultaneously reported the development of an in vitro procedure for the selection of single-stranded nucleic acids to perform a specific function. This in vitro procedure has been called systematic evolution of ligands by exponential enrichment (SELEX). Since its introduction over 10 years ago, in vitro selection has been widely adopted as a tool for the development of research reagents, and shows promise generating diagnostic and therapeutic agents. The SELEX process consists of exponentially selecting and amplifying aptamers from a large library (1015 to 1018) of oligonucleotide molecules.18-21 (Figure 1-1). The starting point for the in vitro selection process is a combinatorial library composed of single-stranded nucleic acids (RNA or DNA) typically containing 20 to 40 randomized positions. Randomization creates an enormous diversity of possible sequences (e.g., four different nucleotides at 40 randomized positions gives a theoretical possibility of 440 or ~1024 different sequences). Since short, single-stranded nucleic acids adopt somewhat rigid structures (imposed by their sequences), such a library contains many molecular shapes or conformations. To identify high-affinity nucleic acid ligands to a given target protein, the starting library of nucleic acids (in practice, ~1014 to 1015

PAGE 17

4 different sequences) is incubated with the protein of interest. Nucleic acid molecules that adopt conformations that allow them to bind to a specific protein are then partitioned from other sequences in the library that are incapable of binding to the protein under the conditions used. 1014 10 15 DNA or RNA sequences Incubate with specific target Figure 1-1.Overview of the systematic evolution of ligands by exponential enrichment process. The initial library of nucleic acids is incubated with the protein target of interest. Molecules that bind to the target protein are then partitioned from other sequences in the library. The bound sequences are then amplified to generate a library enriched in sequences that bind to the target protein. The bound sequences are then removed from the protein and amplified by reverse transcription polymerase chain reaction (RT-PCR) (for RNA-based libraries) or PCR (for DNA-based libraries) to generate a reduced-complexity library enriched in sequences that bind to the target protein. The DNA population is then used to generate a new library of single-stranded molecules (transcription for RNA; strand separation for DNA) that is again subjected to the selection procedure. This process is repeated, until a group of Partition Bound Unbound Repeat process with selected sequences Amplify selected DNA/RNA Protein target

PAGE 18

5 single-stranded molecules are selected whose binding affinities are sufficient for the intended application. Aptamers versus Antibodies Most of the ligand-detection methods known in the bioanalytical sciences are based on antibody binding. Like aptamer-based methods, antibody-based detection methods require maintenance of the ligand-antibody complex, to generate a detectable signal. In addition, antibody methods such as ELISA or competitive radioimmunoassay (RIA), while robust and highly sensitive, have limited applicability because of the require heterogeneous assay conditions: The detection must be done on a solid surface In most applications both a capture antibody and detection antibody are required For ELISA-based protein detection methods, the antibodies must recognize the folded, native structure of the protein that is present in cell or tissue isolates.7;22;23 As an alternative (or complementary) molecule to antibodies, aptamers have several inherent advantages that merit study. Since aptamers consist of a short, single strand of DNA or RNA, they are much easier and economical to synthesize and have a much longer shelf life. In terms of stability, aptamers are chemically robust and are intrinsically able to regain activity after exposure to heat and denaturants, and can be stored for extended periods (more than a year) at room temperature as lyophilized powders. In contrast, antibodies must be stored refrigerated. Aptamers are chemically synthesized and thus can be readily scaled as needed to meet production demand. Whereas difficulties in scaling production currently limit the availability of some biological products and the cost of a large-scale protein production plant is high, a single large-scale synthesizer can produce more than 100 kg of oligonucleotide per year and

PAGE 19

6 requires a relatively modest initial investment.18 One of the most significant advantages of aptamers is that their selection process mimics natural selection, so it is possible, in theory, to develop a highly specific aptamer for virtually any target molecule with the same high affinity similar offered by antibodies. The affinities of aptamers range from dissociation constants of 0.3 to 500 nM, with most aptamers having binding affinities in the range of 1 to10 nM (comparable to or grea ter than the typical Fab fragments obtained through immunization). In comparison to engineering antibodies for particular applications, incorporating site-specific labels or coupling sites into an aptamer is usually a minor procedure. Since the aptamer itself lacks any means of signal transduction on binding a protein, this ease of molecular labeling is crucial for their development into fluorescence probes for protein studies. Aptamers in Bioanalytical Chemistry Aptamer molecules have been developed and used to detect specific analyte molecules in different experimental settings. RNA or DNA molecules must create a specific cavity to enclose and recognize a small molecule. Nonetheless, proteins possess extensive surfaces (with ridges, grooves, projections, and depressions) with several hydrogen-bond donors and acceptors; suggesting that proteins may be excellent targets for aptamer-probe developmental studies.9 The first known nuclei-acid aptamer to a protein that does not normally interact with RNA or DNA was a single-stranded DNA aptamer against thrombin.24 The aptamer for thrombin has been used extensively for multiple applications. In one case25, fluorescently labeled anti-thrombin aptamers attached to a glass surface were used to detect the presence of thrombin proteins in a sample, by detecting changes in the optical properties of the aptamers. In this system,

PAGE 20

7 binding of thrombin to the labeled aptamer is monitored by detecting fluorescent emission of the dye-labeled aptamer on excitation by an evanescent field. McGown and her group at Duke University26 used the same aptamer for affinity capture of thrombin with Matrix Assisted Laser Desorption and IonizationTime of Flight-Mass Spectrometry (MALDI-TOF-MS). The aptamer was covalently attached to the surface of a glass slide that served as the MALDI surface. Results showed that thrombin is retained at the aptamer-modified surface while other proteins, such as albumin, are removed by rinsing with buffer. Upon application of a low-pH MALDI matrix, the G-quartet structure of the aptamer unfolded, releasing the captured thrombin. After TOF-MS analysis, residual matrix and proteins were washed from the surface, and buffer was applied to refold the aptamers, allowing the surface to be reused. The aptamer demonstrated high selectivity against mixtures of thrombin and albumin and of thrombin and mixtures of prothrombin from human plasma. Aptamers can also be used to separate closely related compounds depending on their dissociation constants.27 Deng et al 28 explored the use of a 42-mer DNA aptamer for adenosine/ATP binding, as a weak affinity chromatography stationary phase using biotin-avidin coupling. Their data demonstrated that affinity chromatography is useful for characterizing the binding of aptamers to multiple targets. In addition, they evaluated the aptamer stationary phase for separation of cyclic-AMP, NAD+, AMP, ADP, ATP, and adenosine. Their data demonstrated that an aptamer immobilized by biotin-avidin coupling can selectively retain and separate related target compounds by weak affinity chromatography. The reported data suggested that it may be possible to develop stationary phases that selectively retain and separate classes of compounds based on

PAGE 21

8 aptamers and that affinity chromatography is a valuable and convenient technique for characterizing the affinity and selectivity of aptamers. Other nucleic acid-based detection schemes have exploited the ligand-sensitive catalytic properties of some nucleic acids, such as ribozymes. For example, Robertson and Ellington29demonstrated that a ribozyme that acquires a ligase activity on ligand binding can be used to detect a ligand by monitoring the ligation of a small, labeled, second oligonucleotide to the ribozyme. In a complementary approach, 30;31 labeled allosteric ribozymes that undergo cleavage upon binding to a ligand have been used to detect ligands by monitoring the release of the label from the ribozyme. However, the majority of the detection techniques have the disadvantage that the ligand-activated ribozyme is irreversibly modified in the course of generating a signal. Thus, ribozymes can be used only once in an assay. Furthermore, signal generation is slow with ribozymes and can take from 1 minute to 1 hour or more. Manalis32 developed a method for label-free protein detection using a microfabricated cantilever-based sensor that was functionalized with DNA aptamers to act as receptor molecules. The sensor used two adjacent cantilevers that constituted a sensor/reference pair and allowed direct detection of the differential bending between the two cantilevers. In this case one cantilever was functionalized with aptamers selected for Taq DNA polymerase while the other was blocked with single stranded DNA. They reported that the polymerase aptamer binding induced a change in surface stress, which caused a differential cantilever bending dependent on the ligand concentration. Protein recognition on the sensor surface was specific and had a concentration dependence that is similar to that in solution.

PAGE 22

9 Another class of aptamer probe is based on monitoring the change in fluorescence anisotropy on protein binding.21;25;33 Anisotropy probes are designed by labeling an aptamer with a single fluorophore molecule. When the labeled aptamer is bound with its target protein, the rotational motion of the fluorophore becomes much slower as a result of the larger molecular weight of the aptamer-protein complex. As a consequence of the binding event, a significant increase in fluorescence anisotropy can be monitored using plane-polarized light. Anisotropy-based aptamer probes serves as an alternative mechanism of signal transduction for aptamer-protein binding which results in on and off type signals. Some advantages of anisotropy-based aptamer probes versus analogous antibody probes are a result of their respective sizes. Since most aptamers are substantially smaller than antibodies, the relative increase in molecular weight and fluorescence anisotropy will be much larger on binding with proteins for aptamer probes. However, the sensitivity of aptamer anisotropy probes is generally not as good as fluorescence resonance energy transfer-based probes. For Platelet Derived Growth Factor (PDGF-BB) detection, an aptamer was converted into an anisotropy probe through 5-labeling with fluorescein.33 Addition of PDGF-BB (nM concentration) to an equimolar solution of the labeled aptamer increases the fluorescence anisotropy more than 2-fold. The binding of protein to the aptamer clearly results in a large molecular-weight complex, thereby slowing down the rotational diffusion of the fluorescein label. The binding reaction is quick, with equilibrium achieved in less than 20 s, including mixing time. Controls using free fluorescein dye with PDGF-BB show no effect on anisotropy, validating that the observed changes are a direct result of the aptamerPDGF-BB binding. All the reported applications of aptamers

PAGE 23

10 demonstrate that with the use of aptamer-based molecules it is possible to detect analyte binding in both solution (homogeneous) and on solid supports (heterogeneous). Aptamers are required to work along with a signal transduction mechanism to report the binding event to the target molecule. High affinity and selective aptamer probes can be further exploited by incorporating a highly sensitive signal transduction mechanism within the synthetic DNA sequence that would allow their use for single-step detection schemes without the need of separating the aptamer-target complex from the solution. Fluorescence Methods for Signal Transduction Fluorescence is a highly sensitive detection method. Figure 1-2 shows a typical Jablonski diagram where S0, S1 and S2 stand for the singlet, first and second electronic states, respectively, while T1 stands for triplet state. Following light absorption, a fluorophore is excited to a higher vibrational level. By a process called internal conversion, the molecules in condensed phases rapidly relax to the lowest vibrational level of S1. Internal conversion is generally complete prior to emission; hence, fluorescence emission generally results from the lowest-energy vibrational state of S1. S0 S1 S2 T1 FLUORESCENCE INTERSYSTEM CROSSING INTERNAL CONVERSION hhhh PHOSPHORESCENCE Figure 1-2. Typical Jablonski diagram

PAGE 24

11 Upon excitation into higher electronic and vibrational levels, the excess energy is quickly dissipated leaving the fluorophore in the lowest vibrational level of S1. Since this is a rapid relaxation, emission spectra are usually independent of the excitation wavelength. One of the general characteristics of fluorescence is that it exhibits what is known as the phenomenon of Stokes Shift. This phenomenon reveals that the energy of the emission is typically less than that of absorption; hence, fluorescence typically occurs at lower energies or longer wavelengths. The intensity of fluorescence can be decreased by a wide variety of processes known as quenching. Fluorescence Quenching The process of decreasing fluorescence intensity is termed quenching and it usually happens through two major mechanisms.34 The first mechanism, known as collisional quenching, takes place when a fluorophore is excited by incident light and as a consequence goes into an excited state and stays there for a short period of time (nanoseconds). The time it stays in the excited state is called fluorescence lifetime. In this case, the quencher must diffuse to the fluorophore during the lifetime of the excited state. Upon contact, the fluorophore returns to the ground state, without emission of a photon. Collisional quenching is described by the Stern-Volmer as shown in Equation 2-1; where F0 and F are the fluorescence intensities in the absence and presence of quencher, respectively, kq is the bimolecular quenching constant, 0 is the lifetime of the fluorophore in the absence of the quencher, and [Q] is the concentration of the quencher. F0/F=1+kq0[Q]=1+KD[Q] (2-1) The second mechanism of fluorescence quenching, named static quenching, is characterized by the formation of a nonfluorescent complex between a fluorophore and a

PAGE 25

12 quencher. When fluorophore-quencher complex absorbs light, it immediately returns to ground state without emission of a photon. Fluorescence Resonance Energy Transfer (FRET) Fluorescence resonance energy tranfer is an interesting fluorescence-related phenomenon which relies on the radiationless transfer of energy from a donor fluorophore to an acceptor fluorophore. The distance dependent transfer of energy is a physical process that depends on the spectral overlap of the donor and acceptor spectra and proper dipole alignment of the two fluorophores. FRET is considered to be one of the few tools available for measuring nanometer scale distances and changes in distances, both in vitro and in vivo. In FRET, a donor fluorophore is excited by incident light, and if an acceptor is in close proximity, the excited state energy from the donor can be transferred. This leads to a reduction in the donors fluorescence intensity and excited state lifetime, and an increase in the acceptors emission intensity. One of the requirements for FRET is that the two molecules need to be very close, typically less than 10 nm. In fact, Lakowics34 showed that the efficiency of this process (E) depends on the inverse of the distance, between donor and acceptor, to the sixth power as shown in Equation 2-2, where E is the energy transfer efficiency, r is the distance between the donor and the acceptor, and Ro is the Forster distance at which half of the energy is transferred (typically 20 to 60 ), and depends on the spectral characteristics of the dyes and their relative orientation. When non-fluorescent molecules are used as acceptors, the result of the energy transfer is quenching of the donor fluorescence. E = R06/R06+r6 (2-2)

PAGE 26

13 ABSORPTION ENERGY TRANSFER hh DONOR ACCEPTOR Figure 1-3. Fluorescence resonance energy transfer diagram. Whereas normally an excited fluorophore returns back to the ground state with the emission of a photon, FRET results in the excitation of the nearby acceptor fluorophore that in turn emits a photon when it returns to the ground state. The occurrence of FRET is characterized by a decrease in observed donor emission, and a simultaneously sensitized (increased) acceptor emission. Fluorescence resonance energy transfer is a distance-dependent phenomenon that can be applied to the development of aptamer probes to study intermolecular and intramolecular relationships in biophysical systems and cell biology. The development of an efficient aptamer probe is not only dependent on the oligonucleotide sequence and conformation but also on the selection of the fluorophore and the quencher pair. With the high diversity of existing fluorophores and the growing number of the alternatives quenchers, the aptamer probes can be modified to fulfill specific experimental needs. Several reports have been published with updated information regarding the efficiencies of FRET for commonly used fluorophore-quencher pairs and their effects on duplex stability.35 Various approaches, including the well studied molecular beacons, have exploited the ability of fluorescent compounds to absorb energy and transfer it to nearby molecules for the study of molecular interactions.36

PAGE 27

14 Molecular Beacons Molecular beacons (MBs) 36 are synthetic oligonucleotide probes that possess a stem-and-loop structure, designed for specific recognition of DNA or RNA targets. The basis for target recognition is the hybridization of the nucleic acid to its complementary target. The single-stranded loop portion of the MBs has a sequence complementary to the target DNA and can report the presence of specific target nucleic acids. The stem typically has five to seven base pairs which are complementary to each other but unrelated to the target. Signal transduction in MBs is accomplished by resonance energy transfer. A fluorophore (donor molecule) and a quencher (acceptor molecule) are covalently linked to the two ends of the stem. The stem keeps the two moieties in close proximity, causing the fluorescence of the fluorophore to be quenched by energy transfer. When the probe encounters a target DNA molecule, the molecular beacon undergoes a spontaneous conformational reorganization, leading to the formation of a hybrid that is longer and more stable than that of the stem, that forces the stem apart and a consequent fluorescence restoration. The conformational state of a molecular beacon is thus directly reported by its fluorescence: in the closed state, the molecular beacon is not fluorescent; in the open state, when the fluorophore and the quencher molecules are apart, it emits intense fluorescence. Different molecular beacons can be designed by choosing loop sequence and length. Also, the quencher and the fluorophores can be changed according to the application. Molecular Beacons for Non-Specific Protein Detection Although molecular beacons were originally designed to bind and recognize specific nucleic acids, the probes can also lead to increased fluorescence on binding to certain proteins. Since the binding of proteins to DNA or RNA molecules can readily

PAGE 28

15 Figure 1-4. Working principle of molecular beacon. Before hybridization with target nucleic acid sequences the MB takes on a stem-loop structure which maintains the close proximity of the fluorophore (F) and quencher (Q) moieties. The resulting fluorescence is minimal due to static quenching of the fluorophore molecule. After introducing target nucleic acid sequences, the loop sequence of the molecular beacon hybridizes to the target and unstabilizes the stem hybrid. Consequently, the two moieties spatially separate and result in the restoration of the fluorescence. disturb the conformation of the nucleic acid, it is expected that this binding would result in spatial separation of the MB fluorophore and quencher. As an example37, an E.coli single stranded DNA binding protein (SSB) was used to demonstrate the protein recognition capability of MBs. This DNA-binding protein is used in DNA replication, recombination, and repair. Using a conventional fluorescence spectrofluoremeter, the investigators detected as low as 20 nM of SSB using a MB labeled with tetramethylrhodamine (TMR) and dimethylaminophenylazobenzoic acid (DABCYL). The measured fluorescence intensity changes over time revealed that the SSB-MB interaction is rapid, reaching equilibrium within 10 s. The MB-based SSB assay is not, however, particularly specific. In fact, SSB leads to a fluorescence enhancement nearly equal to that of the complementary DNA, but other proteins

PAGE 29

16 (including histone and Rec A) can also bind with the MB and cause a fluorescence intensity increase. The results demonstrated that while MBs were sensitive and somewhat selective to DNA-binding proteins, they were not specific enough to be capable of distinguishing a particular protein. In order to apply the principle of MBs for real-time detection of protein biomarkers, a more selective protein recognition mechanism is required. Aptamers and FRET for Protein Detection In addition to exploiting the inherent conformational changes that aptamers undergo, aptamers have been engineered similar to MBs such that the addition of an analyte results in a conformational change and simultaneous reduction or increase in a fluorescent signal. Utilizing FRET between fluorophore and quencher moieties for signal transduction of protein binding is one of the more popular designs. Depending on the relative positions of the fluorophore and quencher before and after protein binding, probes can result in either enhanced or reduced fluorescence on binding. It is important to mention that the design of optimized probes is not always a trivial process. A maximum change in fluorescence on binding is often achieved by gaining knowledge about the conformations of the free aptamer and the aptamer-protein complex. Ideally, when this structural information is available, one can strategically incorporate the fluorophore and quencher moieties on the aptamer probe in such a way that the transition from bound to unbound conformations causes a dramatic change in their relative positions. An example of a FRET based quenching aptamer probe is the one developed using a thrombin aptamer.37 The first reported aptamer for thrombin contains a 15-nucleotide consensus sequence.38 When bound to thrombin, the aptamer exists primarily in its quadruplex form containing two G-quartet structures, but in free solution, it can adopt

PAGE 30

17 either conformation, dependent in part on the ionic strength and temperature. This conformation shift provides the basis for an aptamer probe. By labeling the two ends of the aptamer with a fluorophore and quencher pair and extending the aptamer by one base on each end, aptamer binding of thrombin would force the quencher adjacent to the fluorophore, resulting in a substantial decrease in fluorescence. Another method of detecting binding of a target protein to an aptamer has also been described which relies on the use of fluorescence-quenching pairs whose fluorescence is sensitive to changes in secondary structure of the aptamer upon ligand binding.39 However, target protein-mediated changes in secondary structure were engineered into the aptamer molecule via a laborious engineering process in which four to six nucleotides were added to the 5' end of the aptamer that was complementary to the bases at the 3' end of the thrombin binding region. In the absence of thrombin, this structure forms a stem loop structure, while it forms a G-quartet structure in the presence of thrombin. Fluorescent and quenching groups attached to the 5' and 3' end signal this change. A related class of aptamer probes is the two fluorophore FRET probes.40 Binding of target protein can be detected by monitoring the fluorescence of a second fluorophore (F2) directly or preferably by the ratio of fluorescence of both fluorophores (F2/F1). Ratiometric detection may provide enhanced sensitivity, with reported detection limits in the low pM regime. All of the design considerations applicable to the quenching aptamer probes are also crucial to this class of aptamer probes. Detailed structural information of the aptamer and aptamer-protein complex allows optimized positioning of the fluorescence donor and acceptor.

