Investigation of Conjugated Polyelectrolytes and Their Protein Sensing by Fluorescence Correlation Spectroscopy

MISSING IMAGE

Material Information

Title:
Investigation of Conjugated Polyelectrolytes and Their Protein Sensing by Fluorescence Correlation Spectroscopy
Physical Description:
1 online resource (180 p.)
Language:
english
Creator:
Wu, Danlu
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Chemistry
Committee Chair:
Schanze, Kirk S
Committee Members:
Castellano, Ronald K
Cao, Yun Wei
Powell, David H
Batich, Christopher D

Subjects

Subjects / Keywords:
biosensor -- conjugated -- correlation -- diffusion -- fluorescence -- fret -- polyelectrolytes -- quenching -- spectroscopy
Chemistry -- Dissertations, Academic -- UF
Genre:
Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Conjugated polyelectrolytes (CPEs) are water-soluble polymers characterized by p-conjugated backbones with ionic side groups. By possessing favorable optical properties, charge interaction ability and solvent-dependent self-assembly, this class of polymers has been studied for chemical and biological sensor applications. A DNA intercalator biotin-tetramethylrhodamine (biotin-TMR) is found to be capable of intercalating into a helical conjugated polyelectrolyte. Efficient fluorescence resonance energy transfer (FRET) from the polymer to the TMR chromophore is observed. It can be disrupted by mixing of biotin-TMR with avidin prior to the addition of the polymer. A discontinuous sensing strategy is developed for avidin with sensitivity as low as 100 pM. This project is further studied by a 543 nm laser coupled fluorescence correlation spectroscopy (FCS), which can provide insight regarding diffusion behavior and size change of molecules. A remarkable increase in the diffusion time of the poly-1/biotin-TMR complex in the presence of avidin is observed. This change is attributed to the formation of large supramolecular polymer aggregates, giving rise to a sensitive detection method for avidin with detection limit An FCS system coupled with 405 nm blue laser, which can directly track the diffusion of CPE, is successfully built up in house. A detailed construction of such type of FCS along with troubleshooting, optimizing and calibrating are fully described. This FCS setup is employed in the investigation on an array-based protein sensing project. A series of CPEs is developed as a probe array and exposed to seven proteins. The diversity in the final molecular sizes and diffusion times of the CPE probes is revealed through FCS. The chemometric linear discriminant analysis (LDA) is employed to differentiate different protein groups and identify unknown protein samples. The high discriminant/recognition accuracy (~93%) verified the feasibility of this new sensor array. In the last project, a CPE (mPPESO3py) containing meta-linked pyridine rings in the backbone is developed as a Pd2+ sensor. The polymer shows a great affinity and selectivity for the Pd2+ ion for that the quenching efficient of Pd2+ on polymer emission greatly surpasses that of the other metal ions. The FCS study discovers that aggregation of mPPESO3py is induced when several multivalent metal ions, except Pd2+, are added into the system. While shorter diffusion time is observed for CPE/Pd2+. The quenching mechanism of Pd2+ on mPPESO3py is dominated by charge transfer instead of aggregation.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Danlu Wu.
Thesis:
Thesis (Ph.D.)--University of Florida, 2012.
Local:
Adviser: Schanze, Kirk S.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-06-30

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044900:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 INVESTIGATION OF CONJUGATED POLYELECTROLYTES AND THEIR PROTEIN SENSING BY FLUORESCENCE CORRELATION SPECTROSCOPY By DANLU WU A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLME NT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Danlu Wu

PAGE 3

3 To my parents

PAGE 4

4 ACKNOWLEDGMENTS I would avail myself of this opportunity to express my heartfelt thanks to those people who have ever helped, supported and accompanied with me d uring my journey toward this important milestone in my life First of all, I would like to express the sincere appreciation to my advisor, Dr. Kirk Schanze for his expert valuable guidance and encouragement. His knowledgeability and enthusiasm to science stimulate my creativity and independence, making me enjoy wandering in the science wonderland I also would like to take this opportunity to record my sincere thanks to all the former and current members of the S chanze group. When I first started experiments Dr. Eunkyung Ji and Dr. Katsu Ogawa kindly taught me how to use instruments and how to design an analytical experiment. Dr. Chen Liao helped me to get used to work in the Schanze group. Dr. Fude Feng and Dr. Zhen Fang always shared their knowledge and experience of scientific research with me and helped me to overcome difficulties. Dr. Yan Chen from Dr. Weihong Tan s group kindly trained me on the fluorescence correlation spectroscopy and also shared her exper ience in instrument construction. Dr. Dongping Xie, Zhuo Chen and Zhenxing Pan helped me with sample characterization and instrument s using. I am also very grateful to Sile Hu from Department of Computer and Information Science and Engineering for being so supportive in LDA programming. It is also dedicated to Patrick Wieruszewski Randi Price and Russell Winkel for helping with my English writing In addition, thanks to Dr. Anand Parthasarathy Dr. Galyna Dubinina Subhadip Goswami Jie Yang and Shanshan Wang who have provide d help s and suggestion s for my Ph.D. study and research I would like to express my sense of gratitude to all my committee members: Dr. Ronald Castellano Dr. Christopher Batich Dr. Charles Cao and Dr. David Powell f or

PAGE 5

5 their time and helpful suggestions Thanks to Dr. Weihong Tan, Dr. Ronald K. Castellano and Dr. Christopher Batich for t h eir kindness in writing strong recommendation letter s for me. In addition, Dr. Christopher Batich who taught me much knowledge in polymer character ization when I spent my one year master study in Department of Materials Science and Engineering still gave me many helps after I transfer r ed to Department of Chemistry Finally, my family especially my parents to whom I am extremely grateful and indebt ed, always pr ovide s me the best for my life Without their unceasing support s and love s I would not stand here and be who I am today. Again, I place on record, my deepest gratitude to my parents for making my dream s come true

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF FIGURES ................................ ................................ ................................ ........ 10 LIST OF ABBREVIATIONS ................................ ................................ ........................... 16 ABSTRACT ................................ ................................ ................................ ................... 22 C H A PT ER 1 INTRODUCTION ................................ ................................ ................................ .... 24 Conjugated Polymers ................................ ................................ .............................. 24 Conjugated P olyelectrolytes (CPE) ................................ ................................ ......... 25 Fluorescence Quenching and Stern Volmer Plot ................................ .................... 27 Fluorescence Resonance Energy Transfer ................................ ............................. 31 Amplified Quenching and Molecular Wire Effect ................................ ..................... 32 Aggregation of Conjugated Polyelectrolytes ................................ ........................... 35 Environmental Effect on the C onformation of CPE ................................ ................. 39 Sens ing Assay Development of CPEs ................................ ................................ .... 43 Small Ions/Molecules Sensing ................................ ................................ .......... 43 Metal i on s ensing ................................ ................................ ....................... 44 Small m olecule s ensing ................................ ................................ ............. 46 DNA S ensing ................................ ................................ ................................ .... 48 Protein/ E nzyme A ctivity S ensing ................................ ................................ ...... 51 Protein s ensing by CP E b ased s ensor ................................ ....................... 52 Enzyme activity sensing ................................ ................................ ............. 55 Non specific I nteraction of CPEs ................................ ................................ ............ 60 Sensor Array ................................ ................................ ................................ ........... 61 Linear Discriminant Analysis ................................ ................................ ................... 64 Fluorescence Correlation Spectroscopy (FCS) ................................ ....................... 68 Basic Principle of FCS ................................ ................................ ...................... 68 Application of FCS ................................ ................................ ............................ 69 Overview of This Dissertation ................................ ................................ ................. 70 2 CONSTRUCTION OF A FLUORESCENCE CORRELATION SPECTROSCOPY .. 72 Theory of FCS ................................ ................................ ................................ ........ 72 Construction of FCS System ................................ ................................ ................... 77 Laser Optimization and Alig nment ................................ ................................ .... 77 Blue ( 405 nm ) d iode Laser ................................ ................................ ......... 77 Fiber c oupling s patial f ilter and f iberport ................................ .................... 79 Dichroic Mirror Cube and Laser Alignment ................................ ....................... 80 Objective Lens and Focusing ................................ ................................ ........... 82

PAGE 7

7 Signal Collection and Det ection ................................ ................................ ........ 84 Potential Interference ................................ ................................ ....................... 85 Calibration ................................ ................................ ................................ ........ 86 Examination of FCS System ................................ ................................ ................... 88 Summary ................................ ................................ ................................ ................ 92 Materials and Methods ................................ ................................ ............................ 93 Materials ................................ ................................ ................................ ........... 93 FCS Component and Measurement. ................................ ................................ 93 3 THE APPLICATION OF DYE LIGAND INTERCALATED HELICAL CONJUGATED POLYELECTROLYTE ON PROTEIN SENSING .......................... 95 Background ................................ ................................ ................................ ............. 95 Results and Discussion ................................ ................................ ........................... 96 FRET Study of Helical CPE/dye l igand with Protein ................................ ......... 96 Photophysical properties of Poly 1 and biotin TMR ................................ ... 96 FRET from Poly 1 to TMR ................................ ................................ .......... 98 Addition of avidin to Poly 1/biotin TMR ................................ .................... 100 Titration of preformed avidin/biotin TMR to Poly 1 ................................ ... 101 FCS Study on the Poly 1/biotin TMR/avidin System ................................ ...... 103 Diffusion behavior of four types of molecules/complex ............................ 105 Mechanism for formation of supramolecular aggregation ........................ 10 8 Control experiment ................................ ................................ ................... 109 Avidin sensing strategy ................................ ................................ ............ 113 Conclusion ................................ ................................ ................................ ............ 115 Experiments and Materials ................................ ................................ ................... 116 Materials ................................ ................................ ................................ ......... 116 Preparation of Poly 1 / biotin TMR complex ................................ ..................... 117 Negative Control Experiment by Using BSA ................................ ................... 117 Instrumentation ................................ ................................ .............................. 118 FCS measurement ................................ ................................ ................... 118 Fluorescence s pectroscopy. ................................ ................................ .... 118 UV Vis measurement ................................ ................................ ............... 118 4 STUDY OF CONFORMATION CHANGE OF CPES INDUCED BY PROTEINS AND DEVELOPMENT OF SENSOR ARRAY FOR PROTEINS BY FCS ............. 119 Background ................................ ................................ ................................ ........... 119 Results and Discussion ................................ ................................ ......................... 121 Properties of Six CPEs and Seven Proteins ................................ ................... 121 FCS Results and Discussion ................................ ................................ .......... 124 Protein Sensing ................................ ................................ .............................. 130 Linear discriminant analysis of FCS diffusion times for protein/CPE mixtures ................................ ................................ ................................ 130 Unknown sample test ................................ ................................ ............... 134 Summary ................................ ................................ ................................ .............. 137 Material and Experiment ................................ ................................ ....................... 138

PAGE 8

8 Materials ................................ ................................ ................................ ......... 138 Bradford Protein Assay Procedure ................................ ................................ 139 Instrumentation ................................ ................................ ............................... 140 FCS m easurement ................................ ................................ .................. 140 UV Vis measurement ................................ ................................ ............... 140 5 STUDY OF INTERACTION OF META LINKED POLY(PHENYLENE ETHYNYLENE) SULFONATE CONTAINING PYRIDINE WITH METAL IONS .... 141 Background ................................ ................................ ................................ ........... 141 Results and Discussion ................................ ................................ ......................... 143 Photophysical Properties of mPPESO 3 py ................................ ...................... 143 Solvent induced photophysics change ................................ ..................... 143 Palladium ion induced photophysics change ................................ ........... 146 Photophysic s change of mPPESO 3 py with various metal ions ................ 148 FCS S tudy on the mPPESO 3 py with Various M etal I ons ................................ 150 Summary ................................ ................................ ................................ .............. 153 Experimental ................................ ................................ ................................ ......... 154 Materials ................................ ................................ ................................ ......... 154 Instrumentation ................................ ................................ ............................... 154 Absorption and Emission measurement ................................ .................. 154 FCS measurements ................................ ................................ ................. 155 6 CONCLUSION ................................ ................................ ................................ ...... 156 The Application of Dye Ligand Intercalated Helical Conjugated Polyelectrolyte to Protein Sensing ................................ ................................ ............................. 156 New Fluorescence Correlation Spectroscopy and Application on Protein Sensor Array Development ................................ ................................ ............................ 157 Study o f m eta Linked Poly(Phenylene Ethynylene) Sulfonate Containing Pyridine Quenched b y Metal Ions ................................ ................................ ...... 158 Outlook for Application of FCS in CPE ................................ ................................ .. 158 A P P E N D I X A TABLES OF TRANING DATA ................................ ................................ .............. 160 B TABLE OF TEST DATA ................................ ................................ ........................ 164 LIST OF REFERENCES ................................ ................................ ............................. 166 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 180

PAGE 9

9 LIST OF TABLES Table page 3 1 Photophysical properties of TMR Poly 1 and biotin TMR in 10 mM phosphate buffer solution (pH 7.4) ................................ ................................ .... 98 3 2 Diffusion data for four species in phosphate buffer (10 mM, pH 7.4) ............... 106 4 1 Basic information of CPEs ................................ ................................ ............... 122 4 2 Basic information of proteins ................................ ................................ ........... 124 4 3 Eigenvalues with their percentage of each LDA operation training matrix ....... 133 A 1 Training matrix of Log ( d / 0 ) of six CPE sensor array (P1 P6) against seven proteins ................................ ................................ ................................ 160 B 1 Unknown sample test matrix of Log ( d / 0 ) of 6 CPE sensor array (P1 P6 ) against various proteins ................................ ................................ ................... 164

PAGE 10

10 LIST OF FIGURES Figure page 1 1 Structures of commonly used conjugated polymers. ................................ .......... 25 1 2 Normalized absorption and emission spectra of perylene in benzene with corresponding electronic energy diagram. ................................ .......................... 27 1 3 The pH dependent absorption and emission spectra of fluoresce in. .................. 28 1 4 Mechanism and Stern Volmer plot of A) collisional quenching and B) static quenching (right). ................................ ................................ ................................ 30 1 5 Schematic representation of the FRET spectral overlap integral and energy diagram. ................................ ................................ ................................ ............. 31 1 6 Schematic illustration of amplified fluorescence quenching of a conjugated polymer by MV 2+ ................................ ................................ ................................ 32 1 7 Illustration of the molecular wire. ................................ ................................ ........ 33 1 8 Absorption and fluorescence spectra of MPS PPV in water in the presence and absence of MV 2+ ................................ ................................ ......................... 34 1 9 Structure of some polymers that used in the text. ................................ ............... 34 1 10 Stern Volmer plots of a series of PPE CO 2 with [MV + ]. ................................ ...... 35 1 11 A) Absorption and fluorescence spectra of PPE NH 3 with increasing [PPi]. B) S tructure of PPE d NH 3 and PPi. ................................ ................................ ......... 36 1 12 Illustration of amplified quenching by oppositely charged quencher in A ) non aggregated CPE B ) aggregated CPE. ................................ ................................ 37 1 13 Structure of MV 2+ and MBL PPV and Stern Volmer plot for MBL PPV quenched by MV 2+ ................................ ................................ ............................. 38 1 14 Quenching of 10 M PPE CO 2 emission by MV 2+ in water ( ) and in methanol with 0 M ( ), 2.5 M ( ), 5.0 M ( ), 7.5 M ( ) CaCl 2 ................................ ................................ ................................ ................. 38 1 15 Absorption (left) and fluorescence (right) spectra of PPE SO 3 in MeOH ( ), (1:1) H 2 O MeOH ( -), and H 2 O ( ). ................................ ................................ .. 40 1 16 Solvent effect on the aggregation state of CPE. ................................ ................. 40

PAGE 11

11 1 17 filling model showing the conformational equilibrium for OPE of length n = 18. ..................... 42 1 18 A) Normalized emission spectra of OPE in various solvents. Insert: absorption spectra B) Structure of OPE and cartoon for coi l helix transition process. ................................ ................................ ................................ .............. 42 1 19 Schematic illustration of Hg 2+ ind uced agglutination of the h PPE CO 2 /papain complex. ................................ ................................ ................................ 44 1 20 A) Sensi ng strategy for aptamer mediated CPE based K + sensor. B) A bsorption spectra and visible colors of polythiophene mix ed with different ions. ................................ ................................ ................................ .................... 46 1 21 A) I nteraction between p BV 2+ and sugar B) Fluorescence recovery titration curves against three sugars. ................................ ................................ ............... 47 1 22 Photographs, possible structure and UV V is absorption spectra changes for the PT based DNA sensor ................................ ................................ .................. 48 1 23 Schematic illustration of DNA detection by CPE labeled molecular beacon. ...... 50 1 24 Chemical structures of quenchers or ligands ................................ ..................... 53 1 25 tether ... 53 1 26 Schematic description of the specific detection of human thrombin by use of ssDNA thrombin aptamer and cationic polymer ................................ ............. 54 1 27 based sensors ........................ 55 1 28 A) Structures of polymer, BpPPESO 3 and substrate, 10CPC, and reaction scheme for hydrolysis of 10CPC by PLC. B) Mechanism of PLC turn off assay. ................................ ................................ ................................ ................. 59 1 29 Mechanism of ALP turn off assay and p hotographs of solutions illuminated with near UV light illustrat ing the polymer fluorescence under the different conditions of the assay. ................................ ................................ ...................... 60 1 30 Schematic presentation of the cell detection assay by CPEs including signal response pattern and canonic score plot. ................................ ........................... 63 2 1 Working principle for FCS ................................ ................................ .................. 73 2 2 The excitation volume in A) Z direction and B) X Y planar as generated by a diffraction limited objective lens. ................................ ................................ ......... 74 2 3 evelopment of an autocorrelation curve. ................................ ............................. 74

PAGE 12

12 2 4 Ellipsoid like excitation volume formed by the objective lens. ............................ 75 2 5 Schematic diagram of the FCS setup described in the text. ............................... 78 2 6 Photo graphs for A) excitation part of FCS setup and B) fiber coupling spatial filter ................................ ................................ ................................ .................... 79 2 7 Cross section s for laser beam in each optimization step. ................................ ... 80 2 8 A) Inner structure of fluorescence microscope and B) photo graph for a part of the setup. ................................ ................................ ................................ ............ 81 2 9 Diagram of the dichroic mirror cube. ................................ ................................ ... 81 2 1 0 Effect of objective media A) oil immersion objective B) water immersion objective ................................ ................................ ................................ ............ 83 2 1 1 Photo graphs for signal detection and correlation components of FCS setup. .... 84 2 1 2 Confocal volume and optics. ................................ ................................ ............... 85 2 1 3 Black cover and breadboard for FCS setup. ................................ ....................... 86 2 14 Structure of molecules that used in calibration and their acronyms. ................... 88 2 15 Correlation curves for TMR in water. ................................ ................................ .. 89 2 1 6 Effect of molecular weight (MW) on the correlation curves of molecules A) using 590 nm emission filter B) using 500 n m emission filter. ........................... 91 2 1 7 Plot of diffusion coefficient of f our standard samples as a function of their molecular weight ................................ ................................ ................................ 92 3 1 Structure of polymers and dye ligand compound. ................................ .............. 97 3 2 Absorption (Abs) and emission (E m) spectra of Poly 1 and biotin TMR. ............ 97 3 3 Normalized emission spectra for titration of 0 0.3 M biotin TMR into 15 M P oly 1 in 1 mM phosphate buffer solution, pH = 7.4. ................................ .......... 98 3 4 ex = 320 nm) of biotin TMR (2 25 nM) in the absence ( ) and presence ( --) of Poly buffer. ................................ ................................ ................................ ................. 99 3 5 Fluorescence anisotropy spectra for biotin TMR with and without Poly 1. ......... 99 3 6 Fluorescence spectra of Poly 1 solution ( ) upon addition of biotin and avidin ( --). ................................ ................................ ................................ 100

PAGE 13

13 3 7 Normalized emission spectra for Poly 1 mixed with preformed avidin/biotin TMR at various [avidin]/[biotin TMR]. ................................ ............................... 101 3 8 R atio of intensities at 590 nm and 450 nm after addition of pre mixed biotin avidin complex at various avidin concentrations in phosphate buffer ............... 102 3 9 B inding of preformed avidin/biotin TMR complex to Poly 1 as a functio n of added avidin concentration. ................................ ................................ .............. 103 3 10 Basic setup for FCS with 543 nm Laser. ................................ .......................... 104 3 11 Background detection for A) phosphate buffer (10 mM, pH 7.4) B ) P oly 1 ( 1 M) in phosphate buffer (10 mM, pH 7.4) ................................ ........................ 104 3 12 Normalized correla tion functions of biotin TMR P oly 1/ biotin TMR biotin TMR /avidin, and Poly 1 / biotin TMR /avidin ................................ ....................... 105 3 13 Normalized correlation functions of biotin TMR Poly 2/biotin TMR Poly 2/biotin T MR /avidin and biotin TMR /avidin ................................ ....................... 106 3 14 Photon counting rate (fluorescence fluctuation) during the detection time (1200 s) for P oly 1 / biotin TMR and Poly 1 / biotin TMR /avidin ........................... 107 3 15 Proposed m echanism of protein induced aggregation ................................ .... 108 3 16 AFM images for A) pure Poly 1 B) Poly 1/biotin TMR C) Poly 1/biotin TMR/ avidin Line scans for D ) pure Poly 1 E ) Poly 1/biotin TMR F ) Poly 1/biotin TMR/ avidin ................................ ................................ ........................... 109 3 17 Normalized correlation functions of biotin TMR Poly 1/biotin TMR and Poly 1/bi otin TMR/ avidin ................................ ................................ ........................... 110 3 1 8 Normalized correlation functions for ligand fr ee TMR with and without avidin .. 111 3 1 9 Normalized correlation curves for Poly 1/TMR and Poly 1/TMR/avidin in 10mM phosphate buffer A ) pH 7.4 B ) pH 10.5 ................................ .................. 112 3 20 AFM images for A) Poly 1/avidin and B) Poly 1/TMR/avidin Line scans for C) Poly 1/avidin and D) Poly 1/TMR/avidin. ................................ .......................... 112 3 21 Normalized correlation functions for biotin TMR Poly 1/biotin TMR and Poly 1/biotin TMR /BSA ................................ ................................ ............................. 113 3 22 Normalized correlation functions for biotin TMR Poly 1/biotin TMR and Poly 1/biotin TMR /avidin ................................ ................................ .................. 114 3 23 Au tocorrelation FCS curves for Poly 1/biotin TMR, Poly 1/TMR/avidin and Poly 1/biotin TMR/avidin ................................ ................................ .................. 115

PAGE 14

14 4 1 Chemical structures of six CPEs. ................................ ................................ ..... 122 4 2 Normalized absorption and emission spectra for six CPEs. ............................. 123 4 3 FCS curves for P4 without protein ( ) and with a vidin ( ) LYZ ( )HRP ( ) ) in 5 mM HEPEs buffer pH 7.2. ........... 125 4 4 F luctuation profile s for A) P4 without and with seven proteins and B) their enlargement ................................ ................................ ................................ ..... 126 4 5 Log ( d / 0 ) response 2D bar pattern of six CPEs mixing with seven proteins. 127 4 6 Log ( d / 0 ) response 3D column pattern of six CPEs mixing with seven proteins. ................................ ................................ ................................ ........... 128 4 7 Flowchart for protein sensor array development. ................................ .............. 130 4 8 Flowchart of multiple LDA operation for training known samples. .................... 132 4 9 LDA discriminant spaces/ plots for the diffusion time response patterns ........... 133 4 10 Training results for multiple LDA operation of diffusion time response for six CPE probes against seven proteins ................................ ................................ 134 4 11 Flowchart for multiple LDA operation for testing unknown samples. ................ 135 4 1 2 Test results for multiple LDA operation of diffusion time response for six CPE probes against forty two unknown protein samples. ................................ ......... 136 5 1 Structure of mPPESO 3 py ................................ ................................ ................. 142 5 2 A) A bsorption spectra for mPPESO 3 py i n solvent mixture with different component volume ratio. B) Ratiometric plot of A L /A S versus t he percentage of water in methanol ................................ ................................ ....................... 144 5 3 A) Mechanism of solvent and metal ion induced formation of helical structures. B) Mechanism of multi valence metal ion induced crosslinking of CPE. ................................ ................................ ................................ ................. 145 5 4 A) E mission spectra for mPPESO 3 py in solvent mixture with different component volume ratio. B) A bsorbance ratio of two bands versus the percentage of water in MeOH ................................ ................................ .......... 146 5 5 A) A bsorption spectra of 15 M mPPESO 3 py titrated with Pd 2+ B) T he a bsorbance ratio of two bands versus the concentration of Pd 2+ in aqueous solution ................................ ................................ ................................ ............ 147

PAGE 15

15 5 6 Emission spectra for 15 M mPPESO 3 py with various [Pd 2+ ] in aqueous solution ................................ ................................ ................................ ........... 148 5 7 Stern Volmer plots for mPPESO 3 py with various metal ions in water .............. 149 5 8 Stern Volmer constant for various metal ions in water. ................................ .... 149 5 9 Bar graph for fluorescence quenching ratios I 0 /I at 680 nm of mPPESO 3 py with different metal ions at various concentration s in aqueous solution ......... 150 5 10 A) FCS correlation curves and B ) fluctuation profiles for 15 M mPPESO 3 py without (red) and with 40 M Pd 2+ (green) or Cr 3+ (blue) ................................ 151 5 11 Diffusion time ratio for mPPESO 3 py with different amount of Pd 2+ measured for 30 min in aqueous solution. ................................ ................................ ........ 152 5 1 2 Diffusion time ratio of mPPESO 3 py before and after the addition of metal ions. ................................ ................................ ................................ ................. 153

PAGE 16

16 LIST OF ABBREVIATION S S pecific gravity V iscosity of the solvent Average fluorescence intensity 10C PC P hosphatidylcholine 2D Two dimension 3D Three dimension A A bsorbance Abs Absorption ADP A denosine di phosphate AFM A tomic force microscopy A L A bsorbanc e at the longer wavelength band ALP A lkaline phosphatase AMP A denosine mono phosphate APD S in gle photon counting avalanche photodiode A S A bsorbance at the shorter wavelength band ATP A denosine triphosphate B iotin TMR B iotinylated rhodamine, 5 (and 6) tetramethylrhodamine biocytin BpPPESO 3 Biphenyl sulfonated poly(para phenylene ethynylene) BSA B ovine serum albumin Bz FVR pNA N benzoyl P he Val Arg p nitroanilide hydrochloride hydrate C Concentration CaM C almodulin CNC C harge neutral complex CO 2 C arboxylate

PAGE 17

17 CP C onjugated polymer CPE C onjugated p olyelectrolyte Cyt c Cytochrome c D Diffusion coefficient Dabcyl 4 (4 ( D imethylamino) phenylazo) benzoic acid DAG Degraded product of 10CPC d laser D iameter of laser beam DLS D ynamic light scattering DMSO D imethylsulfoxide DNA Deoxyribonucleic acid d pupille D iameter of lens pupil dsDNA D ouble s tranded DNA EB E thidium bromide ELISA E nzyme linked immunosorbent assay Em Emission ET Energy transfer F(t) F luorescence intensity at time FCS F luorescence correlation spe ctroscopy FRET F luorescence resonance energy transfer FWHM F ull width half maximum G( ) Autocorrelation function GFP Green fluorescence protein GOx G lucose oxidase from Aspergillus niger HEPES 4 (2 Hydroxyethyl) 1 piperazineethanesulfonic acid

PAGE 18

18 HK3 H exokinase from saccharomyces cer evisiae, type III h PPE CO 2 h omo C arboxylate substituted poly(phenylene eth yn ylene) HRP P eroxidase from horseradish type I I F luorescence intensity with quencher I 0 F luorescence intensity without quencher ICP MS I nductively coupled plasma mass spectrometry k Boltzmann s constant K assoc A ssociation constant for formation of the fluorophore quencher complex K b Binding constant K D D ynamic constant K pNA L Lys p nitroanilide dihydrobromide k q B imolecular quenching rate constant K SV Stern Volmer constant LDA L inear discrimination analysis LED L ight emitting diode LYZ Lysozyme MB M olecular beacon MeOH Methanol mPPE SO 3 py S ulfonated PPE containing meta linked pyridine rings in the backbone M PS PPV Sulfonated poly(phenylene vinylene) MV + M onovalence methyl viologen MV 2+ M ethyl viologen MW M olecular weight MWCO M olecular weight cutoff