PAGE 31

18 Molecular Beacon Aptamer (MBA) Development for a Model Protein System Based on previous studies that showed that the high affinity and selectivity of aptamers can be successfully combined to develop aptamer probes, it is of interest to further investigate and develop aptamers for biomolecule studies. In this case, PDGF-BB was chosen based on the significance of the protein in biological systems and the fact that an aptamer had already been selected for this protein. Platelet derived growth factor was discovered as a major mitogenic factor present in serum but absent from plasma. The biological function of PDGF-BB is to stimulate the division and proliferation of the cells through binding to receptors on the cell surface. Studies showed41-44 that PDGF is not one molecule but three, each a dimeric combination of two distinct but structurally related peptide chains designated A and B. Dimeric isoforms PDGF-AA, AB and BB are differentially expressed in various cell types and their effects are mediated through two distinct receptors, named and Platelet derived growth factor is considered a cancer biomarker because it is expressed at low or undetectable levels in normal cells, but it is found to overexpressed in many tumor cells and the autocrine and paracrine effects of PDGF-BB increase w ith the degree of malignancy. 45;46 In addition to cancer, PDGF has been implicated in other diseases, including renal disease. There is a hypothesis that PDGF is important to cell transformation processes, tumor growth and progression and thus could be used as a cancer marker for diagnosis. The assumed biological foldings for PDGF-BB are shown in the Figure 1-5. Aptamers for Platelet Derived Growth Factor-BB (PDGF-BB) The first step towards the development of a novel MBA is the selection of an aptamer. In 1996, Louis Green and his group47 used the SELEX procedure for the development of an aptamer for high affinity and selectivity against the B chain of PDGF

PAGE 32

19 BB. The first selection was initiated by incubating approximately 3x1014 molecules of random ssDNA with PDGF-AB in binding buffer. The two best ligands from this group were identified by their relative affinities for PDGF-BB and its molecular variants over a range of protein concentrations. Figure 1-5. Assumed biological foldings of Platelet Derived Growth Factor (PDGF) 44 The biological molecule is the macromolecule that has been shown to be or is believed to be functional. In both cases, the affinity of ligands for PDGF-BB was higher than the affinity for the other PDGF isoforms. The selected aptamers, 41t and 36t, shared a secondary structure motif: a three-way helix junction with a three-nucleotide loop at the branch point.(Figure 1-6) Two of the helices end in highly variable loops at the distal end from the junction, suggesting that the regions distal from the helix junction are not important for high-affinity binding to PDGF. The highly conserved nucleotides are indeed found near the helix junction. The 36t aptamer is composed of 39 DNA bases with a dissociation constant of 0.093nM. The other selected aptamer, 41t, is a 44-mer sequence with a higher dissociation constant of 0.129 nM. Based on theoretical determinations, both sequences appear to have several possible conformations before and after binding of the protein. Using this information, the possibility exists to develop a new aptamer based detection scheme for PDGF. Also, given that the protein has multiple isoforms, this makes the PDGF system particularly

PAGE 33

20 attractive to explore because we can evaluate the different interactions of a single aptamer with a family of proteins that have, in this case, ~60 % identity. In the next few chapters, a systematic investigation of the PDGF-BB protein-aptamer system will be presented. A variety of conditions have been evaluated for the effective recognition of the protein as well as several transduction mechanisms studied for potential use in biological studies. Figure 1-6. Folding of 36t molecular beacon aptamer (MBA) The scope of the work presented here was to conduct a systematic investigation of a novel design for an aptamer-based biomolecule, used to detect and study a target biomarker PDGF-BB. The combination of a nucleic acid molecule with different fluorescence phenomena as signal transduction mechanisms will be investigated. Potential applications of this probe in multiple assay environments and assay formats will be explored and evaluated for homogeneous solution and in cellular samples.

PAGE 34

CHAPTER 2 DEVELOPMENT AND CHARACTERIZATION OF A MOLECULAR BEACON APTAMER FOR PLATELET DERIVED GROWTH FACTOR Introduction Proteins play very important roles in almost all functions of life. Molecular probes for specific and sensitive detection of proteins and their molecular variants are necessary in many biotechnology applications and biomedical studies. Monoclonal antibody-based immunoassays have been used for protein analysis and studies during the past decades. Most immunological methods and many nucleic-acid-based methods involve multiple steps to achieve amplification of the specific signal produced by a target protein. By attaching a fluorophore and a quencher to an aptamer, the high degree of sensitivity afforded by fluorescent signals40;48;49 can be combined with the selectivity of the DNA binding to the target protein to form a single-step molecular beacon aptamer (MBA) assay. Fluorescence changes upon the interaction between a fluorophore and a quencher have been studied using molecular beacons (MBs) for DNA and RNA targets. Both, MBs and MBAs share a common characteristic: they have similar inherent signal transduction mechanisms and allows for target detection and reporting without removing for probe-target complex from the sample solution (detection without separation). Some differences between the two probes include the fact that MBs are mostly used for DNA and RNA detection and are based on the hybridization of the loop sequence to the target sequence. MB probes interact with their complement primarily through hydrogen bonding at 21

PAGE 35

22 various sites, depending on the length of the hybrid. In contrast, the use of MBAs is mainly used for the detection of protein targets and relies on a mixture of forces: hydrophobic, ionic and hydrogen bonding, at only a few specific sites. The basis for differences in secondary structures between the free and bound forms of MBAs for protein targets is not well understood, however interactions with different proteins are likely to result in secondary structure changes that could be used to produce FRET signals.40 Platelet derived growth factor is a dimeric protein for which several natural molecular variants are known, and at least two natural variants of cell-membrane receptors have been described with different specificity for the PDGF variants.41-43 The expression of variants of PDGF and the PDGF receptors have been implicated in malignancy and developmental abnormalities.50;51 The use of MBAs for protein targets is in its infancy and the basis for differences in secondary structures between the free and bound forms of MBAs for protein targets is not well understood. Theoretical models predicted that when the aptamers selected for PDGF-BB are free in solution, several stable conformations are possible. Based on the previously described MBA for thrombin which forms a G-quartet after binding to the protein, the possibility existed for the PDGF aptamer to undergo a conformational change after binding to the protein as a result of stabilizing one of the predicted conformations. Initially, it was unclear whether the stem structure would be open or closed prior to protein binding. Therefore, the first PDGF-BB probe was designed as an anisotropy probe in an effort to explore the application of an aptamer anisotropy probe for protein analysis. The PDGF-BB aptamer was labeled with a single fluorescein dye at the 5' end.33 PDGF-BB in the nM range, when added to the

PAGE 36

23 aptamer solution, caused an anisotropy increase due to the larger molecular weight of the complex. The results were our initial attempts toward developing a PDGF-BB probe. Given the possibility of conformational changes that could occur due to binding of the protein, a distance dependent signal transduction mechanism was incorporated into the aptamer molecule. The initial hypothesis was that a FRET-based aptamer probe could be designed for the PDGF aptamer. In this chapter, the result of a systematic study of the chemical and physical properties of a fluorescence quenching DNA aptamer probe will be presented. Materials and Methods Synthesis and Purification of MBAs All DNA molecules were obtained from Geno-Mechanix, LLC (Gainesville, FL) and the general synthesis protocols are described in this section. The DNA molecules were synthesized using the standard phosphoramidite chemistry in an Expedite 8909 automated DNA synthesizer on controlled pore glass beads, deprotected in ammonium hydroxide and purified by gel filtration. Unmodified DNA aptamers were purified by ion-exchange HPLC followed by gel filtration to remove salts. The standard MBA was synthesized using DABCYL immobilized on controlled pore glass beads (BioSearch Technologies, Novato, CA) and fluorescein (6-FAM) was added at the 5-end using 6-carboxyfluorescein phosphoramidite (BioSearch Technologies, Novato, CA). The dually modified DNA was deprotected at 55C for 6 hr with ammonium hydroxide (50%) in methanol. The deprotected DNA was first subjected to gel filtration followed by two cycles of reverse phase HPLC using two different hydrophobic resins with triethylammonium acetate (pH 6.0, 0.05 M) buffer and eluted with an acetonitrile

PAGE 37

24 gradient. The final purified material was subjected to gel filtration to remove solvents, dried under vacuum and stored dry at -20C until use. MBAs with tetramethylrhodamine (TMR)-6-FAM pair was synthesized and purified as described above using TMR-immobilized on controlled pore glass (BioSearch Technologies, Novato, CA), except that deprotection was carried out in potassium carbonate (0.1M) in methanol (80%) for 4 hr at 55C before processing the deprotected dual labeled DNA. For the synthesis of MBA with Black Hole Quencher 2 (BHQ2)Texas Red pair, synthesis was done using BHQ2 immobilized on controlled pore glass (BioSearch Technologies, Novato, CA) and a 5-modification with a functional amine group using a C3 spacer. Following deprotection at 55C with ammonium hydroxide/methanol and gel filtration, the DNA was reacted with the N-hydroxysuccinimidyl ester of Texas Red (Molecular Probes, Eugene, OR) in sodium borate buffer (pH 8.2, 0.1 M) and the labeled material was again subjected to gel filtration prior to the two cycles of reverse phase gel filtration as described above. For the synthesis of MBA labeled with Cy3 and Cy5, the DNA was synthesized with a functional amine group immobilized on controlled pore glass beads and Cy5 was added to the 5-end using Cy5 phosphoramidite (Amersham Bioproducts, Piscataway, NJ). Following deprotection with potassium carbonate (0.1M)/methanol and gel filtration, the DNA was reacted with N-hydroxysuccinimidyl ester of Cy3 (Amersham Bioproducts, Piscataway, NJ) and the dual labeled material was purified as described for Texas Red labeling above. All the MBA stock solutions were prepared at 100 M concentration in Tris/HCl buffer (pH 7.5, 20 mM) with NaCl (20 mM), and stored at -20C in aliquots. The DNA aptamer sequence in the standard 36tMBA used in most of the studies

PAGE 38

25 described here is the one designated 36t by Green et.al (5-CACAGGCTACGGCAC GTAGAGCATCACCATGATCCTGTG-3) and contains a 6 base complementary sequence at the 5 and 3ends that can form a double stranded stem in a closed conformation.47 Recombinant human PDGF-BB, PDGF-AB, and PDGF-AA were purchased from R&D Systems (Minneapolis, MN). PDGF and its isoforms were reconstituted in 4 mM hydrochloric acid with at least 0.1% bovine serum albumin (BSA). Other recombinant human growth factors used in the selectivity experiments, epidermal growth factor (EGF), insulin-like growth factor-I (IGF1), were bought from Roche (Indianapolis, IN). Human hemoglobin (HEM), porcine lactic dehydrogenase (LDH), myoglobin (MYO), chicken lysozyme (LYS), and human gamma-thrombin (THR) were purchased from Sigma (St. Louis, MO). Other biomolecules used include: BSA, (New Englands Biolabs, MA), thrombin (Haematologic, Inc, VT), ovalbumin (US Biological, Sawmpscott, MA) and, glycogen (Fisher Biotech, USA).The binding buffer used for all the characterization experiments was 20 mM Tris-HCl (pH 7.5) with 20 mM sodium chloride. Standard Fluorescence Quenching Assay All the characterization experiments for the MBA measurements were carried out in a 100 l cuvette (Starna Cells). Standard fluorescence quenching assays were performed by incubating the desired concentration of protein target with a 50 nM solution of the MBA in binding buffer at room temperature in a final volume of 100 l. All fluorescence measurements were monitored using a Fluorolog-Tau-3 spectrofluorometer (Jobin Yvon Inc., Edison, NY) equipped with a thermostat accurate to 0.1C. The sample cell was a 100 l cuvette (Starna Cells,Atascadero, CA). The fluorescence emission of 6

PAGE 39

26 FAM was monitored at the excitation maximum, 480 nm, and the fluorescence intensity was measured at the emission maximum, 520 nm. Both excitation and emission slits were varied to yield the best signals. The results are reported as mean values of triplicates. The effect of temperature on MBA fluorescence was monitored using a Fluorolog-Tau-3 spectrofluorometer (Jobin Yvon Inc, NY) equipped with a thermostat accurate to 0.1C. The sample cell was a 100 l cuvette (Starna cells, Atascadero, CA). The fluorescence intensity of a 50 nm MBA solution in the binding buffer was monitored before and after the addition of 200 nM PDGF-BB. In order to adapt the fluorescence quenching assay to high throughput format, a Tecan Saphire (Durhan, NC) fluorescence microplate reader was used with 96-well flat bottom microtiter plates (Nalge Nunc International, Rochester, NY). All experiments were carried in a final volume of 100 L, and the excitation and emission maxima were selected according to the fluorophores used. Results and Discussion Aptamer probes have significant appeal in the development of new methodologies for purification, labeling, and inhibition and characterization of proteins; however, their use as molecular probes has been under development and advances made only in the last 5 years.8;52 One of our major goals for this research was to develop a single-step assay which could analyze protein concentrations of distinct biomarkers in biological samples. The initial efforts focused on the protein, PDGF-BB, for which two potential DNA aptamer sequences for probe development have been selected.47 To characterize the MBA and determine if the probe could operate in a typical biological sample, a variety of chemical and physical parameters were evaluated.

PAGE 40

27 Choice of Aptamer Sequence for MBA Development Based on the predicted secondary structures, the first step toward developing a FRET-based aptamer probe for PDGF was to determine if a secondary structure change could be observed on binding to PDGF. To accomplish this task, two MBAs which are based on the aptamers previously characterized,47designated 36t and 41t, were evaluated (Table 2-1). Each aptamer was converted into an MBA by attaching DABCYL at the 3'-end and 6-FAM at the 5'-end. Using thermodynamic software calculations, the aptamers were predicted to form a helix-loop conformation with a single stranded region at the 5' and 3' end as a possible conformation with a reasonable free energy in the absence of PDGF.53;54 As a result of labeling the two ends of the aptamer, the fluorescence intensity of the free aptamer sequence is expected to be fairly high since the fluorophore and quencher pair will be spatially separated. Upon the addition of PDGF-BB, if protein binding induces a conformational change in the stem structure of the aptamer a change in the fluorescence intensity should be observed. In a fluorescence versus time experiment, both aptamers were first monitored for approximately 300 s and then PDGF-BB was added at four-fold excess (Figure 2-1). Upon PDGF-BB addition the fluorescence signal decreases and reaches equilibrium rapidly. The quenching efficiencies were considerably different between the two aptamer sequences. The reported47 binding affinities demonstrated that the 36t aptamer has a higher affinity for PDGF consistent with the fact that this aptamer showed a higher fluorescence quenching. The labeling of the aptamer molecule with a fluorophore and quencher at each end of the oligonucleotide sequences should not affect the binding capabilities of the aptamers since it was reported47that the regions distal from the trinucleotide loop of both sequences are not critical for binding. Since FRET is dependent on the distance between the fluorophore and the quencher, it

PAGE 41

28 Table 2-1. Percentage of quenching of different consensus sequences of 50 nM of MBA for PDGF aptamer with 200 nM of PDGF Name Sequence % quenching 36t CACAGGCTACGGCACGTAGAGCATCACCATGATCCTGTG 94 41t TACTCAGGGCACTGCAAGCAATTGTGGTCCCAATGGGCTGAGTA 55 02000004000006000008000000100200300400500600700800Time ( s ) Fluorescence Intensity (A.U.) 41T 36T Figure 2-1. Fluorescence intensity changes versus time of different aptamer sequences. may be possible that preferential binding of PDGF stabilizes, or closes, the stem thus resulting in a decreased FRET response. Based on the observed decrease in the fluorescence signal, the 36t and 41t aptamer sequences appear to have a conformation change on binding to the protein. The PDGF FRET aptamer is proposed to have a stem-open conformation prior to binding, due to the high fluorescence signal that was observed (See Figure 2-2). Prior to binding one or several structural conformations could be possible, and PDGF most likely stabilizes the closed-stem structure for the 36t probe. Since the 41t aptamer had similar fluorescence prior to binding and quite different from after binding, most likely the structure of the 41t

PAGE 42

29 aptamer after binding is not a stem-closed structure but a partially-closed structure. Also, it could be possible that due to the decreased affinity of the 41t sequence for PDGF, the equilibrium fluorescence would be less in comparison to the 36t control aptamer. However since the experiments were done in 4-fold molar excess of PDGF, this is not likely to be the case because most of the aptamer probes should be bound. Another possibility for the observed difference could be the modifications themselves. If single fluorophore labeled DNA impurities existed within the solution, one might expect that the degree of quenching would be less. As a result other synthetic batches were tested and still similar results were obtained. PDGFA B Figure 2-2. Proposed mechanism of PDGF-BB induced FRET responses for the 36t probe. When the MBA is free in homogeneous solution, multiple conformations are possible. Upon binding of PDGF-BB, the probe conformation shifts from the opened (A), or partially opened (B), to a closed conformation. The fluorescence signal, therefore, is high prior to binding; and on stem formation, the signal is quenched. Evaluation of Additional Fluorophore Quencher-Pairs Since the 36t aptamer probe was clearly superior to the 41t in producing a quenching response, the 36t probe was selected to further develop and characterize the assay. One of the advantages of the MBA assay is that the use of FRET as a signal transduction allows for a single step detection of the target without having to isolate the MBA-protein complex from the solution. With FRET, alterations in the emission of

PAGE 43

30 fluorescent labels are measured. An important consideration in the design of aptamer probes for protein binding assays is the efficiency of energy transfer between the fluorophore and quencher used to label the probes. The versatility of the assay would be significantly improved if various known fluorophore-quencher pairs were found to be more effective for PDGF detection. Measurements of quenching efficiencies of different fluorophore-quencher pairs can be used to aid in the design of different kinds of MBAs. In the next section, several MBA molecules carrying known FRET pairs were studied and compared to the standard MBA labeled with 6-FAMDABCYL pair. The following pairs: Texas Red-BHQ2, Cy5-BHQ2 and 6-FAM-TMR were selected for their high quenching efficiency in other systems. The synthesis and purification of MBAs with FRET pairs is described in the Experimental Section and tested in the presence of a four-fold molar excess of PDGF-BB in the standard fluorescence quenching assay. As shown in Figure 2-3, all the selected FRET pairs were quenched when PDGF was added, albeit to different degrees. Two of the most commonly used quenchers, DABCYL and TMR, were paired against 6-FAM. Both quenchers may affect the sensitivity and flexibility of FRET assays. DABCYL is a dark quencher that absorbs broadly without emitting light; consequently, when paired with 6-FAM it can reduce more than 90% of the original fluorescence intensity of the MBA with high sensitivity and low background fluorescence. On the other hand, TMR is not a dark quencher and contributes to an overall increase in background because of its own native fluorescence. When both quenchers are paired against the same fluorophore for the MBA, DABCYL is superior in terms of overall fluorescence quenching against 6-FAM.

PAGE 44

31 0.00.20.40.60.81.0FAM-DABCYLFAM-TMRTexasRed-BHQ2CY5-BHQ2Relative fluorescence quenching MBA MBA + PDGF-BB Figure 2-3. Evaluation of fluorescence quenching caused by PDGF-BB with MBAs containing different fluorophore-quencher combinations. PDGF MBA molecules labeled with different fluorophore-quencher pairs were incubated in the standard fluorescence quenching assay buffer (final volume 100 L) with a four-fold molar excess of PDGF-BB (300 nM) to the concentration of the MBA (75 nM). Black Hole quenchers were designed to maximize the spectral overlap with many fluorophores and it does not re-emit the absorbed fluorescence as light, thus decreasing the background fluorescence. BHQ-2 was capable of quenching more than 50 % of the initial fluorescence intensity when paired with two commonly used dyes, Texas Red (TR) and Cy5 with different extents. In this particular case, the Cy-5 labeled MBA resulted in 74% quenching efficiency versus the Texas Red analog with only 54 %, even though previous reports35 predicted that both dyes, Texas-Red and Cy5 should have similar quenching efficiency mediated by contact quenching. Possible reason for the difference in quenching efficiency includes the possibility that the effective closed conformation of the MBA on binding to PDGF-BB is favorable for the FRET efficiency of the Cy5 labeled probe.34 In addition, when the BHQ-2-Cy5 pair is in close proximity, the dye-pair interaction actually increases the stem stability.55;56 Since FRET is a distance dependent

PAGE 45

32 phenomenon, it could be possible to improve the performance of other fluorophore-quencher pairs by changing the location of the label molecules in the MBA.57 The availability of choices in the selection of fluorophore and quencher is an important part of MBA-based assay development because various biological samples might present potential interference at certain excitation-emission spectra that may be effectively addressed by selecting the proper FRET pair. In addition, the development of a multiplex assays for the simultaneous detection of multiple proteins in a single homogeneous incubation may be possible by using two or more MBAs that are selective for different targets and that are labeled with different fluorophore-quencher pairs. The results above support the notion that MBAs are compatible with various fluorophores; however, the aptamer sequence, conformation, and binding principle will all potentially affect the FRET process. Each MBA that is developed would need to be fully characterized in order to obtain the most effective position and dye-pair for labeling. Once the MBA was developed with an appropriate fluorophore-quencher pair, we then investigated the applicability of the MBA for the in vitro detection of PDGF-BB in homogeneous solution by incubating a fixed amount of the 36t MBA with increasing amounts of PDGF-BB in binding buffer. As can be observed in Figure 2-4A, a dose-dependent fluorescence quenching was observed on the addition of PDGF-BB and 1.5 nM of PDGF was detected in repeated experiments. In comparison, an ELISA-based PDGF-BB assay was reported to have a detection limit of ~15 pg/mL (0.6 pM) PDGFBB (Figure 2-4B). The almost 100-fold difference in detection capabilities is understandable in this case.