PAGE 19

19 N Number of fluorophore in confocal volume NA N umerical aperture of objective lens NR 3 + Q uarternary ammonium OPE O ligo(phenylene ethy nylene) P1 P oly(para phenylene ethynylene) with p olyeth yn ylene glycol and poly dendric ammonia side chains P2 Sulfonated meta linked poly(phenylene ethynylene) P3 Sulfonated para linked poly(poly(phenylene ethynylene) P4 C ationic poly( thiophene phenylen e ) with bis alkylammonium side groups P5 C ationic poly( thiophene phenylene ) with quaternary ammonia side chains P6 C ationic P oly(para phenylene ethynylene) with bis alkylammonium side groups PA P olyacetylene PANI P olyanaline p BV 2+ B oronic acid functi onalized benzyl viologen PE CO 2 O ligomer carboxylate d para phenylene ethynylene PEG P olyeth yn ylene glycol PET Photoinduced electron transfer PF P olyfluorene PF P CO 2 C arboxylate d p oly(fluorene co phenylene) p I I soelectrical point PLC P hospholipase C PLD2 P hospholipase D from arachis h ypogaea (peanut), type II p NA p nitroanilide

PAGE 20

20 PO 3 P hosphonate P oly 1 Sulfonated meta linked poly(phenylene ethynylene) Poly 2 Sulfonated para linked poly(poly(phenylene ethynylene) PPE P oly(para phenylene ethynylene) PPE CO 2 C arboxylate d poly(para phenylene ethynylene) PPE d NH 3 C ationic poly(para phenylene ethynylene) with dendric ammonia side chains PPE PEG d CO 2 P oly(para phenylene ethynylene) with p olyeth yn ylene glycol and poly dendric ammonia side chains PPE PEG d CO DNA DNA conjugated PPE PEG d CO 2 PPE PO 3 P hosphonate substituted poly(para phen ylene ethynylene) PPE SO 3 Sulfonated poly(para phenylene ethynylene) PPi P yrophosphate PPP P oly(para phenylene) PPV P oly(para phenylene vinylene) PPy P olypyrrole PRU R epeat unit concentration PT P olythiophene Q Quencher QTL Q uencher tether ligand R Radius of a sphere R H H ydrodynamic radius Rho Arg Emissive p eptide derivative rhodamine Rho Arg 2 Non emissive p eptide derivative rhodamine Ru(bpy) 2 (dppz) 2+ Ruthenium complex (bpy = 2,2' bipyridine and dppz = dipyrido[3,2 a:2',3' c] phenazine)

PAGE 21

21 Ru(ph en) 3 4 R uthenium complex ((phen)=4,7 bis(4 sulfophenyl) 1,10 phenanthroline) SM Single mode SNP S ingle nucleotide polymorphism SO 3 S ulfonate ssDNA S ingle strand DNA SV Stern Volmer T T emperature TCSPC Time Correlated Single Photon Counting TMR Tetramethylrhodamine UV Vis Ultra violet and vis ible V V olume of molecule V eff E ffective detection volume ex Excitation wavelength F(t) Deviation of the fluoresce nce intensity from at time t M olar absorptivity or extinction coefficient fl Quantum yield max Wavelength of maximum emission peak D ecay lifetime of fluorophore 0 Initial diffusion time D Diffusion time of a fluorophore d 0 Diffusion t ime of pure CPE Structure parameter r T ransversal or waist radius of confocal volume z L ongitudinal radius of confocal volume

PAGE 22

22 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial F ulfillment of the Requirements for the Degree of Doctor of Philosophy INVESTIGATION OF CONJUGATED POLYELECTROLYTES AND THEIR PROTEIN SENSING BY FLUORESCENCE CORRELATION SPECTROSCOPY By Danlu Wu December 2012 Chair: Kirk S. Schanze Major: Chemistry Conj ugated polyelectrolytes (CPEs) are water soluble polymers characterized by conjugated backbones with ionic side groups. By possessing favorable optical properties, charge interaction ability and solvent dependent self assembly, this class of polymers has been studied for chemical and biological sensor applications. A DNA interc alator biotin tetramethylrhodamine (biotin TMR) i s found to be capable of intercalating into a hel ical conjugated polyelectrolyte Efficient fluorescence resonance energy transfer (FRET) from the polymer to the TMR chromophore i s observed It can be disrup ted by mixing of biotin TMR with avidin prior to the addition of the polymer. A dis continuous sensing strategy i s developed for avidin with sensitivity as low as 100 pM. This project i s further studied by a 543 nm laser coupled fluorescence correlation spe ctroscopy (FCS) which can provide insight regarding diffusion behavior and size change of molecules A remarkable increase in the diffusion time of the poly 1/biotin TMR complex in the presence of avidin i s observed. This change is attributed to the forma tion of large sup ramolecular polymer aggregates giving rise to a sensitive detection method for avidi n with detection limit < 100 pM

PAGE 23

23 A n FCS system coupled with 405 nm blue laser, which can directly track the diffusion of CPE, i s successfully built up in house. A detailed construction of such type of FCS along with troubleshooting optimizing and calibrating are fully described. This FCS setup i s employed in the investigation on an array based protein sensing project. A series of CPEs i s developed as a pr obe array and exposed to seven proteins. The diversity in the final molecular size s and diffusion time s of the CPE probes i s revealed through FCS. The chemometric linear discriminant analysis (LDA) i s employed to differentiate different protein group s and identify unknown protein samples. The high discriminant/recognition accuracy (~93%) verified the feasibility of this new sensor array. In the last project, a CPE (mPPESO 3 py) containing meta linked pyridine rings in the backbone i s developed as a Pd 2+ senso r. The polymer show s a great affinity and selectivity for the Pd 2+ ion for that the quenching efficient of Pd 2+ on polymer emission greatly surpasses that of the other metal ions. The FCS study discovers that aggregation of mPPESO 3 py i s induced when severa l multivalent metal ions, except Pd 2+ are added into the system While shorter diffusion time i s observed for CPE/Pd 2+ The quenching mechanism of Pd 2+ on mPPESO 3 py is dominated by charge transfer instead of aggregation.

PAGE 24

24 CHAPTER 1 INTRODUCTION Conjugated Polymers In chemistry, the conjugation means an overlap of one orbital with another across an intervening sigma bond (in larger atoms d orbitals can be involved). 1 A conjugated system is a system comprising a region of overlapping orbital bridged with interjacent single bonds. The involved molecules generally have lower overall energy and higher stability. Those molecules may be cyclic, acyclic, linear or mixed, allowing their electrons delocalize across all the adjacent aligned orbitals. 2 Among the large number of conjugated systems, conjugated polymers (CPs) with altern ating sin gle and multiple bonds have attracted an overwhelming interest in laboratories around the world. In 1967, the first conjugated polymer polyacetylene (PA, Figure 1 1 3 launching the investigation on this novel conductive material. Various conjugated polymers have been generated, such as poly(para phenylene) (PPP), 4 poly(para phenylene vin ylene) (PPV), 5 poly(para phenylene ethynylen e) (PPE), 6 polythiophene (PT), 7 polypyrrole (PPy), 8 polyan i line (PANI) 9 and polyfluorene (PF) 10 ( Figure 1 1 ). One of the milestones is that in 2000, Hideki Shi rakawa, Alan MacDiarmid ( University of Pennsylvania ) and Alan Heeger ( University of California at Santa Barb a ra ) were awarded the Nobel Prize for the discovery of conducting polyacetylene. 11 By possessing various well designed functional groups, the CPs have favorable electrical, optical or magnetic properties, which offer them great potential in application s including light emitting diodes (LED), 5, 12 field effect transistors, 13, 14 solar cells 15 and chemosensors. 16 Among those application s chemosensin g is experiencing a rapid growth in the pas t few

PAGE 25

25 years. A large part of this is driven by the need in the fields of medical diagnostics, environmental monitoring and toxicological analysis. Generally speaking, a sensor in analytical chemistry is commonly d efined to be a chemical indicator that produces a signal indicative of the presence of a target analyte. 17 Consequently, CP has been demonstrated to be a versatile chemosensor, since their color (colorimetric), 18 emission (fluorometric), 16 conductivity ( conductometric ) 19 or redox potential ( potentiometric ) 20 changes upon analyte binding. The diversity or variety in the CP sensor type is a result of amplified signal response due to the efficient coupling between optoelectronic segments 21, 22 and the rapid transport of electronic excitations governed by the nature of conjugation. 16 Figure 1 1 Structures of commonly used conjugated polymers. Conjugated P olyelectrolytes (CPE) Conjugated p olyelectrolytes (CPEs) are water soluble conjugated polymers characterized by conjugated backbones with various ionic side groups, such as

PAGE 26

26 sulfonate (SO 3 ) 23, 24 carboxylate (CO 2 ), 25 phosphonate (PO 3 ) 26 and quarternary ammonium (NR 3 + ) 27, 28 either branched 25, 28 or linear. 23, 24, 27, 29 CPEs possess not only attractive electrical or optical properties, but also charge interaction ability and sol vent dependent self assembly T hese features satisfy the requirement of sensor appl ied in the biological system or aqueous solutions. The intrinsic amphiphilic nature of CPEs offers them various conformations in aqueous solution. For example, the meta lin ed PPE with more than 4 repeat units tends to form helical structure in aqueous solution. 23, 30 para Linked PPEs with linear ionic groups undergo spontaneous aggregation in water 24, 29, 31 while some CPEs with high charge density, e.g., with branched ionic side chains, re main single polymer chain in water. The charge and structure nature of CPE facilitates the ability to control the distance and the strength of interaction be tween CPE and other ionic species. For instance, a rhodamine labeled ligand compound with net positive charge was found to intercalate into the helix of an anionic meta linked PPE, forming a ligand functionalized complex. 30 Pyrophosphate (PPi) was demonstrated to induce aggregation of a monodisperse cationic poly(phenylene ethynylene) with den d ri ti c polyamine side chains through binding betw een the PPi and amine ligands in water. 28 CPEs are able to form complexes with oppositely charge d surfactants via Coulombic attraction, which result s in dramatic and tunable changes in both geometric conformation and optical properties of CPEs. 32 It was also found from the optical aspect that proteins can modify the s 33, 34 The change in the chemical or physical properties of CPE s induced by other species provides a platform for sensor applications of th ese material s

PAGE 27

27 Figure 1 2 Normalized absorption and emission spectra of perylen e in benzene with corresp onding electronic energy diagram. A dapted wi th p er m issi on from Lakowicz, J. R 35 Fluorescence Quenching and Stern Volmer Plot Fluorescence is the light emitted by a fluorophore after absorbing energy from excitation electromagnetic radiation, e.g., light. It occurs as the molecu les relax from higher electronic singlet states. At each of the electronic energy levels there are several vibrational energy levels. The fluorophores can stay in a lower vibrational energy level through vibration al relaxation. 35 A typical absorption and emission spectra o f an organic fluorophore with corresponding energy level diagram, is shown in Figure 1 2 The excited molecules usually relax to a lower vibrational state so the ir fluorescence bands are shifted to wavelengths that are longer than the original band This shift is termed the S tokes shift. A large S tokes shift can help to separate the excitation light and emission light. Usually, the average time a molecule spends in the excited state prior to return to the ground state is called lifetime ( ) The lifetime can be detected through a t ime c orrelated s ingle p hoton c ounting (TCSPC) coupled fluorescence lifetime spectrometry and it has been used to analyze the physical state of the fluorophore as well as related

PAGE 28

28 chemical reaction. 35 The way to characterize the em ission efficiency of a fluorophore is to measure its quantum yield ( fl ) which is the number of emitted photons relative to the number of absorbed photons. 35 The absorption and emission spectra of some species change upon the change of surrounding chemical environments. F or instance, the spectra of fluorescein, which is a synthetic organic compound available as a dark orange/red powder soluble in water, is pH sensitive over the range of 5 to 9 as shown in Figure 1 3 36 By making use of this property, its absorption and emission efficiency at a certain wavelength can be ca refully tuned to achieve some particular purpose which is applied in Chapter 2 Figure 1 3 The pH dependen t absorption and emission spectra of fluoresce in. R e p ri n te d w ith pe r mi ssion from Invitrogen 36 F luorescence quenching, which refers to any process (fluorescence resonance energy transfer, electron transfer, intersystem crossing due to heavy atom effect, etc.) that reduces the fluorescence intensity of a fluorophore, is an important process that has been well utilized in many applications, especially in biosensor s Quenching can occur from many diffe rent mechanisms, among which, two mechanisms dynamic and static quenching are quite important and commonly seen. Dynamic quenching, also termed collisional quenching, occurs when the excited state fluorophore is deactivated upon colliding with some other m olecule i.e., quencher, in solution It is a diffusion

PAGE 29

29 limited process since the quencher must come into contact with the flu o rophore obtain ing the energy of emitting photon within the lifetime of fluorophore at excited state. T h is results in the fluorop hore return ing to the ground state wit hout emission of a photon. S tatic quenching occurs due to formation of a non fluorescent ground state complex by the fluorophore and the quencher. Quenching occurs in the ground state (Figure 1 4 ). 35 To qualify or quantify a quenching process, a Stern Volmer (SV) plot and its slope, named Stern Volmer constant is normally employed. The SV equation is expressed as follow s, (1 1) w here and are the fluorescence intensity without and with que ncher at a specific wavelength, usually the emission maximum of the fluorophore is the quencher concentration. In the simplest case, the plot is a straight line that has a n intercept at with i.e., In the situati on of dynamic quenching mechanism, where is the bimolecula r quenching rate constant (unit ) and is the decay li fetime of the fluorophore (unit s). is so called dynamic constant (unit ) In t he other case of static quenching mechanism, where the latter term is the association constant for formation of the fluorophore quencher complex. There is a simple way to distinguish dynamic and static quenching process by obse rving the slope change of SV plot upon increasing experimental temperature. Due to the diffusion dependent property of dynamic quenching, an increas e in temperature will enhance quenching the higher temperature the more collision occurs the thus increases. While in the other side, higher temperature will cause dissociation of

PAGE 30

30 fluorophore quencher complex, resulting in fluorescence recovery. A decrease in the slope of SV plot is obse rved (Figure 1 4 ). Sometimes an upward curved SV plot appears e specially, when a large amount of quencher is added into the system. This non linear plot is the results of combination of static and dynamic quenching effect. It can be described as follow s : (1 2) T his phenomenon is frequently observed for CPE quenching study, where the effect is 37, 38 is obtained by applying Equa tion 1 1 in fitting of the SV plot at low where the plot is still linear Figure 1 4. Mechanism and Stern Volmer plot of A) collisional quenching and B) static quenching (right). F 0 and F are the same as I 0 and I described in the text. Reprinted w ith permission from Lakowicz, J. R 35 Generally, reflects the quenching ability of the quencher on fluorophore, or, from a sensing aspect, it represent s the sensitivity of a sensor/probe (fluorophore) against the analyte (quencher). A larger indicate s that the sensor exhibits greater response when expos ed to the analyte. Another way to think of the Stern Volmer

PAGE 31

31 constant is that its inverse (1/ ) is the concentration of analyte (quencher) produc ing 50% quenching on sensor molecule Figure 1 5. Sche matic representation of the FRET spectral overlap integral and energy diagram. A dapted wi t h pe r mis sion from Lakowicz, J. R 35 Fluorescence Resonance Energy Transfer One of the important quenching processes is fluorescence resonance energy transfer (FRET), also named F rster reso nance energy transfer. This process occurs whenever the emission spectrum of a fluorophore, called the donor, overlaps with the absorption spectrum of another molecule, called the acceptor (Figure 1 5 ). A donor chromophore, initially in its electronic exci ted state, transfers energy to an acceptor chromophore at ground state through nonra diative dipole dipole coupling with little emission from the donor. If t he acceptor is fluorescent, the emission belonging to the acceptor is observed. The efficiency of FR ET is dependent on the inverse sixth power of the intermolecular distance, making it sensitive to detect the molecular proximity and

PAGE 32

32 interaction between donor and acceptor. 39 This strategy has been fully utilized in the biological sensor development, 40 such as molecular beacon DNA probe. 41 Amplified Quenching and Molecu lar Wire Effect proposed by Zhou and Swager. 42 Their classic paper published in 1995 shows that a PPE featuring a number of crown ether type functional groups, is quenched by the methyl viologen (MV 2+ Figure 1 6 ) 50 100 fold more efficiently compared to a small molecule featuring only one repeat unit with one crown ether of PPE (Figure 1 6 ). The value for the polymer was ~10 5 M 1 whereas that for the small molecule counterpart was ~10 3 M 1 The authors explained that the conjugate d polymer resembles a molecular wire, along which the exciton, a quasiparticle that delocalizes at the excited state is smoothly diffusing until it encounters the first electron accepter MV 2+ that binds efficiently to the crown ether portion. So that one single MV 2+ can adequately quench a large number of exciton producers, creating the amplif ied quenching event. This also effect as shown in Figure 1 7 Figure 1 6 Schematic illustration of amplified fluorescence quenching of a conjugated polymer by MV 2+ Reprinted w i t h permission from Zhou, Q. et al 42

PAGE 33

33 Figure 1 7 Illustration of the molecular wire. A dapted with per m iss ion from Zhou, Q. e t al 42 A few years later, Whitten and co workers discovered tha t very efficient amplified quenching occurred to an anionic poly(phenylene vinylene) CPE (MPS PPV Figure 1 8 ) when mix ed with the oppositely charged MV 2+ electron acceptor. 43 The ion pairing between the CPE and quencher ion mediates the photoinduced electron transfer, so that the exciton transporting within the CPE single chain, or clusters due to aggregation, can be efficiently and immediately trapped by the electron deficient MV 2+ resulting in quenching amplification. The reported is 1.7 10 7 M 1 which is nearly six orders of magnitude greater than that for stilbene, which has the similar structure with that of PPV monomer. Subsequent to the Whitten paper, a number of groups quickly showed that the amplified quenching eff ect is general when fluorescent CPEs are quenched by oppositely charged ions. 24, 37, 44 In 2008, Schanze and coworkers 31 conducted an investigation on the effect of length of CPE (PPE CO 2 Figure 1 9 ) on the quenching efficiency of a monovalen t quencher methyl viologen (MV + ) in methanol. They synthesized a series of PPE CO 2 with various repeat unit number (n = 7, 13, 35, 49,

PAGE 34

34 108) and th e se polymer s are b elieved to be molecular ly dissolve d in the methanol. The quenching study shows that for low and moderate CPE polymerization degree, the systematically increases as polymers chain extend saturating at a value of ~ 40 polymer repeat units (Figure 1 9 ) which suggests that a singlet exciton is able to effectively migrate a PPE chain with length of ~ 80 phenylene ethynylene units during its lifetime. Figure 1 8 Absorption and fluorescence spectra (excited at 500 nm) of MPS PPV (1.7 10 5 M in monom er repeat units) in water in the presence (dotted line) and absence (solid line) of MV 2+ (1 10 7 M). Reprinted w i th permission from Chen, L. H. et al 43 Figure 1 9. Structure of some polymers that used in the text

PAGE 35

35 Figure 1 10 Stern Volmer plots of a series of PPE CO 2 with [MV + ]. Reprinted w ith permission from Zhao, X. Y. et al 31 Aggregation of Conjugated Polyelectrolytes One important property of CPE s is their self assembly in various environments. Due to its highly hydrophobic r igid backbone and hydrophilic ionic side chains, CPE s tend to aggregate in polar solvents via stacking interaction between the phenyl groups while re maining soluble. Once aggregated, the absorption and emission spectra are changed because the inter chain conjugation has much lower energy state than that of the intra chain conjugation. A red shi ft in both absorption and emission bands is observed. In 2010, Zhao and Schanze 28 reported that PPi, a anionic biomolecule, can greatly induce aggregation of a cationic PPE with dendri ti c ammonia side chains (PPE d NH 3 Figure 1 11 B ). The absorption of PPE was dominated by a 430 nm instead of 400 nm (Figure 1 11 A left ). A ttenuation in the emission intensity and a change from a sharp fluorescence peak at 430 nm to a br oad band around 520 nm indicate that the pol ymer 11 A right ). A blue to green color transition, which correspond s to the switch from non

PAGE 36

36 aggregation to aggregation state can be directly observed by eye under the U V illumination (insert in Figure 1 11 A ). Figure 1 11 A) Absorption and fluorescence spectra of PPE NH 3 with increasing [PPi]. Inset: fluorescence emission of the solution before and after the addition of PPi. B) S tructure of PPE d NH 3 and PPi. Reprinted with permission from Zhao X. Y., et al. 28 It is well evident that aggregation of CPE s will significantly influence the fluorescence quenching process. Whitten and co worker s 43 found that MPS PPV is ion pair ed with the quencher MV 2+ to form a crosslink ed structure, which allow s the inter chain exciton migrati on and trapping. Due to the dual function of the viologen in the quenching process, the quenching efficiency of MV 2 + is greatly enhanced. Upon calculation, a single quencher MV 2+ can quench ~1000 repeat unit s of MPS PPV. Based on the amplification factor reported by Schanze (~ 80 phenylene ethynylene repeat unit

PAGE 37

37 per quencher) for a single polymer chain, the huge value 1 000 was the result of the combined effect of intra chain and inter chain electron transfer (Figure 1 1 2 ). Figure 1 1 2 Illustration of amplified quenching by oppositely charged quencher in A ) non aggregated CPE B ) aggregated CPE. Reprinted wi th permission from Tan, C. Y. et al 45 A superlinear phenomenon on the SV plot was reported by Heeger and Bazan. 37, 38 When a large amount of MV 2+ was added into the polymer solution, an upward deviation was observed in the high [quencher] range in SV plot (Figure 1 1 3 ). They explained t of 46 T hat is when a quencher is within the quenching sphere, the quenching of fluorophores by the pairing between the fluorophores and quencher through electrostatic attraction enhance s the local concentration of quencher around the large space that the polymer occupies. So of of the polymer, leading to the superlinear quenching process. The higher t he quencher concentration, the more pronounced this phenomenon.

PAGE 38

38 Figure 1 1 3 Structure of MV 2+ and MBL PPV and Stern Volmer plot for MBL PPV quenched by MV 2+ with [MV 2+ ] from 110 8 to 1 10 7 M. Inset: the same plot extended to higher MV 2+ concentration s (from 1 10 8 to 4.3 10 7 M). Reprinted w it h permission from Eftink, M. R. et al 34 Figure 1 1 4 Quenching of 10 M PPE CO 2 emission by MV 2+ in water ( ) and in methanol with 0 M ( ), 2.5 M ( ), 5.0 M ( ), 7.5 M ( ), or 10.0 M ( ) CaCl 2 Reprinted wi t h permission from Jiang, H. et al 47 The superlinear SV plot was also found in the aggregated polymer quenching system as reported by Schanze an d coworkers in 2006. 47 It was un covered that addition

PAGE 39

39 of a certain amount of Ca 2+ into the anionic carboxylate CPE (PPE CO 2 ) in methanol could greatly enhance the quenching efficiency of MV 2+ The key mechanism is Ca 2+ could induces aggregation due to its capability for effectively cross linking PPE CO 2 chains by complexing with the carboxyl side groups belonging to two adjacent polymer chains. The superlinear property of SV plot was observed in the plots and it is more pronounced when more calcium ions participate (Figure 1 1 4 ). Environm ental Effect on the C onformation of CPE From above description, it is easy to understand that the solvent has much effect on the conformation change of CPE. In 2002, Schanze and cowork er s 24 reported that an anionic PPE (PPE SO 3 Figure 1 9 ) exis ts as single polymer chain in methanol, where its fluorescence properties (sharp emission peak, short lifetime, high quantum yield) are very similar to those it exhibits in good solvents (Figure 1 1 5 ). However, the quantum yield decreases substantially, ac companied by the appearance of a new broad emission band with low intensity and longer wavelength in a H 2 O/MeOH (1:1) solvent. In pure water, PPE SO 3 feature s a broad peak in the fluorescence spectrum and a relative ly low quantum yield. The longer lifetime i s observed at longer wavelength. The entire change suggests th at aggregation occur s in the system. The very broad band indicate the inter chain interactions induced by the polar solvent, i.e., H 2 O, which is presumably due to stacking between rigid poly mer backbone, leading to a relatively three dimension (3D) structure of the CPE. This statement can be supported by the red shift and narrowing of the absorption peak as the portion of water in the mixture increas es consistent with the features appeared in the absorption spectrum when a polymer has a higher structure order and longer conjugation length. 48 50

PAGE 40

40 Figure 1 15. Absorption (left) and fluorescence (right) spectra of PPE SO 3 in MeOH ( ), (1:1) H 2 O MeOH ( -), and H 2 O ( ). Fluorescence spectra were measured with excitation at 400 nm and normalized according to relative quantum yield. Reprinted w ith permission from Tan, C. Y., et al 24 Figure 1 1 6 Solvent effect on the aggregation stat e of CPE. A dapte d w ith p er mi ss io n from Tan, C. Y. et al 45

PAGE 41

41 In th e polar solvent, the hydrophobic backbone of CPE has the tendency to stack The ordered structure has lower surface energy, so that a well arranged aggregation is favorable for CPE to remain soluble and stable in the water. A more optimized situation will be the polymer chains align with their long axes parallel, keeping the phenyl rings nearly co planar (Figure 1 1 6 ). The structure is stabilized by the stacking interaction and polar functional groups. The above examples are based on the linear linked P PE. How about the solvent effect on the structure of meta linked PPE? In the late 1990s, Moore and co workers 51 53 demonstrated that meta linked oligo(phenylene ethynylene)s (OPEs) that are substituted with oligo(et hyleneoxy) side chains self assemble into a helical conformation in a poor solvent such as acetonitrile. The structure of a n OPE and a simulation model created by computer explaining the transition of its conformation in different solvents are shown in Fig ure 1 1 7 This conformational transition is further studied by fluorescence and absorption spectra. The changes in the spectra for the OPE in solvent titration experiment (vary the composition of polar and non polar solvents or vary the polarity of solvent ) resemble that of quenching or aggregation process, which features excimer like band in fluorescence emission (Figure 1 1 8 ). The observed phenomena are expected to aris e molecular stacking interaction as shown in the bottom cartoon of Figure 1 1 8 Moore and co workers also demonstrate d that formation of the helical conformation is a cooperative effect, requiring a minimum number of 10 repeat units (one phenylene and one eth yn yle ne unit) in the OPE chain before the process becomes thermodynamically favored. 51

PAGE 42

42 Figure 1 1 7 filling model showing the conformational equilibrium for OPE of length n = 18. Side chains are not shown for clarity. Reprinted w ith permission from Prince, R. B. et al 51 Figure 1 1 8 A) Normalized emission spectra of OPE in various solvents. Insert: absorption spectra B) Structure of OPE and cartoon for coil helix transition process. Reprinted w i th permission from Lahiri, S. et al 53

PAGE 43

43 Other than solvent type, the pH of solution also affe ct s the conformation of CPEs. Schanze and coworkers 26 have investigated the photophysic al changes of an anionic phosphona te substituted PPE (PPE PO 3 Figure 1 9 ) in the solution with various pH As the pH decreased from 12.0 to 7.5, the absorption spectrum undergoes a red shift by 35 nm. At pH 11.0, the fluorescence spectrum has a maximum peak at 447 nm, which becomes very broad and red shift s to 518 nm at pH 7.5. The quantum yield also decreas es from 0.05 to 0.03. All the phenomena indicate that the CPE undergoes aggregation due to the decrease in pH which would result in partial protonation of the phosphate side chains Th is process consequently reduces the overall negative charge density on the polymer backbone and diminishes the repulsion between different chains that hinder s polymer polymer association. Sens ing Assay Development of CPEs In the pa st decade CPEs have been developed as sensitive fluorescence based sensors 16, 54 for metal ions, 55, 56 small molecules, 57 and biomacromolecules ( p rotein s 43, 58 enzymes, 59 62 DNA 63 66 ). The high sensitivity of this type of sensor benefit s from the intrinsic fluorescence signal amplificatio n that result s from the electron ic delocalization and inter or intra chain exciton migration. In general, there are several mechanisms for CPE based sensing, including photoinduced electron transfer (charge transfer mechanism), energy transfer ( FRET ) an d conformational change. Small Ions/Molecules Sensing For environmental and biological interests, small molecules like heavy metal ions, small organic molecules which are usually toxic hazards or dangerous require highly sensitive and real time detection. The following section will provide detailed examples for metal ions and small molecule s sensing by CPEs.

PAGE 44

44 Metal i on s ensing Figure 1 1 9 Schematic illustration of Hg 2+ induced agglutination of the h PPE CO 2 /papain complex. Reprinted w ith permission from Kim, I. B. et al 67 From the above discussion, we can see that electron deficient ions can act as an efficient quencher due to the electron transfer quenching process. Meanwhile, many metal cations, especially multivalen t ions are capable of crosslinking adjacent CPE chains by chelation or multi binding interaction, which spontaneously induces aggregation and alters the photophysical properties of a CPE. Based on this strategy, many CPE based metal ions sensors have been successfully designed and developed. For example, in 2005, Bunz and coworkers designed a highly sensitive and selective sensor for lead i ons in water by making use of the multivalent interactions between the carboxylate group on the CPE with the target ion. The presence of Pb 2+ is signal by the amplified quenching of a homo carboxylate substituted poly(phenylene ethy ny lene) ( h

PAGE 45

45 PPE CO 2 Figu re 1 9 ), which is dissolved as single polymer chains in aqueous solution due to the high negative charge density on the side chains. A set of different divalent metal ions including Ca 2+ Zn 2+ Hg 2+ Mg 2+ Pb 2+ Cu 2+ and Mn 2+ were tested individually on th e polymer in pH 7.2 buffer system. Pb 2+ has the overwhelming values ~8.810 5 M 1 with at least 10 fold higher sensitivity than other ions. This high quenching efficiency is attributed to the electron transport properties of the CPE and the simultaneous bonding between the cation and carboxylates on polymer c hains. Then, the same CPE was applied in another sensing project for mercury ion by the same research group In this sensing strategy a positively charged protein papain, known to bind to Hg 2+ through sulfhydryl groups, 67 is added into the polymer solution and form an electrostatic complex with h PPE CO 2 (Figure 1 1 9 ). Only in the presence of Hg 2+ the co mplexes are more prone to agglutinate and even precipitate at high [Hg 2+ ], resulting in weak or no emission for h PPE CO 2 The biomolecule mediated strategy was also employed in the development of a potassium ion sensor. In 2004, Leclerc and coworkers 68 demonstrated that a CPE/DNA based aptamer complex has superior sensitivity towards K + in aqueous solution. They synthesized a cationic CPE, a polythiophene (PT) which is able to couple with anionic DNA to form a more planar, potentially aggregated complex. The aptamer was intentiona lly selected so that it can form a quadruplex in the presence of K + After addition of CPE to the aptamer/K + system, the polythiophene displayed a conformation wrapping over the adduct formed by quadruplex aptamer and K + (Figure 1 20A) The absorption peak locates in between that of free and aptamer coupl ed CPE in the spectra (Figure 1 20 B ). The chromatic change corresponding to the conformational

PAGE 46

46 change on CPE can be observed directly by eye. This smart strategy has been further applied by Wang and coworke rs 69 in a ne w ssDNA mediated K + sensor by introducing in the FRET mechanism, which has been reported in 2005. Figure 1 20 A) Sensing strategy for aptamer mediated CPE based K + sensor. B) A bsorption spectra and visible colors of polythiophene mix ed with different i ons. Reprinted w ith permission from Ho, H. A et al 68 Small m olecule s ensing Besides metal ions, many CPE based sensors have been designed for other small molecules, especially biomolecules. W ater solub ility of CPE s makes them applicable in the biological system. In 2002, Schanze a nd coworkers 70 developed a sugar sensor by making use of the high quenching ability ( = 2.8 10 7 M 1 ) of a boronic acid functionalized benzyl viologen, p BV 2+ on an anionic CPE PPE SO 3 ( Figure 1 9 ). Upon the addition of sugar molecules t he strong association between p BV 2+ and PPE SO 3 i s hindered since the sugar molecules can re act with the viologen to produce a charge neutral bisboronate derivative that barely has quenching ability on the CPE (Figure 1 21 A ). A remarkable fluorescence recovery i s observed for CPE. Particularly, a 50 fold

PAGE 47

47 increase in intensity i s detected for the CPE upon the addition of 10 mM D fructose, which is significantly higher than that of D galactose and D glucose (Figure 1 21 B ). ed by Schanze and co workers 71 The metal ion Cu 2+ i s coupled with the PPE CO 2 ( Figure 1 9 ) to form a quenched system at the beginning. Upon the addition of PPi, a more stable PPi/Cu 2+ complex i s created, which sequesters the Cu 2+ from t he CPE, turning on the fluorescence of polymer. Figure 1 21 A) I nteraction between p BV 2+ and sugar B) Fluorescence recovery titration curves against three sugars. Reprinted w i th permission from DiCesare, N. et al 70 In a ddition to the examples described above, there are many other type s of sensors developed for small molecule s including adenosine triphosphate (ATP), 72 hydrogen peroxide(H 2 O 2 ), 73 antioxidants, 74 Fe(CN) 6 4 44 and Ru(phen) 3 4 ((ph en)=4,7 bis(4 sulfophenyl) 1,10 phenanthroline). 44

PAGE 48

48 DNA S ensing Deoxyribonucleic acids (DNAs) play essential roles in genetic and epigenetic researches. As a special information carrier with stab le and well ordered macromolecular structure, DNA is recognized as a key targe t for the diagnosis of cancer and disease as well as the de tection of bacteria and virus. To develop in vitro DNA sensors or probes, the general way for recognition of target DNA sequence lies on hybridization between DNA sequences, which is signaled typic all y through electrical, magnetic or optical responses. Figure 1 22 Photographs, possible structure and UV V is absorption spectra changes for the PT based DNA sensor : a ) PT alone, b) PT/ssDNA duplex, c) PT/dsDNA triplex, d) PT/ssDNA plus a complementar y target with a two base mismatch, and e) PT/ssDNA plus a complementary target with a one base mismatch after five minutes. Reprinted w ith permission from Ho, H. A. et al 75 CPE based DNA probe particularly when it is labeled on the DNA or a DN A sequence has been well developed in the pas t decade Among the research groups working toward CPE based DNA sensing, Leclerc group is

PAGE 49

49 the pioneer who initially develop ed ultrasensitive polythiophene ( PT ) based DNA sensors that achieve a zmol sens itivity 75 Drive n by electrostatic attraction, the cationic PT undergoes a conformation al transition from a random coil to a tw isted planar conformation complexing with single strand DNA (ssDNA), forming a so The 143 nm red shift in absorption spectrum affords the CPE colorimetric sensing ability: the yellow red transition can be directly distinguished by naked ey e. Interestingly, if the complementary target ssDNA is further added into the duplex system at high temperature, the color of sample turn s back to yellow and a blue shift of the absorption peak is observed after 5 minute incubation. This change is attribut e d to the formation of coil. When two base mismatch or one base mismatch compl e mentary ssDNA i s added to the duplex solution, a color of orange, in between red and yellow, is observed, which correspond s to the intermediate state of tight and loose binding for the complex due to incomplete hybridization. All those changes are quantitatively detectable through the absorption spectra as shown in Figure 1 2 2 They further inco rporated the FRET mechanism in the design by conjugating the dye molecule on to the end of probe ssDNA. 75 Distin ction between the triplex and duplex can be achieved by comparing the FRET efficiency under two circumstances: more emission of the dye molecule can be observed for triplex which is possibl y due to the higher qua ntum yield of PT in non aggregate d state. Non complementary or mismatched ssDNA is also distinguishable and this turn on sensor has a detection limit as low as 3 zM toward target ssDNA. Th e strategy that combine s PF with FRET also inspired Bazan and Wang. They have successively reported a number of applications

PAGE 50

50 such as detection of DNA mismatch, single nucleotide polymorphism (SNP), DNA methylation, gene regulation, etc. 76, 77 Moreover, Bazan and co work ers 78 improve d the method to a two step energy transfer by introducing in ethidium bromide (EB) which can intercalate into the helix dsDNA and become e missive Thus the dye molecule on the probe ssDNA acts as a convertor for energy dipole moments, bridging the polymer and EB. Hence, the non specific interaction between non complementary or mismatched DNA and duplex probe will not cause interference Fi gure 1 23. Schematic illustration of DNA detection by CPE labeled molecular beacon. Reprinted w ith permission from Yang, C. Y. J. et al 79 A covalent conjugat e between a molecular beacon (MB) and CPE was first reported by Tan and Schanze. 79 As illustrated in Figure 1 2 3 this sensing strategy employs a sulfonated PPE as fluorescent probe and Dabcyl (Dabcyl = 4 (4 (dimethylamino) phenylazo) benzoic acid) as q uencher It has a rapid response (at second time scale) to the presence of target DNA, which simultaneously stretches the hairpin like DNA strand and disrupts the quenching process, in the regular hybridization buffer solution. The a dvantage of this sensor lies in its resistance to both ion ic strength and surfactant effect by increasing which, t he nonspecific interactions, such as electrostatic and hydrophobic interactions between the target DNA and CPE chain, are

PAGE 51

51 effectively diminished. The molecular beac group. 80 The anionic CPE structure was optimized by introducing pendant polyethynylene glycol ( PEG ) groups to avoid aggregation of the CPE under higher ionic strength conditions and increase the quan tum yields in aqueous solution. Dabcyl labeled DNA sequence s w ere conjugated to both end s of the CPE to generate a device with a stoichiometric ratio of [ Dabcyl ]:[ CPE ] as 2:1. This method also shows its capability of distinguishing single base mismatched DNA from target D NA at the same concentration. Protein/ E nzyme A ctivity S ensing Proteins are essentially involved in almost every process within cells and the constitution of organisms. A large number of proteins are various types of enzyme s that catalyze tens of thousands of reactions that are vital to maintaining life. Some proteins also are important components in muscle, blood, skin, cartilage, and bones. Others participat e in cell signaling, molecular recognition, cellular communication, and gene expression. 81 Therefore, identification or recognition of proteins is highly important to medical diagnostics or clinical research. 82 85 A great n umber of methods or strategies have been developed in the past decades including electrochemistry, 86 Raman, 87 flow cytometry, 88 fluorescence immunoassay, 89 and mass spectrometry. 90 However, the requirement for sophisticated instrumentation and proficient manipulation limit s their broader application. The method combining sensitivity, simplicity and eco nomy is still highly demanded. CPE based protein sensing methods have drawn much a ttention throughout the past years. 16, 54, 91 94 The following discussion will be divided to two sensing categories:

PAGE 52

52 protein and enzyme sensing. Each category has models representing the different mechanisms of ener gy transfer, electron transfer and conformation change Protein s ensing by CPE b ased s ensor tether i s among the earliest CPE based protein sensing assays by making use of the electron transfer superquenching property of CP E s 43, 54 It was well discussed in the previous section that MV 2+ is a remarkably efficient CPE quencher A biotin MV 2+ which has a flexible tether in between biotin and MV 2+ ( Figure 1 24 ), was used in the sensing strategy as shown in Figure 1 2 5 This ligand quencher conjugate display s net positive charge and form s the complex with anionic MPS PPV (structure shown in Figure 1 8) via electrostatic attraction, quenching the fluorescence of MPS PPV with similar quench ing efficiency as that of non labeled MV 2+ Avidin, which has four biotin binding sites with extremely high binding affinity ( ~ 10 15 M), 95 is added into the MPS PPV/biotin MV 2+ system. The stronger interaction between biotin and avidin allows the formation of a more stable complex, causing the former CPE/quencher complex to fluorescence i s observed even at lo w concentration of avidin (100 nM). This strategy paved a new way for development of CPE based protein sensing method s by using the QTL model. The de s been further extended in other sensor s by many researcher s. 60 62 Besides the QTL based sensing model, another sensing strategy that utilize s the direct quenching property of some special proteins has emerged. 58, 68, 96, 97 Usually, t hose proteins have an electron deficien t center, such as heme moiety, which acts as an electron acceptor center like metal ions. Although the direct sensing methods lack the

PAGE 53

53 exclusive specificity as antigen antibody type, they are much simpler and labor saving by avert ing the need for additional ligand design or modification. Figure 1 24. Chemical structures of quenchers or ligands Figure 1 2 5 tether se nsing. Reprinted w it h permission from Chen, L. H. et al 43

PAGE 54

54 Figure 1 2 6 Schematic description of the specific detection of human thrombin by use of ssDNA thrombin aptamer and cationic polymer Reprinted w ith permission from Ho, H. A. et al. 68 Analog ous to the sensing strategy for K + discussed in th e previous section, a sensing method for the human thrombin utilizing the same polythiophene and ssDNA thrombin aptamer was developed by Ho and Leclerc in the same report. 68 As shown in the Figure 1 2 6 Path A, the thrombin aptamer form s a quadruplex structure upon binding to the human thrombin. The electrostatic attraction betw een the cationic polythiophene and anionic DNA forces the polymer to wrap over the quadru p lex, altering the conformation of polymer. De aggregation of the polymer can be easily monitored through its photophysical properties change a blue shift in absorptio n spectrum and an emission intensity increase in fluorescence spectrum. The high selectivity is demonstrated with no signal detectable when non specific protein or

PAGE 55

55 ssDNA is added The same strategy has been applied to detect the protein calmodulin (CaM) la ter by Nilsson and coworkers. 98 Enzyme activity sensing One of the most commonly used sensing strategies for enzymes incorporates the well defined QTL model into the sensing scheme. By coupling an enzyme substrate with the quencher and mixing with CPEs, the presence of a target enzyme can be signaled by fluorescence quenching (turn off) or dequenching (turn on) modes. 60 62 Figure 1 based sensors Reprinted w ith permission from Pinto, M. R. et al. 61 In 2004, Schanze and co worker s approaches to detect protease activity. 61 Commercial available fluorescence quencher

PAGE 56

56 tether substrates were used in two strate gies. Two p nitroanilide ( p NA) labeled peptide, L Lys p nitroanilide dihydrobromide (K pNA, Figure 1 24 ) and N benzoyl Phe Val Arg p nitroanilide hydrochloride hydrate (Bz FVR pNA, Figure 1 24 ), are the substrates of two enzyme peptidase and thrombin, res 2 7 upper ), since the two p NA labeled substrates display net positive charge they ion pair with the anionic sulfonate d PPE (PPE SO 3 Figure 1 9 ). The p NA moiety strongly quench es the fluorescence of the poly mer even at nanomolar concentration through the amplif ied quenching mechanism. The addition of peptidase or thrombin to the system cause s peptide hydroly zed cleaving the p NA apart from the peptide, and leaving the p NA with no charge. The fluorescence of the polymer i s recovered due to the lack of association between th e neutral quencher and polymer. 24), comprises another anionic CPE PPE CO 2 ( Figure 1 9 ) with a non emissive peptide derivative rhodamine substrate (Rho A rg 2 Figure 1 24 ). The mixture is first excited at 400 n m and emit s strongly around 470 nm. The I ntroduction of proteolytic enzyme papain, which catalyze s the hydrolysis reaction of the peptide, separate s the rhodamine and peptide segments restoring the e missive property of rhodamine. The decreasing of the fluorescence of the polymer as well as the appearance of a second fluorescence peak at 515 nm i s attributed to the singlet singlet energy transfer from PPE CO 2 to Rho Arg ( Figure 1 24 ) w hich has an emis sion peak (Figure 1 27 bottom). The detection limit for this enzyme i s evidenced to be 3.5 nM. This strategy successfully introduced in the fluorescence resonance energy transfer ( FRET ) paving another way for enzyme sensing. In addition, several groups have developed a new

PAGE 57

57 FRET based enzyme sensing strategy by just simply substituting the quencher 99 101 In sum, t hese two sensing strategies establish a novel platform for sensitive detection of enzyme s motivating the development of a variety of similar CPE based enzyme cleavage involved sensing methods. 60, 62, 102, 103 Another attractive strateg y appearing recently is natural enzyme substrates based model. 102, 104, 105 Instead of sophisticated designing or labeling a substrate, it applied the naturally existing substrate directly into sensing assay. By utilizing the interaction between the substrate and a common quencher or CPE itself, the photophysical urn response. Very recently, Liu and co worker reported a turn on assay by using the natural substrate Cytochrome c (Cyt c ) in conjunction with an anionic p oly(fluorene co phenylene) ( PF P CO 2 ), to detect protein trypsin activity 102 N aturally cationic Cyt c is a n electron deficient heme containing protein, holding superquenching property. Thus it can be regarded as a natural quencher ligand and utilized in the turn on mode following the first path in Figure 1 2 7 The enzyme trypsin was reported to di gest Cyt c ( isoelectric point, p I = 10.5) into more than 15 fragments 106 among which the metal containing heme peptide fragment, responsible for the quenching effect in Cyt c has a reduced p I = 7.1 (based on an online peptide p I calculator 107 ). In the pH 8.9 buffer system, the positively charged Cyt c can form a complex simultaneously with anionic PF P CO 2 eff iciently quenching the fluorescence with = 1.32 10 7 M 1 After the addition of trypsin, the Cyt c i s broken down, producing t he metal containing heme

PAGE 58

58 peptide fragment with negative charges The complexation of the quencher and CPE i s subsequently disturbed, resulting in the fluorescence recovery for PF P CO 2 which can be easily observed by eye. T h e sensitivity for this sensor is found to be 1.7 nM. The selectivity of this assay was demonstrated by detecting three other enzymes, lysozyme, alkaline phosphatase ( ALP ) and thrombin, under the same condition and no recovery response i s observed for all of them. B y taking advantages of natural superquenching property of Cyt c complicated design for the substrate is not needed and the procedure bec o me s s impler. Meanwhile, t he interference and uncertainty brought into the catalytic reaction by the substrate modification are eliminated. Conformation change based strategy also attracts much attention due to its non invasiveness without need for covalent lab eling or molecule modification. In 2009, Schanze and co works report ed a fluorescence turn off assay for phospholipase C (PLC) by making use of the reversible aggregation process of anionic CPE (BpPPESO 3 Figure 1 2 8 A ). 103 The BpPPESO 3 aggregate s in the aqueous solution, featuring with a broad, structureless band in the emission spectrum. The addition of the natural substrate of PLC, phosphatidylcholi ne (10CPC), cause s the increase in CPE fluorescence intensity. Th e zwitterion phospholipid 10CPC has the surfactant property that helps de aggregate the polymers by forming lipid polymer complex. However, as enzyme PLC presents, 10CPC is degraded to phosph orylcholine and a long chain (DAG) without charge. As the product molecules lose the surfactant function, the polymers are not able to maintain non aggregation state an d their fluorescence intensity turns off subsequently (Figure 1 2 8 B ).

PAGE 59

59 Figure 1 2 8 A) Structures of polymer, BpPPESO 3 and substrate, 10CPC, and reaction scheme for hydrolysis of 10CPC by PLC. B) Mechanism of PLC turn off assay. Reprinted w i th permission from Liu, Y. et al 103 In the previous section, copper ion has been proved to be an excellent quencher for anionic CPE. Consequently, in many reports, Cu 2+ has particularly involved in the indirect sensing assay for enzymes activity. 71, 104, 105, 108 In those methods, substrates or reaction products, such as PPi and amino acid, are capable of capturing or complexing with Cu 2+ keeping the Cu 2+ away from the fluorescent CPE and making the signal tur n alkaline phosphatase ( ALP ) was reported by Schanze and coworker. 71 The addition of PPi to the pre quenching system of PPE C O 2 ( Fi gure 1 9 ) with Cu 2+ readily induced fluorescence recovery. Then as the ALP is added into the system of PPE CO 2 /Cu 2+ /PPi, the hydrolysis process of

PAGE 60

60 PPi is triggered, concomitant with the releasing of Cu 2+ back to the PPE from the PPi Cu 2+ complex Thi s process is signaled by the intensity (Figure 1 2 9 ). The same idea has been applied in another sensing method for adenylate kinase activity by employing ATP, ADP, and AMP as substra tes, which was also reported in 200 9. 93 Figure 1 2 9 Mechanism of ALP turn off assay and p hotographs of solutions illuminated with near UV light illustrat ing the polymer fluorescence under the different conditions of the assay. Reprinted w i th permission from Liu, Y. et al 71 Non specific I nteraction of CPEs CPE based protein sensing methods have many advantages including sensitivity, simplicity and economy. However, it has its own shortcomings. One of the challenges is tha t too much reliance on the electrostatic or hydrophobic interaction s may bring in uncertainty and unwanted non specific interaction between non target species with probe, causing false signals. There are a number of reports regarding non specific effects p ublished and reviewed. 33, 109 111 Some solutions have been proposed such as pre formed CPE containing charge neutral complex (CNC) by mixing cationic CPE and

PAGE 61

61 anionic polymer in 1:1 ratio to obtain neutrality. At the expense of sensitivity, this type of sensor can be readily quenched by both ca t ionic and anionic target analytes eliminating the non specific effect s 109 Other solutions for solving non specific interaction including employing CPE coated microspheres 60, 62 or grafted colloids 112 instead of CPE itself as the sensing probe, so that the aggregation state change will be less influenced by interfe ring species in the solution. While non spe cific interaction may be considered as a negative factor in sensor design, it also can be employed to develop new sensors. As aforementioned, the electrostatic or hydrophobic interaction between CPE s and biomolecules, such as DNA s and protein s can induce conformation change of CPE s The strategy has already shown the power in a variety of application s for metal s proteins and DNA sensor s development as reported. 33, 34, 68, 75 This strategy is much attractive becaus e it is laborless, economic, and simple without sophisticated design on probe. Although one CPE may lack the specific detection of an analyte by non specific interaction, an assembly of more than one polymer has the ability to provide informative and suffi cient signal response for analyte recognition More details and information will be displayed in the next topic sensor array. Sensor Array In recent years, as increasingly demanded by genetic analysis, clinical diagnosis, environmental analysis, and homela nd defen s e, the sensor that is able to distinguish or recognize more than one target analyte attracts much attention. The traditional specific recognition through lock and key mode requires sophisticated sensor/probe design and sufficient chemical/biologic al knowledge. Moreover its high specificity, i.e., only one or one type of analyte can be detected at one time, hinders its universalization. An

PAGE 62

62 alternative approach that emphasizes differential receptor analyte binding interactions has emerged in the pas t two decades. 113 116 The basic concept is chemical producing human senses by mimicking working principle of nose or tongue. T he probe elements in the sensor arrays are analog ous to specialized sensory cells in the nose or mouth and the created pattern recognition systems are like the recognition portion in for differentiat ing odors and flavors. Specifically, the re sponse generated by the sensor array should be discernibly different for different analyte s exhibiting fingerprint characteristics. To conduct a detection by the sensor array, three basic stages are involved. Firstly, data/signals need to be collected fro m a relatively large pool of qualified samples with known category/type so as to build a database of reference. Secondly, a differential tool/method needs to be set up and trained so that the above samples can be assigned to their own group with relatively high accuracy. Finally, the technology can recognize new samples with little category information by comparing their signal fingerprint to those stored in the database. Thus qualitative or quantitative analysis can be performed. This type of sensor has be en proved to be highly useful for a wide variety of chemicals sensing, e.g., metal ions, 117 volatile agents, 118 aromatic amines, 119 amino acids, 120 carbohydrates, 121 and proteins, 122 which are quite important in quality control, production process/development, crime prevention and security, human health and environmental monitoring. Several array based biological sensors have been developed by taking advantage of the optical perturbations of conjugated polymers driven by analyte induced quenching or aggregation. 97, 110, 123, 124 For instance, Bunz and Rotello 124 developed a cell sensing

PAGE 63

63 strategy us ing the fluorescence signal change from a group of CPEs. As shown in Figure 1 30 those polymers with various pendant charged residues display distinct response s i.e., the ratios of final and initial fluorescence of the CPEs, when mix ed with different typ es of cells individually. A chemometrics technique called linear discrimina nt analysis (LDA), is applied to process the result ing data matrix A canonical pattern is created and each sample is well classified. This array based sensing system is demonstrat ed can differentiate between cell types as well as discern cancerous from noncancerous mammalian cells. The robustness of the sensor array was further tested using unknown samples. An accuracy of 80% for identify three types of cells is obtained by usin g o nly four CPEs as the probes. Figure 1 30. Schematic presentation of the cell detection assay by CPEs including signal response pattern and canonic score plot. Reprinted w i th permission from Bajaj, A. et al 124

PAGE 64

64 CPE based sensor arrays for efficient protein recognition have been increasingly reported and similar procedures as described above ha ve bee n applied and high accurac ies ha ve been obtained (97% for 17 proteins as reported by Bunz and Rotello 97 ). Unlike the conventional specific markers design, the CPE sensory recognition elements do not need to be attached covalently as they self asse mble onto the surface of cells or proteins via non specific interaction. Without proficient manipulation and strong biological background, an ordinary technician can accomplish the test smoothly. Although having so many outstanding strengths, those novel s ensor arrays still need further development in more complicated biological environments to eliminate potential interference and improve their feasibility in the real wor l d. Linear Discriminant Analysis There are many tools available for the analysis of dat a from an array of chemical sensors. Basically, two methods are appropriate for classification purpose: if only independent variable information is required (unsupervised mode), principal component analysis is suitable; while for those that need dependent variable information (supervised mode), such as analyte classes, the linear discriminant analysis (LDA) can be applied. 125 In this dissertation, a set of data with known categories is needed to be classified in Chapter 4 where a protein sensor array is developed. LDA is more suitable for data processing and analysis and so that only LDA will be emphasized in this section. LDA is used to separate classes of objects or assign new objects to appropriate classes by analyzing the vari ance s between classes and within classes. 126 As o ne type of chemometrics, LDA tests multivariate differences between classes, optimizes the dimensionality of class, calculate s the relative importance or contribution factor for each variable and classifies known and unknown objects into groups. The discri minants are

PAGE 65

65 linear combinations of the measured variables, typically, in a sensor array, sensor responses. Consider that a sensor array with n variables (probes) is used to differentiate k classes of analytes f or each mixture or treatment, i.e., one analy te with one probe, m replicates are performed so we have m sets of data for each mixture Let represent the individual observation for each measurement, so is the observation for n th probe at m th measurement for a certain analyte. For i th class, the data matrix abstracted from each measurement for this class of analyte can be written as follows, (1 3) Take an average over the m sets of data for each probe analyte mixture, (1 4) Where is the average of j th row in each submatrix T he whole data matrix ( ) is: (1 5 ) Let be the total average matrix: (1 6 ) T he between classes sums of squares matrix B representing the variance between classes is expressed as: (1 7 ) The covariance matrix can be computed as

PAGE 66

66 (1 8 ) So that the combined covariance matrix is: (1 9 ) The within class sums of squares matrix representing the variance within class is displayed as: (1 10 ) The key objective for discriminant analysis is to figure out the best way to let to be the maximum, i.e., the distance between classes relative to the varia nce with in classes is maximum. It can be achieved by calculating the eigenvectors s of and scaling them such that (1 1 1 ) w here denotes the th eigenvector ( ) : (1 1 2 ) The eige nvalue can be treated as the contribution factor, i.e., how important of its eigenvector for transmitting the data information. Usually, we use the percentage of the value took over total eigenvalue to rep resent its importance. For a given n dimensional sample vector that belongs to a ce r tain class it will be projected to a new discriminant space by the operation In the new space, it is believed to be better separate d from the samples that belo ng to other classes while getting close r to the samples that belong to the same class. So that it will have a new coordinates calculated by Normally, s discriminant functions are available ( s = min ( n k 1)), so that there are s sets of eige nvalues and eigenvectors. For the sample

PAGE 67

67 can be graphically display ed in a 2D plane or 3D space. For the three most important discriminant functions, usually carrying more 80% information/variation in total can be used to create the canonic plot although the samples may not be 100% discriminated based on the plot. When referring to the unknown sample test, the essential rule is to assign the unknown sample to the nearest class after it undergo es the same operation to the identical discriminant s pace. Assume is the vector for the unknown sample, if for all (1 1 3 ) is classified to the group of l This assignment is quantitatively perf ormed in terms of comparing the distance between the spot of unknown sample and the center of the cla sses in the discriminant space. As the data for the unknown samples must undergo the operation that set up for the standard/training data, i.e., they will be projected to the same discriminant spaces the sensing probes as well as the discriminant method must be carefully evaluated and a high accuracy for classification is preferred when handling the training data. If a low accuracy is obtained, it means the current probes or method are not sufficient to discriminant the classes, which will lead to incorrect identification of unknown sample. Therefore a modification or replacement of probes/method needs to be considered. The last step for the analysis will b e verification, that is, use a well defined method to test the unknown sample and validate the result of LDA. The method varies from one sensor to another, relying on the properties of the signal and samples themselves.

PAGE 68

68 Fluorescence Correlation Spectroscop y (FCS) Basic Principle of FCS Investigation of conformation or aggregation state change of CPE under various circumstances and the interaction between CPE and other molecules has been extensively conducted mainly through f luorescence spectroscopy or UV V i s spectroscopy. Researches utilize the color (absorption) or glow (fluorescence) changes to conjecture the inner situation including inter or intra molecular interaction. Other techniques including X ray scattering, 127, 128 light scattering, 129 fluorescence anisotropy, 130 fluorescence correlation spectroscopy 131, 132 are also involved in the exploration. Among them, fluorescence correlation spectroscopy (FCS) is emerging as a powerful tool that can achieve single m olecular analysis. FCS is a statistics based analytical technique, first introduced in the early 1970s by Madge, Elson and Webb 133 which monitors the spontaneous fluctuations of fluorescent intensity of diffusing molecules within a small excitation volume (~femtoliter). Various processes including Brownian d iffusion, chemical reaction or flow contribute to the fluorescence intensity fluctuations. By applying a n autocorrelation function G ( ), the raw fluctuations data are converted to a decay curve, which can be analyzed by an appropriate fitting model. The important dynamic and kinetic information for a particular molecule or particle can be obtained, thus the related chemical reaction, ph ysical interaction between molecules or change in chemical environment is revealed The FCS is somewhat similar to another correlation related technique dynamic light scattering (DLS). However, instead of recording the light from all the particles showing up in the excitation volume, it only focuses on the fluorescent species and this specificity can be further enhanced by selecting an appropriate emission filter.