PAGE 46

33 010000200003000040000500000255075100PDGF (nM)Fluorescence Intensity (A.U.) 0100002000030000400005000002468 A 0.00.51.01.52.002004006008001000PDGF (pg/mL)Absorbance B Figure 2-4. Calibration curves for PDGF-BB. (A)Dose-dependent fluorescence quenching of the MBA 36t-MBA on addition of PDGF-BB. Increasing concentrations of PDGF-BB were incubated in the standard fluorescence quenching assay buffer (final volume 100 uL) with the 36tMBA labeled with Dabcyl quencher at the 3end and 6-FAM at the 5-end (fluorescence measurements were made at the excitation and emission maxima for 6-FAM) (B) Calibration curve obtained with PDGF-BB ELISA kit (RnDsystems)

PAGE 47

34 In the MBA assay, the fluorescence response is produced as the result of a single (or possibly two as will be shown in later chapters) fluorophore molecule being quenched in the presence of a single PDGF-BB molecule. Since the ELISA assays utilizes an enzyme to effectively amplify the signals for each bound protein, clearly the sensitivity and detection limits of such assays will be better. However, ELISA does present the difficulty of being extremely time consuming (~7 h for one set of analysis). The time factor is essentially attributed to the washing and incubation steps which allow for binding of target molecules and the removal of unbound species. Since the MBA bioassay involves only the mixing of two solutions, the MBA solution and the analyte solution, and a short binding time, the analysis time is much shorter (~30 min for the MBA assay). As a result, in terms of sensitivity, if one can sacrifice analysis time and the detection limits and sensitivity are not an issue, then the MBA assay will provide a simple single step capability for screening purposes. Incubation Conditions for MBA Bioassay Effects of temperature. A critical factor in the MBA-based bioassay is the difference in distance between the quencher and the fluorophore in the free MBA (open conformation) and the target-bound MBA (closed conformation). Physical factors that affect DNA conformation and the distance between the FRET pairs are expected to influence the dynamic range of the assay. It is unclear as to the exact mechanism for PDGF binding to the 36t control sequence; however, it is believed that the opened, or partially open, conformation is bound by the protein and then the stem is consequently stabilized. At higher temperatures, the free MBA would preferentially assume a more open conformation and increase the fluorescence signal. Conversely, at lower

PAGE 48

35 temperatures, the equilibrium would be shifted to a more closed conformation. Both instances could potentially influence the binding of PDGF by either inhibiting it or increasing its ability to bind. In order to assess the effect of temperature on the MBA probe and binding, we measured the fluorescence intensity of the MBA at different temperatures under conditions used in the standard fluorescence quenching assay and then compared 25, 30 and 37C for binding to the PDGF aptamer. 03000006000009000001200000150000018000001525303745526067Tem p erature ( C ) Fluorescence Intensity (A.U.) no PDGF-BB with 200 nM PDGF-BB Figure 2-5. Effect of increasing temperature on fluorescence signal of free MBA in solution. A solution of MBA (75nM) in standard fluorescence quenching assay buffer as described in Materials and Methods was subjected to a gradual increase in temperature (5C per minute), held at the indicated temperatures for 5 min and the fluorescence signal was measured. The results presented in Figure 2-5 show that as the temperature increases, the initial fluorescence intensity increases presumably resulting from a larger separation of the arms of the predicted three-way helix junction where the six-base pair double stranded region would be formed. The largest fluorescence quenching was observed for

PAGE 49

36 the MBA incubated at 37C, which is the standard physiological temperature. The initial fluorescence intensity does not significantly increase above 37C. Since the aptamer probe can potentially exist in solution as multiple conformations (i.e., random coil, partially open, and closed structures), this particular assay offers the unique opportunity to observe not only the specific closing of the probe due to PDGF binding but nonspecific interactions that could result in the opening of the probe. As a result, experiments were performed at 30C to monitor events that induce conformational changes that may result in higher fluorescence intensities. This temperature also allows to monitoring PDGF-BB induced quenching. Effects of pH. Since the 6-FAM/DABCYL dye pair was chosen for assay development, the pH sensitivity of fluorescein could potentially affect the overall performance of the FRET based assay. As a result, the effect of pH on the fluorescence intensity of the probe without target protein was investigated. The fluorescence intensity of a 75 nM solution of the MBA in binding buffer used in the standard fluorescence quenching assay was also measured in the microplate reader except that the pH was varied between 5.0 and 10. The scope of this experiment was limited by the pH dependence of fluorescence intensity of fluorescein. The results (Figure 2-6) showed that within a pH range of 6.5 to 9.5, the fluorescence intensity of the MBA is within 10% of that observed at pH 7.5. The pH-dependent intensity profile of the MBA correlated with the pH-dependent spectra of fluorescein;58 consequently, the observed trend is believed to be due to the fluorescence characteristics of the dye molecule rather than conformational changes leading to quenching. Even though fluorescein is a highly pH dependent fluorophore, the optical properties should not affect the measurements of the FRET based

PAGE 50

37 010000200003000040000500004.55.56.57.58.59.510.5pH Fluorescence Intensity (A.U.) Figure 2-6. Effect of increasing pH on fluorescence signal of free MBA in solution. Fluorescence intensity measurements of a solution of MBA (75nM) in binding buffer with varying pH conditions. assay because the pH conditions during the experiments are similar to the optimal range for fluorescein.47 Effects of ionic strength. Certain monovalent and divalent cations commonly encountered in biological specimens are known to affect DNA conformation. A series of experiments to study the effects of varying concentrations of two divalent cations (magnesium and calcium) and two monovalent cations (sodium and potassium) on the fluorescence intensity of the free MBA in solution were performed. The results are presented in Figures 2-7 and 2-8. Each of the divalent cations caused a concentration-dependent decrease in the fluorescence intensity (Figure2-7) and at more than 1 mM

PAGE 51

38 00.20.40.60.811.20.00.51.01.52.02.53.0ion concentration (mM)Relative Fluorescence Intensity Mg2+ Ca2+ Figure 2-7. Effect of divalent cations on fluorescence signal of free MBA in solution. Increasing concentrations of divalent cations (CaCl2 and MgCl2) were added to a solution of MBA (75nM) in the standard fluorescence quenching assay buffer and the fluorescence signals were measured. concentration, the extent of quenching was comparable to that obtained with an excess of PDGF-BB. The results were not surprising since divalent cations stabilize secondary structures in DNA molecules and would thus shift the equilibrium to a closed conformation of the MBA.Monovalent cations sodium and potassium caused only a small initial decrease in fluorescence intensity (Figure 2-8) and even at concentrations of more than 50 mM the extent of quenching by the ions was less than 13% as compared to more than 90% quenching obtained with a 4-fold molar excess (300 nM) of PDGF-BB in a standard fluorescence quenching assay. Since typical combined concentrations of Na1+ and K1+ in cell culture medium is between 150 mM to 160 mM, desalting columns can be used to remove monovalent and divalent cations and prevent potential contributions towards quenching of the probe.

PAGE 52

39 00.20.40.60.811.20100200300400500600ion concentration (mM) Relative Fluorescence Intensity K+ Na+ Figure 2-8. The effect of monovalent cations on fluorescence signal of free MBA in solution. Increasing concentrations of monovalent cations (NaCl and KCl) were added to a solution of MBA (75nM) in the standard fluorescence quenching assay and the fluorescence signals were measured. Selectivity of the MBA-Based Fluorescence Quenching Assay Once the aptamer was selected and tested against PDGF for fluorescence quenching on binding, the selectivity of the MBA was examined. We incubated 20 nM MBA with 100 nM of either PDGF-BB or one of several extracellular proteins (bovine serum albumin, hemoglobin, lactate dehydrogenase, lysozyme, myoglobin, and thrombin) or unrelated growth factors (such as epidermal growth factor and insulin-like growth factor-1). The results shown in Figure 2-9 indicated that only pure human PDGFBB causes a marked reduction in fluorescence signal. All other proteins tested failed to cause any significant change in fluorescence, even at 10-fold higher concentrations than that used for PDGF-BB.

PAGE 53

40 Figure 2-9. Binding selectivity of the MBA. Relative fluorescence signals of incubation mixtures containing MBA (20 nM) and one of the following proteins (200 nM): bovine serum albumin (BSA), hemoglobin (HEM), lactate dehydrogenase (LDH), lysozyme (LYZ), myoglobin (MYO), thrombin (THR); or one of the growth factors, epidermal growth factor (EGF), basic fibroblast growth factor (bFGF), insulin-like growth factor-1 (iGF1), or PDGF-BB. In a subsequent experiment, some of the previously tested biomolecules including, growth factors (EGF), common proteins that are abundant in biological specimens (thrombin, ovalbumin and serum albumin) and glycogen (an abundant natural polysaccharide in tissues) were tested individually in a dose dependent manner. Dose-response curves were obtained by incubating increasing amounts of each biomolecule in a standard fluorescence quenching assay. The results are presented in Figures 2-10A and 2-10B. The small dose-dependent decrease in fluorescence due to thrombin and EGF indicated a slight interaction with the MBA probe. In comparison to similar

PAGE 54

41 02000040000600000510biomolecule concentration (g/mL)Fluorescence Intensity (A.U.) 15 BSA Ovalbumin Glycogen A 02000040000600000255075100protein concentration (nM)Fluorescence Intensity (A.U.) EGF Thrombin PDGF-BB B Figure 2-10. Fluorescence quenching displayed by the MBA in response to increasing concentrations of PDGF and other biomolecules. In a typical fluorescence quenching assay containing MBA (75 nM) in Tris (pH 7.5, 20 nM) and NaCl (20nM) as described in Materials and Methods, fluorescence signal was measured following the addition of increasing concentrations of various biomolecules: [A] BSA, ovalbumin, glycogen; and [B] PDGF-BB, epidermal growth factor, thrombin.

PAGE 55

42 concentrations of PDGF, the final extent of quenching was substantially lower than the quenching observed for PDGF. Conversely, minor changes in fluorescence intensity were observed for unrelated, commonly occurring biomolecules (bovine serum albumin, ovalbumin and glycogen, up to 12 ug/mL). Since only PDGF was capable of producing a large change in fluorescence signal in comparison to other proteins tested individually at the respective concentrations, the assay should be selective and sensitive for PDGF detection in the presence of other proteins commonly found in biological specimens. Fluorescence Quenching Assay Distinguishes Molecular Variants of PDGF Green et al. reported47 that the PDGF aptamers selected for binding to PDGF-BB by using isotropic labeling bound to three molecular variants of PDGF (Figure 2-11), namely BB, AB, and AA, albeit with different affinities, presumably because of the amino acid sequence homology (60%) between the A and the B chains. We tested the three common PDGF variants for their effects on the fluorescence quenching assay. Serial dilutions of each protein were incubated with 10 nM MBA in binding buffer and the dose-response curves for fluorescence quenching were compared. Parallel sets of experiments were conducted with varying amounts of the MBA. The calibration curves obtained with 10 nM MBA are shown in Figure 2-12. The results presented in Figure 2-12 indicated that the slopes and the concentration of protein required to attain 50% of the maximum quenching are distinct for the three molecular variants of PDGF. Although this data indicates that the MBA response shown for PDGF-AB is more similar to that for PDGF-BB (in comparison to the response for PDGF-AA), the difference between PDGF-AB and PDGF-BB is much more easily distinguishable than that reported method.47

PAGE 56

43 A A A B B B Figure 2-11. Cartoon representing PDGF-BB and its molecular variants (PDGF-AA and PDGF-AB) and its receptors alpha () and beta (). PDGF-AB 070000140000210000280000350000051015202530PDGF (nM)Fluorescence Intensity (A.U.)Denatured PDGF PDGF-BB PDGF-AA PDGF-AB Figure 2-12. Dose-response curves of PDGF variants: fluorescence signals of MBA for PDGF-AA (), PDGF-AB (), PDGF-BB (), and denatured PDGF-BB (). The concentration of the MBA was 10 nM. When PDGF-BB is reduced by dithiothreitol and denatured with sodium dodecyl sulfate, the resulting protein fails to cause any fluorescence quenching. Also, based on the slopes of the linear portion of the serial dilution curves it is clear that the binding affinity of each isoform is different for the aptamer probe. The results collectively indicated that the fluorescence quenching assay is not merely dependent on the primary sequence of the

PAGE 57

44 protein chains, but is capable of distinguishing conformational characteristics or the folding of highly related protein molecules. Conclusions For the development of a molecular beacon aptamer probe that will successfully report the binding event to a target protein in a single step, a signal transduction mechanism must be incorporated into the recognition molecule. Theoretical calculations using commercially available software showed that when the selected aptamers for PDGF are free in homogeneous solution, one possible conformation was the spatial separation of the 5' and 3' end.53;54 Taking advantage of the inherent structural properties of the aptamer, a distance dependent signal transduction mechanism, such as FRET, was incorporated to create a probe for the detection and study of PDGF-BB. Once the aptamer binds PDGF-BB, the distance between the ends of the aptamer molecule was reduced, resulting in the formation of a 6-base long stem. Also, the results obtained with varying incubation conditions described, in this chapter, support the notion that MBA-based FRET may be used in assays for protein detection in biological specimens since the FRET assay is effective at physiological pH and temperature. Divalent cations show interference with the assay by causing quenching in the absence of PDGF target probably due to a stabilizing effect on the stem of the probe. The results offer us new insights into, and justify the need for, future development and optimization of this one step MBA-based FRET assay. It was also demonstrated that three highly related molecular variants of PDGF (AA, AB, and BB dimers) can be distinguished from one another. In comparison to ELISA, the quenching assay developed is less sensitive, detection limits are higher, and has a higher throughput. Also, based on product literature and experimental controls for the ELISA assay, cross reactivity of the

PAGE 58

45 PDGF-BB antibody is prevalent for the AA isoform, which is often the case for antibody assays. As a result, depending on the information and degree of scrutiny needed from a given experiment one method may be preferred over the other. The use of fluorescence quenching as a measure of binding between the aptamer probe and the target protein eliminates the potential false signals that may arise in traditional fluorescence enhancement assays as a result of degradation of the aptamer by contaminating nucleases. Utilizing the conformational change of the aptamer to signal protein binding may be applied to essentially all proteins, especially those that do not naturally bind to DNA. Finally, other DNA aptamers with desirable selectivities for chosen target proteins can be synthesized and coupled to different fluorophore-quencher combinations to allow simultaneous detection of multiple target proteins in the same solution at different excitation/emission wavelengths or, conversely, a multiple-well microtiter plate can be used for many distinct samples for monitoring the same protein.

PAGE 59

CHAPTER 3 IMPLICATIONS OF USING A MBA IN A FRET BASED ASSAY FOR PDGF STUDIES IN BIOLOGICAL SAMPLES Introduction The development of a bioanalytical method includes the characterization of all procedures and parameters that demonstrate that a particular method can be used for quantitative measurements of analytes (in this case PDGF). The performance characteristics of the method should be suitable, reproducible and reliable for the intended analytical applications. In the previous chapter, detailed characterization experiments revealed that the developed MBA can detect nanomolar amounts of PDGF in a homogeneous solution, and that the optimal conditions for the MBA-PDGF complex formation are compatible with the conditions typically found in a biological sample. The next step in the development of the MBA-based bioassay will be focused on the implications of using this aptamer probe to study and detect PDGF in cellular samples. The aptamer selected for PDGF, was reported to have a 700 fold higher affinity for PDGF versus other random oligonucleotide sequences.47 This characteristic is especially important for the detection of PDGF in the presence of a complex biological sample. Nonetheless, in a biological sample such as cell culture medium, many other species may be present that could potentially interfere with the binding and the subsequent monitoring of the fluorescence change on binding. It has been reported, that PDGF and PDGF-like growth factors can be over-expressed and secreted into the media of various types of cancer cells; including MCF-7,46MDA-MB-231,50 and, PC3 cancer cell lines.59;60 In 46

PAGE 60

47 addition to PDGF, high levels of other proteins typically encountered in cell cultures may non-specifically bind to the probe, induce conformational changes, and as result produce false positive quenching. Also, many species in biological samples may present optical interferences for fluorescence measurements, due to absorption and auto-fluorescence. The above issues and others need to be carefully considered and explored before attributing the total signal change solely to PDGF binding. As a result, two directions were concurrently taken. One specifically designed to address non-specific interactions of the aptamer and the other addresses signal differentiation of PDGF responses and cellular interferences. This chapter will be aimed towards the determination of the robustness of the fluorescence quenching assay against other species present in the biological samples. Specifically, the incorporation of sample processing prior to carrying out the assay to remove potential biological interferences was investigated to minimize the effect of the factors mentioned above. The first part will include studies with simulated biological samples containing PDGF. For the second part, cancer cell conditioned media will be used in the first attempt to detect PDGF in biological samples using a MBA. Materials and Methods Instrumentation Fluorescence was monitored using a Fluorolog-Tau-3 spectrofluorometer (Jobin Yvon Inc., Edison, NY). The sample cell was a 100 l cuvette (Starna Cells, Atascadero, CA). The fluorescence emission of 6-FAM was monitored at the excitation maximum, 480 nm, and the fluorescence intensity was measured at the emission maximum, 520 nm. Both excitation and emission slits were varied to yield the best signals. For high throughput format assay, a Tecan Saphire (Durhan, NC) fluorescence microplate reader

PAGE 61

48 was used with 96-well flat bottom microtiter plates (Nalge Nunc International, Rochester, NY). All absorbance measurements were carried out in a Cary 300 UV-Vis spectrophotometer (Varian, Palo Alto, CA). Preparation of a Simulated Biological Sample A simulated biological sample (SBS) was prepared by combining Dulbeccos Modified Eagles Medium (DMEM, Mediatech, Hendon,VA) with serum proteins from fetal bovine serum(Invitrogen, Carlsbad, CA). SBS will be used as an intermediate between a homogeneous solution and a complex biological specimen, such as cell culture media, to investigate the applicability of using MBAs for protein detection. Total protein content was determined using the Bradford Protein Assay Kit61 (Biorad, Hercules, CA). Cell Culture and Preparation of Conditioned Media The cell lines used were human breast carcinoma MDA-MB-231, and murine BALB/c-3T3 fibroblasts (American Type Culture Collection,Manassas, VA), were grown in DMEM supplemented with 1 % Gentamicin (Sigma, Saint Louis,MI) and either 10 % fetal bovine serum for MDA-MB-231 or 10 % calf serum (Invitrogen,Carlsbad,CA) for BALB/c-3T3. Prostate cancer cells, PC-3(American Type Culture Collection, Manassas,VA), were grown in R12K medium (American Type Culture Collection, Manassas,VA) supplemented with 10 % fetal bovine serum and 1 % Gentamicin. The cells were incubated in humidified air containing 5 % CO2 at 37C. For the collection of serum-free conditioned media, the culture medium of confluent cultures was replaced with serum-free DMEM when the cell monolayers reached 80-90 % confluence. After approximately 60 minutes, the medium was replaced with fresh serum-free DMEM and the cells were incubated for an additional 24 hours. The serum-free CM was collected and clarified by centrifugation. Acetic acid was added to the supernatant to a final

PAGE 62

49 concentration of 0.1 M and the material was lyophilized. The lyophilized powder was resuspended in 0.1 M acetic acid, clarified by centrifugation at high speed in a microfuge for 5 minutes, and the supernatant containing the soluble proteins was subjected to size exclusion chromatography utilizing a NAP-5 (Amersham Biosciences, Piscataway, NJ) column packed with Sephadex G-25 in 0.1 M acetic acid. The excluded volume fractions with the highest absorbance at 280 nM were collected and lyophilized again. The protein preparation was resuspended in binding buffer (Chapter 2) to maintain appropriate pH conditions and used as the stock solution from which dilutions were made for incubation in the fluorescence assay. Results and Discussion Use of MBA in Biological Sample A SBS sample was used to determine the potential application of this one-step fluorescence-quenching assay for PDGF-BB detection in the presence of low levels of serum proteins. In the first case, the sample was lyophilized and split in two equal portions, one of which (Sa) was additionally supplemented with 100 nM recombinant human PDGF-BB (the other was labeled Sb and did not contained PDGF). Each portion (Sa and Sb) was resuspended in 0.1M acetic acid and the protein components were collected using a gel filtration column packed with Sephadex G-25 to remove salts and small molecules. As was mentioned on Chapter 2, high levels of monovalent and divalent cations resulted in background quenching and need to be removed prior analysis. Fluorescence quenching activity of the two preparations was compared by adding 50 nM MBA. As shown in Figure 3-1a, Sa caused a marked reduction in fluorescence, comparable to that observed with pure human PDGF-BB (see Figure 2-4A), and Sb caused a small fluorescence decrease.33 Analytical denaturing and reducing

PAGE 63

50 polyacrylamide gel electrophoresis of Sa and Sb showed that the low protein compositions of the two samples were indistinguishable from each other, which may indicate that the differences in fluorescence quenching presented by both samples was due to the PDGF-induced conformational change. Nonetheless, 20 % of quenching in the nonspiked sample may indicate nonspecific binding of other proteins in the sample. It is conceivable that as a result of binding to the MBA a reduced fluorescence is observed from the sample, as seen in Chapter 2 for thrombin and EGF. The results do, however, de monstrate that, when in the presence of similar total protein concentrations, the MBA has the potential to qualitatively distinguish the presence of nanomolar qua ntities of PDGF-BB in SBS. The next sample to be examined for PDGF detection was conditioned media from cultured cells. Conditioned media (CM) is a general term to describe media in which cells have been cultivated for a period of time. It contains many mediator substances that were secreted by the cells which were produced in this medium. The mediator substances contain growth factors, such as PDGF, and may promote the growth of new cells. A human breast carcinoma cell line, MDA-MB-231, has been reported to secrete PDGF-BB in culture medium.50 Serum-free CM was collected from human HTB cells and normal murine BALB/3T3 fibroblast cells in culture. The samples were collected and processed as described in Materials and Methods and serial dilutions of protein preparations from each cell line were incubated with a fixed amount of the MBA and fluorescence measurements were obtained to construct a dose-response curve. Parallel sets of dose-response curves were obtained by using different amounts of the MBA for each such set.