PAGE 69

69 Application of FCS FCS has shown its large potential in many applications, 134 136 e.g., study of translational 137 139 and rotational diffusion, 140, 141 protein folding 142 ligand macromolecule binding, 133, 143 146 hybridization reactions 147 in bulk solution, on surfaces or in cells. 148 152 Although FCS has been primarily employed to the analysis of biological systems, application of FCS on polymers has gained increasing interest throu ghout the past decade due to its high sensitivity and ability to effect single molecule analysis. For example, Van Rompaey and coworkers studied interaction between a dye labeled biomolecule and a polymer formed pharmaceutical carrier 153 Bonn and coworkers focused on the aggregation behavior of dye labeled diblock copolymer poly(2 alkyl 2 oxazoline), 154 Laguecir and coworkers investigated the conformational behavior of the dye labeled poly(acrylic acid) affected by size and pH. 155 Most of those applications utilized a specific dye as a tracer or directly labeled the polymer with a fluorop hore. Normally, those dye molecules or fluorophores can be excited by a laser line with a specific wavelength, typically, 543 nm for rhodamine, and 488 nm for Alexa 488. Therefore, the diffusion behavior of the fluorescent molecules or particles can be mon itored and the chemi cal environment can be deduced. Because of their inherent fluorescence CPEs can be obser ved directly using FCS. Avoid ance of the tedious dye labeling process on the target molecule makes the study of conformation al or diffusion al chang es of the CPE via FCS simpler and more direct For example, Jayakannan and coworkers have systematically studied the influence of chain length and molecular weight distribution on the diffusion dynamics of CPE at the single molecular level by FCS. 156 Cotlet and coworkers investigated the solvent polarity effect on chain conformation using FCS measurement s 157 By employing a FCS coupled

PAGE 70

70 with a time correlated single photon counting ( TCSPC ) device Masuo and coworkers determined the relati onship between the probability of single photon emission and the spatial size of the CPE chains. 158 Most of these applications utilized a lase r beam with a wavelength H owever, most of the commonly used CPEs have close to 400 nm. Consequently, an FCS equipped with a 405 nm blue diode laser specifically for CPE detection was constructed in our lab. As far as we know, only a few papers focus on the use of FCS with a short wavelength laser, specifically 405 nm. 159, 160 Even fewer technical data or details can be found in the literature for th e construction of such a FCS system Details about the construction will be d iscussed in the second chapter. Overview of This Dissertation This dissertation is aim ed at investigating the interaction between the CPEs and other molecules including dye ligan d compound s proteins and metallic ions as well as their sensor application for metal ions and proteins by using various optical technologies, especially, fluores cence correlation spectroscopy. I n the second chapter, t he theory of FCS including its autoco rrelation function, basic equations, single species fitting as well as the relationship between the diffusion time and molecular weight is introduced. Then it is followed by the d etails in the construction of FCS setup with 405 nm diode laser. The functio n of each component of the setup laser alignment and system optimization are fully described. E xamination of the setup and the test results further validate the feasibility of the instrument Chapter 3 is composed of two parts. In the first part, a tetram ethylrhodamine labeled biotin ( biotin TMR ) intercalated helical CPE probe i s developed as an avidin sensor by fluorescence spectroscopy through the technique of fluorescence resonance

PAGE 71

71 energy transfer In the second part, an FCS system coupled with 543 nm l aser, particularly for TMR study, i s employed to further study the above system T he formation of large aggregates in solutions containing helical CPE/TMR biotin with avidin i s demonstrated, giving rise to a new avidin sensing strategy The detection limit less than 100 nM ma kes this method among the most sensitive avidin s ensors that have ever been reported. In the fourth chapter, s tudy of conformation change of CPEs induced by proteins as well as the development of array based sensing of proteins through the FCS system with 405 nm laser is conducted. A systematic study on the interaction between six different types of CPE and seven target proteins are carried out The result ing 2D bar or 3D column response graph displays a unique signal response pattern fo r each protein, which i s sufficient to build up a sensor array for proteins. The discriminant functions and canonic score plots obtained through linear discriminant analysis are and employed for further unknown sample test. The high accuracy of 93% for ide ntifying unknown samples well prove s the feasibility of this novel protein sensor array Finally, a s tudy of meta l inked p oly(phenylene ethynylene) s ulfonate c ontaining p yridine ( mPPESO 3 py ) q uenched by m etallic i ons is conducted in Chapter 5 The high quen ching efficiency for Pd 2+ over other metallic ions on mPPESO 3 py i s demonstrated by comparing their Stern Volmer plots, giving rise to a new Pd 2+ sensing system. Then FCS data reveals that some multivalent ions, such as Cr 3+ Fe 3+ can crosslink mPPESO 3 py a nd induce the aggregation. However, aggregation is not the dominant element in the fluorescence quenching process of mPPESO 3 py

PAGE 72

72 CHAPTER 2 CONSTRUCTION OF A FLUORESCENCE CORRELA TION SPECTROSCOPY As discussed in Chapter 1 fluorescence correlation spectro scopy ( FCS ) has already shown its power in detecting chemical and physical change of fluorescen t molecules, particularly conjugated electrolytes ( CPEs ) A n FCS system equipped with a 405 nm blue diode laser specifically for CPE study was constructed in our lab. As far as we know, only a few papers focus on the use of FCS with a short wavelength laser beam, specifically, 405 nm. 159, 160 Even fewer technical data or details can be found in the literature for the const ruction of such a n FCS system. In the current chapter, the details for construction, optimization and calibration of a n FCS system coupled with a 405 nm diode laser will be provided. Theory of FCS This section follows closely the classical paper reported by Haustein and Schwille. 161 Figure 2 1 illustra te s the working principle for FCS. The measurement is accomplished by focusing an excitation laser beam onto the sample through an objective lens to form an ellipsoid like femtoliter volume (Figure 2 2 ), and then collecting the fluctuating emission signals within the excitation volume. The autocorrelation function, defined as (2 1) is used to characterize the temporal fluctuation that is typical species. in Equation (2 1) represents the fluctuation of the fluorescence signal as the deviations from the temporal average of the signal at time t. helps co n vert the r aw data to a decay curve, which represents the similarity between the signal and replicate of the same signal but shift ed with a time lag

PAGE 73

73 (Figure 2 3 ). 136 In the early time period, typically < 10 6 s f or most systems, the correlation curve is much nois y due to the after pulse effect of avalanche photo diodes (APD), which is one of the major detector non idealities that affect the output. 162 There are some methods for after pulse effect handling, such as coupling in one more detector pa ir in the system for cross correlation, 163, 164 or using a mathematic al model to fit the after pulse 165 However, in this dissertation, only the information provided by the FCS curve without after pulse part, i.e., > 10 6 s for small molecules and > 10 5 s for large molecules is considered. Figure 2 1 Working principle for FCS

PAGE 74

74 Figure 2 2 The excitation volume in A) Z direction and B) X Y planar as generated by a diffraction limited objective lens. R e p rin te d wit h pe r miss ion from Wilhelm, S. 166 Figure 2 3. Development of an autocorrelation curve. A dapted with pe r mi ssi on from Schwille, P. et al 136 The three dimensional ( 3D ) and two dimensional ( 2D ) fitting models are commonly used for fitting the correlation curve of a single com ponent sys tem. The 3D equation is written as:

PAGE 75

75 (2 2) In E qu ation 2 2, the structure parameter, equates to where is the longitudinal radius a nd is the tra nsversal or waist radius of the confocal volume (Figure 2 4 ) ; N is the average number of fluorescent molecules in the detection volume ; is the average time of fluorescent molecules diffusing in the well defined detection volume, which i s charac teristic of a specific molecule or particle. When a 2D model is obtained: (2 3) Figure 2 4 Ellipsoid like excitation volume formed by the objective lens. In genera l, two important parameters can be obtained through the fitting : the diffusion time and number of fluorescence molecules inside the excitation volume N which is the invers e of G (0), i.e., the reciprocal value of G ( ) when approach es to 0 (Figure 2 1 ). According to the diffusion concept, small molecules move more rapidly through the excitation volume than larger ones. So the FCS curves of larger molecules will reflect the longer diffusion time and a significant retardation of the diffusion of small fluo rescent molecules can be expected when they interact with larger macromolec ules.

PAGE 76

76 While the increase in concentration of the molecules will cause the decrease of 1/ G (0) accompanied with decrease of the correlation curve amplitude (Figure 2 1 ). The relations hip of to the molecular diffusion coefficient D (m 2 s 1 ) is given by: (2 4) The waist radius is obtain ed from its rearranged equation (2 5) where is the diff usion coefficient o f the standard calibration dye. The effective detection volume is obtained from the concentration of analyte C and the number of diffusing particle s inside the detection volume, N : (2 6) where N A can be simply obtained from below: 35 (2 7) Usually, the 2D model is used to initially fit the standard dye sample for calibration. By using equations 2 4, 2 5, 2 6, and 2 7, the structure parameter can be calcula ted. As long as the experimental conditions (excitation wavelength, excitation power, cover glass thickness, solvent and immersion medium, emission filter and position of optic al elements) are the same during the measurement, the detection volume theoretic ally does not change and neither does Once the value of is obtained, the 3D model with the can be employed further to fit the other samples. The translational diffusion coefficient D of a molecule is related to its size by the Stokes Einstein equation (2 8)

PAGE 77

77 where k is Boltzmann s constant ; T is the temperature ; is the viscosity of the solvent ; and R is the hydrodynamic radius. This equation can be used to estimate the size of diffusing particles by assuming that the particles are spherical with radius R R is rel ated to the molecular weight ( MW ) of the molecule with a specific gravity by 167 (2 9) where V is the molecular volume. Thus we have (2 10) These equations show that the radius R and diffusion coefficient D are weakly dependent on the molecular weight. By combining E quation 2 8 and 2 10, we have: (2 11) This relationship is useful for estimati ng the MW of a spherical particle from its diffusion coefficient. Construction of FCS System A n FCS system coupled with a 405 nm diode laser sour ce was constructed in the lab. The basic setup is displayed in Figure 2 5 Details about the function of each component and the instrument construction including alignment, optimization and calibration will be described next Laser Optimization and Alignme nt Blue ( 405 nm ) d iode Laser The continuous wave laser beam ideally satisfies high quality and stability requirements for FCS light source. Since most of our CPEs have high absorbance around 400 nm, a packaged, blue violet diode laser (405 nm), is used i n our system. A near Gaussian distributed and symmetric circular laser beam is critical for achieving high performance in FCS measurements. However, the laser beam from the packaged

PAGE 78

78 diode laser has asymmetric dimensions and divergence in the X and Y direct ions due to the asymmetr y of laser export slit, which strongly affect s the formation of a well shaped confocal volume in the sample. Additional assemblies including a fiber coupling spatial filter, a single mode fiber, and a fiberport collimator are employ ed to reshape, optimiz e and collimate the laser beam. Figure 2 5 Schematic diagram of the FCS setup described in the text. Black dash line represents the outline of fluorescence microscope.

PAGE 79

79 Fiber c oupling s patial f ilter and f iberport Figure 2 6 Phot o graphs for A) excitation part of FCS setup and B) fiber coupling spatial filter F igure B is r e p rin te d w ith per mi s sio n from Thorlab s 168 A fiber coupling spatial filter system (Figure 2 6 B ) i s used for producing a clean, spatially uniform, Gaussian beam. In our setup this spatial filter i s positioned after a pair of angle adjustable mirrors (Figure 2 6 A ). By carefully tuning the two adjusters on each mirro r, original laser beam is aligned straight into the center of a size adjustable iris in the front of the spatial filter Then the input intensity noisy laser beam is first roughly filtered through the s mall iris. By slowly adjusting the X, Y adjusters on t he focusing optic translation mount in the middle of the spatial filter the laser beam i s directed to the center of an aspheric lens that attached to the optic al tube on the middle mount. A fiber that is mounted on the Z translator i s carefully positioned at the foc al point of the aspheric lens allowing laser beam to be focused into the pinhole on the fiber. This optic al fiber, which is a 405 nm single mode (SM), act s as a second spatial filter, allowing only the Gaussian profile of laser beam with wavele ngth 405 nm to transmit.

PAGE 80

80 Figure 2 7 Cross section s for laser beam in each optimization step. A beam of light c o me s out through the end of the fiber, where a fiberport collimator is well connected (Figure 2 8 B ). Incorporated with another aspheric lens, t he fiberport not only collimates but also expand s the laser beam. The cross sections for the laser beam in each optimized step are displayed in Figure 2 7 for comparison. A remarkable improvement in the laser beam quality is easily observed. Dichroic Mirro r Cube and Laser Alignment After the filtration and expansion, a collimated, horizontally straight and clean laser beam with a circular and symmetrical cross section, i s obtained. The final diameter for laser beam i s ~4.4 mm, which is more than four times larger than that of the original beam (~1 mm). This beam slightly underfill s the back aperture of the objective lens. The collimated laser beam then enter s the back port of an Olympus IX70 inverted microscope (Figure 2 8 ), reaching a 405 nm dichroic mirro r incorporated in the filter cube. This dichroic mirror i s designed so that it reflect s more than 90% of the incident light, whose wavelength is ~ 405 nm, while transmitting more than 90% of the light with

PAGE 81

81 wavelength above 405 nm. With an incidence angle of 45, the dichroic mirror help s to redirect the incident laser beam upwards to the water immersion objective lens (Figure 2 9 ). Figure 2 8 A) Inner structure of fluorescence microscope and B) photo graph for a part of the setup. F igure A is r e pr i n t ed with p er m i ss ion from O lympus 169 Figure 2 9 Diagram of the dichroic mirror cube.

PAGE 82

82 The beam axis needs to be aligned so that the reflected axis is coincident with that of objective lens. Before screwing in the microscope objective, the alignment needs to be c hecked by applying two piece s of paper adhesive tape on the top of round objective lens holder and the round hole in center of the sample holding plate If the laser beam has been well aligned, the reflected excitation spot s are centered on both round circ les. After the objective i s placed, the laser beam coming out should be locat e d on the center of the hole in the sample holding plate as well. Objective Lens and Focusing A small spot volume at the foc al point in the sample is critical for achieving sing le molecul e detection and reducing background noise in FCS measurements The factors that affect the full width half maximum (FWHM) of the diffraction limited excitation spot formed by the objective lens, can be found in the following equation: 170 (2 12) Where is the wavelength of light; NA is the numerical aperture of the objective lens used; and are the diameters of the laser beam and lens pupil, respectively. T he E quation 2 12 shows that shorter beam wavelength, higher objective NA or larger laser beam diameter would lead to smaller focus. Theoretically, once the types of laser and objective lens are fixed, only expanding the laser beam would shrink the foc al v olume Thus a beam expander is a necessity for most FCS systems. However, although overfilling the back aperture of the objective lens by the laser beam would form the smallest spot size with diffraction limited size, the intensity distribution inside the focal volume is complicated and may not be described by a simple model. A more desirable Gaussian intensity distribution in a small volume i s achieved by

PAGE 83

83 underfilling the back aperture of the objective, although trading off with a larger excitation volume. 171 Figure 2 1 0 Effect of objective media A ) oil immersion objective B ) water immersion objective R e p rin ted wi t h per m iss ion from Olympus 172 As one of the most critical components of the system, objective lens should be selected carefully. A higher NA of the objective lens is desirable for produ cing a small beam waist (Equation 1 12) as well as efficiently collecting signals. Most of the oil immersion objective lenses have NA as high as 1.5. However, a mismatch of the refractivity of the medium and sample would cause deviation of the light rays. 172 For instance, using an oil immersion obj ective lens (refracti ve index ~ 1.5 for oil medium) in aqu eous sample (refractive index ~ 1.33) detection may limit the position of focus point quite close to the bottom glass of sample container would jeopardize the quality of an FCS measurem ent (Figure 2 1 0 ). Moreover the working distance of the lens become s smaller as its NA becomes lager Since most of our measurements are done in aqueous solution, a 60 water immersion objective lens with NA = 1.2 is suitable. According to the working di stance of our objective lens, the excitation volume i sample interface 173 which is

PAGE 84

84 Signal Collection and Detection The fluorescence i s collected by the same objective and separated from the incident laser line through the dichroic mirror. After passin g a long pass 420 nm filter in the bottom of the dichroic mirror cube (Figure 2 9 ), the emission signal exit s the microscope through reflection by a 50/50 optic splitter (Figure 2 8 A ). A band pass filter i s installed next to the outport (Figure 2 1 1 A). F igure 2 1 1 Photo graphs for signal detection and correlation components of FCS setup. The filtered fluorescence i s focused onto a pinhole, with on one end of a multimode optical fiber (Figure 2 1 1 A ). The combination of the confocal optics components collimator, dichroic mirror, objective lens and pinhole help reduce out 1 2 ). 174 An XYZ travel translation stage i s used to position the pinhole precisely (Figure 2 1 1 A ) The desirable detector for FCS is a sensitive single photon counting avalanche

PAGE 85

85 photodiode (APD) module, which has high quantum efficiency ( 50%) between 500 nm and 700 nm which is the typical emission range for CPE. In this device, all the photon signal s are converted to electronic signal s and are amplified (Figure 2 1 1 B ). Then the signals are fed into a correlator (Figure 2 1 1 D ) where the autocorrelation function is calculated The best result i.e., smoothest autocorrelation curve, i s obtained when th e pinhole i s correctly positioned. Figure 2 1 2 Confocal volume and optics. confocal volume passes through the pinhole. Others (yellow and blue) are blocked by the optics. R e pri nte d wi th per mi ssio n from fcsxpert.com. 174 Potential Interfe rence To obtain higher signal to noise ratio and protect the detector, performance of FCS experiment in the dark is a requirement. Since the fluorescence fluctuation of the molecules is our measuring target, any external interference, such as vibration, needs to be avoided. Therefore, a stable heavy breadboard table, a black cover in the detection part and a quiet working place are preferred (Figure 2 1 3 ). The laser power is controlled around 1 mW through software in computer (Figure 2 1 1 C ). Below that va lue, lack of intensity will affect the formation of the ellipsoid like confocal volume as well as the signal detection resulting in a poor FCS curve.

PAGE 86

86 However, if the laser power intensity is too high, say 10 times higher, photobleaching or photon saturati on occurs which also affect s the quali ty and accuracy of the results. Figure 2 1 3 Black cover and breadboard for FCS setup. Photobleaching is a problem which will cause an erroneously fast diffusion due to the f ading of the fluorescence occur s before e xpected. In FCS, the photobleaching is eas ily observed on the fluctuation profile. The photobleaching induced decreas e in the fluorescence intensity cause s a distortion on the FCS curve U sually a second FCS curve with longer diffusion time w ill show up. The recording of the fluctuation should be restarted until the stable fluorescence is obtained. It is rather important to monitor the fluorescence intensity change be fore saving the final FCS data. Calibration FCS can provide reliable information about dif fusion coefficient and concentration values when the calibration is done properly A stable and bright dye with a known coefficient is required for cali brati ng FCS system. P ossessing high quantum efficiency

PAGE 87

87 and large absorption cross section, the calibrati on dye must also have high photostability, which enables it to withstand the enormous power in the laser focus. In traditional FCS system, fluorescein, r h odamine and cyanine are the most popular calibrating dye In our system, the dye must have absorbance peak extend to 405 nm, and have emission spectrum co vering 450, 500, 600, or 650 nm. In this dissertation, two emission filter s at 500 nm and 600 nm are used for most of the CPEs having emissions cover either of the two wavelengths With emission at ~ 500 n m, f luorescein is considered to be a good calibrating dye Although its absorbance at 405 is relatively low, fluorescein is a pH sensitive molecule (Figure 1 2 ). A minor increas e of pH to 8 will enhance the light absorption. Due to its high photostability and low container adsorption, 175 fluorescein tolerates a relatively long focusing time ( > 10 min) under the laser beam For 600 nm emission system, although tetrameth ylrhodamine (TMR) suffers from low absorbance around 400 nm, however, the relatively high quantum yield enable s it to emit relatively high intensity a t 600 nm. The 2D model ( E quation 2 3) i s first applied to fit the FCS curve for a standard dye The value for structure parameter i s obtained and then it is substituted in the 3D model ( E quation 2 2) to characterize the other FCS curves. Calibration of system with standard dye should be done routinely before measurement and a recalibration is needed if the confocal volume is possibly changed for some reason, e.g., laser turn ed off or the optic elements realigned For a stable system, the brightness or count rate is almost consistent for a standard dye with fixed concentration. Based on this, the routine meas urement of dye sample not only helps calculating the size of confocal volume, but also keeps track on the quality of system.

PAGE 88

88 Examination of FCS System V arious test experiments were conducted to exam ine the performance of the new instrument. The chemical structures of the compounds and polymers used in the work are shown in Figure 2 14 along with the acronyms that are used in the text. TMR and fluorescein in aqueous solution w ere used for calibration of the system with emission filters centered at 590 nm and 5 0 0 nm respectively T he effective volume obtained for the confocal volume is ca 0.5 1 femtoliter with z in micrometers and r in sub micrometers. Based on the calculation method described above, t he structure parameter s are found to be 11 for emission at 590 nm, and 45 for 500 nm emission Figure 2 1 4 Structure of molecu les that used in calibration and their acronyms

PAGE 89

89 Figure 2 1 5 Correlation curves for TMR in water. From top to bottom the TMR concentrations were 5, 10, 25, 50, and 100 nM. The insert shows the measured average number of molecules. A series of TMR sample s with var ying concentration s was tested with emission detection at 590 nm and the results are shown in Figure 2 1 5 As expected, due to the inverse relationship between G (0) and the number of molecules in the confocal volume, a clear decrease of the corre lation curve amplitude is observed as sample concentration is increased The inset plot indicates that the number of molecules N, equal to 1/ G (0), is linear with the increase in concentration A second test of the FCS setup was conducted by measuring the d iffusion time of samples with different MW. In the 590 nm emission system (Figure 2 1 6 A ), the experimentally observed diffusion time for TMR (386 Da) with a known diffusion coefficient 2.8810 10 m 2 s 1 is 45.0 3.5 Further FCS measurement s w ere carried out on three samples, biotin TMR (869 Da), TMR labeled DNA chain (~19,300 Da), and biotin TMR/avidin ( [TMR]/[avidin] = 1:10, ~67,000 Da). The resulting diffusion times are 75.4 0.6 16 5 2 an d 294 12 s respectively Based on E quation 2 4, the measured diffusion coefficients for each species are (1.72 0.17)10 10 (0.79

PAGE 90

90 0.09)10 10 and (0.44 0.05)10 10 m 2 s 1 The diffusion coefficient of biotin TMR (1.72 0.17)10 10 m 2 s 1 is clo se to the value of 1.6710 10 m 2 s 1 as previously reported by our group using a different FCS system (Chapter 3 ) 176 The diffusion coefficient of avidin/biotin TMR (0.44 0.05)10 10 m 2 s 1 compares well with the value 0.410 10 m 2 s 1 reported for streptavidin 177 178 and 0.3910 10 m 2 s 1 for avidin / biotin TMR ([avidin]:[biotin]=1:4, 69,000 Da) 176 For those molecules/complex, the diffusion time increases with the increasing of the molecular wei ght. Assuming that they exhibit approximate spherical structure and the conditions for the highly diluted aqueous solvents (viscosity and temperature) are the same based on the E quation 2 8 the hydrodynamic radi i R H s which can be interpreted as an effe ctive radii, are estimated to be 0.85 1.42 0.01 3. 12 0.04 and 5.55 0.03 nm for TMR, biotin TMR and biotin TMR/avidin, respectively. The hydrodynamic radius of avidin with biotin TMR 5.55 0.03 nm is larger than the value 4 nm reported for avidin b iotin complex in the literature, where the complex was modeled as a steric sphere having the same molecular volume. 179, 180 Consider ing the non spherical nature of avidin and the fact that R H is the apparent size of the dynamic hydrated/solvated particle, the difference in the data is within acceptable limits In the 5 00 nm emission system (Figure 2 1 6 B ), fluorescein (332 Da) with diffusion coefficient 3.00 10 10 m 2 s 1 has the experimental diffusion time 23.5 3.3 The o ligomer PE CO 2 (2,302 Da), PPE PEG d CO 2 ( ~ 1 1 ,000 Da ), and PPE PEG d CO DNA (~49,000 Da) were also investigated Their diffusion times from shortest to longest as shown in F igure 2 are, 39 2 4 0, 54.8 3.3 112 3 The corresponding diffusion coefficients are ( 1.80 0.24) 10 10 (1.29 0.14)10 10 and ( 0.63 0.09 ) 10 10 m 2 s 1

PAGE 91

91 Although it i s difficult to model accurately the conformation of these polymer chain s however, we can still obtain an idea of their size by making the same assumption described above and applying E quation 2 8 again. The result ing R H s are 0.8 2 1.3 7 0.18 1.91 0.19 and 3. 9 1 0.07 nm for fluore s cein, PE CO 2 PPE PEG d CO 2 and PPE PEG d CO DNA, respectively. Figure 2 1 6 Effect of molecular weight (MW) on the co rrelation curves of molecules A ) usin g 590 nm emi ssion filter B ) using 500 nm emission filter. T he black solid lines are the single species fitting curves. A plot of diffusion coefficient versus molecular weight is displayed in Figure 2 1 7 A simple fitting model (red line) indicates an approximately inverse cub ic root relationship between the diffus ion coefficient ( D ) and the molecular weight ( MW ) (2 13) W hen compared to E quation 2 11 the difference in the power value ( 0.30 vs 0.33) is result ed from a rough assumption of spherical shape we made for the rigid rod polymer. However, the results here are still consistent with t he former conclusion a higher molecular weight and larger particle size lead to longer diffusion time. Since the testing

PAGE 92

92 results for the FCS system with 405 nm blue laser are reasonable, this instrument is reliable to be used in the following projects. Figure 2 1 7 Plot of diffusion coefficient of four standard samples as a function of their molecular weight (emission filter 500 nm). Red line is th e fitting curve. Summary An FCS with 405 nm laser as light source has been successfully constructed and the details about the alignment, optimization and calibration of the setup are provided The examination of the newly built FCS setup further proved the feasibility of this instrument. By coupling with 405 nm laser, the application of FCS is successfully exten ded to the shorter wavelength region With assistance from FCS, more insight and information regarding the CPEs properties from a microscopic motion aspect can be obtained which is not supposed to be achieved via conventional technologies.

PAGE 93

93 Materials and M ethods Materials Oligomer PE CO 2 181 and PPE PEG d CO 2 182, 183 were synthesized and prepared according to the reported procedures. PPE PEG d CO 2 was conjugated with ssDNA with A10CCCAATCACTAA PEG d CO DNA using the synthesis method in the lit erature. 184 The loading ratio was ~5 ssDNAs per polymer chain. M olecular weight s of oligomer PE CO 2 PPE PEG d CO 2 and PPE PEG d CO DNA are ca 2,3 0 0, 11,000 and 49,000 Da respectively Tetramethylrhodamine (TMR) was purchased from AnaSpec, Inc. Fluorescein was purchased from Fisher. B iotin TMR ( 5 (and 6) Tetramethylrhodamine biocytin) was purchased from Invitrogen TM. Avidin was purchased from Sig ma. DNA TMR was prepared according to the literature. 185 All sample solutions were prepared using water distilled and purified by a Millipore purification system (Millipore Simplicity Ultrapure Water System). Buffer solutions were prepared with reagent grade materials (Fisher). All polymer concentrations are reported as the polymer repeat unit concentration (PRU). Concen trated s tock solutions of bio tin TMR and avidin were prepared in 10 mM phosphate buffer (pH 7) to obtain the desired concentrations. Fluorescein was prepared in 10 mM phosphate buffer (pH 8 ). FCS Component and Measurement. The measurements were performed i n the setup constructed in house (Figure 2 6 ). The Olympus IX70 epi fluorescence microscope platform and objective lens ( 60 NA 1.2, water immersion ) were purchased from Olympus The 405 nm diode laser (CUBE) was purchased from Coherent The 405 nm single mode fiber 50 m inner diameter multi mode optical fiber spatial filter system (KT110), fiberport collimator 500 20 nm and 590 20 nm band pass filter and XYZ stage were purchased from

PAGE 94

94 Thorlabs The dichroic mirror cube was purchased from Chroma. T he avalanche photodiode (APD, SPCM AQR 14 FC) was purchased from Perkin Elmer The c hambered coverglass (Thermo Scientific, Nunc, Lab Tek) was purchased from Fisher. The correlator (Flex02 12D) was purchased from correlator.com.