PAGE 64

51 A B Figure 3-1. Use of molecular beacon aptamer in biological samples.(A) Fluorescence signals of incubation mixtures of MBA (50 nM) with serum proteins without (Sb) or supplemented with (Sa) PDGF-BB, as compared to the MBA alone. (B) Dose-response curves/calibration curves for conditioned media and PDGF standard samples: fluorescence signals of mixtures of MBA (10 nM) with serial dilutions of protein preparations of conditioned cell media from HTB cells (triangles) or BALB cells (squares). Serial dilutions of a 500 nM solution of PDGF-BB (circles) were used as a control. Vd is the dilution factor.33

PAGE 65

52 For each set, a control dose-response curve for fluorescence quenching was obtained with serial dilutions of a solution of pure human PDGF-BB. The results obtained with 10 nM MBA are presented in Figure 3 -1B.33 Typically, several factors may indicate the presence of PDGF or PDGF-like molecules when the dose-response curves from the CM samples are compared with the one obtained using standard PDGF-BB. The factors include a concentration-dependent fluorescence intensity decrease, the similarity of the slope, and the final extent of quenching. For HTB conditioned media, the final extent of quenching as well as the dose-dependent decrease in fluorescence is comparable to the standard PDGF-BB. In the case of BALB conditioned media, the final extent of quenching is lower than the one obtained with HTB cells a nd standard PDGF-BB. It is important to mention here that initially BALB cells were selected as a negative control since it has not been reported that this particular cell line over-expresses PDGF. The observation that serial dilutions of BALB conditioned media produced a dose dependent quenching in some way similar to the standard PDGFBB reveals that this cell line may have measurable levels of PDGF (or PDGF-like mo lecules) or high levels of interfering proteins that bind non-specifica lly to PDGF and produce a co nformational change of the MBA. Although the final extent of quenchi ng for both samples was different indicating that the total amount of PDGF is different in each case, the use of MBA for PDGF quantitation and detection in cancer cell media samples may not be reliable due to the presence of interfering species in the biological environment that may lead to false positive results. An additional contribution to the change in fluorescence intensity may be due to filtering effects in the conditioned media that masked the true fluorescence signal.

PAGE 66

53 Implications of Total Protein Content on the MBA Assay The MBA appear to have the ability to qualitatively discriminate between cellular samples that contain PDGF; however, in order to verify the reliability of the results, the potential interfering compounds that could exist within cellular samples needs to be explored. For protein assay development, it is critical to determine the tolerable amount of protein that would allow a reliable fluorescence measurement. Protein aggregation and non-specific protein binding may have significant implications in the formation of the MBA-PDGF complex. Two samples of protein mixtures containing a total of 10 g/mL and 100 g/mL were prepared in binding buffer to study solely the protein effect on the MBA binding by combining the following representative biomolecules: bovine serum albumin, albumin, ovalbumin, lyzozime, glycogen, myoglobin, thrombin and a few growth factors including: insulin-like growth factor (IGF), epidermal growth factor (EGF) and vascular endothelial growth factor (VEGF). In the first case, a solution containing 10 g/mL of total protein was incubated with 50 nM of MBA in binding buffer and compared with a solution of 10 g/mL of pure PDGF-BB incubated with 50 nM MBA. The results presented in Figure 3-2 revealed that 10 g/ml of a protein mixture was capable of causing a 17 % quenching of the MBA. PDGF was added to the sample the fluorescence intensity drastically decreased, up to 96 %, indicating that PDGF is necessary for a complete quenching of the MBA at this protein concentration. In a separate experiment, the total protein content of the sample was increased by an order of magnitude. With 100 g/mL in the sample, 76 % of the

PAGE 67

54 0100200300400500600700MBA10 ug/mLPDGF-BB10 ug/mLserumproteins100 ug/mL serumproteins100 ug/mLserumproteins andPDGF-BBFluorescence Intensity (A.U.) Figure 3-2. Molecular beacon aptamer fluorescence quenching caused by a protein mixtures containing 10 g/mL. fluorescence of the MBA is quenched indicating that non-specific binding may be responsible for the decrease in fluorescence intensity. Nonetheless, the addition of 10 g/mL of pure PDGF-BB in the solution further reduces the fluorescence intensity of the aptamer probe. Based on the above results, a plausible explanation for the observed fluorescence quenching in cellular samples is non-specific binding. Thus, the dose dependent fluorescence decrease observed in cellular samples can not only be attributed to the presence of PDGF but to a combination of specific and nonspecific binding of proteins to the aptamer probe. Another alternative includes that the the MBA may be recognizing similar binding sites in other growth factors that may have a degree of identity with PDGF. In order to be able to use the PDGF aptamer in an assay for protein detection, alternative strategies for sample clean-up before analysis needed to be developed.

PAGE 68

55 Sample Pre-Treatment The development of recognition molecules for any type of assay should be experimentally convenient, sensitive and easy to perform in a short period of time. The developed MBA meets the characteristics for homogeneous solution analysis. Unfortunately, the response of the aptamer may be compromised by the presence of excess proteins in solution. Having the protein of interest in a biological environment is somewhat difficult for reliable detection by DNA-based methods. Based on the data presented in the last section, a mixture with total protein content higher than 100 g/mL in a 50 nM MBA solution may lead to false positive results due to non-specific binding. For this reason, the protein of interest needed to be isolated or the sample itself needs to be partially cleaned-up in order to obtain more reliable results. The ability to study proteins in mixtures with complete control of their environment (salts, temperature, pH, total protein content, etc.) is critical for the development of an in vitro assay using the PDGF-MBA. To simplify the sample before analysis, one must begin with the starting material (in this case cellular CM) and fractionate it using any one of a large number of physical or biochemical approaches: centrifugation, solid precipitation, binding to affinity columns or separation by sizing columns (gel filtration chromatography). In this case, size exclusion chromatography was used. After the cell medium was collected, clarified by centrifugation and lyophilized, it was run through a commercial NAP-5 or in a G-100 column to desalt the samples and remove low molecular weight protein components. In gel filtration chromatography, the stationary phase consists of porous beads with a well-defined range of pore sizes. The stationary phase for gel filtration is said to have a

PAGE 69

56 fractionation range, meaning that molecules within that molecular weight range can be separated. Proteins that are small enough can fit inside all the pores in the beads and are said to be included. The small proteins have access to the mobile phase inside the beads as well as the mobile phase between beads and elute last in a gel filtration separation. Proteins that are too large to fit inside any of the pores are said to be excluded. They have access only to the mobile phase between the beads and, therefore, elute first. Proteins of intermediate size are partially included meaning they can fit inside some but not all of the pores in the beads. The proteins will then elute between the large ("excluded") and small ("totally included") proteins. For our experiments we tested two different packing materials: Sephadex G-25 (PDGF will be included in the pores) and Sephadex G-100 (PDGF will be excluded from the pores). A control SBS was prepared to mimic conditioned media collected from tissue cultured cells that are commonly used to study the expression of proteins by normal and deceased cells. The sample contained DMEM supplemented with serum proteins and no PDGF-BB. It was processed by lyophilization to reduce the volume, diluted to 1 mL with 0.1 M acetic acid, loaded into a NAP-5 colu mn equilibrated with acetic acid, and collected at 0.5 mL fractions with 0.1 M acetic acid running buffer. After lyophilization and reconstitution in the binding buffer, each fraction was analyzed using the standard fluorescence quenching assay. Each fraction contained ~16.5 g/ml on average of serum proteins, as calculated with the Bradford assay. The results for the control SBM (Figure 3-3 A) showed that Fractions 6 to 16 caused different extents of quenching even though PDGF-BB had not been added.

PAGE 70

57 0100002000030000400005000024610121416MBAFraction numberFluorescence Intensity (A.U.) A 0200004000060000MBAFraction 1250 ng PDGF250 ng PDGFFluorescence Intensity (A.U.) B Figure 3-3. Fluorescence quenching caused by PDGF in the presence of proteins in a simulated biological specimen. Protein fractions were obtained from a Sephadex G-25 gel filtration of acetic acid soluble materials in a simulated biological specimen as described in Materials and Methods and each fraction was resuspended in 130 ul of a stock buffer Tris (pH 7.5, 10mM) with NaCl (20 mM). (A) 65 ul of each fraction indicated on the X-axis was incubated in a standard fluorescence quenching assay (final volume 100 ul) with the MBA (75nM) and the resulting fluorescence was measured. (B) Indicated amounts of PDGF-BB were added to a dilution of Fraction 12 with serum-derived proteins and incubated in a standard fluorescence quenching assay.62

PAGE 71

58 Based on repeated experiments conducted with PDGF-BB in buffer solution, it was determined that fractions 11 to 13 should contain PDGF-BB if added to the control SBS. Given that in fraction 6 there is ~50% quenching without the presence of PDGF-BB, gel filtration does appear to minimize the interferences by removing them from the PDGF-BB containing factions. To investigate this possibility and establish the limit of detection for the PDGF assay in the presence of serum proteins, varying amounts of PDGF-BB were added to fraction 12 of the control SBS. The results (Figure 3-3 B) showed a detectable decrease in fluorescence on the addition of as low as 50 ng of PDGF-BB.62 This data shows that the current assay can detect as little as 50ng/16.5 g of serum proteins. This leads us to believe that for MBA applications to be effective in biological samples the model system chosen for any protein of interest, should adequately express the protein at high enough levels to allow for the detection of the target in the presence of typical amounts of serum proteins. We further tested the reliability and effective range of operation for the fluorescence quenching assay by trying to improve the sample clean-up procedure using a Sephadex G-100 gel filtration column. For the G-100 packing material, the solid phase has larger pores and, as a result, retains PDGF in the column longer. This property should improve the exclusion of interfering impurities from PDGF containing fractions. Also, a direct comparison between an ELISA analysis and MBA assay was conducted. The ELISA not only allowed for the identification of the distinct fractions for which PDGF was eluted, but also, enabled us to compare the quenching results with the PDGF-BB responses in the presence of serum proteins. For the next experiment, a buffer sample containing only 50 g/mL PDGF was used. A Sephadex G-100 gel filtration column was

PAGE 72

59 equilibrated with 0.1 M acetic acid and the sample loaded. Then, 0.5 mL fractions were collected as the sample was eluted. To determine the fraction(s) that contained PDGF-BB, each one of the collected 0.5 mL fractions was tested using a commercial ELISA kit and PDGF was found in fractions 19 to 22. Using this information, we ran a SBS spiked with PDGF-BB in the Sephadex G-100 gel-filtration column and tested fractions 11 to 25 with the fluorescence quenching assay. Similar to the previous results obtained with the NAP-5 column (Figure 3-3A), several fractions exhibited different degrees of quenching when analyzed with the fluorescence quenching assay (Figure 3-4). The quenching assay indicated a significant amount of quenching in the fractions 17 to 22 after a 2-fold dilution of the original sample which was required for assay analysis. To determine the amount of PDGF in each fraction of the spiked SBS sample, fractions 7 to 18, 19 to 20, and 21 to 22 were pooled and analyzed using ELISA. PDGF-BB was found in samples 19 to 22 as expected based on the spiked ELISA experiments (Table 3-1). Even though fractions 17 to 18 did not contain PDGF-BB based on the ELISA experiment, a large degree of quenching was still observed for the MBA assay. The results further suggest the need for sample pre-treatment in order to minimize the interfering proteins such as those found in fractions 17 to 18. Unfortunately, it is still not understood as to whether the amount of quenching is due to PDGF versus non-specific binding to the probe. Interestingly, it was also observed that when the total protein concentration in individual fractions that did not contain PDGF from the two gel-filtration columns studied here was higher, the extent of quenching was higher. This finding points out the possibility that dilutions could be

PAGE 73

60 performed to minimize the extent of perceived interferences. A control sample that did not contain PDGF was made and processed as before. No PDGF was found in fractions 19-22 using ELISA. Since fractions 21-22 contained the highest amount of PDGF-BB, a side-by-side comparison of the PDGF and non-PDGF containing fractions was conducted with the MBA assay (Figure 3-5). Table 3-1. Analysis for PDGF-BB spiked samples after fractionation on a G-100 Sephadex column using enzyme-linked immunosorbent assay (ELISA) Eluted Fraction # g/mL PDGF 7-18 0.8 19-20 2.4 21-22 3.0 01000020000300004000050000111213141516171819202122232425MBAMBA+ PDGF-BBfraction numberFluorescence Intensity (A.U.) Figure 3-4. Fluorescence intensity of fractions of simulated biological sample (SBS) collected from G-100 column. Each fraction was 0.5 mL. After lyophilization samples were resuspended in binding buffer for the fluorescecence quenching assay. MBA concentration was 50 nM. Both fractions were adjusted to the same total protein concentration and the fluorescence quenching was monitored with the MBA assay. As shown in figure 3-5, the measured florescence intensity of the combined fractions spiked with PDGF-BB was more than the un-spiked control. As the dilution factor increases, the corresponding

PAGE 74

61 fluorescence intensities increased for both samples. This indicated that the cause of quenching in the control sample was capable of being diluted. To determine if PDGF could interact with the aptamer to cause complete quenching of the probe in the samples, an excess of PDGF-BB (250 ng) was spiked into the control samples after processing. The aptamer appeared to respond and bind to this addition of PDGF even in the presence of the serum proteins regardless of the dilution factor. The results were consistent with the above results and told us that with dilutions and gel-filtration PDGF-BB could be monitored in the presence of serum proteins. The amount required to obtain full quenching of the probe would, however, be significant considering the sample would need to be diluted. Upon a 4-fold dilution of the total proteins, the aptamer appeared to be unaffected by the serum proteins and capable of distinguishing between the control sample with and without PDGF. 010000200003000040000500002 fold4 fold6 fold8 fold10 foldbufferMBAMBA +PDGFFluorescence Intensity(A.U.) spiked with pdgf-bb no pdgf-bb added pdgf-bb added to spl 21 Figure 3-5. Fluorescence intensity of dilutions of SBS. Green columns represent fraction spiked with an additional 200 nM PDGF-BB post processing. MBA concentration is 50 nM.

PAGE 75

62 Conclusions The results presented in this chapter were used to explore the applicability of the PDGF-MBA probe to monitor PDGF secretion from cancer cells in cell media. Initially, the assay showed the ability to detect PDGF in spiked biological samples containing low levels of serum proteins. The conditioned media from two cell lines, which express different levels of PDGF, was analyzed with the assay. For HTB conditioned media, the final extent of quenching as well as the dose-dependent decrease in fluorescence illustrated a noticeable increase in the PDGF concentration in comparison to the BALB conditioned media. The MBA appeared to have the ability to qualitatively discriminate between cellular samples that contain PDGF; however, higher levels of serum proteins may result in false positive quenching. Simple size exclusion chromatography allowed for more reliable results to be obtained and showed that 50 ng of PDGF could be detected in the presence of ~16.5 g/mL of serum proteins. Further improvements in the sample processing could be explored; however, signal differential methodologies could provide more reliable results.

PAGE 76

CHAPTER 4 LIGHT SWITCHING APTAMER PROBE FOR PLATELET DERIVED GROWTH FACTOR USING RATIOMETRIC AND TIME-RESOLVED FLUORESCENCE MEASUREMENTS Introduction Fluorescent techniques offer excellent choices as signal transduction mechanisms to report binding events because most samples remain intact under normal excitation conditions and the sensitivity afforded by fluorescence detection. Fluorescence resonance energy transfer (FRET),40;62-64 anisotropy 25;33and lifetime-based measurements are examples of some fluorescent methods that, combined with a selected aptamer sequence, can be used for bioassay development. All signal transduction techniques have their individual strengths for different experimental needs. Nonetheless, FRET and anisotropy suffer from some limitations that could hamper their effective applications in complex biological samples. For instance, although fluorescence anisotropy only requires singly labeling of one dye molecule on each aptamer sequence, it entails non-standard instrumentation and data interpretation. FRET or fluorescence quenching based probes quantify target concentrations with changes in fluorescence intensity, but the two methods are sensitive to the solution environment and protein content as was showed in the last chapter. More importantly, depending on the selection of fluorophores, directly applying them to biological samples in their native environments may be difficult due to fluorescence interferences. The major goal of the experiments discussed in this chapter was to explore the possibility of designing a DNA probe that has the ability to differentiate between PDGF 63

PAGE 77

64 responses and the biological background responses. Typically, protein analysis in their native environments has two significant contributors to background signals. The first one is the probe itself. For example, in a quenching based FRET molecular probe for protein studies, the probe may suffer from incomplete quenching, resulting in significant background. Moreover, in a native biological environment, there are many potential sources for non-specific quenching of the probe as discussed in the previous chapter. This greatly limits exploring the full potential of FRET-based MBAs for protein detection. Concurrent with the experiments presented in this chapter, the effect of extracellular proteins in cellular samples was also addressed in Chapter 3 and a sample processing procedure to aid in the removal of most interfering proteins. Nonetheless, the effect of background signal from auto-fluorescing components of the cell media still needed to be investigated. Several molecular species exist in a biological environment, some of which will yield a strong fluorescence background signal when light is applied to the dye molecules. If the auto-fluorescence intensity of some of the cell media components is comparable to or larger than the fluorescence intensity of the fluorescence of the MBA, the fluorescence quenching becomes masked and the analysis of the fluorescence signals becomes difficult. Great efforts have been made to solve this problem 65; however, effective solutions are limited. The interesting spectroscopic properties of pyrene will be employed and combined with time-resolved measurements as an alternative approach to FRET. 34;66-68 Photophysics of Pyrene Excimers Pyrene is a four-ringed polycyclic aromatic hydrocarbon (PAH) that gives monomers and excimer emissions depending on its concentration. An excimer is a

PAGE 78

65 dimmer which is associative in an electronic excited state and dissociative in its ground state. The pyrene excimer was the first species of this type to be discovered.69;70 Figure 4-1. The chemical structure of pyrene The formation of a pyrene excimer requires an electronically excited pyrene encountering with a second pyrene in its ground electronic state. The excimer fluorescence of pyrene can be described by the following equilibrium: M* + M D* Molecular absorption occurs to produce an excited monomer (M*), which then interacts with its neighbor (M) to form the excimer (D*). When excimer formation takes place in any given pyrene-containing system, it can be easily monitored by a steady-state fluorescence spectrum: the broad, featureless emission centered at 480 to 500 nm is extremely easy to recognize, even when extensive monomer emission occurs; since the monomer fluorescence takes place in the 380 to 400 nm wavelength range.67 Another good example of a dye that exhibits the excimer phenomena is BODIPY, whose monomer emits at 520 nm, and excimer has emission of 620 nm.71;72 The formation of the excimer is useful to probe spatial arrangement of some molecules. Similar to FRET, the distance dependent property of excimer can be used as a unique signal transduction in the development of aptamer-based probes that change their secondary structures on binding of its target, such as the aptamers selected against, PDGF,33;62;63 cocaine,73thrombin,39;74 and HIV1 Tat protein.75The emission wavelength

PAGE 79

66 switching property solves the background signal problem that occurs with FRET molecular probes. This finding encouraged us to apply the use of pyrene molecules in a new aptamer probe design to develop oligonucleotide probes. Time-Resolved Measurements Time-resolved fluorescence spectroscopy is a well-established technique for studying the emission dynamics of fluorescent molecules, i.e. the distribution of times between the electronic excitation of a fluorophore and the radiative decay of the electron from the excited stated producing an emitted photon. The temporal extent of this distribution is referred to as the fluorescence lifetime of the molecule. This technique is widely used in fluorescence spectroscopy because it often contains more information than is available from steady-state data. With steady-state measurements some of the molecular information is lost during the averaging process. Many macromolecules can exist in more than a single conformation and the decay time of a bound probe may depend on the conformation. The intensity decay may reveal two different decay times indicating the presence of more than one conformational state. The steady state intensity will only reveal an average intensity dependent on a weighted average of two decay times. There are additional reasons for using time-resolved experiments. For example, in the presence of energy transfer, the intensity decays may reveal how acceptors are distributed in space around the donors. Also, time resolved measurements reveal whether quenching is due to diffusion or to complex formation with the ground state fluorophores. In fluorescence, much of the molecular information content is available only from the time-resolved measurements.34 Fluorescence decay kinetics gives a complete picture of the fluorophore and its interactions within the microenvironment. Complex intermolecular interactions can be

PAGE 80

67 revealed by lifetime measurements made across an emission spectrum which has little structure. Time correlated single photon counting (TCSPC) is a commonly used method for the detection of the fluorescence lifetimes. TCSPC is a digital counting technique that relies on the concept that the probability distribution for emission of a single photon after an excitation event yields the actual intensity against time distribution of all the photons emitted as a result of excitation. By sampling the single photon emission following a large number of excitation flashes, the probability distribution can be constructed. As a result of the above spectroscopic properties afforded by lifetime measurements together with the use of a pyrene-based aptamer probe, it was of interest to evaluate its potential for use with biological samples. Both steady-state and time-resolved fluorescence measurements were conducted to provide in vitro monitoring of biological samples with MBAs. In this approach, the MBA was labeled with pyrene molecules, similar to what has been reported in using pyrene for molecular beacons.66 The optical properties of pyrene were exploited, including that with the same excitation, the excimer emitted at a longer wavelength than that for the monomer. Although excimer formation is a very attractive characteristic of pyrene for protein detection in solution, the excimer phenomena alone can not solve the problem of signal from the multiple species in the biological environment. One feature of the pyrene excimer is that it has a long lifetime67 compared with other potential fluorescent species. The lifetime of the pyrene excimer could be as long as 100 ns, while that for most of the species that contribute to the biological background signal is shorter than 5 ns. This property should allow the separation of the biological background signal from that of the excimer signal using time-resolved fluorescent measurements.

PAGE 81

68 Experimental Section All the aptamer-based probes listed in Table 4-1 were synthesized in-house and purified with RP-HPLC. Reagents for the synthesis of DNA were purchased from Glen Research (Sterling,VA). The pyrene labeled aptamer used for this set of experiments, Pyr-MBA, is a shorter version of the original 36t aptamer designed specifically to reduce pyrene background emission due to the close proximity of the dye molecules at each end of the probe. Since the stem sequence is not important for the high affinity binding to PDGF-BB47, in order to reduce the background emission, the stem was shortened gradually to identify an aptamer sequence that was in fully open conformation in the absence of target while reserving good binding affinity. All aptamers were labeled with pyrene at both termini. SGL-MBA is single labeled with pyrene at its 5 end. SCR-MBA is a 39-mer scramble oligonucleotide sequence with pyrene molecules at both ends. Table 4-1. Pyrene-labeled oligonucleotide sequences used in this study. Sequence Name Sequence Pyr-MBA Pyrene-AGGCTACGGCACGTAGAGCATCACCATGATCCT-Pyrene SGL-MBA Pyrene-AGGCTACGGCACGTAGAGCATCACCATGATCCT SCR-MBA Pyrene-GGAACGTAATCAACTGGGAGAATGTAACTGACTGC-Pyrene Pyrene butylic acid was purchased from Sigma (St Louis, MO). Recombinant human PDGF-BB, PDGF-AB, PDGF-AA, and, TNFwere purchased from R&D Systems (Minneapolis, MN) and dissolved in 4 mM HCl with 0.1% BSA and then diluted in 20 mM Tris buffer (pH 7.5) before use. Other recombinant human growth factors, including recombinant human epidermal growth (EGF) and insulin-like growth factor 1 (IGF-1), were from Roche (Indianapolis, IN). Bovine serum albumin (BSA), human hemoglobin (HEM), horse myoglobin (MYO), chicken lysozyme (LYS), human

PAGE 82

69 thrombin (THR) and other chemicals were from Sigma. The binding buffer used for all the characterization experiments was 20 mM Tris-HCl (pH 7.5) with 20 mM sodium chloride. To prepare a simulated biological sample (SBS), Dulbeccos Mofication of Eagles Medium (DMEM) (Mediatech, Inc, Hendon, VA) was supplemented with fetal bovine serum (Invitrogen, Carlsbad, CA) and used for the detection of PDGF in biological samples. Instruments An ABI3400 DNA/RNA synthesizer (Applied Biosystems, Foster City, CA) was used for DNA synthesis. Probe purification was performed with a ProStar HPLC (Varian, Walnut Creek, CA) where a C18 column (Econosil, 5u, 250.6 mm) from Alltech (Deerfield, IL) was used. UV-Vis measurements were performed with a Cary 300 UV-Vis spectrophotometer (Varian, Palo Alto, CA). Steady-state fluorescence measurements were performed on a Fluorolog-Tau-3 spectrofluorometer (Jobin Yvon, Inc., Edison, NJ). For emission spectra, 349 nm was used for excitation. Time-resolved measurements were made with a single photon counting instrument (OB900, Edinburgh Analytical Instrument), where a nitrogen flash lamp was used as the excitation source (=337 nm). Synthesis and Purification of Pyrene Labeled MBA Several methods have been reported for oligonucleotide labeling.66;76 Unfortunately, for dually labeling pyrene on both nucleic acid termini, the reported procedures proceeded in low yields partially due to solubility incompatibility of pyrene derivatives and oligonucleic acids. A new solid phase coupling method that allow multi

PAGE 83

70 labeling organic dyes to DNA sequences with coupling yields as high as 80 % was used to synthesize all the pyrene labeled aptamers.77 Results and Discussion Light Switching Excimer Aptamer Probe Design As a proof of principle, the excimer signaling approach was used to develop a probe for PDGF-BB. An excimer switching aptamer probe was developed by labeling both ends with pyrene molecules that can form excimers. When the dual-pyrene-labeled aptamer probe is free in solution without the target protein, both pyrene molecules will be spatially separated, based on the FRET experiments in Chapter 2, and only the monomer emission peaks (at 375 nm and 398 nm) will be observed. The binding of the aptamer probe to protein brings the pyrene molecules at 3' and 5' ends close together, allowing the excimer to form. Thus, the emission peak around 485 nm will be observed. The change in emission color serves as a rapid process for qualitation and the excimer fluorescence intensity can be used for highly sensitive real-time quantitation of PDGF in homogeneous solutions. Figure 4-2. Use of pyrene excimer to probe PDGF-BB. PDGF-BB aptamer was labeled with pyrene molecule at both ends. The two pyrenes will be away from each other because of the open structure of the aptamer. After binding to PDGF-BB, the aptamer adapts a close conformation, bringing two pyrenes close to each other. As a consequence, pyrene excimer forms and green light emits.

PAGE 84

71 Light Switching Aptamer Probe for PDGF-BB Detection A light switching aptamer probe, named Pyr-MBA (sequence shown in Table 4-1), was prepared by labeling a 33-nucleotide sequence, a shorter version of the 36t MBA with pyrene molecules at both 3' and 5' ends.47 Figure 4-3 shows the fluorescence emission spectrum from a solution containing 100 nM of Pyr-MBA. Two emission peaks at 375 and 398 nm corresponds to pyrene monomer emission. No significant excimer emission was observed from this solution (Figure 4-3). Upon addition of 50 nM PDGF-BB into the Pyr-MBA solution, an aptamer-protein target complex was formed leading to a secondary structure where the 3' end sequence hybridizes to the 5' end sequence, forming a stable stem. This stem brings both pyrene molecules together, resulting in excimer emission at 485 nm (Figure 4-3). 400450500550600050000100000150000200000250000300000350000400000450000 Fluorescence Intensity (A.U.)Wavelength (nm) Pyr-MBA Pyr-MBA + PDGFBB Figure 4-3. Fluorescence emission spectra of Pyr-MBA with and without PDGF-BB in binding buffer. As shown in Figure 4-3, protein-bound probe gives emission peaks, two monomer peaks at 375 nm and 398 nm respectively, and excimer peak at 485 nm. This allows for ratiometric measurements. By taking the intensity ratio of the excimer peak to either one

PAGE 85

72 of the monomer peaks, one could effectively eliminate signal fluctuation and minimize impact of environmental quenching on the accuracy of measurement. We further explored excimer formation by designing two control DNA sequences, SGL-MBA and SCR-MBA. The sequences were prepared to confirm that the observed excimer emission was a result of the aptamer-protein binding. The first sequence SGL-MBA was the same sequences of the Pyr-MBA aptamer labeled with pyrene at the 5' end. Green et al. reported47 that PDGF-BB has two binding sites; the phenyalanine-84 residue in the B-chain of PDGF forms a point of contact with a specific nucleotide residue in loop at the helix junction of the selected aptamer However, the addition of PDGF-BB into the SGL-MBA solution did not change the emission spectrum of the solution, indicating that two pyrene molecules need to be in close proximity. This suggests that the distance between adjacent molecules was not enough for the excimer formation and emission. This data also showed that PDGF-BB itself has no measurable effect on the optical properties of pyrene molecules. The second sequence, SCR-MBA, is a random DNA sequence with its 3' and 5' ends labeled with pyrene. This sequence was used as a control aptamer with no or low affinity for PDGF-BB to demonstrate that the excimer emission is a direct result of the formation of the Pyr-MBA/PDGF-BB complex. This dually labeled scramble sequence, did not give any excimer emission after the addition of PDGF-BB. Both results indicated that the excimer emission showed in Figure 4-3 was indeed from the aptamer conformation change on specific binding to PDGF-BB.

PAGE 86

73 e501001502000.00.20.40.60.81.01.21.41.61.82.02.22.4 F485nm / F378nmt (s) pyr-MBA SGL-MBA SCR-MBA PDGF-BB addition Figure 4-4. Real-time response of excimer/monomer ratio. It showed that the binding of PDGF-BB aptamer to PDGF-BB takes place within seconds. It affords a rapid protein assay. Sensitivity and Selectivity of the Probe The dose-dependent excimer formation was determined to evaluate the sensitivity of the pyrene-labeled aptamer probe (Figure 4-5). A linear response was observed with the addition of increasing amounts of PDGF-BB in concentrations ranging from 0 to 40 nM. This result confirmed the formation of the excimer due to the spatial proximity of the two pyrene molecules (and the appropriate geometrical arrangement required for excimer emission) at the ends of the MBA. According to the measurements for blank samples, the limit of detection was calculated an about 200 pM were consistently detected in homogeneous solution. Moreover, the high sensitivity together with the emission wavelength switching and detection without separation nature of this probe, enabled detection of the presence of 40 nM PDGF-BB just with the naked eye (Figure 4-6). A clear green color was observed when PDGF-BB was added to 100 nM of excimer probe solution.

PAGE 87

74 360380400420440460480500520540560050000100000150000200000250000300000350000400000 Fluorescence Intensity(A.U.)Wavelength (nM) 0nM 2nM 4nM 6nM 8 nM 10nM 12nM 14nM 16nM 20 nM 24nM 30nM 35nM A 0510152025303540450.00.20.40.60.81.01.21.41.61.8 F 485nm / F 378nmPDGF-BB (nM) B Figure 4-5. Dose-dependent excimer formation on additions of PDGF-BB to a Pyr-MBA. A)in binding buffer, B)calibration curve.

PAGE 88

75 Figure 4-6. Visual detection of PDGF-BB. Solution of the 100 nM pyr-MBA under UV light (left), and 100 ul of 100 nM pyr-MBA with 40 nM of PDGF-BB (right). To detect PDGF-BB in the presence of other biomolecules the selectivity of the pyr-MBA probe is critical. Similar to the FRET-based MBA, the pyrene labeled MBA was mixed with excess extra-cellular proteins such as albumin, hemoglobin, myoglobin, and lysozyme. Even at 10 times the PDGF-BB concentration, the proteins did not produce significant changes in signal (Figure 4-7). LSMHMGMYGTHRBSAPDGF BB0.00.20.40.60.81.0 Relative Fluorescence Quenching (A.U.) Figure 4-7. Response of excimer probe to different proteins. Concentration of PDGF-BB was 50 nM, while the concentration of other proteins was ten times higher (500 nM). Furthermore, we tested the selectivity of the excimer probe for proteins and peptides potentially coexisting with PDGF-BB in biological samples. The results showed this shorter version of the original MBA probe retain its high selectivity for PDGF-BB in

PAGE 89

76 homogeneous solution. The responses to epidermal growth factor (EGF), vascular endothelial growth factor (VEGF), or insulin-like growth factor 1 (IGF-1) were minimal. PDGF-AA and PDGF-AB, both of which have shown lower response to the aptamer sequence in the FRET system, gave lower fluorescence-signa l response in this case as well (Figure 4-8).47 EGFIGFVEGFPDGFAAPDGFABPDGFBB0.00.20.40.60.81.0 Probe response (A.U.) Figure 4-8. Response of 50 nM Pyr-MBA to 500 nM of different growth factors. Bovine serum albumin (BSA) belongs to the class of serum proteins called albumins, which make up about half of the protein in plasma and are the most stable and soluble proteins in plasma. It is very common for laboratories developing immunoassays, mostly due to its availability, solubility and the numerous functional groups present for coupling to haptens. For PDGF-BB reconstitution, a buffer containing BSA as a carrier protein was used and as result BSA will be present throughout. To rule out any possible contribution of BSA to the measured fluorescence emission, increasing amounts of BSA were incubated with the pyr-MBA. In this case, high concentrations of BSA did not alter

PAGE 90

77 the optical properties of pyr-MBA; the measured excimer emission is negligible as can be observed from the data presented in Figure 4-9. 05000100001500020000375425475525575Wavelength (nM)Fluorescence Intensity (A.U.) 0.1 mg/mL 0.15 mg/mL 0.2 mg/mL 0.25 mg/mL 0 mg/mL Figure 4-9. Increasing amounts of BSA incubated with 50 nM Pyr-MBA. Direct Quantitative Detection of PDGF-BB in Cell Medium To be applicable for a potential bioassay development, the MBA probe must tolerate interferences typically present in a biological sample. In the previous chapter we showed that when the MBA is incubated with a protein mixture containing high levels of cellular proteins, false-positive quenching signals were obtained. On the other hand, when the protein levels are monitored and adjusted to acceptable levels, the MBA binds to PDGF-BB with high affinity and responds by a decrease in fluorescence intensity attributable to the formation of the PDGF-BB-MBA complex. In this chapter, the interference presented by autofluorescencing components in the media will be investigated. For the purposes of simplicity and to investigate solely the effect of auto-fluorescence cell media components, a simulated biological specimen (SBS) was made

PAGE 91

78 by combining cell medium with low levels of serum proteins. Figure 4-10 shows the spectra of 200 nM Pyr-MBA in a Tris-HCl buffer solution and cell medium (DMEM). 3504004505005506006500100000200000300000400000500000600000700000800000900000 Fluorescence Intensity (A.U.)Wavelength (nm) 200 nM Pyr-MBA in buffer 200 nM Pyr-MBA + 50 nM PDGFBB in buffer 200 nM Pyr-MBA in DMEM DMEM 200 nM Pyr-MBA + 50 nM PDGF-BB in DMEM Figure 4-10. Fluorescence spectra of simulated biological sample. (DMEM, 200 nM Pyr-MBA in DMEM, 200 nM Pyr-MBA and 50 nM PDGF-BB in DMEM, 200 nM Pyr-MBA in Tris-HCl buffer, and 200 nM Pyr-MBA and 50 nM PDGF-BB in Tris-HCl buffer) In binding buffer, the MBA showed an intense excimer emission which was observed when the target protein was added to the probe solution. Unfortunately, intense background fluorescence, contributing from some species typically encountered in the cell medium such as riboflavin, nicotinamide, pyridoxine, tryptophan, tyrosine as well as phenol red, was observed, which masked the signal response from the probe and made it indistinguishable. This result indicates that simple steady-state fluorescence measurements are questionable for direct detection of PDGF-BB in a biological sample. As mentioned before, the MBA labeled with pyrene molecules not only has the excimer formation capability but also the advantage of a long fluorescence lifetime. Most of the background fluorescence has a lifetime of less than a few nanoseconds(~10 ns), while the monomer and excimer emission of pyrene have much longer lifetimes. Since

PAGE 92

79 we were interested in studying the possibility the pyrene molecules for signal differentiation based on lifetime, the total amount of protein was maintained at acceptable levels (~10 g/mL) for the PDGF-BB aptamer. Also, based on the steady state fluorescence spectra (Figure 4-10) there is considerable background signals at this concentration produced by SBS at the wavelength of interest which impede the ability for pyrene to be useful in standard fluorescence measurements. Therefore, there is a need to study this format by alternative fluorescence measurements and explore the potential of using the lifetime characteristics of pyrene. Time correlated single photon counting measurement of the Pyr-MBA and Pyr-MBA-protein complex solutions were investigated to characterize their lifetimes and identify the nature of excimer formation. The results obtained by TCSPC suggested that the lifetimes of both pyrene monomer and excimer were around 40 ns. This is one magnitude longer than lifetimes of most organic fluorophores and fluorescent components in cell medium and cells, which should allow temporal resolution of the excimer signal from intense background fluorescence from cell medium. Three cell media samples were then subjected to SPC measurements: simulated biological sample (SBS), 200 nM Pyr-MBA in SBS, and 200 nM MBA-Pyr-MBA and 50 nM PDGF-BB in SBS. The data reported on Figure 4-11 and 4-12 showed that the fluorescence emission from both pyrene monomer and excimer can be differentiated from background fluorescence. Although pyrene monomer has a longer lifetime, the difference in intensity between Pyr-MBA in SBS and SBS itself is small (~3 times). This may be because many components of the SBS fluoresce at this region and their concentrations are higher than the MBA concentration.

PAGE 93

80 0255075100125150175200110100100010000100000 Fluorescence Intensity (A.U.)Time (ns) SBS SBS + Pyr-MBA SBS + Pur-MBA + PDGFBB Figure 4-11. Monomer time-resolved spectra collected at 398 nm (monomer emission) for a sample of simulated biological medium (SBS), 200 nM Pyr-MBA in SBA and 50 nM Pyr-MBA and PDGF-BB in SBS. 0255075100125150175200110100100010000 Fluorescence Intensity (A.U.)Time (ns) SBS SBS + Pyr-MBA SBS + Pyr-MBA + PDGFBB Figure 4-12. Excimer time-resolved spectra collected at 480 nm (excimer emission) for a sample of simulated biological medium (SBS), 200 nM Pyr-MBA in SBA and Pyr-MBA and 50 nM PDGF-BB in SBS. Time resolved emission spectra revealed changes in emission spectra on a nanosecond time scale (Figure 4-13). Due to its short fluorescence lifetime, the fluorescence from the SBS decayed rapidly, getting to 0.1 % of its original signal 40 nanoseconds after the excitation pulse reached maximum intensity. In contrast, the excimer emission decayed slower and retained reasonably high emission intensity even

PAGE 94

81 after 40 ns of decay. The highest signal to noise ratio at 480 nm was observed from 40 to about 100 ns after the excitation pulse reached maximum intensity. Before 40 ns, there was still significant background fluorescence from the SBS while the signal to noise ratio dropped below 3 after 100 ns of decay. In contrast, the excimer emission from a mixture of protein and probe in SBS decayed much slower (Figure 4-13). The emission remained after 40 ns of decay. The temporal separation of intense background from weak signals is evident in Figure 4-14, where time-resolved fluorescence emission spectra of SBS samples are compared with steady-state fluorescence emission spectra. For steady-state measurements no resolved peak around 485 nm was observed when PDGF-BB was added to Pyr-MBA in the SBS sample due to a significant amount of background signal. However, in time-resolved emission spectra the long lifetime emission peak at 480 nm was well resolved. This peak corresponded to excimer emission from protein-bound probe, which was supported by two observations: its intensity varied with changes of protein concentration, and no such peak was observed in either spectrum of cell medium or cell medium solution containing PDGF-BB and a single-pyrene-labeled aptamer sequence (SGL-MBA). Thus, this characteristic emission peak could be used to examine the presence of PDGF-BB in biological fluids.

PAGE 95

82 Figure 4-13. Fluorescence intensity as a function of wavelength at different times. SBS (green), 200nM Pyr-MBA in SBS (red), and 200nM Pyr-MBA with 50nM PDGF-BB in cell SBS (blue) at 25C. Excitation=337nm. Time=0 ns correspond to the time at which the excitation pulse reaches maximum intensity. Detection of PDGF-BB in Simulated Biological Sample With the single photon counting technique, the Pyr-MBA probe can qualitatively detect target protein in homogeneous solution and in a simulated biological sample containing low levels of potential interfering proteins. When the concentration of PDGF-BB was increased, the fluorescence intensity of the excimer peak in time-resolved emission spectrum increased accordingly. Figure 14-15 shows the decays of 200 nM

PAGE 96

83 aptamer probe Pyr-MBA in the cell medium with increasing concentrations of PDGF-BB at 480 nm emission. 3603804004204404604805005205405605806000.00.20.40.60.81.01.2 Signal (counts)Wavelength (nm) Steady State: Pyr-MBA in SBS Steady State: Pyr-MBA + PDGFBB in SBS Time-resolved: Pyr-MBA in SBS Time-resolved: Pyr-MBA + PDGFBB in SBS Figure 4-14. Steady-state and time-resolved fluorescence spectra of SBS samples. Steady state fluorescence emission spectra of 200 nM Pyr-MBA in SBS and 200 nM Pyr-MBA with 50 nM PDGF-BB. Time-resolved fluorescence spectra of 200 nM Pyr-MBA in SBS, and 200 nM Pyr-MBA with 50 nM PDGF-BB in SBS. Time resolved spectra were taken over 60 ns to 79 ns after the excitation pulse reaches the maximum intensity. Fluorescence intensity from the response of each solution could be calculated by integrating photons emitted over an optimized time window. Photons emitted between 40 and 100 ns were counted and integrated for each concentration to construct a calibration curve. The resulting fluorescence intensity is proportional to the PDGF-BB concentration (Figure 14-14). This linear response of fluorescence intensity to PDGF-BB

PAGE 97

84 in cell medium demonstrates the feasibility of detection of the target proteins in simple biological samples containing low levels of proteins. 020406080100120140160180200110100100010000 Signal (counts)Time (ns) 100nM 75nM 50nM 25nM 0nM Medium A 020406080100050000100000150000200000 Signal (counts)PDGF-BB (nM) B Figure 4-15. Fluorescence decays of 200 nM Pyr-MBA in simulated biological specimen. (A) Simulated biological sample with various concentrations of PDGF-BB and response of fluorescence intensity to the change of protein concentration. Excitation= 337 nm. The collection time for each decay was 3000 seconds. (B)Photons emitted between 40 and 100 ns after excitation pulse reached maximum were counted to construct the calibration curve.