PAGE 95

95 CHAPTER 3 THE APPLICATION OF DYE LIGAND INTERCALATED HELICAL CONJUGATED POLYELECTROLYTE ON P ROTEIN SENSING Background I n previous work reported by our group, we have demonstrated from an optical spectroscopic study that a meta linked poly(phenylene ethynylene) ( P oly 1 Figure 3 1 ) can self assemble into a helical conformation stabilized by the stacking interactions between the phenyl rings with the ionic side chains extending to the surrounding polar solvent 23 Due to the structural similarity between Poly 1 and double stranded DNA (dsDNA) (e.g., the stacking helical bac kbone is analogous to stacked base in dsDNA; while the negatively charged side chains resemble the phosphate groups in dsDNA), some DNA intercalators featuring planar aromatic structures exhibit identical spectroscopic effects upon binding to P oly 1 as they intercalate to dsDNA. 186 188 For example, mixing Ru(bpy) 2 (dppz) 2+ with Poly 1 leads to large enhancement of the luminescence from the Ru complex metal to ligand charge transfer state. The phenomeno n is similar to that of intercalation of the Ru dppz ligand into the double heli x in dsDNA. 187 Avidin, a protein obtained from egg white, is toxic to many organisms due to its ability to deplete biotin, an essential vitamin (Vitamin H), from their environment s 189 The binding between a vidin and biotin is extraordinarily stable and essentially irreversible. 189 This association has been exploited as a versatile tool for broad applications in biochemistry, 190, 191 including enzyme li nked immunosorbent assay (ELISA), 192, 193 cell surface labeling, 194 and affinity purification. 195

PAGE 96

96 This chapter focuses on the interaction between helical Poly 1 and a dye ligand compound, biotin TMR (biotinylated rhodamine, 5 (and 6) tetramethylrhodamine biocytin Figure 3 1 ) as well as the influence on their interaction brought by avidin. The fluorescence resonance energy transfer ( FRET ) is observed when biotin TMR and Poly 1 are mixed together, and it is not interrupted when avidin i s present. However, when a pre formed avidin/biotin TMR is added into the Poly 1 buffer solution, disruption of energy transfer is observed, which gives rise to a novel FRET sensor for avidin with a detection limit for avidin of 100 pM. Interestingly, a sy stematic investigation on the fluorescence spectra of Poly 1 mixed with pre formed avidin/biotin TMR at various [avidin]:[biotin] ratios reveals the so called phenomenon steric constraint A n FCS with a 543 nm laser, which typically excites TMR, is assiste d to investigate the system. The intercalation of TMR to Poly 1 and the formation of supramolecular aggregat es by Poly 1/biotin TMR in the presence of avidin are demonstrated. The formation of aggregat es also has been evidenced by atomic force microscopy ( AFM) which well explains why the direct addition of avidin to polymer/biotin TMR solution cannot interrupt the FRET. Another new avidin sensor is proposed bas ed on the FCS curve changes with sensitivity < 100 pM. Results and Discussion FRET Study of Helic al CPE/dye ligand with Protein Photoph ys ical p roperties of Poly 1 and b iotin TMR As shown in Figure 3 2 in aqueous phosphate buffer, Poly 1 absorbs in the near UV (323 nm) and it exhibits a broad, structure less emission band with max ~ 450 nm, which aris es due to interactions between the phenylene ethynylene units that are in close proximity in the helix. The TMR chromophore in biotin TMR absorbs at max ~ 554

PAGE 97

97 nm (more photophysical data are summarized in Table 3 1). The emission of Poly 1 and absorpti on of biotin TMR overlap well as shown in Figure 3 1, satisfying the prerequisite for FRET. The electrostatic attraction, as well as the hydrophob ic interactions between anionic Poly 1 and net positively charged biotin TMR, brings these two potential donor and acceptor in close proximity, typically < 0.1 nm, where FRET is permitted. 35 Poly 1 Biotin TMR Poly 2 Figure 3 1. Structure of polymers and dye ligand compound Figure 3 2 Absorption ( A bs) and emission ( E m) spectra of Poly 1 and biotin TMR.

PAGE 98

98 Table 3 1. Photophysical properties of TMR Poly 1 and biotin TMR in 10 mM phosphate buffer sol ution (pH 7.4) Reprinted with permission from Wu, D. L. et al 176 ( nm ) ( nm ) ( 10 4 M 1 cm 1 ) TMR 548 575 8.39 0.03 Poly 1 32 3 450 3.29 0.07 B iotin TMR 55 4 5 8 0 7. 04 0 .1 2 FRET from Poly 1 to TMR Figure 3 3 Normalized emission spectra for titration of 0 0.3 M biotin TMR into 15 M Poly 1 in 1 mM phosphate buffer solution, pH = 7. 4. Figure 3 3 illustrates the changes in fluorescence when various concentrations of biotin TMR are titrated to an aqueous buffer solution of 15 M Poly 1. The fluorescence of Poly 1 at 450 nm is quenched, and a strong emission band from TMR appears at ma x ~ 590 nm. Importantly, under the same excitation conditions, the fluorescence from biotin TMR alone is very weak (Figure 3 4 ), indicating that the FRET takes place from Poly 1 to TMR Fluorescence anisotropy spectra (Figure 3 5 ), used to characterize the molecular motion via emission polarity, show that in the presence of polymer, the rotational motion of biotin TMR is restricted in a certain orientation with anisotropy

PAGE 99

99 values varying from 0.05 to 0.15, indicating the biotin TMR locates in a confined area presumably the helix of Poly 1, where depolarization is difficult to occur. Figure 3 4 ex = 320 nm) of biotin TMR (225 nM) in the absence ( ) and presence ( --) of P oly 1 (15 buffer Reprinted with permission from Ji. E., et al 30 Figure 3 5 Fluorescence anisotropy spectra for biotin TMR with and without Poly 1. Reprinted wit h permission from Ji. E. 196

PAGE 100

10 0 Addition of a vidin to Poly 1/biotin TMR As reported by Whitten and co workers, in a PPV/MV 2+ biotin quenching system ([PPV] = 1.7 10 5 M, [MV 2+ biotin ] = 2 10 6 M), the addition of only 2 10 7 M avidin can restore the fluorescence of PPV by 50%, due to the interruption on the charge transfer path from exciton to quencher. 43 So in this section, the influence avidin brings to the Poly 1/biotin TMR system, especially to the FRET process, is under investigation. In Figure 3 3 as the [biotin TMR] reaches 0.30 M, the FRET effect is approaching saturation. So here, the solution of Poly 1/biotin TMR with [Poly 1] = 15 M and [biotin TMR] = 0.225 M is used for avidin titration. As shown in Figure 3 6 t he fluorescence of biotin TMR is slightly quenched by avidin; however, recovery of Poly 1 fluorescence does not occur. This result suggests that avidin is unable to disrupt the FRET from Poly 1 to TMR indicating that the rhodamine chromophore remains highly interacting with, possibly intercalat ing into Poly 1 F igure 3 6 Fluorescence spectra of Poly 1 solution ( ) upon addition of biotin TMR ( ) and avidin ( --). [Poly 0.065 ex = 320 nm. Reprinted with permission from Ji. E. 196

PAGE 101

101 Tit ration of p reformed a vidin/biotin TMR to Poly 1 Figure 3 7 Normalized emission spectra for Poly 1 mixed with preformed avidin/biotin TMR at various [avidin]/[biotin TMR]. [Poly 1] = 15 M, [biotin TMR] = 225 nM. Titration is done in 1 mM phosphate buffer solution, pH 7.4, room temperature. By contrast, a discontinuous titration of pre mixed avidin/biotin TMR to the polymer solution was conducted. Initially, v arious amounts of avidin were mixed with identical amount of biotin TMR in vials. After ~10 min incubation, the mixtures were added into 15 uM Poly 1 individually with final [biotin TMR] fixed at 0.225 M Figure 3 7 shows the emission spectra of Poly 1 solutions with varying [avidin] :[biotin TMR] in the range of 0 10. Interestingly, as [avidin]:[biotin TMR] increases, the emission of TMR first reduces gradually followed by an increase and then decreases again. A similar trend in spectra change occurs with 5 M Poly 1 mixed with pre formed avidin/biotin TMR ( final [biotin TMR] = 100 nM). The emission spectra are converted to a graph where the ratio of the fluorescence intensity at 590 nm and 450 nm (I 590 /I 450 ) is plotted as a function of log

PAGE 102

102 [ avidin ] (Figure 3 8 ). Impressively, the mi nimum FRET occurs at [avidin]/[ biotin ] = 0.25 for both systems with different polymer concentration indicating that the 1:4 avidin biotin TMR complex is least able to interact with and undergo FRET from Poly 1 This ratio value corresponds to the four bin ding sites on avidin for biotin, 189 revealing the fact that steric constraints may prevent intercalation of the avidin bound biotin TMR to Poly 1 Figure 3 8 Ratio of intensities at 590 nm and 450 nm after addition of pre mixed biotin avidin complex at variou s avidin concentrations in phosphate buffer (1 mM, ex = 320 nm. For [ biotin TMR ] = 100 nM, the analytical detection limit of avidin is 100 pM. Reprinted with permission from Ji, E., et al 30 Figure 3 9 illustrates the possible mechanism for the complicated dependence of FRET on the [avidin] / [ biotin TMR ] ratio. First, for [avidin]/ [biotin TMR] < 0.25, there are not enough binding s ites for biotin in avidin so excess biotin TMRs are able to intercalate to Poly 1, maintaining partial FRET response Increase in [avidin] decreases the available biotin TMR thus the FRET events are reduced to a minimum However, when [avidin]/ [biotin T MR] > 0.25, all of the biotin is bound to avidin, but because the

PAGE 103

103 protein is in excess, the avidin /biotin TMR complexes appears with less than four biotin TMR binding ligands A less bulky avidin /biotin TMR complex may be able to more effectively interact with Poly 1 perhaps allowing intercalation of the TMR chromophore thus slightly recovering the FRET signal. Figure 3 9 Binding of p reformed avidin/biotin TMR c omplex to Poly 1 as a f unction of a dded a vidin c oncentration Reprinted with permission from Ji, E., et al 30 FCS Study on the Poly 1/biotin TMR/avidin System In order to obtain more details and insights about this system, we empl oyed FCS in the study. The FCS system we used in this work was built by the Dr. Weihong Tan s group at the University of Florida, Department of Chemistry (Figure 3 10 ). The system couples with a 543 nm green laser, which excites the TMR dye. In this sectio n, all of the FCS measurements are focusing on the diffusion behavior of biotin TMR and its complex. Although Poly 2 ) and

PAGE 104

104 FCS correlation signal (Figure 3 11 B ), it can be m olecules, so that any diffusion behavior change of the polymer can be followed by monitoring the fluctuation of the TMR Figure 3 10 Basic setup for FCS with 543 nm Laser Figure 3 11 Background detection for A) phosphate b uffer (10 mM, pH 7.4) B ) P oly 1 (1 M) in phosphate buffer (10 mM, pH 7.4) Reprinted with permission from Wu, D. L., et al 176

PAGE 105

105 Diffusion b ehavior of f our t ypes of m olecules/ c omplex Figure 3 1 2 Normalized correlation functions of biotin TMR P oly 1/ biotin TMR biotin TMR /avidin, an d Poly 1 / biotin TMR /avidin with [ avidin ]/[ biotin TMR ] = 0.25 in phosphate buffer solution (10 mM, pH 7.4). The black solid lines are single species fitting curves. Reprinted with permission from Wu, D. L. et al 176 T he diffusion time of aforementioned molecules and complex were first measured Using u ltrac entrifugation, the Poly 1/biotin TMR complex was prepared with a molecular weight between 30 kDa and 10 kDa The FCS experiment al results are displayed in Figure 3 12 and Table 3 2 B iotin TMR mixed with Poly 1 (Poly 1/biotin TMR) exhibits a significantly longer diffusion time ( 0.79 ms) compared to free biotin TMR molecules (0.29 ms) in neutral buffer solution In contrast, a para linked poly(poly(phenylene ethynylene) (Poly 2, Figure 3 1), which features a similar repeat unit structure as Poly 1, but adopt s non helical in buffer solution, was mixed with biotin TMR in a control experiment. The diffusion time for Poly 2/biotin TMR (0.27 ms) is almost the same as the value observed for free dye molecules (0.29 ms) (Figure 3 1 3 ). The result clearly shows that t he slower diffusion for Poly 1/biotin TMR is caused by intercalative binding

PAGE 106

106 instead of electrostatic attraction between Poly 1 and biotin TMR. When biotin TMR binds to a larger molecule of avidin (MW = 66 kDa) at a stoichiometric concentration, a longer d iffusion time of 1 21 ms (Figure 3 12 ) is obtained Figure 3 1 3 Normalized correlation functions of biotin TMR Poly 2/biotin TMR Poly 2/biotin TMR /avidin and biotin TMR /avidin with [ avidin ]/[ biotin TMR ] = 0.25 in phosphate buf fer (10 mM, pH 7.4). Reprinted with permission from Wu, D. L. et al 176 Table 3 2 Diffusion data for four species in phosphate buffer (10 mM, pH 7.4) Reprinted with permission from Wu, D. L. et al 176 Item B iotin TMR Poly 1 / biotin TMR B iotin TMR /avidin ( [ avidin ] /[ biotin ] = 0.25 ) Poly 1 / biotin TMR /avidin ( [ avidin ] /[ biotin ] = 0.25 ) Diffusion t ime D (ms) 0.29 0.02 0.79 0. 08 1.21 0. 20 11.50 3.40 Diffusion coefficient D ( 10 10 m 2 / s) 1.67 0.14 0.61 0. 07 0.39 0.0 8 0.0371 0.0164 Estimated MW ( k Da) 0. 869 10 .0 30 .0 69.6 NA Calculated MW ( k Da) 0.869 18.2 3.2 66.3 14.8 (79 .3 15.2) 10 3

PAGE 107

107 Figure 3 1 4 Photon counting rate (fluorescence fluctuation) during the detection time (1200 s) for P oly 1 / biotin TMR and Poly 1 / biotin TMR /avidin ( [ avidin ]/[ biotin TMR ] = 0.25 ) in phosphate buffer (10 mM, pH 7.4). Reprinted with permis sion from Wu, D. L. et al 176 The diffusion coefficient ( D ) scales approximately as the inverse of the cube root of the molecular weight ( MW ) 35 Due to the inversion relationship between diffusion ti me ( ) and D ( E quation 2 4), we have ( 3 1 ) The approximate cube root dependence of the on MW provides a way to roughly estimat e the molecular weight of each species T he FCS based calculation results displayed in Table 3 2 are consistent to the values estimated on the basis of the ir chemical structure A significantly longer diffusion time is observed for Poly 1 /biotin TMR in the presence of the avidin with [ avidin ]/ [biotin TMR] = 0.25 suggesting the formation of very large aggregates. The fluorescence time trajector y ( c ount rate) of Poly 1/biotin TMR/avidin show s pronounced spikes on a lower baseline (Figure 3 1 4 ), which are attributed to the large particles formed by the protein and polymer, passing through t he excitation volume at a slow rate It must be noted that the large molecular

PAGE 108

108 weight (79.3 15.2) 10 3 k Da calculated for the diffusing particle based on E quation 3 1 is only a rough estimation due to complex structure and size heterogeneity of the larg e aggregates. However, it still can be evidence for the large size of the aggregati on, in clear contrast with the much smaller molecular weight of the Poly 1/biotin TMR complex in the absence of avidin Mechanism for f ormation of s upramolecular a ggregation The possible mechanism for this phenomenon is illustrated in Figure 3 15 In aqueous solution, more than one biotin TMR binds to helical Poly 1 by intercalation of funct four biotin binding sites is added biotinylated biotin binding. The explanation for the cross linked aggregates with avidin as the brid ge, bears some analogy to the one reported by Sleiman and co workers that the biotin Ru dppz intercalated dsDNAs are cross linked through streptavidin functionalized gold nanoparticles. 197 Figure 3 15. Proposed m ech anism of protein induced aggregation Reprinted with permission from Wu, D. L. et al 176 This interpretation of the findings is further corroborated by AFM images of dry films deposited from dilute solutions of Poly 1, Poly 1/biotin TMR and Poly 1/biotin TMR/avidin measured in air individually As illustrate d in Figure 3 1 6 deposited

PAGE 109

109 particles for Poly 1 and Poly 1/biotin TMR are well spread out with comparatively small heights ~4 and 12 nm, respectively. In sharp contrast, the Poly 1/biotin TMR/avidin produces much fewer and considerably larger particles (h eight ~200 nm) in a comparable area, which further suggests the formation of large aggregates. Figure 3 1 6 AFM images for A) pure Poly 1 B) Poly 1/biotin TMR C) Poly 1/biotin TMR/ avidin ( [ avidin ]/[ biotin TMR ] = 0.25 ) Line scans for D ) pure Poly 1 E ) Pol y 1/biotin TMR F ) Poly 1/biotin TMR/ avidin ( [ avidin ]/[ biotin TMR ] = 0.25 ). Different colors represent different particles. Reprinted with permission from Wu, D. L. et al 176 Control e xperiment Considering that avidin ( p I > 10) 189 is positively charged in pH 7.4 buffer solution a series of control experiments wer e conducted to detect the possible non specific interaction between avidin and the anionic polymer The first experiment was carried out on Poly 1/biotin TMR with and without avidin in 10 mM pH 10.5 buffer where avidin is overall negatively charged A sli ghtly shorter diffusion time (~ 0.69 ms) is observed for Poly 1/biotin TMR. As discussed in the Chapter 1 due to the partial protonation of the

PAGE 110

110 ionic side chains CPE subject s to an aggregation behavior as pH decreases from 12 to 7. So here, the shorter di ffusion time presumably corresponds to the smaller Poly 1/biotin TMR complex resulting from the de aggregation of Poly 1 in a higher pH solution. Nevertheless a considerabl y larger diffusion time (~ 8.71 ms) is observed with addition of avidin, indicating the large supramolecular aggregate is still formed, regardless of the charge type of avidin (Figure 3 1 7 ). Figure 3 1 7 Normalized correlation functions of biotin TMR Poly 1/biotin TMR and Poly 1/biotin TMR/ avidin with [ avidin ]/[ biotin TMR ] = 0.25 in p hosphate buffer solution (10 mM, pH 10.5). The black solid lines are single species fitting curves. Reprinted with permission from Wu, D. L. et al 176 Further control experiments were carried out to test whether the biotin avidin interaction is essential to the aggregation behavior of the complex. This was a chieved by replacing biotin TMR with TMR that is still able to intercalate into the Poly 1 but has no binding interaction with avidin (Figure 3 1 8 ). Parallel FCS experiments were conducted in both pH 7.4 (Figure 3 19A) and pH 10.5 (Figure 3 19B) solutions. Not

PAGE 111

111 surprisingly, t he addition of avidin to the two Poly 1 /TMR assaying solution s does not cause a significant change in the FCS curves, except a slight increase in the diffus ion time for the Poly 1/TMR/avid in in pH 7.4 buffer solution (~ 0.89 ms) (Figure 3 1 9 A ) which arises from non specific interaction between the polymer and protein However, the effect is negligible when compared to that of the binding between avidin and biotin The AFM images of deposited Poly 1/avidin and Poly 1/TMR/avidin are shown in Figure 3 20 Consistent with FCS results, no large particles are observed. The conclu sion can be derived from th e above finding s that the non specific interaction between avidin and polymer is not strong enough to enable the formation of huge aggregatio ns and cause the substantial change in diffusion time. Since the pH dependent non specific interactions are weak and negligible, additional work regarding the avidin assay was performed in buffers with pH 7.4. Figure 3 1 8 Normal ized correlation functions for ligand free TMR with and without avidin in 10 mM phosphate buffer Reprinted with permission from Wu, D. L. et al 176

PAGE 112

112 Figure 3 1 9 Normalized correlation curves for Poly 1 /TMR and Poly 1 /TMR/avidin in 10mM phosphate buffer A ) pH 7.4 ([Poly 1] = 200 n M, [TMR] = 13.6 nM, [avidi n] = 3.4 nM) B ) pH 10.5 ([Poly 1] = 200 nM, [TMR] = 9.2 nM, [avidin] = 2.4 nM). Reprinted with permission from Wu, D. L. et al 176 Figure 3 20 AFM images for A) Poly 1/avidin and B) Poly 1/TMR/avidin ([avidin]/[TMR] = 0.25) Line scans for C) Poly 1/avidin and D) Poly 1/TMR/avidin. Different colors repre sent different scans. Reprinted with permission from Wu, D. L. et al 176 In a negative control experiment, t he same amount of protein BSA ( p I = 4.7 33 ), which has a similar molecular weight of 66 kDa with avidin but has no binding sites for

PAGE 113

113 biotin was added into the Poly 1/biotin TMR in pH 7.4 buffer. No significant change is oberved in the FCS curves (Figur e 3 21 ). All the findings shown above prove that the aggregation/crosslinking in Poly 1/biotin TMR /avidin system is the result of the intercalation of TMR into the polymer and the binding between biotin and avidin. Figure 3 21 N ormalized correlation functions for biotin TMR Poly 1/biotin TMR and Poly 1/biotin TMR /BSA with [BSA]/[biotin TMR] = 0.25 in phosphate buffer (10 mM, pH 7.4). Black solid lines are single species fitting curves. Reprinted with permission from Wu, D. L. e t al 176 Avidin sensing strategy The avidin induced a ggregation of P oly 1/ biotin TMR gives rise to the sensing strategy of avidin. Concentration dependent experiments were performed by adding varying amounts of avidin into the Poly 1/biotin TMR solution (Figure 3 22 [Poly 1] = 102.8 nM, [biotin TMR] = 7 nM ) When [ avidin ]/[ biotin TMR ] is lower than 0.1, no significant change in the diffusion time is observed. However, a noticeable distortion in the correlation curve is observed when [ avidin ]/[ biotin TMR ] reaches 0.1 As [avidin] / [ biotin TMR ] increasing from 0.1 to 100 the correlation curves shift largely to a

PAGE 114

114 longer diffusion time Only at a ratio of 0.25 the well defined correlation function is obtained indicating a relative ly homogeneous distribution in the size of aggregation possibly due to the 4:1 st oichiometry between biotin and avidin. For other ratio s especially at higher [ avidin ]/[biotin TMR] the correlation curves become complicated and cannot be simply fit by E quation 2 2, suggesting an inhomogeneous distribution of the aggregate size. Figu re 3 22 Normalized correlation functions for biotin TMR ( ), Poly 1/biotin TMR ( ), and Poly 1/biotin TMR /avidin with [ avidin ] /[ biotin TMR ] equal to 0.0 1 ( ), 0.02 ( + ), 0.1 ( ), 0.25 ( ), 0.5 ( ), 1 ( ), 10 ( ) and 100 ( ) in phosphate buffer (10 mM, pH 7 .4) Reprinted with permission from Wu, D. L. et al 176 Based on the analysis above, for our system, the critical [avidin]/[biotin TMR] when avidin is detectable is 0.1. A Poly 1/biotin TMR complex with [Poly 1] = 14.68 nM and [biotin TMR] = 1 nM was further detected in the presence of 100 pM avidin. A corre lation curve for Poly 1/TMR/avidin with the same concentration of Poly 1 and

PAGE 115

115 avidin was also recorded for monitoring the non specific interaction between Poly 1 and avidin. As shown in Figure 3 23 there is a slight non specific interaction between Poly 1 and avidin ( Poly 1/biotin TMR ~ 0.79 ms, while Poly 1/TMR/avidin ~ 0.85 ms), however, the effect is quite limited. A relatively large shift for Poly 1 / biotin TMR/avidin is well detectable, even at such a low concentration A detection limit less than 100 pM for avidin is expected for our sensing system, which is lower than most of the values reported for avidin sensing. 26, 198 201 Figure 3 2 3 Aut ocorrelation FCS curves for Poly 1/biotin TMR, Poly 1/TMR/avidin an d Poly 1/biotin TMR/avidin in phosphate buffer (10 mM, pH 7.4). [Poly 1] = 14.68 nM, [ biotin TMR ] = 1 nM, [avidin] = 0.1 nM are fixed. Reprinted with permission from Wu, D. L. et al 176 Conclusion biotin fluorescent sensor is first deve loped The sensor response is based on FRET between an intercalated dye and the helical CPE, and interruption of this process by the

PAGE 116

116 preformed avidin biotin interaction. A systematic investigation on the fluorescence spectra of CPE mixed with pre formed av idin/biotin dye reveals the so called phenomenon steric constraint. R el ying on molecular recognition/hydrophobic interactions, a FRET based discontinu ous biosensing method is developed with a detection limit as low as 100 pM for the target protein. The int ercalative interaction between dye biotin and helical CPE as well as the formation of sup ra molecular aggregation of helical CPE/dye biotin/avidin is further demonstrated by FCS and AFM images. A novel avidin sensing system is developed with signal response based on the diffusion behavior change of the CPE/dye biotin in the presence of the target molecule avidin as [avidin]/[biotin] > 0.1. The sensitivity is < 100 pM, which is better than most of other methods reported. The two new biosensing strategies des cribed in this chapter provide platform s for establishment of a highly sensitive biosen sing system. Experiments and Materials Materials The synthesis of Poly 1 and Poly 2 i s described in the literature. 23, 24 B iotin TMR ( 5 (and 6) t etramethylrhodamine biocytin ) was purchased from Invitrogen. Avidin was purchased from Sigma. All sample solutions were prepared using water that was distilled and purified by a Millipore purification system (Millipore Simplicity Ultrapure Water System). Buffer solutions were prepared with reagent grade materials (Fisher). All concentrations of polymers were provided in the polymer repeat unit concentration (PRU). Concentrated stock solutions of Poly 1, Poly 2, biotin TMR and avidin were pr epared in buffer to obtain the desired concentrations. All assays were conducted in 1 mM or 10 mM phosphate buffer (pH 7.4 or pH 10.5). Centrifugal filter units were bought

PAGE 117

117 from Millipore. Membranes with both 10,000 and 30,000 m olecular w eight c utoff (MWCO ) were used. Preparation of Poly 1 / biotin TMR complex The Poly 1/biotin TMR complex was prepared by ultra centrifuging the mixture of 100 M Poly 1 and 10 M biotin TMR in two centrifugal filter units with different molecular weight cutoff (MWCO, 30 kDa and and 10 kDa) membranes under 14,000 x g All of the polymer chains with MW higher than 30 kDa or less than 10 kDa, as well as free unboun d biotin TMR molecules, were centrifuged away. The remaining concentrated solution, which was collected for experiments, contains Poly 1/biotin TMR complex with the MW in the range of 10 kDa 30 kDa. Two buffer solutions, 10 mM phosphate buffers (pH 7.4 o r pH 10.5), were used for sample preparation. Concentrations of two components ([Poly 1] = 9.6 M, [biotin TMR] = 0.65 M) in the ( 3 2 ) where A is absorbance. is the molar absorptivity or extinction coefficient in unit of M 1 cm 1 The values for for each component are displayed in Table 3 1. l in unit of cm is the path length of the cuvette in whi ch the sample is contained c is the concentration of the compound in solution, expressed in M Negative C ontrol E xperiment by U sing BSA Protein bovine serum albumin ( BSA p I = 4.7 33 ), was added into the Poly 1/biotin TMR ([ Poly 1 ] = 102.8 nM, [ biotin TMR ] = 7 nM) in pH 7.4 buffer to get [BSA] = 7 nM The diffusion time becomes smaller ( ~ 0.68 ms ) when comparing to the diffusion time 0.79 ms of Poly 1/biotin TMR complex

PAGE 118

118 Instrumentation FCS measurement FCS measurements were performed using the homemade setup shown in Figure 3 7 The FCS was developed from an Olympus IX70. A 543 nm HeNe laser, specific for rhodamine dye, was employed as the excitation source light, which was focused onto the sample though Olympus 60x numerical aperture 1.2 wa ter immersion objective. The fluorescence is collected by the same objective, separated from the excitation light by a dichroic mirror, then split by a 50/50 cube splitter and sent into an avalanche photodiode (SPCM AQR 14 FC, Perkin Elmer) through a 50 m inner diameter optical fiber after passing through a 590 20 nm band pass filter. Chambered Cover Glasses (Fisher) were used as the container for samples in FCS measurement. In each FCS experiment the fluorescence fluctuations were measured for 10 20 m in. Free tetramethylrhodamine (TMR, D = 2.88 10 10 m 2 s 1 ) 143 was used for calibration. Autocorrelation was processed by a hardware correlator (ALV 5000/EPP, ALV GmBH, Langen, Germany). Fluorescence s pectroscopy. Fluorescence spectra were recorded on a Photon Technology International (PTI) fluorometer and corrected by using correction factors generated with a primary standard lamp. UV V is m easurement UV Vis spectra were measured in quartz cuvettes with 1 m m light path on a UV V is spectrophotometer (Shimadzu, UV 1800).