PAGE 98

85 Expression of PDGF-BB from Cells Several reports have indicated that PDGF-BB can be over-expressed by various types of cancer cells, including MCF-7, MDA-MB-231 and PC3 cultured cell lines.46 Unfortunately, the literature is ambiguous as to the specific amounts of PDGF-BB that can be expressed under any given condition. This is largely limited to the fact that most of methodologies currently used provide only qualitative information and often not just specific to PDGF-BB expression.78;79 As a result, ELISA was employed to determine specific amounts of expression that could be expected from PC-3 tissue cultured cells. PC-3 cells are adherent prostate cancer cells and on incubation with TNFover-express PDGF-BB. TNF(also called cachectin) is a major immune response modifying cytokine produced primarily by activated macrophages. TNFalso induces the expression of other autocrine growth factors, increases cellular responsiveness to growth factors and induces signaling pathways that lead to proliferation. TNFacts synergistically with EGF and PDGF-BB in some cell types. In order to determine the levels of PDGF-BB present in conditioned media, ELISA tests were performed on different samples. The levels of PDGF-BB lwere measured from samples which were cultured to confluency and spiked with Recombinant Human TNFto induce PDGF-BB production (Table 4-2). The total protein concentration was determined by the Bradford Assay. Since the addition of TNFto PC-3 cells in culture should increase the levels of PDGF-BB in the conditioned media, it was first important to find out if we could monitor PDGF-BB induction. PC-3 cells were cultured in 35 mm cell culture dishes in a final volume of 2 mL. Twenty mg of TNFwas added to each cell

PAGE 99

86 dish and the conditioned media was collected at 0, 3, 6, 9, 12 and 24 hours after induction. Table 4-2. Amount of protein collected in 24 hours Incubation time (hrs) PDGF-BB (pg/mL) PDGF-BB (pM) Total protein concentration (g/mL) 0 < 15 <0.6 5893 3 < 15 <0.6 6915 9 16.3 0.7 6787 12 42.9 1.7 6934 24 197 7.9 7340 In this case it can be observed that TNFinduced the expression of PDGF-BB in PC-3 cells. Non induced samples have less than 15 pg/mL of PDGF-BB. On the other hand, after 9 hours of incubation, an increase in PDGF-BB levels can be observed, up to 197 pg/mL after 24 hours. The total protein concentration of all the samples was measured and it ranged from approximately 5,900 g/mL to more than 7000 g/mL. A second set of samples was prepared with the idea of increasing the amount of PDGF-BB present per sample. Changes included: incubation in a larger flask, with a small volume (enough to cover the surface of the flask), and with a higher amount of TNF-. A large surface area of adherent cells in a small volume should allow for a higher amount of PDGF-BB collection. The second change in the sample handling includes a concentration step by lyophilization (25 ml were reduced to 5 mL). In this case, PC-3 cells were cultured in five 75 cm2 flasks with 5 mL of cell media in each. Cells were grown to confluency and 25 mg of TNFwas added to each flask and the conditioned media was collected after 24 hours for one set and after 36 hours for the following. The total volume of conditioned media collected was 25 mL. After lyophilization the volume was reduced to 5 mL. The results from the ELISA and Bradford assays are shown in

PAGE 100

87 Table 4-3. It is illustrated that the amount of PDGF-BB is indeed higher with the modified sample collection method, doubling in a period of 12 hours. Table 4-3. Amount of protein collected in 24 and 36 hours Incubation time (hrs) PDGF-BB (pg/mL) PDGF-BB (pM) Total protein concentration (g/mL) 24 865 35 9334 36 1775 71 11934 The two sample collections methods showed an increase in PDGF-BB production after induction with TNF-, but the levels of PDGF-BB are lower than the current limits of detection of the pyrene and the FRET assay. In addition, even if the levels of PDGF-BB were detectable, the total amount of protein per sample is about 2 orders of magnitude higher than the allowed protein concentration determined for the quenching assay. Since the interfering non-specific interactions would be similar for the Pyr-MBA and FRET MBA (essentially, non-specific binding to the aptamer), it is reasonable to assume that the total protein concentration would limit both assays. If the total amount of protein commonly found in serum exceeds 100 g/mL before sample processing, non-specific interactions will result in false positive signals for this particular aptamer as shown in Figure 3-2. A longer incubation time may be a possibility for increasing PDGF-BB production, but after 36 hours of incubation, cells are 100% confluent and as result they begin to die. Also, the total amount of total protein concentration increased as well. Conclusions The method of excimer light switching is an excellent signal transduction for aptamer probe development. Pyr-MBA is able to detect protein in homogeneous solution and simple biological samples containing low, or acceptable levels of proteins for the aptamer. The generation of the excimer emission requires the conformation change of the

PAGE 101

88 aptamer brought about by complexation with a target protein to bring two pyrene molecules together. This stringent requirement prevents false positive signals when the probe is digested by nucleases. Another advantage of using this approach is that it potentially allows for ratiometric measurements to be obtained, which could minimize environmental errors and afford a more accurate detection. With this approach, a sensitive and selective PDGF-BB aptamer probe has shown a detection limit of about 200 pM in homogeneous solution. With the time-resolved single photon counting technique, the pyr-MBA was able to detect PDGF-BB in a simulated biological sample in the presence of low levels of serum proteins. For homogeneous solution experiments, the detection-without separation and sensitivity of this approach may find useful applications to construct biosensors for protein function studies in vitro with other aptamers for oligonucleotide probes. By using a combined approach, 4-12 mol of PDGF-BB were detected in a few seconds in vitro. With the use of time resolved measurements and the long fluorescence lifetime of pyrene relative to the lifetime of the cell media components, the fluorescence due to the excimer formation can be distinguished from autofluorescence of some of the cell media. Unfortunately, the direct application of the this particular Pyr-MBA for PDGF-BB detection may be hindered by the fact that the levels of PDGF-BB is lower than the limits of detection of the current assay and in addition the total protein concentration is high enough to produce false positive signals due to non specific binding. Both problems need to be addressed carefully in addition to the selection of sensitive signaling mechanisms for future MBA assay applications. The findings for the MBA and PDGF-BB interaction can not be generalized to every aptamer-target system and it may be possible that the

PAGE 102

89 same concepts and methods developed here can be successfully applied for the detection and studies of other protein biomarkers. Further exploration of lifetime measurements can yield information on the molecular microenvironment of a fluorescent molecule. Factors such as ionic strength, hydrophobicity, oxygen concentration, binding to macromolecules and the proximity of molecules that can deplete the excited state by resonance energy transfer can all modify the lifetime of a fluorophore used to label an aptamer based probe. Measurements of lifetimes can therefore be used not only for protein studies but also add an additional dimension to assay development. Furthermore, with instrumentation improvements that would allow for more high-throughput analysis, pyrene-based molecular probes could be adapted for multiplexing and multi-protein analysis.

PAGE 103

CHAPTER 5 FLUORESCENCE RESONANCE ENRGY TRANSFER STUDIES FOR MOLECULAR BEACON APTAMERS: UNDERSTANDING PROTEIN INTERACTION AND VERSITILITY OF PROBES Introduction Given the clear importance and drive to develop aptamers for pharmaceutical leads, detection reagents, and proteomic tools, understanding the structure and behavior of aptamers under practical conditions is critical to developing aptamer based prototypes that can ultimately be optimized and have an impact in industry.21;80-84 Thus, it is necessary to conduct systematic analyses of aptamers to understand any given aptamer-protein interaction, what factors are involved in this interaction, how the modification of the existing conditions affect the total change in fluorescence quenching on biding and if there is any universal rules that govern aptamer functions in assays, much like that of antibodies. Since their inception, aptamers have been shown to work well in homogeneous solution; however, in recent years the focus has changed towards the development of new aptamer based technologies.8 One example of such an application is the development of the most advanced therapeutic aptamer, pegaptanib from Macugen, an aptamer directed against VEGF.85;86 This anti-VEGF-aptamer, a PEGylated short 2'Fand 2'-methoxy modified RNA oligonucleotide was developed for the treatment of exudative age-related macular degeneration, the leading cause of irreversible loss of vision among Americans over the age of 55. Clinical phase trials showed that more than 80% of the patients show stable or improved vision after treatment with Macugen. As a result of these applications, fundamental studies on aptamer behaviors in more complex 90

PAGE 104

91 systems are important. As seen in previous chapters, the fact that any given aptamer probe can be used for sensitive detection of a target molecule (i.e. protein) in optimized homogeneous solution does not necessarily mean that the probe can be directly adapted for use in a complex biological matrix. Our initial efforts focused on the design of the probe and solving two of the major issues for using the PDGF-BB aptamer in biological samples. Even though this aptamer system appeared to be incompatible with the chosen cellular expression systems, several more general areas of interest could be studied with the current FRET-probe system that could be extended to a more general knowledge base for other aptamer detection methods. In this chapter, several assay formats are described and demonstrate that other strategies can be used and, if successful, optimized for the use of aptamers in different applications. Specifically, oligonucleotide backbone modifications, effects of the homologous RNA aptamer on FRET responses, utilization of multiple binding sites to generate signal transduction, and a two step restriction enzyme assay will be discussed. Although our studies could be generally applied to other aptamer systems, we targeted our experiments towards improving the existing FRET format for PDGF-BB. Results and Discussion Backbone-Modification of DNA The exchange of one non-bridging oxygen atom with a sulfur atom in the DNA phosphorothioate modification is a conservative change, as th e negative charge of the phosphate group is retained and the size of the sulfur atom is only slightly larger than the oxygen atom (Figure 5-1). The phosphorothioate internucleotidic linkage is considerably more stable than the phosphodiester bond toward nucleases degradation, and this attribute makes it practical for studies where cells would be involved.87 Oligonucleotide

PAGE 105

92 phosphorothioates can be synthesized by two methods: the first one is chemical synthesis utilizing a sulfurizing reagent in the oxidation step and the second is enzymatic incorporation. When oligonucleotides are employed for protein detection in cellular samples or in vivo therapeutic purposes such as the antisense oligonucleotide-mediated inhibition of gene expression this property (nuclease resistance) becomes especially important. Figure 5-1. Phosphorothioate DNA backbone modification. Standard phosphorothioate backbone (left). The diagram on the right shows were a sulphur atom replaces one of the non-bridging oxygen atoms in the phosphate group. Since the normal phosphodiester backbone of synthetic DNAs is vulnerable to nucleases present in biological samples, the 36t-MBA modified with a phosphorothioate backbone DNA was investigated to determine if an MBA probe could be modified and still remain functional. The phosphorothioate backbone-modified DNA was synthesized using the standard sulfurizing agent, Beaucage Reagent, to achieve stepwise sulphurization instead of the iodine oxidation step in the standard synthetic cycle. The backbone modified DNA was synthesized using a DABCYL moiety at the 3-end and a phosphoramidite fluorescein at the 5'-end. The resulting MBA was then purified using standard by reverse phase HPLC. Retention time for phosphorothioates will be longer than for corresponding phosphodiester analogues. As a result, purification is improved;

PAGE 106

93 however, partial modifications, or incomplete modifications, would still be difficult to remove from the synthesis product. The impurities will mostly attribute to partial cleavage of the probe, but not significantly affect the general open versus closed fluorescence of the probe because cleaved probes would not result in quenching. 01000020000300004000050000Thio-MBAFluorescence Intensity/A.U. withoutPDGF-BB 50 nMPDGF-BB Figure 5-2. Fluorescence quenching produced by a phosphorothiate modified MBA. Seventy five nM of the backbone-modified MBA showed a detectable quenching, approximately 75 % as compared to 90% observed with unmodified MBA, when incubated with a 300 nM 4-fold molar excess of PDGF-BB in a standard fluorescence quenching assay (Figure 5-2). The results presented in Figure 5-2 indicated that the DNA backbone modification to achieve nuclease resistance did not significantly inhibit the binding and the conformational change of the MBA in response to PDGF-BB; but the overall quenching efficiency was slightly lower. Possible reasons for a decrease in the overall fluorescence response of the MBA include the fact that some modifications may

PAGE 107

94 disrupt some of the secondary structures that are important for binding. In another respect, the negative charge of the phosphodiester group is more important for protein interactions than interactions with other nucleic acids, since specific contacts between the protein residues and the phosphodiester groups could be critical for binding. Nonetheless, since the backbone modified MBA should be more stable than the standard 36tMBA to nucleases encountered in biological specimens the use of the modified MBA is promising.87 MBAs Created from Homologous RNA and DNA Aptamers Display Significant Differences in Eliciting FRET in Response to PDGF-BB Homologous RNA and DNA sequences display significant differences in the stability of the secondary structures in solution. RNA is made of four different building blocks, the ribonucleotides. The pyrimidine base thymine is modified in that it lacks a methyl group and the resulting uracil takes its place in base pairing. The ribose is fully hydroxylated form making it less stable than DNA. Together, the presence of uracil in place of thymine, and the 2'-OH in the ribose constitute the two chemical differences between RNA and DNA. Other properties that distinguish DNA versus RNA molecules include: RNA is mostly a single stranded molecule, with some sections that may double back on themselves resulting in double stranded regions; RNA is generally shorter; single stranded RNA is more susceptible to degradation than DNA; and the RNA is less easily repaired due to the lack of a secondary strand to correct mistakes. Similar to DNA, oligoribonucleotide synthesis is also performed on solid supports through the use of appropriately protected phosphoramidites. Continued development of 2-OH protecting groups, the use of efficient activators, and purification methods has greatly improved RNA synthesis and nuclease resistance.

PAGE 108

95 The use of RNA-aptamers is appealing for the studies of small molecules and proteins because it is capable of recognizing a large variety of targets and in many cases RNA aptamers show impressive specificity. With the advances in oligo synthesis methods and with the potential advantages that an RNA aptamer may present for protein detection, a series of experiments were targeted towards switching the DNA-aptamer for PDGF-BB into an RNA-based aptamer. Currently, no reports exist that have investigated whether or not an RNA version of any existing DNA aptamer will bind to the target protein with the same affinity as the equivalent DNA. Therefore, the FRET responses displayed by MBAs created from the DNA aptamer designated 36t47 and an RNA aptamer with the same nucleotide sequence (for the RNA aptamer all the T nucleotides were replaced with U as shown in Table 5-1) were compared. Table 5-1. Aptamers synthesized for PDGF-BB detection Name Sequence 36tDNA MBA 5FAM-CACAGGCTACGGCACGTAGA GCATCACCATGATCCTGTG-Dab3 36tRNA MBA 5FAM-CACAGGCUACGGCACGUAGAGCAU CACCAUGAUCCUGUG-Dab3 A theoretical model for the RNA equivalent of the 36tDNA-MBA show that folding of the aptamer is similar to the standard 36t MBA: a three-way helix junction with a three-nucleotide loop at the branch point. For this set of experiments both aptamer beacons (DNA and RNA based) contained a DABCYL quencher at the 3-end and a 6-FAM dye at the 5-end due to the previous results (Figure 2-3) which indicated, that for this aptamer probe, DABCYL/6-FAM was the most effective pair. Standard fluorescence quenching assays (as described in Material and Methods on Chapter 2) were performed in parallel with the DNA-MBA and RNA-MBA using a two-fold and a four-fold molar excess of PDGF-BB. As reported in Chapter 2 for the standard 36t DNA-MBA, PDGF

PAGE 109

96 BB causes fluorescence quenching in a dose-dependent manner and that between equimolar to 2-fold molar excess of PDGF-BB, the extent of quenching in the standard fluorescence quenching assay is approximately 90%.62 Figure 5-3. Theoretical model of the 36tRNA-MBA. The 3' end was labeled with DABCYL as the fluorophore and the 5'end was labeled with fluorescein as the quencher The RNA equivalent of the MBA was directly compared to the standard 36tMBA and the results are shown in Figure 5-4. Both 2-fold and 4-fold molar excess yielded more than 90% quenching with the DNA-MBA; however, with the RNA-MBA, the extents of observed quenching were significantly lower% at 2-fold and 54% at 4-fold molar excess of PDGF-BB. The higher quenching efficiency of the DNA versus the RNA molecules may be due to the fact that the 2'-OH group is known to play a role in determining helical parameters and helix stability. As a result, for MBAs it may also contribute to tertiary interactions that stabilize aptamer structure and interact directly with the ligand, thereby decreasing, or possibly increasing, an aptamers ability to bind and recognize the target molecule.88

PAGE 110

97 90%91%21%54 % 0300000600000900000DNA-MBADNA-MBA +100 nMPDGFDNA-MBA +200 nMPDGFRNA-MBARNA-MBA +100 nMPDGFRNA-MBA +200 nMPDGFFluorescence Intensity/A.U. Figure 5-4. Comparison of FRETresponse of 36t-DNA-MBA and the homologous 36t-RNA-MBA with PDGF-BB. Two-fold and 4-fold molar excess of PDGF-BB were incubated with 50 nM of the 36t-DNA-MBA and 36t-RNA-MBA in binding buffer. To further analyze the difference between the DNAand the RNA-MBAs, we next examined the ability of unmodified DNA and RNA aptamers to restore the fluorescence quenching by competing with the DNA-MBA for binding with PDGF-BB. A competition assay using unmodified DNA aptamers at increasing concentrations with the standard DNA-MBA (Figure 5-5) showed the expected disappearance of fluorescence quenching. We then examined the reciprocal abilities of unmodified RNA and DNA aptamers to abrogate fluorescence quenching response of DNA-MBA (6-FAM-DABCYL labeled DNA aptamer) and RNA-MBA (6-FAM-DABCYL labeled RNA aptamer), respectively. Competition experiments on the fluorescence quenching assays were performed as follows in 100 L each: 50 nM of the MBA were incubated with different amounts of the competitor (unlabeled DNA and/or RNA with a similar sequence of the 36t) for 30

PAGE 111

98 minutes in binding buffer, then, 200 nM of PDGF-BB were mixed into the solution and fluorescence measurements were obtained immediately. Fluorescence intensities were measured before and after the addition of PDGF-BB to the sample solution. As shown in Figure 5-5, a standard competition assay using unmodified DNA aptamers in increasing concentrations with the labeled DNA-MBA showed an increasing fluorescence as the amount of unlabelled DNA was increased. 040000080000012000000:11:12:14:110:11:0Ratio of DNA:PDGF-BB Fluorescence Intensity/A.U. Figure 5-5. Competition of FRET-response by unmodified 36t-DNA aptamer with the standard 36t-DNA-MBA. A standard competition assay using increasing concentrations of the unmodified DNA aptamer with 50 nM of the standard DNA-MBA incubated with 200 nM PDGF-BB showed the expected disappearance of fluorescence quenching. Unmodified RNA molecules were not able to effectively weaken the FRET response of the DNA-MBA even at a ten-fold molar excess (compared to PDGF-BB) as can be observed in Figure 5-6. In contrast, the unmodified DNA aptamer, even at a 1:1 molar ratio to PDGF-BB, almost completely reversed the fluorescence quenching observed in the presence of 200 nM PDGF-BB relative to the RNA-MBA concentration. The data clearly demonstrates that the DNA-MBA is able to bind and elicit a FRET

PAGE 112

99 response to PDGF-BB better than the homologous RNA-MBA. In terms of the mechanism for binding, the results indicated that the difference between the RNA and DNA aptamer sequences is significant enough to inhibit one-another. The DNA aptamer binds stronger to the protein and therefore is not competed with, when in the presence of the unlabelled RNA aptamer. This point is reiterated in the competition of the labeled-RNA aptamer with unlabelled-DNA aptamer, where the fluorescence is restored for the probe. As a result, MBAs that are developed for molecular detection will most likely bind regardless of whether the probe consists of RNA or DNA. However, individual experiments for other aptamer probes should be conducted to determine which modification will result in the highest affinity for the target. Interestingly for the PDGF-BB, we did observe binding the probe but clearly the DNA based aptamer is more appropriate for MBA applications. Fluorescence Enhancement Assays with MBAs Since the principle of the MBA-based assay relies on signal transduction resulting from differences in the structure of free versus the protein-bound MBA, it should be possible to modify the assay from a quenching to an enhancement format by selecting an appropriate pair of fluorophores. It is generally believed that an enhancement assay format provides a higher degree of sensitivity than that achieved with a quenching format because the initial fluorescence signal from the probe will be low. Specifically, the sensitivity and dynamic range of MBAs are determined mainly by two parameters: the initial fluorescence intensity when the MBA is its open conformation and the residual fluorescence intensity when the MBAs are in the closed conformation (incomplete quenching).