PAGE 119

119 CH APTER 4 STUDY OF CONFORMATIO N CHANGE OF CPES IND UCED BY PROTEINS AND DEVELOPMENT OF SENSO R ARRAY FOR PROTEINS BY FCS Background In recent years, the interaction s between CPEs and proteins have drawn much attention. 33 34, 97, 109 111, 123, 131 It has been well elucidate d from spectra that the non specific interaction causes significant change in both conformation and photophysical properties of CPEs 33, 34, 97, 110 For insta nce, Bazan and coworkers found that an anionic sulfonate d PPV mixed with small amount s of various proteins, including positively charged avidin and tau as well as negatively charged bovine serum albumin (BSA) and pepsin A, would have a several fold increas e in emission intensity, which was arise from a combination of electrostatic and hydrophobic forces 33 Bunz and coworkers also reported that BSA could enhance fluorescence of a carboxylate substituted CPE s ; W hile a series of proteins, such as histone, lysozyme, myoglobin, and hemoglobin quenched the fluore scence of CPE due to the formation of complex 110 Here, the aggregation state change of CPEs induced by interaction with various proteins is exp lored by focusing on the diffusion dynamics of the resulting polymer/protein aggregates. FCS has already been employed to study the conformation change of large molecules, e.g., Borsch and coworkers used FCS to monitor protein folding or unfolding transiti ons 142 Schwille and cowork ers conducted research on the fluorescence fluctuations of green fluorescence protein (GFP) by FCS and revealed the relationship between structural changes of GFP and its fluctuations in emission, making the probing local pH possible. 202 Waldeck and coworkers have uncovered the three regimes in the hydrodynamic radius changing of the complexes formed by an anionic CPE and a surfactant via FCS. 131 Our group has recently systematically investigated the

PAGE 120

120 aggregation of a dye ligand intercalated helical CPE induced by avidin and developed a nov el avidin sensor. 176 Therefore, FCS is very attractive as a promising tool for monitoring the aggregation state and size change of CPEs. The current chapter explore s how the aggregation state or size of six CPEs changes when exposed to solutions of seven different types of proteins which is accomplished by m easuring their diffusion time through FCS and analyzing their diffusion behavior change. I f the charge type s of CP Es and proteins are opposite, the aggregation of CPEs are easily observed; by contrast, if the CPEs and proteins have the same charge nature, de aggregation occur s for CPEs in aqueous solution. Meanwhile, other factors including charge density protein molecular weight, and polymer backbone structure also influence the final physical state of CPEs. As a result, the final diffusion time for each C PE protein mixture varies and the diffusion time response pattern created by the six CPE array for a typical protein is unique, which can be utilized for protein recognition and distinction Subsequently, a sensor array comprising six CPE probes with vario us charge properties, structure characteristics and molecular scales i s developed for seven proteins which also have various isoelectric point (pI), molecular weight and structure specificity Each type of protein can be well classified via l inear d iscrim inant a nalysis (LDA) of the FCS diffusion times Combination of sensor array with LDA has been applied in many sensing strategies. 97, 124, 203, 204 In this project multiple LDA operation s are employed for training data matrix, and creating a series of canonical plots as standard patterns for classifying different proteins. Then the technique readily identifie s a series of unknown protein sa mples with recognition accuracy 93 % One superior

PAGE 121

121 advantage for this sensing approach is noninvasive and labor saving. Unlike the conventional sensors requiring specific markers design, the CPE probes do not need to be attached covalently but undergo self assembl y via non specific interaction T his novel protein sensor array will m ake contribution to the medical diagnostics or clinical research 82 85 where detection of more than one protein in one single technical setting at one time is highly preferred. Results and Discussion Properties of S ix CPEs and Seven Proteins A set of six CPEs consisting of three with anionic and three with cationic ionic charge were chosen for exploring their interactions with proteins. The CPE structures with their abbreviat ions are shown in Figure 4 1 and their est imated molecular weights are listed in Table 4 1. A series of normalized absorption/emission spectra for those CPE is displayed in Figure 4 2 The first anionic CPE P1 is a poly(phenylene ethynylene) ( PPE ) with p oly(ethylene glycol) (PEG) hydrophilic sid e chains and dendri ti c carboxyl ate groups. The bulky, highly charged ionic functional groups are capable of keeping the single chain state of P1 by reduc ing the hydrophobic interchain interactions and increasing the electrostatic re pulsion between polyme r chains The fluorescence spectrum of P1 which features sharp emission peak around 470 nm, evidences the dominan ce of non aggregated chains in the solution. Two other anionic CPEs P2 and P3 have similar structure featuring PPE backbone and sulfonate d side groups. The only difference in structure is P2 is meta linked whereas P3 is para linked on the phenyl ring of the backbone. In water, P2 self assemble s into a helical conformation in aqueous solution attributed to the intra molecular stacking i nteraction in their backbone s. 23, 51 P3 can form lamellar aggregates in the aqueous

PAGE 122

122 solution 24, 26, 127, 205, 206 Both P2 and P3 have broad emission band as shown in Figure 4 2 Figure 4 1. Chemical structures of six CPE s. Table 4 1. Basic information of CPEs ( diffusion times are measured in 5 mM HEPEs buffer, pH 7.2, room temperature by FCS) P1 P2 P3 P4 P5 P6 Estimated MW (kDa) 11.0 40.0 ~100.0 10 70 NA 7 70 d (10 5 s) 8.4 0.9 29.4 3.0 101.3 25.9 21.7 3.9 27.3 3.1 24.0 8.6 The cationic CPEs P4 and P6 are less aggregated than P 3 due to their relatively long cationic bisalkylammonium si de groups that provide a steric barrier between the conjugated chain s 2 9 This statement can be proved by comparing their emission spectra (Figure 4 2), where the peaks with shorter wavelength corresponding to the higher emitting energy in non aggregated states, are obvious for P4 and P6 but invisible for P3 While P5 display ing related broad emission band can also be explained as its shorter

PAGE 123

123 side groups are not capable to prevent aggregation. Due to their various molecular weight s and conformation s this set of six CPEs display s different diffusion behavior reflected by the F CS diffusion time as summarized in the bottom row of Table 4 1. I t is not possible to apply gel permeation chromatography to measure the absolute molecular weights (MW) for P2 P6 due to their amphiphilic nature; however, other techniques have been used to estimate their MWs. 23, 24, 29 The results are displayed in Table 1 with t he factor that the CPEs of higher molecular weight s generally display longer diffusion time s Figure 4 2 Normalized absorption and emission spectra for six CPEs.

PAGE 124

124 Seven proteins were used in this work and their properties are summarized in Table 2 along with the acronyms that are used herein Avidin and LYS have isoelectric point ( p I) > 7, exhibiting positive charge in neutral solution; PLD2, HK3, 207 BSA, and GOx have p I < 7, so they are negative in neutral solution. H owever, for the protein HRP, due to its complexity it is hard to determine its p I value. 208 Those proteins also feature different molecular weight s varying from 14 200 kDa and distinct structur al characteristics. Table 4 2. Basic information of proteins Abbreviation Protein MW (kDa) pI Avidin Avidin 66 10 LYZ Lysozyme 14 11.0 HRP Horseradish Peroxidase, type I 44 3 9 PLD2 Phospholipase D, type II 200 4.65 HK3 Hexokinase Type III 54 PI :5.25 PII: 4 BSA Bovine Serum Albumin 66 4.7 GOx Glucose Oxidase 160 4.2 FCS R esults and D iscussion To quantitatively test the interaction bet ween the CPEs and proteins, all the experimental conditions including concentration s of b oth C P E s a nd p rot ei n s ion strength and pH of buffer solution (pH = 7.2) are fixed. The concentration s of proteins w er e determined through Bradford protein assay, a dye binding assay i n which a differential color change of a dye occurs in response to various weight concentrations of protein 209 Each of the s ix CPEs are mixed with seven proteins individually in 5 mM HEPEs buffer with final [CPE] = 500 nM (in repeat units) and [ protein ] = 2 g/mL. Twenty replicates are prepared for each CPE protein pair and a l l the samples are submitted f or FCS measurement in sequence.

PAGE 125

125 Figure 4 3 FCS curves for P4 without protein ( ) and with a vidin ( ) LYZ ( )HRP ( ) HK3 ( ) BSA ( ) PLD2 ( ) GOx ( ) in 5 mM HEPEs buffer pH 7.2. Black lines are the single species fitting curves. Figure 4 3 illustrates a typical FCS measurement results for the cationic CPE P4 with and with out seven proteins. Based on the FCS curves, the mixtures of P4 with a vidin ( d = 2 0.7 10 5 s) or LYZ ( d = 22.5 10 5 s), whose p I value > 10, have approximately the same diffusion times as pure P4 ( d = 21.7 10 5 s) By contrast, the other 5 proteins, whose p I averagel y < 7, to various degrees, induce the aggregation of P4 and a d ramatic increase in the diffusion times The order for the diffusion rate of P4 /proteins is: HRP > BSA > HK3 > PLD2 > GOx with increasing d = 1.95, 5.66, 7.80, 13.0 0 and 219.00 ms, respectively. Due to the polydispers e nature of the aggregates, the FCS c urves are the combination of several single species curves with different diffusion times. Nevertheless, the single species fitting equation (Equation 2 2) is still

PAGE 126

126 applied and an average d i f f u s i o n t i me for each CPE/protein aggregate is thereby obtained. As stated a bove, HRP and HK3 are the mixtures of several isozymes. Their FCS curves are more complicated and difficult to be fit by a s ingle species fitting equation. Figure 4 4 F luctuation profile s for A) P4 without and with seven proteins and B) their e nlargement ([ P4 ] = 500 nM, [protein] = 2 g/mL in 5 mM HEPEs buffer pH 7.2)

PAGE 127

127 The fluorescence fluctuation profiles for each mixture are displayed in Figure 4 4 Higher and broader peaks are corresponding to the large aggregates passing through t he excitation volume which lead to longer diffusion times as shown in Figure 4 3 Those time dependent profiles also provide the evidence for the heterogeneity or size multi di stribution of aggregated CPEs. Figure 4 5 Log ( d / 0 ) response 2 D bar pattern of six CPE s mixing w ith seven protein s Bar height is the average value of twenty re p licat e s for each CPE protein pair The FCS results for all the samples are shown as a bar graph in Figure 4 5 in terms of l og ( d / 0 ), where d and 0 are the diffusion time of CPE with and w ithout protein, respectively. The error bars represent the calculated standard deviation for 20 replicates. Figure 4 6 is the 3D column graph for the FCS results. The row s with the same color are belonging to the same protein and the charge type for each s pecies is embodied in their abbreviation s by color: negative charge is black, positive charge is red and the protein HRP with mixed p I is blue. There are several clear trends that emerge upon inspection of the data in Figures 4 5 and 4 6. First, it is clea r that the diffusion time increases (i.e. log ( d / 0 ) > 0) for the opposite ly charged CPE/protein pairs, e.g., LYZ/ P1 or PLD2/ P4 due to the attraction induced formation of polymer protein aggregates. In

PAGE 128

128 sharp contrast when the charge of the protein and CPE are the same, especially for the pairs of anion ic CPE and protein with p I < 7, such as PLD2/ P2 or HK3/ P3 the mixtures exhibit shorter diffusion time compared to the CPE alone (i.e., log ( d / 0 ) < 0 ). Apparently the Coulombic repulsion between two ani onic macromolecules disrupts the aggregation of CPEs. Figure 4 6 Log ( d / 0 ) response 3D column pattern of six CPE s mixing with seven protein s Column height is the average value of twenty replicat e s for each mixture. For the cationic CPE with positivel y charged proteins, e.g., LYZ/ P5 or P6 some aggregation is apparent with average l og ( d / 0 ) varying in the range of 0 0.1. This is likely due to the hydrophobic interaction. 109 The much larger error bar s observed for HRP/CPE pairs ( e.g., HRP/ P5 ) is p resumably due to the complexity in the mixture of protein isozymes. However, based on the average log ( d / 0 ) values for HRP with various CPEs, HRP is able to induce aggregation when mixed with cationic CPEs, while disrupting aggregation when mixed with anionic aggregated CPEs. Therefore, HRP is displaying more anionic characteristics in neutral system

PAGE 129

129 Besides electrostatic or hydrophobic interaction, other factors may also influence the final response. For example, the log ( d / 0 ) for anionic P2 with cationic LYZ is much larger compared to that of P2 with avidin. W e attribute this phenomenon to the hi gh charge density of LYZ, which has the smallest molecule weight and highest p I value. However, when considered P4 mixed with a series of anionic proteins, the molecule weight of proteins seems have impact on the final diffusion time change. T he value of l og ( d / 0 ) increases roughly as the increase of the molecular weight of the protein: HRP (44 kDa) < HK3 (54 kDa) BSA (66 kDa) < PLD2 (200 kDa) < GOx (160 kDa). However, this trend does not apply to other types of oppositely charged CPE/protein mixtures. In addition to the influence of the protein, the properties of polymer s also influence the trends in the diffusion time changes for the CPE protein mixtures For example, the thiophene containing CPEs ( P4 and P5 ) appear to form larger aggregates compared to P6 which contains only phenylene repeat units. In particular, when P4 is mixed with oppositely charged proteins (BSA, HK3, PLD, GOx), the resulting log ( d / 0 ) values are larger than those for P6 with the same proteins This finding is consistent with the previous reports that the thiophene containing CPEs appear to have a larger hydrophobic character compared to the phenylene analogues 210 The bond linking style in ba ckbone may also enhance the interaction between CPE and proteins. For instance, the diffusion time for P2 a helical CPE is greater than that of P3 which ha s linear backbone, mix ed with oppositely charged proteins (LYZ, a vidin), even though P3 has higher MW than P2 does. Consequently, the final signal response is affected by charge type, charge density, molecular weight and structure property of both proteins and CPEs. It is difficult to de convolute each part just relying on the FCS measurement. We

PAGE 130

130 are p leased to find that the six CPEs are displaying unique signal response pattern for each protein, which is much beneficial for processing distinguish and identification of proteins in the next sensory development. Protein Sensing Linear discriminant analysi s of FCS diffusion times for protein/CPE mixtures In order to provide more insight into the structure property relationships for the CPE/protein mixtures, we carried out studies aimed at subjecting the FCS results to linear discriminant analysis (LDA the details of the theory and procedures for LDA can be found in the literature). 125, 126 As outlined below, this work lead s to the development of a novel CPE FCS based method allowing the identification of a protein in an unknown sample ( Figure 4 7 ) As can be seen below, a single LDA operation on the entire set of data was unable to afford a high degree of accuracy due to the reason that the single LDA was originally developed for two class problems and it is sub optim al if multiple classes are considered. 211 Thus, we applied a sequence of LDA steps afterwards to multiple subgroups of proteins and generated subspaces that have higher overall classification power. 212 Figure 4 7 Flowchart for protein sensor a rray development.

PAGE 131

131 The details about the multiple LDA process can be found in Figure 4 8 Initially, w e used the full set of log ( d / 0 ) values to construct a matrix consisting of ( 6 CPE s) ( 7 p roteins ) ( 20 replicates ) for LDA analysis (Table A 1 in Appendix A). The eigen vectors which maximiz e the ratio of between class variance to the within class variance are obtained through LD A implemented as a script in Matlab. Then the three most significant eigen vectors (carrying 82. 5 9.3 1 and 5. 77 % of the discriminant information, respectively) are used to plot sample data in a 3D discriminant space as shown in Figure 4 9A. The 6 7 20 samples are presented with different colors denoting the different proteins. In principle, each protein should occup y a specific region in the 3D space and the different proteins should be well separated from each other. As displayed in Figure 4 9A and Fi gure 4 10, three protein groups o f avidin, BSA and GOx are well classified with individual accuracy 95%. However, mingling occurs among the categories of HRP, PLD2, HK3 and BSA with error to be 4/20, 7/20, 7/20, and 5/20, respectively, resulting in the tota l classifi cation accuracy 83%. A second LDA operation is applied using only the sub sets of data that belong to HRP, PLD2, HK3 and BSA A new discriminant space specifically for these four proteins is created and the improvement in the separation of groups can be easily observed (Figure 4 9B). The errors are reduced to be 2/20, 2/20, 2/20, and 1/20 for HRP, PLD2, HK3 and BSA, respectively, with an increased total accuracy 94% for seven proteins (Figure 4 10). More LDA operations are continuously applied for those groups with relatively lower individual training accuracy (<95%). As shown in Figure 4 9C, D and Figure 4 10, after four times LDA operations, the total classification accuracy for the 6 7 20 training matrix reaches 98%, much higher than the res ult of a single LDA

PAGE 132

132 operation, i.e., 83%. By building up multiple subspaces instead of a single large space, the protein discriminant method is well established (4 sets of eigenvalue s are summarized in Table 4 3). Figure 4 8 Flowchart o f multiple LDA o peration for training known samples

PAGE 133

133 Figure 4 9 LDA d i sc ri m i n ant s pac e s / plot s for the diffusion time response patterns obtained with six CPE probes sensor array against A ) seven proteins ( a vidin, LYZ, HRP, PLD2, HK3, BSA, GOx), B ) four proteins (HRP, PLD2, HK3, BSA ), C ) three proteins (HRP, PLD2, HK3), D ) two proteins (HRP, PLD2). Twenty replicates for each mixture. Table 4 3 Eigenvalues w i t h their percentage of each LDA operation training matrix LDA operation 1 2 3 Total 1 st 1.7293 (82.5%) 0.195 1 (9.31%) 0.1209 (5.77%) 2.0963 2 nd 0.2026 (64.1%) 0.0593 (18.7%) 0.0543 (17.2%) 0.3162 3 rd 0.1975 (76.8%) 0.0597 (23.2%) 0.2572 4 th 0.1772 (100%) 0.1772

PAGE 134

134 Figure 4 10 Training results for multiple LDA operation of diffusion time response for si x CPE probes against seven proteins ( twenty replicates for each probe target pair) Unknown sample t est After s ucceeding in classifying known samples, the detection and identification of unknown protein samples (Table B 1 in Appendix B) were further studie d Forty two artificial protein samples prepared by a second researcher are undergoing the same protocol as described above, including determining concentration through Bradford protein assay, mixing proteins with six CPE probes individually, adjusting the ir

PAGE 135

135 concentration s to [CPE] = 500 nM and [protein] = 2 g/mL, conducting FCS measur ements and generat ing final data matrix ( Figure 4 7 ). Figure 4 11 Flowchart for multiple LDA operation for testing unknown samples. As processed by first level LDA operation, the entire unknown samples are projected to the d iscriminant space built up in the above training process. Based on the theory of LDA, 125, 126 in the created 3D discriminant space, the Mahalanobis distances of each unknown spot in the space to the centroid of each class are calculated and these

PAGE 136

136 unknown spots are assigned and identified according to the shortest Mahalonobis distance. Based on the classification accuracy obtained in the training process, the samples that assigned to the groups of avidin, LYZ and GOx are believed to be identified with accuracy > 95%. The sub set data belonging to those three proteins are removed from the entire matrix, and the rest matrix is submitted for a second level LDA operation and processed in a similar manner ( Figure 4 11 ). As four time LDA has been operated, all of the unknown samples get identification. Verification is assisted by the researcher who prepares unknown samples. Error report is shown in the Figure 4 12 Total identification accuracy is improved from 88% (failure t est: 1 for HRP, 1 for HK3, and 3 for BSA) for a single LDA operation to 93% (failure test: 1 for HRP and 2 for BSA) for a multiple LDA operation (final failure test samples are marked with star in Table B 1). Figure 4 1 2 Test results for multiple LDA op eration of diffusion time response for six CPE probes against forty two unknown protein samples. Based on the LDA results discussed above, we can find that proteins with p I > 7, e.g., avidin and LYZ, are easily separated from the rest groups with

PAGE 137

137 classific ation/identification accuracy ~ 100% for both known and unknown samples, followed by GOx ( p I = 4.2) with classification/identification accuracy 95% and 100% for known and unknown samples, respectively. HRP, PLD2, HK3 and BSA are relatively hard to be distin guished from each other, which are probably due to their similarity in charge properties, resulting in similar interaction with each CPEs and requiring further LDA processes. The complexity in the p I value for multi isozyme mixture, such as HRP, may also b ring in the difficulty in identification of proteins : the identification accuracy for HRP is relatively low. The separation of proteins with opposite charge is the easiest to be achieved, suggesting the charge type plays the most significant role in the in teraction between CPEs and proteins as well as the recognition of proteins. While further discriminant between proteins with similar charge properties needs more analysis on the minor difference s in data matrix. Those differences probably arise from molecu lar structure, charge density or molecular weight of both CPEs and proteins. As discussed above, those factors can also make their own contributions to the differentiation of proteins. Summary In sum, a systematical investigation was conducted on the aggre gation state/size change of CPEs induced by non specific interaction between various CPEs and proteins from a molecular dynamics aspect. By employing FCS system, the aggregation/de aggrega tion of CPEs can be reflected on the diffus ion behavior changes Many factors including charge type, charge density, molecular weight and structure of CPEs or proteins contribute to the final conformation al and diffusion al changes of CPEs. Among those factors, charge type plays the essential role. The pattern s of signal res ponses generated by six CPEs are discernibl e for diffe rent proteins, arousing the

PAGE 138

138 development of a new CPE based sensor array for protein. By applying m ultivariate pattern recognition chemometrics LDA, in a multiple operation mode, a series of discriminan t spaces is created and seven different proteins have been successfully classified Forty two unknown samples were further tested and a high identification accuracy 93% was obtained, which verified the robustness and feasibility of this novel sensor array This type of sensing strategy establishes a new protein sensing platform where the proficient manipulation and strong biological background are not required for operator s Further effect can be made upon the optimization of probes including improving the monodispersion of polymers, conjugating more versatile functional groups to the backbone or introducing in new probes Moreover, the development of the sensor array in more complicated biological environments is necessary to eliminate potential interferenc e and enhance their feasibility in the real wor l d Material and E xperiment Materials A vidin from egg white ( a vidin, A9275) l ysozyme from chicken egg white ( LYZ, L6876) p eroxidase from h orseradish, type I ( HRP, P8125) p hospholipase D from a rachis hypogae a (peanut), type II ( PLD2, P0515) h exokinase from s accharomyces cerevisiae, type III ( HK3, H5000) a lbumin from bovine serum (BSA, A2153 ), and g lucose o xidase from A spergillus niger (GOx, G7141 ) were purchased from Sigma. The synthesis procedures and char acterization of P1 182, 183 P2 23 P3 24 P4 29 P5 27 and P6 29 has been previously reported All sample solutions were prepared using water that was distilled and purified by a Millipore purification system (Millipore Si mplicity u ltrapure w ater s ystem). Buffer solutions were prepared with reagent grade materials (Fisher). All concentrations of polymers are provided in polymer repeat unit concentra tion (PRU).

PAGE 139

139 C oncentrated stock solution s of the CPEs and proteins were prepared in buffer to obtain the desired concentrations. All assays were conducted in 5 mM phosphate buffer pH 7. 2 at room temperature. Coomassie b rilliant b lue G 250 was obtained from Sigma. Phosphoric acid and methanol were HPLC grade solvents bought from Sigma. Bradford Protein Assay Procedure Bradford protein assay was conducted follow ing an instruction. 213 Coomassie Blue G (10 mg) was dissolved in 5 mL of methanol and displaying dark blue The solution was added to 10 mL of 85% H 3 PO 4 and diluted to 20 mL with Millipore water. The final reagent was dark red and containing 0 .5 mg/mL Coomassie Bl ue G, 25% methanol, and 42.5% H 3 PO 4 The stock solution was kept in a dark bottle at 4C. The assay reagent was freshly prepared for each measurement by diluting 1 volume of the dye stock with 4 volumes of Millipore H 2 O and it appear e d brown. S ix standard solutions (1 mL each) containing 0, 250, 500, 1000, 1500 and 2000 g/mL BSA (extinction coefficient 43,824 M 1 cm 1 at 280 nm) 214 were prepared as standard samples The UV Vis spectro photometer was set in a point read mode at 595 nm to collect the data. A 4 mL plastic cuvette filled with millipore water was inserted to auto zero the reading at 595 nm. 2.0 mL a ssay reage nt and 0.04 mL of protein standard solution were mixed in a vial and it was gently inversely shaken several times After 10 min standing, the absorbance of each stable sample was measured under room temperature, starting with the lowest protein concentrati on and working up All the absorbance must locate between 0 and 2, otherwise, the outstanding sample should be adjusted or rejected and repeated A plot of a bsorbance at 595 nm vs [ BSA ] as the standard curve was drawn through the instrument. The Bradford a ssay gives a hyperbolic plot for absorbance versus protein concentration, but within a range of relatively low protein concentrations, the hyperbolic

PAGE 140

140 curve can be approximated reasonably well by a straight line. 213 A one order straight line was used to fit the points and a high fitting coefficient above 0.99 was obtained (if not, all the experiment s should be restarted over) The measurement of pr otein samples with unknown concentration was followed the same procedure and their absorbance at 595 nm was obtained. By us ing the standard curve their concentration s were easily figured out Instrumentation FCS m easurement The measurements were perform ed in a setup constructed in house that described in Chapter 2 In each FCS experiment the fluorescence fluctuations were recorded for 1 2 min. Free fluorescein (D = 3.00 10 10 m 2 s 1 ) 215 w as used for calibration. UV V is m easurement UV Vis spectra were measured in 1 cm light path dispos able polystyrene cuvettes (Fisher) on a UV V is spectrophotometer (Shima d zu, UV 1800)

PAGE 141

141 CHAPTER 5 STUDY OF INTERACTION OF META LINKED POLY(PHENYLEN E ETHYNYLENE) SULFONATE CONTAINING PYRIDINE WITH METAL IONS Background T h e strong photoluminescence and amplif ied quenching intrinsic to CPEs enable them to be u ltrasensitive chemosensors for metal ions 16, 54, 91 The addition of oppositely charged small ions into solution s of CPEs has been reported to lead to significant c hanges in the ir photophysical properties 16, 42 Early studies found that some metal ions such as Pd 2+ and Ca 2+ exhibit a n amplified quenching ability towards c arboxylated poly( p phenylene ethynylene) s (PPE) in aqueous solution. 47, 216 The superior sensitiv e photophysical response of CPEs is attributed to a combination of several factors, including delocalization and rapid diffusion of the singlet exciton along the CPEs backbone to the quencher trap site as well as the various interactions between CPEs and ions such as electrostatic attraction or chelation, through which the aggregation of CPEs may be induced 24, 42, 43 Consequently, a series of CPE based sensors for metal ions has been developed. 47, 67, 216 P yridine ha s a strong binding a ffinity towards metal ions and many pyridine substituted macromolecules have been synthesized and developed for metal ion sensors in the past few years. 56, 217 219 Bai and coworkers published a series of paper s rega rding development of pyridine derivative containing conjugated polymer based sensing methods for Pd 2+ or Pt 2+ 217 219 Wang and coworkers also reported a fluorimetric detection method for copper ions using bip yri dyl substituted cationic CPEs. 56 As stated in Chapter 1 and 3 meta linked PPE s ha ve the tendency to form a helical structure in polar solvent s stabilized by their amphiphilicity and intramolecular stacking. Winter and Eisenbach have re ported that meta l inked PPE provides flexibility and spatial

PAGE 142

142 registry for three dimensional molecular binding of analytical targets. 220 By coupling in the pyridine units, t he he lical structure adopted by meta linked CPE can be stabilized and enhanced by silver ion s or H + 221 Recently, Tan and Huang have designed an anionic CPE by combining meta and para linked pyridyl and phen y lene ethyny lene (PE) units into one polymer chain which shows high sensitivity and selectivity for silver ion s 222 Palladium is capable of eliciting a series of cytotoxic effects which may cause severe primary skin and eye irr itations. 223, 224 Among the c onventional analytical techniques used for palladium detection, such as atomic absorption spectrometry, 225 plasma emission spectroscopy, 226 X ray fluorescence 227 and inductively coupled pl asma mass spectrometry (ICP MS) 228 c olor imet ric and fluorimetr ic based methods are expected to be more desira ble due to their simplicity and ultrasensitivity. 217, 218, 229 232 Particularly conjugated polymer ( CP ) based sensor s have attracted much attention. 217, 218, 230 Figure 5 1. Structure of mPPESO 3 py In this chapter, a sulfonated poly (phenylene ethynylene) (PPE) containing meta linked pyridine rings in the polymer backbone (mPPE SO 3 py Figure 5 1 ) is investigated. A solvent induced change in photophysics is observed, suggesting a coil helix conformation transition and an electron donor acceptor complexation. By incorporat ing

PAGE 143

143 the meta substituted monopyridyl units the spatial matching for selective binding is improved. Particularly, the photophysical study indicates that the strong chelation of Pd 2+ with pyridine rings rearranges and sta bilizes the helical conformation of mPPE SO 3 py Moreover, Pd 2+ ion shows overwhelming amplified quenching ability over other metal ions towards mPPE SO 3 py giving rise to a novel type of Pd 2+ sensor that overcomes the water insolubility drawback of CP s FCS stud ies show that upon addition of Pd 2+ to an mPPE SO 3 py solution, a poor correlation of the emission signal and a shorter diffusion time for mPPE SO 3 py are observed. These phenomena are presumably due to the formation of a more compact conformation for mPPE SO 3 py with Pd 2+ via chelation as well as the low quantum yield for mPPE SO 3 py due to amplified quenching which leads to fast fading of photons emitted by a single polymer chain. Longer diffusion time is observed for mPPE SO 3 py mixed with various other multi valence ions, particularly, Fe 3+ and Cr 3+ which is the result of aggregation induced by the binding betwe en metal ions and the nitrogen atoms belonging to adjacent CPE chains Nevertheless electron transfer and high binding affinity between CPE s and met al ions are the dominant factors in the amplified quenching mechanism. Results and Discussion Photophysical Properties of mPPESO 3 py Solvent induced photophysics change In previous work, a meta PPE was reported to adopt a random coil conformation in a good solvent such as methanol (MeOH) In a polar solvent, such as acetonitrile and water, the polymer chain collapse s into a stacking helical conformation to minimize solvent backbone contacts while maintaining favorable solvent/side chain interactions for solvation. 51 53 This conformational transition is reflected in the UV V is

PAGE 144

144 absorption spectrum. Figure 5 2 A shows the spectra for mPPE SO 3 py in MeOH titrated with water. The absorption spectrum of mPPE SO 3 py feature s t wo absorption bands at 3 87 nm and 3 06 nm in MeOH, both of which decreas e and slight ly red shift when the volume percentage of water increases in the solvent mixture. These changes in spectra are similar to that of meta oligo(phenylene ethynylene) when the random coil to helix conformation transition occurs. 51 53 Figure 5 2B plots the ratio of the two absorption peaks (A L /A S A L and A S are the absorbance at longer and shorter wavelength peaks, respectively ) as a func tion of the volume concentration of water in the mixed solvent s A general decrease in the ratio i s observed with increasing content of water suggesting the coil helix conformation transition has more impact on the absorbance at longer wavelength, which i s consistent with the observation reported in the literature. 51 53 Figure 5 2 A) A bsorption spectra for mPPE SO 3 py in solvent mixture with different component volume ratio. B) Ratiometric plot of A L /A S versus t he percentage of water in methanol A L and A S are the absorbance at longer and shorter wavelength peaks, respectively. Note that a broad shoulder peak appears at 450 500 nm and increases with increasing volume fraction of water in the solvent, this band may b e due to a charge transfer absorption, which is similar to the spectra observed in the donor accepter system. 233, 234 T he helix theoretically contains six aromatic members per helical turn with

PAGE 145

145 stacking between two identical aromatic rings, e.g., pyridyl pyridyl. However, during the folding process of the coil helix transition, a twist may occur due to the mismatch in the size of functional groups or steric hindrance bringing electron rich alkoxy substituted phe nyl and electron deficient pyridyl units in close proximity where a donor acceptor interaction between them takes place. As illustrated in Figure 5 3A, the solv o phobicity and stacking drives mPPE SO 3 py to adopt a helical conformation in water; while the pyridyl phenyl interaction further stabilize s the helical structure, although it twists the helical circle. As more water is added into the mixture, more random coils tend to fold into helixes and more pyridyl phenyl pairs are involved in the complexation, thus enhancing the intensity of the shoulder peak in the spectrum. Figure 5 3 A) Mechanism of solvent and metal ion induced formation of helical structures. B) Mechanism of multi valence metal ion induced crosslinking of CPE.

PAGE 146

146 Figure 5 4 A) E mission spectra for mPPE SO 3 py in solvent mixture with different component volume ratio. B) A bsorbance ratio of two bands versus the percentage of water in MeOH The solvent induced changes in emission spectra as a function of solvent are shown in Figure 5 4A In a good solvent, mPPE SO 3 py features two bands with a pronounced high peak at ~ 450 nm and a structureless broad band around 680 nm. The 450 nm band is likely to be the emission from a rel atively localized excited state T he broad band in the low energy region may emanate from an excimer like excited state, attribut ed to the stacking hydrophobic interaction or donor acceptor complexation 51 As more and more water is added into the solution, quenching is much more pronounced for the shorter wavelength band A plot of I 450 /I 680 as a function of percentage of water in the mixture is displayed in Figure 5 4B As expected, the ratio decreases as the volume of water in the composition increase s, which corresponds to the folding process of the polymer chain undergoing a coil helix transition Palladium ion induced photophysics change A titration of Pd 2+ to the mPPE SO 3 py aqueous solution was conducted. The two absorption peaks continue to decrease and red shift, which are ascribed to the electron density change caused by the complexation of metal ions or conformational changes in

PAGE 147

147 the polymer's backbon e as a result of the chelation. 235 The A L /A S keep s decreas ing from 1.12 to 0.91, indicating an advanced folding process that further enhances the helical conformation (Figure 5 5B ) A plausible mechanism is illustrated in Figure 5 3A : when Pd 2+ is present the chelation between two Ns and one Pd 2+ causes more random coil portion to undergo coil helix transformation. Meanwhile, the high binding affinity between N and Pd 2+ makes the spring like helical coil press e d more with Pd 2+ as the clip disrupts the charge transfer phenyl pyridyl interaction and r earrang es the stacking of the aromatic rings, e.g., pyridyls pairing with pyridyl s instead of phenyls The closely packed core with relatively narrow helical angle enhances the stacking hydrophobic interaction, leading to the further decrease and red shift in the absorption peak as observed. The rearrangement in the helical packing structure can be evidenced by the gradual disappearance of the shoulder peak in the region of 4 50 500 nm (Figure 5 5A). Figure 5 5 A) A bsorption spectra of 15 M mPPE SO 3 py titrated with Pd 2+ B) T he absorbance ratio of two bands versus the concentration of Pd 2+ in aqueous solution ( A L and A S are the absorbance at longer and shorter wavelength pea ks, respectively ). The quenching behavior of Pd 2+ on the emission of mPPESO 3 py is displayed in Figure 5 6 ( The sharp peak around 780 nm is the second order scattering fo r the

PAGE 148

148 excitation beam at 390 nm ) A dramatic decrease in the emission peak is observed as the [ Pd 2+ ] increases from 0 to 40 M As [Pd 2+ ] reach es 3 0 M the emission of 15 M mPPESO 3 py is almost quenched and the shoulder band in the absorption spectrum disappears, suggesting the saturation of the metal pyridyl complexation at this concentration. Figure 5 6 Emission spectra f or 15 M mPPESO 3 py with various [Pd 2+ ] in aqueous solution The sharp peak shown up around 780 nm is the second order scattering for the excitation beam at 390 nm Photophysics c hange of mPPESO 3 py with v arious m etal i ons High quenching efficiency of Pd 2+ t owards mPPESO 3 py is evidenced when compared to other metal ions including Cu 2+ Fe 3+ Cr 3+ Fe 2+ Li + Ag + Ca 2+ Zn 2+ and Mg 2+ Figure 5 7 and Figure 5 8 display the Stern Volmer plots and bar graph for the polymer with different metal ions in water. Pd 2+ has the highest = 2.50 10 5 M 1 of all the metal ions and is ~ 4 times greater than that of Cu 2+ which has the second highest The selectivity of mPPESO 3 py for Pd 2+ can be evaluated in the bar graph ( Figure 5 9 ) where 15 M Pd 2+ can greatly quench 15 M of the polymer with I/I 0 ratio >

PAGE 149

149 3 times higher than that of the other metal ions at the same concentration As the concentration of Metal ions increases, the selectivity of mPPESO 3 py for Pd 2+ over other metal ions becomes greater, which paves a promising avenue for Pd 2+ sensor development Figure 5 7 Stern Volmer plots for mPPESO 3 py with various metal ions in water Figure 5 8. Stern Volmer constant for various metal ions in water.

PAGE 150

150 Figure 5 9 Bar graph for fluoresc ence quenching ratios I 0 /I at 680 nm of mPPESO 3 py with different metal ions at various concentration s in aqueous solution Insert: E nlargement of lower I 0 /I region. I 0 and I are the emission intensity of mPPESO 3 py without and with metal ions, respectively. FCS S tudy on the mPPESO 3 py with Various M etal I ons To gain more insight into the intermolecular interaction induced by metal ions, t h e FCS technique has been used to detect the diffusion behavior change of mPPESO 3 py with and without metal ions. The diffus ion time of pure polymer (15 M) in aqueous solution is 1.39 10 4 s (Figure 5 10A) Upon titration with Pd 2+ from 10 M 50 M, the correlation curve s for the polymer become noisier at a fixed photon recording time (2 min) which are poorly fitted with fi tting coefficient ~84% when E quation 2 2 is applied (Figure 5 10A) Recall that the correlation of the fluorescence fluctuation is accomplished by calculating t he similarity between a signal F (t) and a replicate of the same signal but shifted by a time lag F (t+ ). Only when the two set s of signal F (t) and F (t+ ) are identical i.e., produced by the same molecule, that the similarity/correlation of signal will reach the maximum. Once the polymer chain diffuses out of the excitation

PAGE 151

151 volume after its charac teristic diffusion time D the correlation is greatly reduced. However, when the quencher molecule s are introduced its efficient quenching ability largely reduces the number of photons emitted by a single polymer chain or aggregate which makes detection of the signal for the photon coun ting device much more difficult resulting in a poor correlation curve A smooth correlation curve and good fitting still can be obtained when the FCS photon recording time is largely extended, e.g., to 30 min. A decrease in diffusion time is observed as the concentration of Pd 2+ increases (Figure 5 11) There are two possible reasons for t he apparent fast diffusion The first is failure in detecting the second photon signal that is generated by the same polymer chain/aggre gate due to amplified quenching, which produces a false appearance that the polymer chain has already diffused out of the excitation volume. The second is the chelation between pyridine and Pd 2+ compacts the conformation of mPPESO 3 py, which decreases the h ydrodynamic radius and accelerate s the diffusion of the CPE (Figure 5 3A). Figure 5 10 A) FCS correlation curves and B ) fluctuation profiles for 15 M mPPESO 3 py without (red) and with 40 M Pd 2+ (green) or Cr 3+ (blue) in aqueous solutions. Black lines in A are the single species fitting curves.

PAGE 152

152 Figure 5 11 Diffusion time ratio for mPPESO 3 py with different amount of Pd 2+ measured for 30 min in aqueous solution D and D 0 are the diffusion time of mPPESO 3 py with and with out Pd 2+ respectively. More FCS experiments have been done for the CPE with other metal ions. As shown in Figure 5 12, u pon addition of 40 M Cr 3+ 15 M mPPESO 3 py undergoes large aggregation with d / d 0 ~ 7 where d0 and d are the diffusion time of mPPESO 3 py without and with Cr 3+ res pectively A relatively smooth correlation curve can be found in Figure 5 10A (fitting coefficient ~ 99%). Several spikes show up in the fluctuation profile in Figure 5 10 B indicating the formation of large aggregates. Fe 3+ can also induce aggregation with d / d 0 ~ 6.5, followed by Fe 2+ ( d / d 0 ~ 2) and Cu 2+ ( d / d 0 ~1.4). Tripl y charged ions seem have higher capability for a ggregation induction, which is probably due to their ability to bind three pyrid yl moieties belonging to different polymer s to induce in terchain crosslinking as depicted in Figure 5 5 B

PAGE 153

153 Figure 5 1 2 Diffusion time ratio of mPPESO 3 py before and after the addition of metal ions. D and D0 are the diffusion time of CPE with and without metal ion, respectively [mPPESO 3 py] = 15 M, [metal ion] = 40 M. Although aggregation is one of the mechanisms that cause amplified quenching of CPE s 28, 1 03 it is not the dominat ing factor in this system. As shown in Figure 5 1 2 and Figure 5 8 even though Cr 3+ possesses highest cross linking ability, it has a relatively low quenching efficiency. The quenching ability of copper ion s is second best compare d to palladium ion s ; however, it can only induce slight aggregation in CPE s Charge transfer from the electron rich conjugated backbone to the electron deficien t metal ions as well as the high coupling affinity between the pyridine moiety and multi valen t ion s are the dominant effects i n the amplified quenching process Summary In summary, the solvent induced coil helix transition on the meta linked anionic CPE containing monopyridyl units is investigated Changes in the p hotophysical

PAGE 154

154 properties provide evi dence for the conformation al alteration of the CPE as well as the donor acceptor complexation in the helix O n the basis of the spectroscopic study, Pd 2+ induced rearrangement and tight er compact ion in helical rings of the polymer backbone are revealed Th e quenching efficiency of Pd 2+ overwhelm s other metal ions giving rise to a Pd 2+ sensing strategy. T h e FCS study shows that although the triply charged ions (Fe 3+ Cr 3+ ) can induce interchain crosslinking, aggregation is not the dominan t factor in the amp lified quenching behavior. Charge transfer and high binding affinity between CPE s and metal ions play essential roles in the quenching phenomenon Experimental Materials mPPESO3py was synthesized according to previous report 196 S olutions of metal ions were prepared from their chloride salts, except for AgNO 3 and FeSO 4 With the exception of PdCl 2 all metal sample solutions were prepared by using wa ter that was distilled and purif ied by a Millipore purification system (Millipore Simplicity Ultrapure Water System). A stock solution of PdCl 2 was prepared in dimethylsulfoxide ( DMSO ) The polymer stock solution was diluted with water or methanol to a final concentration of oncentrations of polymers are provided in polymer repeat unit (PRU) concentration. Instrumentation Absorption and Emission measurement Absorption spectra were obtained on a Varian Cary 100 UV Vis absorpt ion dual beam spectrophotometer Steady state fluores cence spectra were recorded on a spectrofluorometer from Photon Technology International and corrected by using correction factors generated with a primary standard lamp.

PAGE 155

155 FCS measurements The measurements were performed in a setup constructed in house th at is described in Chapter 2. In each FCS experiment the fluorescence fluctuations were recorded for 5 30 min. Free TMR (D = 2.88 10 10 m 2 s 1 ) 143 w as used for calibration.

PAGE 156

156 CHAPTER 6 CONCLUSION The wat er solub ility and conjugated backbones endow c onjugated p olyelectrolytes ( CPEs ) favorable optical properties, charge interaction ability and solvent dependen t self assembly T his class of polymers has been studied for chemical and biological sen sing applications due to their superior signal amplification and increased sensitivity compared to other molecular dyes Fluorescence correlation spectroscopy (FCS) de veloped from a fluorescence microscope is a powerful single molecule spectroscopic technique, which utilizes the fluctuations in emission (caused by Brownian diffusion, flow, chemical reactions, etc.) from fluorescent molecule s moving in and out of a micr oscopic confocal volume (~femtoliters). By employing FCS, the diffusion behavior, concentration, conformation/size change of fluorescence molecules particularly, CPEs, can be monitored directly, so that a number of events, including chemical reaction, mob ility, molecular interaction, self assembly, or binding kinetics, can be detected. The Application o f Dye Ligand Intercalated Helical Conjugated Polyelectrolyte t o Protein Sensing As describe d in Chapter 3 a DNA intercalator biotin tetramethylrhodamine (b iotin TMR) i s found to be capable of intercalating into a helical conjugated polyelectrolyte ( P oly 1). The efficient fluorescence resonance energy transfer (FRET) from the polymer to th e TMR chromophore i s observed ; however, the addition of avidin to the p oly 1/b iotin TMR intercalation complex result s in no emission recovery of P oly 1. The mixing of biotin TMR with avidin prior to the addition of the polymer can effic iently disrupt the FRET signal The study of the FRET response as a function of [ avidin ]/[ biotin TMR ] affords insight into the interaction of the protein with the

PAGE 157

157 intercalation complex and reveals the so called phenomenon steric constraint. This project i s further studied by FCS coupled with 543 nm laser A remarkable incre ase in the diffusion time of P oly 1 / biotin TMR complex in the presence of avidin i s observed This change is attributed to the formation of large supramolecular polymer aggregates induced by the binding between the biotin unit s of intercalat or s and avidin which g i ve s rise to a novel sens ing strategy for avidin The detect ion limit of <100 pM i s figured out, which is lower than that of most other avidin sensing methods reported. New Fluorescence Correlation Spectroscopy and Application on Protein Sensor Array Development In Ch apter 2 a new FCS system coupled with 405 nm blue laser, which can directly track the diffusion of blue emitting CPE s i s successfully constructed Details about the theory and construction of FCS are fully discussed. This new FCS i s further demonstrated its power and feasibility in Chapter 4 where a CPE based protein sensor array i s developed. Six C PE s with various structures and charge properties are exposed to seven proteins Upon hydrophobic or electrostatic interactions, the aggregation states of CPE s in aqueous solution change, which result s in the change of the diffusion behavior of the CPE s The diversity in structure characteristics, interaction properties and the molecular scales of both polymer s and protein s contribute to the variation of the fi nal molecular size and diffusion behavior of probe s which can be recorded by FCS. The result ing data matrix i s analyzed by linear discriminant analysis (LDA). The 3D signal response pattern created through LDA demonstrate s the discrimination ability of th is sensor array for seven proteins. The success in identification of unknown samples further verifie s the feasibility of this new sensor array

PAGE 158

158 Study o f m eta Linked Poly(Phenylene Ethynylene) Sulfonate Containing Pyridine Quenched b y Metal Ions The changes in photophysical properties of a n mPPESO 3 py which is a CPE containing meta linked pyridine rings in the backbone in various conditions are investigated Based on the spectra, the electron donor acceptor complex is formed between alkox y substitute phenyl and pyridyl during coil helix transition driven by solv o phobicity and stacking interaction. The polymer show s a great affinity and selectivity for the Pd 2+ ion for the quenching efficient of Pd 2+ greatly surpasses that of the other metal ions FCS study further reveals that some multivalent metal ions including Cr 3+ Fe 3 + Fe 2+ and Cu 2+ can induce aggregation of mPPESO 3 py due to the crosslinking between metal ions and Ns belonging to adjacent polymers. Failure in getting smooth correlation curve for polymer with Pd 2+ is due to its compact conformation and low quantum yie ld which significantly affect the photon detection. The quenching mechanism is dominated by the photoinduced electron transfer (PET) process or energy transfer (ET) process instead of conformational change of mPPESO 3 py Outlook for Application of FCS in CP E In this dissertation, the FCS has successfully extended its application in the CPE field, providing insights into the physical state change of CPE and their interaction with other molecules from a molecular motion aspect, which cannot be explored by conv entional technologies. By combining FCS with CPE, a novel platform for development of molecular sensing methods has been establish ed, showing its potential and specificity. As CPE FCS based biomolecule sensing methods are just proposed and initialized f ur ther development can be made by optimizing the CPE probes

PAGE 159

159 including modifying the molecular weight distribution and quantum yield of polymers, conjugating more versatile functional groups to the backbone. Sensing assays for some potential analytes other th an proteins, such as ssDNA and dsDNA, can be designed and tested. Moreover, the development of th ese sensing assays in more complicated biological environments is necessary, so that the potential interference can be eliminate d and their feasibility in the real word can be enhanced. For the FCS setup, a temperature controlling accessory may be installed on the sample holder to achieve CPE based enzyme activity sensing, which requires a temperature higher than room temperature. In addition, if one more set of detector is introduced in, by applying different band pass emission filters, the study of fluorescence resonance energy transfer can be conducted. A cross correlation of the emission signals can also be achieved by just upgrading the correlator. Besides t he application on the biosensors discussed above, some fundamental research can be carried out on the interactions between CPEs and other molecules such as small ions, surfactants or between different CPEs. If an organic solvent resist ed container is used, such as qua rtz sample well, the investigation of polymers can be done in organic solvent s which has many benefits for those conjugated polymers that have poor water solubility.

PAGE 160

160 APPENDIX A TABLES OF TRANING DATA Table A 1 Training matrix of Log ( d / 0 ) of six CPE sensor array ( P1 P6 ) against seven proteins (1 = a vidin, 2 = LYZ, 3 = HRP, 4 = PLD2, 5 = HK3, 6 = BSA, 7 = GOx) with [CPE] = 500 nM and [protein] = 2 g/mL in 5 mM HEPEs buffer, pH 7.2, room temperature. P1 P2 P3 P4 P5 P6 Protein # 1.0 15 0.856 0.886 0.013 0.007 0.183 1 1.105 0.610 0.798 0.026 0.075 0.542 1 1.356 0.762 0.344 0.096 0.080 0.372 1 0.978 0.659 0.901 0.014 0.047 0.006 1 0.911 1.010 0.703 0.025 0.001 0.020 1 0.645 1.173 0.712 0.011 0.049 0.068 1 0.711 0.705 0.710 0.045 0.198 0.121 1 0.778 0.565 0.775 0.054 0.057 0.215 1 0.807 0.765 0.737 0.054 0.003 0.511 1 1.192 0.890 0.680 0.040 0.055 0.512 1 0.572 0.864 0.981 0.110 0.025 0.163 1 0.988 0.699 0.772 0.034 0.102 0.266 1 1.057 0.986 0.782 0.035 0.075 0 .089 1 0.279 0.639 0.627 0.004 0.034 0.162 1 0.236 0.622 0.564 0.021 0.103 0.032 1 0.737 0.926 0.612 0.098 0.021 0.036 1 0.309 0.635 0.601 0.162 0.065 0.244 1 0.289 0.656 0.560 0.116 0.039 0.099 1 0.438 0.831 0.514 0.132 0.154 0.334 1 0 .752 0.787 1.050 0.009 0.134 0.010 1 0.814 3.762 0.290 0.042 0.053 0.656 2 1.072 2.144 0.553 0.036 0.059 0.110 2 1.080 2.755 0.828 0.035 0.495 0.029 2 1.003 2.887 0.653 0.000 0.049 0.004 2 1.473 2.886 0.696 0.089 0.027 0.161 2 0.796 2.298 0. 894 0.011 0.026 0.402 2 0.439 2.125 0.473 0.063 0.044 0.098 2 0.318 2.431 0.913 0.012 0.007 0.116 2 0.850 2.263 1.302 0.014 0.233 0.178 2 1.117 3.221 0.836 0.006 0.253 0.071 2 1.030 3.363 1.171 0.003 0.248 0.542 2 1.234 2.458 1.107 0.046 0.24 7 0.439 2 1.494 3.068 0.365 0.208 0.249 0.020 2 1.733 3.061 1.345 0.047 0.244 0.040 2 1.561 3.735 1.325 0.055 0.377 0.038 2 1.572 2.153 0.811 0.051 0.221 0.003 2 1.176 3.522 0.876 0.062 0.010 0.114 2

PAGE 161

161 Table A 1 Continued. P1 P2 P3 P4 P5 P6 Prot ein # 1.564 3.816 1.273 0.065 0.057 0.106 2 1.621 3.149 0.639 0.071 0.401 0.135 2 0.669 3.263 1.099 0.015 0.028 0.154 2 0.013 0.384 0.107 1.727 0.278 0.605 3 0.045 0.374 0.041 1.799 1.842 0.467 3 0.043 0.468 0.099 0.126 2.301 0.450 3 0. 032 0.401 0.009 1.369 2.201 0.574 3 0.079 0.383 0.124 0.117 0.029 0.333 3 0.043 0.492 0.110 0.311 0.030 0.640 3 0.049 0.378 0.129 1.281 0.098 0.454 3 0.010 0.434 0.124 0.222 0.427 0.494 3 0.040 0.409 0.161 0.251 0.008 0.909 3 0.0 02 0.349 0.120 1.638 0.073 0.921 3 0.003 0.391 0.125 1.016 1.232 0.894 3 0.045 0.373 0.125 0.411 0.047 0.579 3 0.050 0.482 0.047 1.509 0.036 0.594 3 0.057 0.416 0.083 0.048 0.187 0.275 3 0.094 0.426 0.141 0.067 0.039 0.150 3 0.029 0.497 0.106 0.043 0.068 0.117 3 0.026 0.496 0.137 0.121 0.014 0.345 3 0.000 0.545 0.129 0.186 0.052 0.571 3 0.021 0.419 0.010 0.040 0.034 0.728 3 0.069 0.349 0.104 0.088 0.096 0.486 3 0.035 0.261 0.024 1.409 3.191 0.750 4 0.023 0.262 0.032 1.422 2.995 1.101 4 0.026 0.333 0.022 1.149 2.588 0.946 4 0.002 0.389 0.051 1.275 2.019 0.832 4 0.032 0.362 0.103 1.609 2.608 0.837 4 0.004 0.421 0.041 1.499 1.469 0.821 4 0.021 0.350 0.066 1.018 3.190 0.942 4 0.050 0.084 0.071 1.582 3.354 0.701 4 0.023 0.112 0.060 2.090 2.345 0.753 4 0.065 0.158 0.088 1.951 1.804 0.790 4 0.013 0.151 0.088 1.470 1.269 0.807 4 0.040 0.122 0.042 2.012 1.272 0.747 4 0.001 0.064 0.112 2.226 1.230 0.919 4 0.022 0.156 0.118 1.452 3.022 0.809 4 0.009 0.212 0.140 2.065 1.396 0.769 4 0.011 0.307 0.141 1.112 1.739 0.695 4 0.024 0.220 0.132 1.468 1.460 0.753 4 0.029 0.157 0.135 1.688 1.603 1.212 4 0.018 0.145 0.144 1.681 2.945 1.012 4 0.005 0.102 0.075 1.216 3.1 24 0.628 4

PAGE 162

162 Table A 1 Continued. P1 P2 P3 P4 P5 P6 Protein # 0.069 0.269 0.169 1.199 1.788 0.998 5 0.030 0.254 0.130 0.816 1.110 0.972 5 0.083 0.213 0.200 1.360 1.857 0.934 5 0.023 0.286 0.188 0.926 1.312 1.031 5 0.040 0.303 0.191 1. 002 1.320 1.019 5 0.029 0.223 0.308 1.173 0.778 1.587 5 0.019 0.314 0.248 1.614 1.312 0.813 5 0.042 0.330 0.228 0.973 1.193 1.178 5 0.000 0.165 0.327 0.834 0.912 1.160 5 0.086 0.213 0.258 1.220 1.355 1.090 5 0.011 0.194 0.288 1.450 0 .749 0.622 5 0.016 0.255 0.301 0.794 1.544 1.301 5 0.026 0.265 0.105 0.886 0.350 1.390 5 0.000 0.072 0.053 1.796 0.970 1.203 5 0.070 0.143 0.050 1.720 1.628 0.985 5 0.034 0.137 0.039 0.988 1.286 1.107 5 0.081 0.098 0.045 1.095 0.896 1.1 63 5 0.019 0.103 0.095 1.304 1.139 1.323 5 0.047 0.034 0.117 1.789 1.492 0.938 5 0.031 0.145 0.111 0.713 1.575 1.170 5 0.058 0.106 0.239 1.482 1.606 0.882 6 0.027 0.127 0.262 1.124 1.732 0.678 6 0.012 0.217 0.260 1.266 2.590 0.592 6 0.014 0.142 0.244 0.916 1.191 0.730 6 0.031 0.121 0.266 1.502 1.588 0.649 6 0.035 0.125 0.278 0.828 1.493 0.799 6 0.025 0.091 0.113 1.166 2.036 0.665 6 0.008 0.123 0.164 1.229 1.994 0.578 6 0.049 0.108 0.199 0.857 2.194 0.584 6 0.032 0.094 0.113 1.228 0.856 0.684 6 0.036 0.120 0.132 1.234 1.045 0.717 6 0.079 0.148 0.175 1.363 2.103 0.843 6 0.003 0.002 0.202 1.122 0.884 0.816 6 0.014 0.111 0.182 0.625 1.218 0.612 6 0.010 0.050 0.144 0.837 1.343 0.863 6 0.066 0.05 8 0.083 0.998 1.165 0.572 6 0.029 0.109 0.137 0.859 1.528 0.593 6 0.030 0.108 0.150 0.723 1.089 0.595 6 0.030 0.192 0.267 1.260 1.013 0.452 6 0.026 0.161 0.122 1.044 1.097 0.757 6 0.039 0.068 0.033 2.364 0.678 0.455 7 0.053 0.165 0.01 8 2.890 0.776 0.507 7 0.024 0.142 0.051 2.605 0.869 0.483 7