PAGE 113

100 02000004000006000008000001000000MBA+1X rnaDNA-MBA +10X rnaRNA-MBA +1X dnaRNA-MBA +10X dnaFluorescence Intensity/A.U. without PDGF-BB 200 nM PDGF-BB 50 nM DNA MBA50 nM DNA MBA50 nM RNA MBA50 nM RNA MBA200 nM RNA2 M RNA200 nM DNA2 M DNA50 nM DNA MBA50 nM DNA MBA50 nM RNA MBA50 nM RNA MBA200 nM RNA2 M RNA200 nM DNA2 M DNA Figure 5-6. Competition of FRET response by unmodified RNA aptamer with DNA-MBA and by unmodified DNA aptamer with RNA-MBA. Unmodified RNA molecules incubated with 50 nM MBA-DNA at 1:1 and 1:10 ratios were not able to effectively abrogate the FRET response of the DNA-MBA on PDGF-BB binding. In contrast, when the unmodified DNA aptamer was studied at a 1:1 molar ratio to the RNA-MBA,the fluorescence quenching observed in the presence of 200 nM PDGF-BB concentration was almost completely reversed. In principle, the fluorophore should be co mpletely quenched by the quencher in the closed form. In reality, however, the residual fluorescence varies greatly, because of factors such as the selection of the fluorophore/quencher pair, the synthesis of the MBA, and the way the fluorophore/quencher groups are attached to the oligonucleotide. Such residual fluorescence greatly limits the detection sensitivity of the MBAs. This issue can be addressed by attaching two fluorophores (F1 and F2) to the ends of the probe instead of a fluorophore and a quencher. When the MBA is in its open conformation prior to binding, the observed fluorescence of the acceptor fluorophore should be minimal. Therefore, in general a fluorescence enhancement assay will initially have a lower fluorescence than the quenching assay before the protein is added to the assay solution. In

PAGE 114

101 the quenching format the signals changes are observed in the presence of high background signals (in this case the open conformation of the quenching assay) which can limit the sensitivity of the assay. For the synthesis of a Cy3/Cy5 MBA, the DNA was synthesized with a functional amine group immobilized on controlled pore glass beads and Cy5 was added to the 5-end using Cy5 phosphoramidite (Amersham Bioproducts, Piscataway, NJ). Following deprotection with potassium carbonate (0.1M)/methanol and gel filtration, the DNA was reacted with N-hydroxysuccinimidyl ester of Cy3 (Amersham Bioproducts, Piscataway, NJ) and the dual labeled material was purified. All the MBA stock solutions were prepared at 100 M concentration in Tris/HCl buffer (pH 7.5, 20 mM) with NaCl (20 mM), and stored at -20C in aliquots. We used the standard fluorescence quenching assay to compare the response of the standard MBA to one that was labeled with a Cy3 dye at the 3-end and a Cy5 dye at the 5-end. The Cy-labeled MBA is suitable for an enhancement format since Cy3 and Cy5 are well known FRET pairs used for DNA hybridization studies. The donor dye Cy3 can be excited at the absorption maximum (546 nm) and the fluorescence signal can be measured at the emission maximum of the acceptor dye, Cy5 (662 nm) when the MBA is in its closed conformation. A 46% enhancement in the measured fluorescence intensity was observed with 50 nM of the Cy3/Cy5-labeled MBA on the addition of 200 nM of of PDGF-BB. Another potential application of the enhancement FRET assay was to further explore and investigate the probe: protein binding ratio. Photo-crosslinking experiments reported by Green et al.47 suggested that the phenyalanine-84 residue in the B-chain of

PAGE 115

102 PDGF-BB forms a point of contact with a specific nucleotide residue in loop at the helix junction of the closed conformation of a PDGF-BB-specific aptamer. It was hypothesized that if the homodimer PDGF-BB could bind two MBA molecules with appropriate conformational characteristics, then FRET could occur between two aptamers each labeled with one of the two dyes of a FRET pair. This would allow the design of a fluorescence enhancement format for a FRET assay for PDGF-BB that uses two single fluorophore-labeled aptamers. Specifically, the aptamer sequence was labeled either with Cy3 (the donor) at the 3-end or with Cy5 (the acceptor) at the 5-end. A solution was then prepared by mixing an equimolar ratio of the 3-Cy3 and the 5-Cy5 labeled PDGF-BB aptamer preparations. Once two aptamers carrying two different labels bind to the same PDGF-BB molecule, the two fluorophores would be in close proximity. It should then be possible to excite the donor fluorophore, Cy3, at its absorption band and due to FRET its fluorescence will be quenched by Cy5. However, there will be a simultaneous increase in the fluorescence of Cy5 even when the MBA molecule is excited at the absorption band of Cy3. For the fluorescence enhancement assay, increasing amounts of PDGF-BB were incubated with a fixed amount (75 nM) of the Cy3/Cy5-aptamer equimolar mixtures and fluorescence intensity was measured at each different concentration. The sample solution was excited at the absorbance maximum of Cy3 and the fluorescence was monitored at the emission maximum of Cy5.

PAGE 116

103 010000200003000002468PDGF (g/mL)Fluorescence Intensity/ A.U. Figure 5-7. Fluorescence enhancement FRET-assay. Increasing concentrations of PDGF-BB were incubated in the standard fluorescence quenching assay buffer (final volume 100 L) with an equimolar mixture of the PDGF-BB aptamer labeled with Cy3 at the 3'-end and a PDGF-BB aptamer labeled with Cy5 at the 5'-end (fluorescence measurements were made at excitation maximum of Cy3 and emission maximum of Cy5). A dose-dependent increase in Cy5 fluorescence emission on additions of increasing amounts of PDGF-BB is shown in Figure 5-7. The results demonstrate the capability of this mixed aptamer solution in a sensitive enhancement assay consistent with the hypothesis that two or more aptamers bind to a single PDGF-BB molecule with a detection limit in the nanomolar range without optimization. This was the first demonstration of a fluorescence enhancement assay in a single incubation where binding to a target protein molecule of two separate aptamer molecules carrying a donor and an acceptor dye presumably results in FRET due to the close proximity of both aptamer molecules. This is a striking contrast to opening up of a double stranded stem of a molecular beacon to produce fluorescence enhancement in studies on nucleic acid targets.

PAGE 117

104 The ability to perform a FRET bioassay with single labeled DNA molecules have important future implications in the development of novel fluorescence assays for multimeric proteins, protein modifications and protein aggregates by using high affinity DNA aptamers that recognize different domains of the proteins. Two-Step FRET Assays Using Restriction Enzymes Cleavage Sites To further explore the relationship between the signal transduction domain (the fluorophore-quencher labeled stem), and the molecular recognition domain (the helix/loop structure) of the MBA and its influence on the FRET efficiency, several modifications need to be properly engineered and designed. With this in mind, an extension of the use of FRET response by MBA to the target protein in a novel manner based on the principle of stabilization of the stem by specific binding of the target protein PDGF-BB will be presented next. As Green et al. reported,47 the base sequence distal from the loop portion of the PDGF-BB aptamer are not critical for binding to the target protein, therefore, modifications in the stem sequence should not prevent the selective binding to PDGF-BB. To demonstrate this ability, two new MBAs were designed with two minimal variations in the terminal sequences of the 36t aptamer to create a BamHI and a HpaII cleavage site in the 6 base pair stem. BamHI and HpaII are restriction enzymes (endonucleases that recognize specific double stranded DNA sequences and subsequently cleave the DNA in both strands). The enzymes have high specificity for their recognition sites and under optimum reaction conditions, even sites that differ in only one base pair from the canonical or original site are not cleaved, unless large enzyme concentrations are used or the reaction is allowed to proceed for sufficient time.89

PAGE 118

105 MBAs designed to contain 3and 5sequences that can form a 6 base pair double stranded stem with the cleavage site either for BamHI (GGATCC) or HpaII/MspI (ACCGGA) were synthesized with a 3-Dabcyl quencher and a 5-fluorescein dye using standard phosphoramidite chemistry as described in Chapter 2. As shown in Figure 5-8, the bases that compose the stem were changed to create a cleavage site for a specific restriction enzyme. This series of modified MBAs can be used in a simple two-step mechanism: the first step will be the stem stabilization due to the binding of PDGF-BB to the MBA which can be confirmed by fluorescence quenching. The second step will be the addition of the appropriate restriction enzyme to the aptamer-protein complex solution, where the fluorescence of the MBA is restored by restriction enzyme cleavage. With this straightforward mechanism, the binding of the protein target to the aptamer can be confirmed with two separate signal changes. Figure 5-8. Working principle of the enzyme-site modified MBA. In this case, the base sequence stem of the MBA was modified for the appropriate bases that will create a specific site for restriction enzyme cleavage. When an excess of PDGF-BB is incubated with the MBA, the stem will be stabilized and as consequence, a decrease in fluorescence intensity will be observed. On a subsequent, step the specific restriction enzyme will be added to the solution. The enzyme will cleave the double stranded stem, and as result the fluorophore and the quencher will be spatially separated and the fluorescence will be restored.

PAGE 119

106 Effects of Mg on the modified MBA probes In order to determine the relative abundance of MBA conformational state with a double stranded stemthe closed conformationwe designed and tested two MBA variants of 36t sequence where the 5and the 3ends (CACAGG [ . ] CCTGTG) were modified to create recognition sequences for two restriction enzymes HpaII/MspI (ACCGGA [ . .] TGGCCT) and BamHI (GGATCC [ . ] GGATCC). Each MBA was tested for its ability to elicit FRET in a standard fluorescence quenching assay except that the MBA was used at a final concentration of 25 nM. After measuring fluorescence following incubation with PDGF-BB at 100 nM concentration, 1 unit of the appropriate restriction enzyme and 10 l of a 10x reaction buffer recommended for the respective enzyme was added. Fluorescence readings were measured after 30 min incubation with the enzyme at 37C. It should be noted that both reaction buffers contain 10 mM Mg2+, a divalent cation that causes detectable fluorescence quenching with the standard 36t MBA.62 As expected from our previous studies, the addition of Mg2+ did cause detectable quenching in both MBAs (Figure 5-9). The differences in the presence of Mg2+ for the two MBAs most likely reflect the greater stability of the 4 base pair duplex in the 6 base pair stem needed by HpaII as compared to the need for the entire 6 base pair stem for BamHI. In the presence of PDGF-BB, both MBAs displayed quenching under standard conditions of fluorescence quenching assay albeit to different extents; both, however, performed worse than the 36t MBA. Our observations show that the sequence of the stem plays a significant role in the mediation of the FRET response for the PDGF-BB aptamer. Cleavage of the stem by restriction enzymes In order to further assess the presence and the stability of the stem in the presence of PDGF-BB, 25 nM of the MBA

PAGE 120

107 was first incubated with 100 nM of PDGF-BB and then treated with the respective restriction enzymes without Mg2+. With both MBAs the fluorescence quenching observed in response to PDGF-BB was followed by an enhancement upon treatment with the respective restriction enzyme, even in the absence of Mg2+ in the reaction buffer. In both cases the fluorescence readings following the restriction enzyme treatment were comparable to those prior to the addition of PDGF-BB. The results shown in Figure 5-10 demonstrate that PDGF-BB induces the formation of stable double stranded stems that serve as effective substrates for the respective restriction enzymes. In addition to the new insight into the basis of FRET response elicited by the MBAs, the observations provide a novel FRET assay, where the combination of two steps might result into a more reliable target protein recognition than the standard fluorescence quenching assays described to date. 010000200003000040000HPA MBAHPA MBA+ MgHPA MBA+ Mg +HPABAMHIMBABAMHIMBA + MgBAMHIMBA + Mg+ BAMH1Fluorescence Intensity/ A.U. Figure 5-9. Molecular beacon aptamer variants with restriction enzyme cleavable stem sequence response to magnesium. MBA variants of 36t sequence where the 5and the 3ends were modified to create recognition sequence for BamHI and HpaII/MspI. The presence of a stable double stranded stem in the MBA in the presence of PDGF-BB was detected by measuring fluorescence enhancement on treatment with the specific restriction enzyme in the standard reaction buffer recommended for the respective enzymes which contains 10 mM Mg2+. 25 nM of the 36t-DNA-MBA was first incubated with 100nM of PDGF-BB and then treated with the respective restriction enzymes.

PAGE 121

108 It appears that the one with HpaII cleavage site was more responsive both in quenching in response to PDGF-BB and subsequent restoration on the addition of either HpaII or its isoschitzomer, MspI. The difference could be due to the fact that HpaII/MspI cleavage requires only the center 4 bases to stably form a duplex in the 6 base pair stem while BamHI requires the entire 6 base pair stem to be stable. With the use of DNA folding algorithms we obtained theoretical secondary structures for both MBAs that showed that in the case of BamHI modified MBA, the majority of the secondary structures do not have the stable stem required for high FRET efficiency. 010000200003000040000HPA MBAHPA MBA+ PDGFHPA MBA+ PDGF +HPABAMHIMBABAMHIMBA+PDGFBAMHIMBA +PDGF +BAMHIFluorescence Intensity / A.U. Figure 5-10. Molecular beacon aptamer variants with restriction enzyme cleavable stem sequence response to PDGF-BB. MBA variants of 36t sequence where the 5and the 3ends were modified to create recognition sequence for BamHI and HpaII/MspI. The presence of a stable double stranded stem in the MAB in the absence of PDGF-BB was detected by measuring fluorescence enhancement upon treatment with the specific restriction enzyme in the standard reaction buffer recommended for the respective enzymes. 25 nM of the 36t-DNA-MBA was first incubated with 100nM of PDGF-BB and then treated with the respective restriction enzymes in the absence of 10 mM Mg2+ at 37 C. Only one of the four possible structures has the stem and is not a thermodynamically preferred conformation. For the HPAII modified MBA, the majority of the possible secondary structures have the stem structure and have the lowest energy (G=-4.8),

PAGE 122

109 versus only one conformation where the ends are completely separated and G =-3.9. This novel two-step FRET assay provides the foundation for a unique strategy of designing MBAs from high affinity DNA aptamers for protein targets by using terminal sequence modifications for one or more of an extensive array of available restriction enzymes. The ability to monitor subtle changes in the structure of the nucleic acid aptamer probes upon specific binding to a target protein as compared to one of its variants should provide a significant expansion of the use of MBAs as novel tools to study protein molecules. Conclusions The ability to develop affinity based systems of tailor designed characteristics, which can be applied to the detection of proteins offers great opportunities to explore new and dynamic routes for sensor development. Among many possible settings, in this chapter, a backbone modification of the MBA showed fluorescence quenching upon binding to the target protein but with lower affinity, a comparison between homologous DNA-MBA and RNA-MBA was conducted and revealed that indeed both RNA and DNA based aptamers of identical sequence could bind to the target. Interestingly, competition assays revealed that they have different binding capabilities. Thus, depending on which background yields better affinities and fluorescence enhancements, improved probe selectivity and signaling should be explored for any MBA that is subsequently developed. Optimization of the assay may also be achieved by using other combinations of dye-quencher or dye-dye FRET pairs for higher signal-to-background ratios. Also, the distance between the FRET signaling moieties can be adjusted based on the conformation

PAGE 123

110 of the closed form of the bound MBA as predicted by DNA structure modeling software The use of a dye-dye pair for signal enhancement showed that an MBA can be readily modified to satisfy experimental requirements. In addition, modified sequences at the 3and the 5-ends of the MBA that allow the formation of restriction enzyme cleavage sites in the predicted double stranded stem portion were investigated. The MBAs presented themselves as a novel class of molecules that displayed an expected fluorescence quenching (using DABCYL-Fluorescein as the quencher-fluorophore pair) upon the addition of PDGF-BB. Subsequent addition of the restriction enzyme elicited fluorescence enhancement presumably as a result of releasing a very short DNA fragment (2 to 4 base pair) carrying the FRET pair but is unstable under assay conditions. The restriction enzyme cleavage was observed even in the absence of Mg2+ further supporting the notion that PDGF-BB stabilizes the double stranded stem of the MBA. The observed two-step FRET response is unique to MBAs designed for protein targets and may serve as the foundation for the development of novel assays for structure-function analysis of protein molecules. The ability to achieve signal transduction using several known FRET pairs not only indicates the versatility of the MBA-based assay method but also strongly supports the notion that multiple MBAs for selected biomarkers can be created with distinct FRET pairs. There is a need to develop multiplex assays to simultaneously monitor disease biomarkers in a single incubation in homogeneous solutions. With the MBAs that contain specific sites for restriction enzyme cleavage, it may be possible to design a panel of MBA for protein biomarkers, where the binding of the biomarker can be confirmed by the combination of two individual signaling events. It will require a careful choice of

PAGE 124

111 FRET pairs, an analysis of their relative positions in the open versus closed conformations of the aptamer and studies on incubation conditions that will give the largest signal change upon binding and cleavage.

PAGE 125

CHAPTER 6 SUMMARY AND FUTURE WORK Summary Our study showed how a synthetic DNA sequence can be modified into a probe to report the binding of a target protein biomarker platelet derived growth factor (PDGF-BB) in homogeneous solution and in simple biological samples. The developed probe, a molecular beacon aptamer (MBA), was used for the development of a fluorescence assay. The selectivity and sensitivity of this assay method arises from the fact that the signal generation requires not only the binding to the target protein but also specific conformational characteristic of the bound DNA. Our results show that using the MBA in a FRET assay, 10 nM of PDGF-BB were consistently detected in repeated experiments and the conditions required for successfully using MBAs to perform the novel single-step FRET bioassay are compatible with the physiological pH, temperature range and common monovalent cations. We also showed that the two highly related molecular variants of PDGF-BB (PDGF-AA and PDGF-AB) can be distinguished from one another in homogeneous solution, in a single-step assay that can be readily adapted to a microtiter plate assay for high-throughput analysis. The results obtained in the first part of this study are promising for future applications of this assay for the detection of PDGF-BB in biological samples, although processing methods to remove divalent cations and major protein contaminants still need to be optimized. Once the assay was completely characterized in homogeneous solution, a simulated biological sample was used to determine the feasibility and applicability of the MBA for 112

PAGE 126

113 PDGF-BB detection in a more complex biological matrix. Initial experiments showed that, when in the presence of similar total protein concentrations, the MBA has the potential to qualitatively distinguish the presence of nanomolar quantities of PDGF-BB in a spiked simulated biological sample. However, when unspiked samples containing high levels of serum proteins were studied, fluorescence quenching was also observed, indicating that it may be possible that non-specific binding of other proteins may lead to false positive quenching. Further studies were conducted towards the determination of the robustness of this assay against other species present in the biological samples. For this particular PDGF-BB aptamer and assay conditions, the presence of more than 100 g/mL of serum proteins resulted in 76% of quenching in the absence of PDGF-BB. The results indicated that a sample processing method needs to be performed before the FRET-based assay in order to obtain consistent results. Simple chromatographic methods were tested to process the sample prior to the assay to remove potential biological interferences. With dilutions and gel-filtration, PDGF-BB could be monitored in the presence of serum proteins; however, the amount of PDGF-BB required to obtain full quenching of the probe would be significant, considering the sample would need to be processed. Moreover, in a native biological environment, there are other potential sources for non-specific quenching of the probe. This greatly limits exploring the full potential of FRET-based MBAs for protein detection. For this reason, the effect of background signal from auto-fluorescing components of the cell media was investigated. Several molecular species exist in a biological environment, some of which will yield a strong fluorescence background signal when light is applied to the dye molecules. If the auto-fluorescence

PAGE 127

114 intensity of some of the cell media components is comparable to or larger than the fluorescence intensity of the fluorescence of the MBA, the fluorescence quenching becomes masked and the analysis of the fluorescence signals. For this purpose, an alternative signal transduction was implemented within the PDGF-BB aptamer sequence. Specifically, the aptamer was labeled with pyrene molecules at each end. The pyrene-labeled probe was able to detect protein in homogeneous solution and in real time. The generation of the excimer emission requires the conformation change of the aptamer brought about by complexation with a target to bring two pyrene molecules together. One feature of the pyrene excimer is that it has a very long lifetime compared with other potential fluorescent species. The lifetime of the pyrene excimer could be as long as 100 ns, while that for most of the species that contribute to the biological background signal is shorter than 5 ns. This property allowed the separation of the biological background signal from that of the excimer signal using time-resolved fluorescent measurements and for the detection of 200 pM of PDGF-BB in a simulated biological sample with low level levels of interfering proteins. An advantage of using this approach is that it allows ratiometric measurement, which could minimize the environmental effect to afford more precise detection. More importantly, this excimer light switching approach significantly reduces background problems both from the probe itself and from biological species. Both signaling approaches, FRET and excimer emission, have wide applicability for the design and development of MBAs for several reasons. First, there are many aptamers besides PDGF-BB aptamer that undergo conformation changes with target binding event, for example, human -Thrombin aptamer, cocaine aptamer, HIV1 TAT

PAGE 128

115 protein aptamer, and Human Tenasin-C aptamer. Conformational change can be immediately exploited for the design of probes for protein detection. Second, even for aaptamer that does not have obvious target-induced structure change, it can be designed to change its secondary structure as desired upon target binding with rational structure modification. Simple and general approaches for the modification of aptamer sequences into structure switching aptamers for real-time signaling applications can be found in the literature. Third, the signal transduction mechanisms reported here require target-induced structure switching in order to produce a signal change, which prevents false signals caused by nucleases degradation. In addition, by combining the fast turnaround of automatic SELEX technique and novel selection techniques virtually any protein target can have at least one conformation-changing aptamer sequence. Thus, numerous aptamer signaling probes may be developed for different targets. For this project, all the experiments were made exclusively with the PDGF-BB aptamer system. Nonetheless, it is necessary to conduct a systematic analysis of aptamers structures and behaviors to understand the different processes that leads to the formation of stable protein-DNA complexes, what factors are involved in this interaction, how the modification of the existing conditions affect the total change in fluorescence quenching upon binding and if there is any universal rules that govern aptamer functions in assays. With advances in the selection of aptamers for different targets, it is necessary to conduct studies to determine what parameters are important for optimal aptamer performance in assays and detection systems as well as their limitations. Aptamers can be easily modified to meet different experimental requirements for bioassays. We investigated several modifications, including a DNA backbone

PAGE 129

116 modification to create a nuclease-resistant that showed fluorescence quenching upon binding to the target protein but with lower affinity. A comparison between homologous DNA-MBA and RNA-MBA was conducted and revealed that indeed both RNA and DNA based aptamers of identical sequence could bind to the target. Optimization of the assay may be achieved by using other combination of dye-quencher or dye-dye FRET pairs for higher signal and lower background and also by adjusting the distance between the FRET signaling moieties based on the conformation of the closed form of the bound MBA as predicted by DNA structure modeling software. In this case we tested a combination of two-dyes, Cy3 and Cy5, which resulted in a fluorescence enhancement assay. The ability to achieve signal transduction using several known FRET pairs not only indicates the versatility of the MBA-based assay method but also strongly supports the notion that multiple MBAs for selected biomarkers can be created with distinct FRET pairs and used in a multiplex bioassay. There is a need to develop multiplex assays to simultaneously monitor disease biomarkers in a single incubation in homogeneous solutions. The design of a successful MBA for protein biomarkers will require a careful choice of FRET pairs, an analysis of their relative positions in the open versus closed conformations of the aptamer and studies on incubation conditions that will give the largest signal change upon binding. In addition, MBAs were designed to incorporate restriction enzyme cleavage sites at the 3and 5-ends of the PDGF-BB aptamer. Upon addition of PDGF-BB the expected fluorescence quenching was observed. Subsequent addition of the restriction enzyme elicited fluorescence enhancement as a result of releasing a very short DNA fragment (2-4 base pair) carrying the FRET pair the fluorescence was restored. The unique two-step

PAGE 130

117 FRET response towards protein targets could provide new means to develop a novel class of bioassays for structure-function analysis of protein molecules. In conclusion, the molecular beacon aptamer studied here indicates the potential for aptamers to be used in the detection of protein targets in homogenous and simple biological solutions as well as provided significant insight into the design and development of aptamer based assays. Future Work New MBA Development for Other Proteins and Small Molecules One of the main problems that have prevented the widespread use of aptamer so far is the limited number of aptamers that has been developed. Currently, much effort is being put into developing new methodologies for aptamer selection. These include automated SELEX,18 CE-based SELEX,90-92 and cell-based SELEX93;94 and should provide a much broader assortment of aptamers than is currently available. As a result, other aptamer systems besides PDGF-BB can be identified and explored for their potential use in assay development. Also, for aptamers can be developed new protein and small molecule targets of interest and then used for probe design. In this work, we have applied several strategies for probe characterization which allowed for the effective evaluation of the PDGF-BB system. Similar approaches will be followed to determine the applicability of others aptamer sequences. The possibility exists that other aptamer probes will not experience the same drawbacks of the PDGF-BB aptamer system, and given the right aptamer-target combination and experimental scenario improved sensitivity and selectivity could be feasible. Enrichment and Simplification of Complex Samples For biological samples, low levels of target molecule often exist in the presence of excess unrelated biomolecules. One possible area to further explore includes sample

PAGE 131

118 enrichment through the use of magnetic particles. Previously, aptamer labeled magnetic-nanoparticles have been used to preconcentrate analyte prior to analysis with mass spectrometry.27 In the case of PDGF-BB, similar methodologies could be applied to extract and enrich PDGF-BB from cellular samples prior to analysis. To effectively develop such a protocol, non-specific interactions with the aptamer functionalized nanoparticle would need to be explored as well as the efficiency of collection and release of the protein preceding analysis. Pyrene Application to Other Protein Recognition Systems The pyrene signaling mechanism has wide applicability and should be further investigated for other systems for various reasons. First, there are many aptamers besides PDGF-BB aptamer that undergo conformation changes with target binding event, for example, human -Thrombin aptamer,39;74 cocaine aptamer,73 HIV1 TAT protein aptamer,75 and Human Tenasin-C aptamer.93 This conformational change should be used to design new probes for the detection of other biologically important protein molecules. For instance, -Thrombin aptamer labeled with two pyrenes has already been shown77 to yield intense excimer emission that was proportional to different concentrations of thrombin. Second, even for aptamers that do not have obvious target-induced structural changes, the sequence can be designed to change its secondary structure as desired upon target binding with rational structure engineering, as was shown in Chapter 5.64;75;95 This strategy has been well demonstrated by Bayer et al. 95 who used aptamer target binding event to switch aptamer structure to convert it as a gene expression regulator. In combination with the spectral characteristics of pyrene molecules, MBA-type of probes can be exploited to study structural changes of aptamers as well as potentially identify targets in complex samples.

PAGE 132

119 Signal Amplification of Aptamer Recognition through PCR When aptamer molecules are combined with signal transduction mechanisms, they can successfully be applied for protein detection in homogeneous solution. However, the practical application of a single-signaling MBA for protein detection is limited by the presence of large amounts of extracellular proteins and the low levels of PDGF-BB in typical biological samples. Several alternatives can be explored to demonstrate that the appealing properties of aptamers can still be applied for protein studies and detection. Aptamer molecules can be combined with antibodies, nanoparticles and signal amplification methods, such as PCR, for the development of novel probes and assay formats that will allow the detection of low levels of PDGF-BB. The use of magnetic nanoparticles immobilized with antibodies against PDGF-BB could be used to selectively isolate and enrich the PDGF-BB. Assuming different binding domains of the protein, aptamers that are flanked by PCR primer sequences could then be added and allowed to bind to PDGF-BB in buffer. Once washed, the aptamer can be released from the particle and real-time PCR performed to detect the original protein concentration. The high affinity and selectivity of antibodies and aptamers together with the amplification power and sensitivity of PCR will enable the detection the secreted PDGF-BB from cellular samples.

PAGE 133

LIST OF REFERENCES 1. Sidransky, D. Nature Reviews Cancer 2002, 2, 210-219. 2. Barnouin, K. Methods in Molecular Biology (Totowa, NJ, United States) 2004, 261, 479-497. 3. Narayanan, S. R. Journal of Chromatography 1994, 658, 237-258. 4. Tang, N.; Tornatore, P.; Weinberger, S. R. Mass Spectrometry Reviews 2003, 23, 34-44. 5. Gullberg, M.; Fredriksson, S.; Taussig, M.; Jarvius, J.; Gustafsdottir, S.; Landegren, U. Current Opinion in Biotechnology 2003, 14, 82-86. 6. Belford, D. A.; Rogers, M. L.; Francis, G. L.; Payne, C.; Ballard, F. J.; Goddard, C. Journal of endocrinology 1997, 154, 45-55. 7. Engvall, E.; Perlmann, P. Immunochemistry 1971, 8, 871-874. 8. Mukhopadhyay, R. Analytical Chemistry 2005, 77, 114A-118A. 9. Ellington, A. D.; Szostak, J. W. Nature 1990, 346, 818-822. 10. Lee, J. F.; Hesselberth, J. R.; Meyers, L. A.; Ellington, A. D. Nucl.Acids Res. 2004, 32, D95-100. 11. Ciesiolka, J.; Yarus, M. RNA 1996, 2, 785-793. 12. Nieuwlandt, D.; Wecker, M.; Gold, L. Biochemistry 1995, 34, 5651-5659. 13. Tuerk, C.; Gold, L. Science 1990, 249, 505-510. 14. Ringquist, S.; Jones, T.; Snyder, E. E.; Gibson, T.; Boni, I.; Gold, L. Biochemistry 1995, 34, 3640-3648. 15. Pan, W.; Craven, R. C.; Qiu, Q.; Wilson, C. B.; Wills, J. W.; Golovine, S.; Wang, J. F. Proceedings of the National Academy of Sciences of the United States of America 1995, 92, 11509-11513. 16. Morris, K. N.; Jensen, K. B.; Julin, C. M.; Weil, M.; Gold, L. Proceedings of the National Academy of Sciences of the United States of America 1998, 95, 2902-2907. 120

PAGE 134

121 17. Pavlov, V.; Shlyahovsky, B.; Willner, I. Journal of the American Chemical Society 2005, 127, 6522-6523. 18. Brody, E. N.; Gold, L. Reviews in Molecular Biotechnology 2000, 74, 5-13. 19. Lupold, S. E.; Hicke, B. J.; Lin, Y.; Coffey, D. S. Cancer research 2002, 62, 4029-4033. 20. Martell, R. E.; Nevins, J. R.; Sullenger, B. A. Molecular Therapy 2002, 6, 30-34. 21. Jayasena, S. D. Clinical Chemistry 1999, 45, 1628-1650. 22. Butler, V. P., Jr. Journal of Immunological Methods 1975, 7, 1-24. 23. Engvall, E.; Jonsson, K.; Perlmann, P. Biochimica et Biophysica Acta 1971, 251, 427-434. 24. Bock, L. C.; Griffin, L. C.; Latham, J. A.; Vermaas, E. H.; Toole, J. J. Nature 1992, 355, 564-566. 25. Potyrailo, R. A.; Conrad, R. C.; Ellington, A. D.; Hieftje, G. M. Analytical Chemistry 1998, 70, 3419-3425. 26. Dick, L. W., Jr.; McGown, L. B. Analytical Chemistry 2004, 76, 3037-3041. 27. Turney, K.; Drake, T. J.; Smith, J. E.; Tan, W.; Harrison, W. W. Rapid Communications in Mass Spectrometry 2004, 18, 2367-2374. 28. Deng, Q.; German, I.; Buchanan, D.; Kennedy, R. T. Analytical Chemistry 2001, 73, 5415-5421. 29. Robertson, M. P.; Ellington, A. D. Nucl.Acids Res. 2000, 28, 1751-1759. 30. Soukup, G. A.; Breaker, R. R. Current Opinion in Structural Biology 2000, 10, 318-325. 31. Tang, J.; Breaker, R. R. Nucl.Acids Res. 1998, 26, 4214-4221. 32. Savran, C. A.; Knudsen, S. M.; Ellington, A. D.; Manalis, S. R. Analytical Chemistry 2004, 76, 3194-3198. 33. Fang, X.; Cao, Z.; Beck, T.; Tan, W. Analytical Chemistry 2001, 73, 5752-5757. 34. Lakowicz, J. R. Principles of Fluorescent Spectroscopy NY, 1986; 2nd Edition. 35. Marras, S. A. E.; Kramer, F. R.; Tyagi, S. Nucl.Acids Res. 2002, 30, e122-1-e122/8. 36. Tyagi, S.; Kramer, F. R. Nature Biotechnology 1996, 14, 303-308.

PAGE 135

122 37. Li, J. J.; Fang, X.; Schuster, S. M.; Tan, W. Angewandte Chemie, International Edition 2000, 39, 1049-1052. 38. Schultze, P.; Macaya, R. F.; Feigon, J. Journal of Molecular Biology 1994, 235, 1532-1547. 39. Hamaguchi, N.; Ellington, A.; Stanton, M. Analytical Biochemistry 2001, 294, 126-131. 40. Li, J. J.; Fang, X.; Tan, W. Biochemical and Biophysical Research Communications 2002, 292, 31-40. 41. Bergsten, E.; Uutela, M.; Li, X.; Pietras, K.; Ostman, A.; Heldin, C. H.; Alitalo, K.; Eriksson, U. Nature Cell Biology 2001, 3, 512-516. 42. Heldin, C. H.; Westermark, B. Physiological Reviews 1999, 79, 1283-1316. 43. Li, X.; Ponten, A.; Aase, K.; Karlsson, L.; Abramsson, A.; Uutela, M.; Backstrom, G.; Hallstrom, M.; Bostrom, H.; Li, H.; Soriano, P.; Betsholtz, C.; Heldin, C. H.; Alitalo, K.; Ostman, A.; Eriksson, U. Nature Cell Biology 2000, 2, 302-309. 44. Oefner, C. ; D'Arcy, A. ; Winkler, F. K. ; Eggimann, B. ; Hosang, M. EMBO J 1992, 11, 3921. 45. Heldin, C. H.; Johnsson, A.; Wennergren, S.; Wernstedt, C.; Betsholtz, C.; Westermark, B. Nature 1986, 319, 511-514. 46. Peres, R.; Betsholtz, C.; Westermark, B.; Heldin, C. H. Cancer Research 1987, 47, 3425-3429. 47. Green, L. S.; Jellinek, D.; Jenison, R.; Oestman, A.; Heldin, C. H.; Janjic, N. Biochemistry 1996, 35, 14413-14424. 48. Fang, X.; Tan, W. Analytical Chemistry 1999, 71, 3101-3105. 49. Fang, X.; Liu, X.; Schuster, S.; Tan, W. Journal of the American Chemical Society 1999, 121, 2921-2922. 50. Bronzert, D. A.; Pantazis, P.; Antoniades, H. N.; Kasid, A.; Davidson, N.; Dickson, R. B.; Lippman, M. E. Proceedings of the National Academy of Sciences of the United States of America 1987, 84, 5763-5767. 51. Leitzel, K.; Bryce, W.; Tomita, J.; Manderino, G.; Tribby, I.; Thomason, A.; Billingsley, M.; Podczaski, E.; Harvey, H.; Bartholomew, M.; Cancer Research 1991, 51, 4149-4154. 52. Mayer, G.; Jenne, A. BioDrugs 2004, 18, 351-359. 53. Jacobson, A. B.; Zuker M. Journal of Molecular Biology 1993, 233, 261-269.

PAGE 136

123 54. Zuker, M. Nucl.Acids Res. 2003, 31, 3406-3415. 55. Moreira, B. G.; You, Y.; Behlke, M. A.; Owczarzy, R. Biochemical and Biophysical Research Communications 2005, 327, 473-484. 56. Yarmoluk, S. M.; Lukashov, S. S.; Ogul'chansky, T. Y.; Losytskyy, M. Y.; Kornyushyna, O. S. Biopolymers 2001, 62, 219-227. 57. Ueberfeld, J.; Walt, D. R. Analytical Chemistry 2004, 76, 947-952. 58. Haugland, R. P. The Handbook A Guide to Fluorescent Probes and Labeling Technologies; Molecular Probes, Inc: Eugene, OR, 2004; pp 7-9. 59. Tokuda, Y.; Satoh, Y.; Fujiyama, C.; Toda, S.; Sugihara, H.; Masaki, Z. BJU international 2003, 91, 716-720. 60. Ng, S. S. W.; MacPherson, G. R.; Gutschow, M.; Eger, K.; Figg, W. D. Clin Cancer Res 2004, 10, 4192-4197. 61. Bradford, M. M. Analytical Biochemistry 1971, 72, 248-254. 62. Vicns, M. C.; Sen, A.; Vanderlaan, A.; Drake, T. J.; Tan, W. ChemBioChem 2005, 6, 900-907. 63. Fang, X.; Sen, A.; Vicens, M.; Tan, W. ChemBioChem 2003, 4, 829-834. 64. Nutiu, R.; Li, Y. Chemistry--A European Journal 2004, 10, 1868-1876. 65. Gao, X.; Cui, Y.; Levenson, R. M.; Chung, L. W. K.; Nie, S. Nature Biotechnology 2004, 22, 969-976. 66. Fujimoto, K.; Shimizu, H.; Inouye, M. Journal of Organic Chemistry 2004, 69, 3271-3275. 67. Birks, J. B. Photophysics of Aromatic Molecules (Wiley Monographs in Chemical Physics); 1970; 704. 68. Winnik, F. M. Chemical Reviews, 1993, 93, 587-614. 69. Birks, J. B.; Lumb, M. D.; Munro, I. H. Proc.Roy.Soc.Ser.A 1964, 280, 289-297. 70. Birks, J. B.; Dyson, D. J.; Munro, I. H. Proc.Roy.Soc. 1963, 275, 575-588. 71. Dahim, M.; Mizuno, N. K.; Li, X. M.; Momsen, W. E.; Momsen, M. M.; Brockman, H. L. Biophysical Journal 2002, 83, 1511-1524. 72. Pagano, R. E.; Martin, O. C.; Kang, H. C.; Haugland, R. P. Journal of cell biology 1991, 113, 1267-1279.

PAGE 137

124 73. Stojanovic, M. N.; de Prada, P.; Landry, D. W. Journal of the American Chemical Society 2001, 123, 4928-4931. 74. Paborsky, L. R.; McCurdy, S. N.; Griffin, L. C.; Toole, J. J.; Leung, L. L. K. Journal of Biological Chemistry 1993, 268, 20808-20811. 75. Yamamoto, R.; Baba, T.; Kumar, P. K. R. Genes to Cells 2000, 5, 523. 76. Masuko, M.; Ohtani, H.; Ebata, K.; Shimadzu, A. Nucl.Acids Res. 1998, 26, 5409-5416. 77. Yang, C. J., Jockush, S., Vicns, M. C., Turro, N. J., and Tan, W. 2005. (Unpublished Work) 78. Nazarova, N.; Golovko, O.; Blauer, M.; Tuohimaa, P. The Journal of Steroid Biochemistry and Molecular Biology 2005, 94, 189-196. 79. Langerak, A. W.; De Laat, P. A. J. M.; Van der Linden-Van Beurfen; Delahaye, M.; Van der Kwast, T. H.; Hoogsteden, H. C.; Benner, R.; Versnel, M. A. Journal of Pathology 1996, 178, 151-160. 80. Bock, C.; Coleman, M.; Collins, B.; Davis, J.; Foulds, G.; Gold, L.; Greef, C.; Heil, J.; Heilig, J. S.; Hicke, B.; Hurst, M. N.; Husar, G. M.; Miller, D.; Ostroff, R.; Petach, H.; Schneider, D.; Vant-Hull, B.; Waugh, S.; Weiss, A.; Wilcox, S. K.; Zichi, D. Proteomics 2004, 4, 609-618. 81. Famulok, M.; Mayer, G.; Blind, M. Accounts of Chemical Research 2000, 33, 591-599. 82. Famulok, M.; Blind, M.; Mayer, G. Chemistry & Biology 2001, 8, 931-939. 83. Burgstaller, P.; Girod, A.; Blind, M. Drug Discovery Today 2002, 7, 1221-1228. 84. Burgstaller, P.; Jenne, A.; Blind, M. Current Opinion in Drug Discovery & Development 2002, 5, 690-700. 85. Fine, S. L.; Martin, D. F.; Kirkpatrick, P. Nature Reviews Drug Discovery 2005, 4, 187-188. 86. Gragoudas, E. S.; Adamis, A. P.; Cunningham, E. T., Jr.; Feinsod, M.; Guyer, D. R. New England Journal of Medicine 2004, 351, 2805-2816. 87. De Clercq, E.; Eckstein, F.; Merigan, T. C. Science 1969, 165, 1137-1139. 88. Wilson, D. S.; Szostak, J. W. Annual Review of Biochemistry 1999, 68, 611-647. 89. Burrell, M. M. Enzymes of Molecular Biology. [In: Methods Mol. Biol. (Totowa, N. J.), 1993; 16]; 1993; p 370.

PAGE 138

125 90. Mendonsa, S. D.; Bowser, M. T. Journal of the American Chemical Society 2004, 126, 20-21. 91. Mendonsa, S. D.; Bowser, M. T. Journal of the American Chemical Society 2005, 127, 9382-9383. 92. Mendonsa, S. D.; Bowser, M. T. Analytical Chemistry 2004, 76, 5387-5392. 93. Daniels, D. A.; Chen, H.; Hicke, B. J.; Swiderek, K. M.; Gold, L. Proceedings of the National Academy of Sciences of the United States of America 2003, 100, 15416-15421. 94. Wang, C.; Zhang, M.; Yang, G.; Zhang, D.; Ding, H.; Wang, H.; Fan, M.; Shen, B.; Shao, N. Journal of Biotechnology 2003, 102, 15-22. 95. Bayer, T. S.; Smolke, C. D. Nature Biotechnology 2005, 23, 337-343.

PAGE 139

BIOGRAPHICAL SKETCH Marie Carmen Vicns Contreras was born on March 11, 1976 in San Juan, Puerto Rico. She is the youngest of seven siblings and was raised by her parents, Francisco Vicns and Carmen Contreras in Caguas, P.R. Marie attended a private Catholic school, Colegio Catlico Notre Dame, for her secondary schooling. She then attended the University of Puerto Rico at Mayagez and received a Bachelor of Science degree in Chemistry in 1998 and a Master of Science degree in Chemistry in 1999. After that, she worked at Searle (now Pfizer) in Caguas, P.R. as a laboratory technician. To increase her opportunities in the chemical sciences, she moved to Gainesville, Florida in August of 2000 to pursue a Ph.D. in Bioanalytical Chemistry. Marie joined Dr. Weihong Tans research group in 2001. She will receive her Doctor of Philosophy degree in analytical chemistry from the University of Florida in August 2005. 126