PAGE 163

163 Table A 1 Continued. P1 P2 P3 P4 P5 P6 Protein # 0.039 0.132 0.038 3.150 0.710 0.503 7 0.033 0.143 0.090 2.383 0.933 0.484 7 0.066 0.205 0.045 2.823 0.707 0.481 7 0.045 0.143 0.043 2.252 0.734 0.426 7 0.017 0.180 0.054 2.461 0.779 0.624 7 0.058 0.126 0.033 2.942 1.224 0.781 7 0.060 0.008 0.049 2.222 0.992 0.446 7 0.012 0.080 0.023 1.926 0.737 0.441 7 0.026 0.036 0.058 1.892 0.722 0.721 7 0.091 0.041 0.052 1.78 8 0.861 0.443 7 0.038 0.016 0.037 1.564 1.185 0.402 7 0.079 0.024 0.060 3.100 0.697 0.254 7 0.014 0.102 0.078 3.172 0.890 0.243 7 0.085 0.077 0.064 3.291 0.873 0.361 7 0.083 0.116 0.056 2.947 1.314 0.443 7 0.069 0.046 0.056 3.369 1.061 0.4 50 7 0.009 0.018 0.025 3.313 0.829 0.380 7

PAGE 164

164 APPENDIX B TABLE OF TEST DATA Table B 1 Unknown sample test matrix of Log ( d / 0 ) of 6 CPE sensor array ( P1 P6 ) against various proteins (1 = a vidin, 2 = LYZ, 3 = HRP, 4 = PLD2, 5 = HK3, 6 = BSA, 7 = GOx) with [CPE] = 500 nM and [protein] = 2 g/mL in 5 mM HEPEs buffer, pH 7.2, room temperature. (failure tests are denoted with s tar I = Identification, V = Verification) Sample # P1 P2 P3 P4 P5 P6 I V 1 0.035 0.223 0.019 1.341 2.198 0.718 4 4 2 1.727 1.032 0.853 0.139 0.013 0.155 1 1 3 1.763 3.791 0.835 0.039 0.017 0.141 2 2 4 0.009 0.247 0.248 0.774 0.962 1.185 5 5 5 1. 280 2.300 1.193 0.016 0.015 0.069 2 2 6 0.001 0.443 0.237 0.088 0.755 0.807 3 3 7 0.100 0.243 0.188 1.023 1.214 1.297 5 5 8 1.354 1.021 0.677 0.004 0.007 0.043 1 1 9 0.044 0.026 0.020 2.507 0.890 0.630 7 7 10 0.029 0.234 0.012 2.002 1.712 0 .688 4 4 11 0.024 0.001 0.011 2.642 0.966 0.184 7 7 12 0.046 0.123 0.263 0.916 2.191 0.730 6 6 13 0.020 0.270 0.205 1.254 1.256 0.965 5 5 14 2.635 2.235 1.480 0.235 0.165 0.029 2 2 15 0.047 0.411 0.025 1.675 2.751 0.724 4 4 16 0.003 0.39 8 0.279 1.031 1.211 0.415 3 3 17 1.052 1.332 0.724 0.020 0.024 0.490 1 1 18* 0.023* 1.157* 0.163* 1.255* 1.156* 0.772* 3* 6* 19 0.000 0.263 0.171 1.597 1.164 1.167 5 5 20 0.020 0.026 0.014 3.219 0.835 0.056 7 7 21 1.850 3.330 1.218 0.032 0. 078 0.340 2 2 22 1.260 1.041 0.775 0.018 0.090 0.386 1 1 23 0.020 0.400 0.039 1.444 2.493 0.617 4 4 24 0.023 0.455 0.235 1.380 0.050 0.673 3 3 25 0.014 0.142 0.115 0.916 1.332 0.668 6 6 26 0.034 0.002 0.020 2.612 0.926 0.531 7 7 27 0.002 0.413 0.206 0.150 1.104 0.918 5 3 28 0.038 0.113 0.259 1.126 1.270 0.576 6 6 29 0.003 0.147 0.228 1.169 1.448 1.566 5 5 30 0.033 0.021 0.012 2.881 0.914 0.902 7 7 31 0.038 0.218 0.039 1.756 1.428 0.939 4 4 32 0.086 0.150 0.2 44 1.021 1.181 0.816 5 6 33 0.082 0.068 0.224 0.971 1.493 0.808 6 6 34 0.021 0.150 0.261 1.309 0.884 1.173 5 5 35 0.030 0.304 0.012 1.643 1.330 0.948 4 4 36 0.888 1.107 0.704 0.158 0.157 0.135 1 1

PAGE 165

165 Table B 1. Continued. Sample # P1 P2 P3 P4 P5 P6 I V 37 1.266 3.183 0.867 0.009 0.087 0.202 2 2 38 0.064 0.455 0.261 0.091 0.032 0.945 3 3 39 0.029 0.016 0.011 1.497 0.873 0.091 7 7 40 0.993 2.728 1.149 0.080 0.011 0.178 2 2 41 0.010 0.409 0.261 0.202 0.103 0.796 3 3 42 0 .530 0.872 0.747 0.085 0.018 0.126 1 1 *Note : failure test

PAGE 166

166 LIST OF REFERENCES 1. J. Mann, Nature 1986, 319 818 81 8. 2. A. D. McNaught and A. Wilkinson, IUPAC. Compendium of Chemical Terminology. (the "Gold Book") Blackwell Scientific Publications, Oxford, 1997. 3. N. Hall, Chem. Commun. 2003, 1 4. 4. A. C. Grimsdale and K. Mullen, Emissive Materials: Nanomaterials 2006, 199 1 82. 5. A. Kraft, A. C. Grimsdale and A. B. Holmes, Angew. Chem. Int. Ed. 1998, 37 402 428. 6. U. H. F. Bunz, Chem. Rev. 2000, 100 1605 1644. 7. I. F. Perepichka, D. F. Perepichka, H. Meng and F. Wudl, Adv. Mater. 2005, 17 2281 2305. 8. A. O. Patil, A. J. Heeger and F. Wudl, Chem. Rev. 1988, 88 183 200. 9. A. G. MacDiarmid, Angew. Chem. Int. Ed. 2001, 40 2581 2590. 10. U. Scherf and E. J. W. List, Adv. Mater. 2002, 14 477 487. 11. H. Shirakawa Angew. Chem. Int. Ed. 2001, 40 2575 2580. 12. A. Montali, P. Smith and C. Weder, Synth. Met. 1998, 97 123 126. 13. L. Torsi, A. Dodabalapur, L. J. Rothberg, A. W. P. Fung and H. E. Katz, Science 1996, 272 1462 1464. 14. H. Sirringhaus, Adv. Mater. 2005, 17 2411 2425. 15. S. Gunes, H. Neugebauer and N. S. Sariciftci, Chem. Rev. 2007, 107 1324 1338. 16. S. W. Thomas, G. D. Joly and T. M. Swager, Chem. Rev. 2007, 107 1339 1386. 17. B. H. Wang and E. V. Anslyn, Chemosensors: Principles, Strategies and Applications John Wiley & Sons, Inc.,, Hoboken, New Jersey, 2011. 18. Y. L. Tang, F. Teng, M. H. Yu, L. L. An, F. He, S. Wang, Y. L. Li, D. B. Zhu and G. C. Bazan, Adv. Mater. 2008, 20 703 705. 19. E. W. Paul, A. J. Ricco and M. S. Wrighton, J. Ph ys. Chem. 1985, 89 1441 1447. 20. J. Janata and M. Josowicz, Acc. Chem. Res. 1998, 31 241 248.

PAGE 167

167 21. J. Gierschner, J. Cornil and H. J. Egelhaaf, Adv. Mater. 2007, 19 173 191. 22. T. Q. Nguyen, J. J. Wu, V. Doan, B. J. Schwartz and S. H. Tolbert, Scien ce 2000, 288 652 656. 23. C. Y. Tan, M. R. Pinto, M. E. Kose, I. Ghiviriga and K. S. Schanze, Adv. Mater. 2004, 16 1208 1212. 24. C. Y. Tan, M. R. Pinto and K. S. Schanze, Chem. Commun. 2002, 446 447. 25. S. H. Lee, S. Komurlu, X. Y. Zhao, H. Jiang, G Moriena, V. D. Kleiman and K. S. Schanze, Macromolecules 2011, 44 4742 4751. 26. M. R. Pinto, B. M. Kristal and K. S. Schanze, Langmuir 2003, 19 6523 6533. 27. S. Chemburu, E. Ji, Y. Casana, Y. Wu, T. Buranda, K. S. Schanze, G. P. Lopez and D. G. Whi tten, J. Phys. Chem. B 2008, 112 14492 14499. 28. X. Y. Zhao and K. S. Schanze, Chemical Communications 2010, 46 6075 6077. 29. X. Y. Zhao, M. R. Pinto, L. M. Hardison, J. Mwaura, J. Muller, H. Jiang, D. Witker, V. D. Kleiman, J. R. Reynolds and K. S. Schanze, Macromolecules 2006, 39 6355 6366. 30. E. K. Ji, D. L. Wu and K. S. Schanze, Langmuir 2010, 26 14427 14429. 31. X. Y. Zhao, H. Jiang and K. S. Schanze, Macromolecules 2008, 41 3422 3428. 32. L. H. Chen, S. Xu, D. McBranch and D. Whitten, J. Am. Chem. Soc. 2000, 122 9302 9303. 33. S. J. Dwight, B. S. Gaylord, J. W. Hong and G. C. Bazan, J. Am. Chem. Soc. 2004, 126 16850 16859. 34. M. R. Eftink and C. A. Ghiron, Anal. Biochem. 1981, 114 199 227. 35. J. R. Lakowicz, Principles of fluoresce nce spectroscopy Springer, New York, 2006. 36. Fluorescein, Oregon Green and Rhodamine Green Dyes Section 1.5, http://www.invitrogen.com/site/us/en/home/References/Molecular Probes The Handbook/Fluorophores and Their Amine Reactive Derivatives/Fluorescein Oregon Green and Rhodamine Green Dyes.html Accessed August, 6th, 2012. 37. J. Wang, D. L. Wang, E. K. Miller, D. Moses, G. C. Bazan and A. J. Heeger, Macromolecules 2000, 33 5153 5158. 38. D. L. Wang, J. Wang, D. Moses, G. C. Bazan and A. J. Heeger, Langmuir 2001, 17 1262 1266.

PAGE 168

168 39. T. Forster, J. Biomed. Opt. 2012, 17 011002 011002 011011 40. R. Y. Tsien, Annu. Rev. Biochem 1998, 67 509 544. 41. W. H. Tan, K. M. Wang and T. J. Drake, Curr. Opin. Chem. Biol. 2004, 8 547 553. 42. Q. Zh ou and T. M. Swager, J. Am. Chem. Soc. 1995, 117 12593 12602. 43. L. H. Chen, D. W. McBranch, H. L. Wang, R. Helgeson, F. Wudl and D. G. Whitten, P Natl Acad Sci USA 1999, 96 12287 12292. 44. B. S. Harrison, M. B. Ramey, J. R. Reynolds and K. S. Schanz e, J. Am. Chem. Soc. 2000, 122 8561 8562. 45. C. Y. Tan, E. Alas, J. G. Muller, M. R. Pinto, V. D. Kleiman and K. S. Schanze, J. Am. Chem. Soc. 2004, 126 13685 13694. 46. J. M. Frank and S. J. Wawilow, Zeitschrift Fur Physik 1931, 69 100 110. 47. H. Jiang, X. Y. Zhao and K. S. Schanze, Langmuir 2006, 22 5541 5543. 48. M. Levitus, K. Schmieder, H. Ricks, K. D. Shimizu, U. H. F. Bunz and M. A. Garcia Garibay, J. Am. Chem. Soc. 2001, 123 4259 4265. 49. K. A. Walters, K. D. Ley and K. S. Schanze, Lang muir 1999, 15 5676 5680. 50. J. Kim and T. M. Swager, Nature 2001, 411 1030 1034. 51. R. B. Prince, J. G. Saven, P. G. Wolynes and J. S. Moore, J. Am. Chem. Soc. 1999, 121 3114 3121. 52. J. C. Nelson, J. G. Saven, J. S. Moore and P. G. Wolynes, Scien ce 1997, 277 1793 1796. 53. S. Lahiri, J. L. Thompson and J. S. Moore, J. Am. Chem. Soc. 2000, 122 11315 11319. 54. Y. Liu, K. Ogawa and K. S. Schanze, J. Photochem. Photobiol., C 2009, 10 173 190. 55. D. L. Zeng, J. G. Cheng, S. J. Ren, J. Sun, H. L Zhong, E. J. Xu, J. P. Du and Q. A. Fang, React. Funct. Polym. 2008, 68 1715 1721. 56. C. F. Xing, Z. Q. Shi, M. H. Yu and S. Wang, Polymer 2008, 49 2698 2703. 57. R. Zeineldin, M. E. Piyasena, T. S. Bergstedt, L. A. Sklar, D. Whitten and G. P. Lopez Cytom. Part A 2006, 69A 335 341.

PAGE 169

169 58. C. H. Fan, K. W. Plaxco and A. J. Heeger, J. Am. Chem. Soc. 2002, 124 5642 5643. 59. K. E. Achyuthan, T. S. Bergstedt, L. Chen, R. M. Jones, S. Kumaraswamy, S. A. Kushon, K. D. Ley, L. Lu, D. McBranch, H. Mukundan F. Rininsland, X. Shi, W. Xia and D. G. Whitten, J. Mater. Chem. 2005, 15 2648 2656. 60. S. Kumaraswamy, T. Bergstedt, X. B. Shi, F. Rininsland, S. Kushon, W. S. Xia, K. Ley, K. Achyuthan, D. McBranch and D. Whitten, Proc. Natl. Acad. Sci. U. S. A. 20 04, 101 7511 7515. 61. M. R. Pinto and K. S. Schanze, Proc. Natl. Acad. Sci. U. S. A. 2004, 101 7505 7510. 62. F. Rininsland, W. S. Xia, S. Wittenburg, X. B. Shi, C. Stankewicz, K. Achyuthan, D. McBranch and D. Whitten, Proc. Natl. Acad. Sci. U. S. A. 2004, 101 15295 15300. 63. B. S. Gaylord, A. J. Heeger and G. C. Bazan, Proc. Natl. Acad. Sci. U. S. A. 2002, 99 10954 10957. 64. S. A. Kushon, K. D. Ley, K. Bradford, R. M. Jones, D. McBranch and D. Whitten, Langmuir 2002, 18 7245 7249. 65 S. A. Kushon, K. Bradford, V. Marin, C. Suhrada, B. A. Armitage, D. McBranch and D. Whitten, Langmuir 2003, 19 6456 6464. 66. B. S. Gaylord, A. J. Heeger and G. C. Bazan, J. Am. Chem. Soc. 2003, 125 896 900. 67. I. B. Kim and U. H. F. Bunz, J. Am. Ch em. Soc. 2006, 128 2818 2819. 68. H. A. Ho and M. Leclerc, J. Am. Chem. Soc. 2004, 126 1384 1387. 69. F. He, Y. L. Tang, S. Wang, Y. L. Li and D. B. Zhu, J. Am. Chem. Soc. 2005, 127 12343 12346. 70. N. DiCesare, M. R. Pinto, K. S. Schanze and J. R. L akowicz, Langmuir 2002, 18 7785 7787. 71. Y. Liu and K. S. Schanze, Anal. Chem. 2008, 80 8605 8612. 72. C. Li, M. Numata, M. Takeuchi and S. Shinkai, Angew. Chem. Int. Ed. 2005, 44 6371 6374. 73. F. He, Y. L. Tang, M. H. Yu, S. Wang, Y. L. Li and D. B. Zhu, Adv. Funct. Mater. 2006, 16 91 94.

PAGE 170

170 74. Y. L. Tang, F. He, M. H. Yu, S. Wang, Y. L. Li and D. B. Zhu, Chem. Mater. 2006, 18 3605 3610. 75. H. A. Ho, A. Najari and M. Leclerc, Acc. Chem. Res. 2008, 41 168 178. 76. X. R. Duan, L. B. Liu, F. D. F eng and S. Wang, Acc. Chem. Res. 2010, 43 260 270. 77. G. C. Bazan and S. Wang, in Organic Semiconductors in Sensor Applications eds. D. A. Bernards, G. G. Malliaras and R. M. Owens, Springer Berlin Heidelberg, Berlin Heidelberg, 2008, pp. 1 37. 78. Q. H. Xu, S. Wang, D. Korystov, A. Mikhailovsky, G. C. Bazan, D. Moses and A. J. Heeger, Proc. Natl. Acad. Sci. U. S. A. 2005, 102 530 535. 79. C. Y. J. Yang, M. Pinto, K. Schanze and W. H. Tan, Angew. Chem. Int. Ed. 2005, 44 2572 2576. 80. K. Lee, L. K. Povlich and J. Kim, Adv. Funct. Mater. 2007, 17 2580 2587. 81. A. Kessel and N. Ben Tal, in Chapman & Hall/CRC mathematical and computational biology series CRC Press, Boca Raton, FL. 82. D. J. Selkoe, Nature 2003, 426 900 904. 83. R. Aebersold and M. Mann, Nature 2003, 422 198 207. 84. C. J. Johnson, N. Zhukovsky, A. E. G. Cass and J. M. Nagy, Proteomics 2008, 8 715 730. 85. S. Fields, Science 2001, 291 1221 1224. 86. J. Wang, G. D. Liu, M. H. Engelhard and Y. H. Lin, Anal. Chem. 2006, 78 6974 6979. 87. T. Li, L. P. Guo and Z. X. Wang, Biosens. Bioelectron. 2008, 23 1125 1130. 88. C. P. Soman and T. D. Giorgio, Langmuir 2008, 24 4399 4404. 89. C. D. Chin, V. Linder and S. K. Sia, Lab on a Chip 2007, 7 41 57. 90. G. B. Hurst, M. V. Buchana n, L. J. Foote and S. J. Kennel, Anal. Chem. 1999, 71 4727 4733. 91. H. Jiang, P. Taranekar, J. R. Reynolds and K. S. Schanze, Angew. Chem. Int. Ed. 2009, 48 4300 4316. 92. F. D. Feng, L. B. Liu, Q. O. Yang and S. Wang, Macromol. Rapid Commun 2010, 31 1405 1421.

PAGE 171

171 93. K. Li and B. Liu, Polymer Chemistry 2010, 1 252 259. 94. D. T. McQuade, A. E. Pullen and T. M. Swager, Chem. Rev. 2000, 100 2537 2574. 95. R. J. Delange, J. Biol. Chem. 1970, 245 907 916. 96. F. Cheng, G. W. Zhang, X. M. L u, Y. Q. Huang, Y. Chen, Y. Zhou, Q. L. Fan and W. Huang, Macromol. Rapid Commun. 2006, 27 799 803. 97. O. R. Miranda, C. C. You, R. Phillips, I. B. Kim, P. S. Ghosh, U. H. F. Bunz and V. M. Rotello, J. Am. Chem. Soc. 2007, 129 9856 9857. 98. K. P. R. Nilsson and O. Inganas, Macromolecules 2004, 37 9109 9113. 99. L. L. An, Y. L. Tang, F. D. Feng, F. He and S. Wang, J. Mater. Chem. 2007, 17 4147 4152. 100. F. D. Feng, Y. L. Tang, F. He, M. H. Yu, X. R. Duan, S. Wang, Y. L. Li and D. B. Zhu, Adv. Mate r. 2007, 19 3490 3495. 101. F. Pu, D. Hu, J. S. Ren, S. Wang and X. G. Qu, Langmuir 2010, 26 4540 4545. 102. Y. Y. Wang, Y. Zhang and B. Liu, Anal. Chem. 2010, 82 8604 8610. 103. Y. Liu, K. Ogawa and K. S. Schanze, Anal. Chem. 2008, 80 150 158. 104 H. L. Fan, T. Zhang, S. W. Lv and Q. H. Jin, J. Mater. Chem. 2010, 20 10901 10907. 105. L. L. An, L. B. Liu and S. Wang, Biomacromolecules 2009, 10 454 457. 106. J. M. Busnel, S. Descroix, T. Le Saux, S. Terabe, M. C. Hennion and G. Peltre, Electroph oresis 2006, 27 1481 1488. 107. Peptide Property Calculator, http://www.innovagen.se/custom peptide synthesis/peptide property calculator/peptide property calculator.asp Accessed Sep. 16th, 2012. 108. Y. Liu and K. S. Schanze, Anal. Chem. 2009, 81 231 239. 109. D. L. Wang, X. Gong, P. S. Heeger, F. Rininsland, G. C. Bazan and A. J. Heeger, Proc. Natl. Acad. Sci. U. S. A. 2002, 99 49 53. 110. I. B. Kim, A. Dunkhorst and U. H. F. Bunz, Langmuir 2005, 21 7985 7989. 111. F. Huang, X. H. Wang, D. L. Wan g, W. Yang and Y. Cao, Polymer 2005, 46 12010 12015.

PAGE 172

172 112. K. Ogawa, S. Chemburu, G. P. Lopez, D. G. Whitten and K. S. Schanze, Langmuir 2007, 23 4541 4548. 113. I. B. Kim, M. H. Han, R. L. Phillips, B. Samanta, V. M. Rotello, Z. J. Zhang and U. H. F. B unz, Chem. Eur. J. 2009, 15 449 456. 114. J. Wang and B. Liu, Chem. Commun. 2009, 2284 2286. 115. H. P. Li and G. C. Bazan, Adv. Mater. 2009, 21 964 967. 116. A. T. Wright and E. V. Anslyn, Chem. Soc. Rev. 2006, 35 14 28. 117. J. W. Lee, J. S. Lee a nd Y. T. Chang, Angew. Chem. Int. Ed. 2006, 45 6485 6487. 118. N. A. Rakow and K. S. Suslick, Nature 2000, 406 710 713. 119. N. T. Greene and K. D. Shimizu, J. Am. Chem. Soc. 2005, 127 5695 5700. 120. J. F. Folmer Andersen, M. Kitamura and E. V. Anslyn, J. Am. Chem. Soc. 2006, 128 5652 5653. 121. A. Buryak and K. Severin, J. Am. Chem. Soc. 2005, 127 3700 3701. 122. C. C. You, O. R. Miranda, B. Gider, P. S. Ghosh, I. B. Kim, B. Erdogan, S. A. Krovi, U. H. F. Bunz and V. M. Rotello, Nature Nanotech. 2007, 2 318 323. 123. L. L. An, S. Wang and D. B. Zhu, Chem Asian J 2008, 3 1601 1606. 124. A. Bajaj, O. R. Miranda, R. Phillips, I. B. Kim, D. J. Jerry, U. H. F. Bunz and V. M. Rotello, J. Am. Chem. Soc. 2010, 132 1018 1022. 125. P. C. Ju rs, G. A. Bakken and H. E. McClelland, Chem. Rev. 2000, 100 2649 2678. 126. R. A. Johnson and D. W. Wichern, Applied multivariate statistical analysis Prentice Hall, Englewood Cliffs, N.J., 1988. 127. A. F. Thunemann and D. Ruppelt, Langmuir 2001, 17 5098 5102. 128. A. F. Thunemann, Adv. Mater. 1999, 11 127 130. 129. L. A. Bastardo, V. M. Garamus, M. Bergstrom and P. M. Claesson, J. Phys. Chem. B 2005, 109 167 174. 130. F. Stabo Eeg, M. Lindgren, K. P. R. Nilsson, O. Inganas and P. Hammarstrom, Che mical Physics 2007, 336 121 126.

PAGE 173

173 131. H. J. Yue, M. Y. Wu, C. H. Xue, S. Velayudham, H. Y. Liu and D. H. Waldeck, J. Phys. Chem. B 2008, 112 8218 8226. 132. M. Y. Wu, P. Kaur, H. J. Yue, A. M. Clemmens, D. H. Waldeck, C. H. Xue and H. Y. Liu, J. Phys. Chem. B 2008, 112 3300 3310. 133. D. Magde, W. W. Webb and E. Elson, Physical Review Letters 1972, 29 705 708. 134. O. Krichevsky and G. Bonnet, Rep. Prog. Phys. 2002, 65 251 297. 135. J. R. Lakowicz, Topics in fluorescence spectroscopy Plenum Press New York, 1991. 136. P. Schwille and E. Haustein, Fluorescence correlation spectroscopy. An introduction to its concepts and applications, www.biophysics.org/education/schwille.pdf 137. P. Kask, P. Piksarv, M. Pooga, U. Mets and E. Lippmaa, Biophys. J. 1989, 55 213 220. 138. P. F. Fahey, D. E. Koppel, L. S. Barak, D. E. Wolf, E. L. Elson and W. W. Webb, Science 1977, 195 305 306. 139. P. F. Fahey and W. W. Webb, Biochemistry 1978, 17 3046 3053. 140. D. E. Koppel, D. Axelrod, J. Schlessinger, E. L. Elson and W. W. Webb, Biophys. J. 1976, 16 1315 1329. 141. J. Widengren, U. Mets and R. Rigler, Chemical Physics 1999, 250 171 186. 142. M. Borsch, P. Turina, C. Eggeling, J. R. Fries, C. A. M. Seidel, A. Labahn and P. Graber, Febs Letters 1998, 437 251 254. 143. J. Schuler, J. Frank, U. Trier, M. Schafer Korting and W. A. Saenger, Biochemistry 1999, 38 8402 8408. 144. E. Van Craenenbroeck and Y. Engelborghs, Biochemistry 1999, 38 5082 5088. 145. H. Schurer, A. Buchynskyy, K. Korn, M. Famulok, P. Welzel and U. Hahn, Biol. Chem. 2001, 382 479 481. 146. T. Wohland, K. Friedrich, R. Hovius and H. Vogel, Biochemistry 1999, 38 8671 8681. 147. M. Kinjo and R. Rigler, Nucleic Acids Res. 1995, 23 1795 1799. 148. J. Borejdo, Biopolymers 19 79, 18 2807 2820.

PAGE 174

174 149. H. Pick, A. K. Preuss, M. Mayer, T. Wohland, R. Hovius and H. Vogel, Biochemistry 2003, 42 877 884. 150. R. Rigler, U. Mets, J. Widengren and P. Kask, Eur. Biophys. J. Biophy. 1993, 22 169 175. 151. R. M. Rigler, Ulo, Proc. Soc. Photo. Opt. Instrum. Eng. 1993, 1921 239 248. 152. M. Eigen and R. Rigler, Proc. Natl. Acad. Sci. U. S. A. 1994, 91 5740 5747. 153. E. Van Rompaey, N. Sanders, S. C. De Smedt, J. Demeester, E. Van Craenenbroeck and Y. Engelborghs, Macromolecules 2000 33 8280 8288. 154. T. B. Bonne, C. M. Papadakis, K. Ludtke and R. Jordan, Colloid. Polym. Sci. 2007, 285 491 497. 155. A. Laguecir, S. Ulrich, J. Labille, N. Fatin Rouge, S. Stoll and J. Buffle, Eur. Polym. J. 2006, 42 1135 1144. 156. A. V. R. Murth y, M. Goel, S. Patil and M. Jayakannan, J. Phys. Chem. B 2011, 115 10779 10788. 157. Z. H. Xu, H. H. Tsai, H. L. Wang and M. Cotlet, J. Phys. Chem. B 2010, 114 11746 11752. 158. S. Masuo, N. Nishi, M. Hosugi, S. Machida and A. Itaya, Chem. Lett. 2010, 39 780 782. 159. A. B. Andrews, R. E. Guerra, O. C. Mullins and P. N. Sen, J Phys Chem A 2006, 110 8093 8097. 160. S. Ghosh, U. Mandal, A. Adhikari and K. Bhattacharyya, Chem Asian J 2009, 4 948 954. 161. E. Haustein and P. Schwille, Annu Rev Biophys Biomol Struct 2007, 36 151 169. 162. E. Overbeck, C. Sinn, I. Flammer and J. Ricka, Rev. Sci. Instrum. 1998, 69 3515 3523. 163. H. C. Burstyn and J. V. Sengers, Physical Review A 1983, 27 1071 1085. 164. J. X. Zhu, D. J. Durian, J. Muller, D. A. Wei tz and D. J. Pine, Physical Review Letters 1992, 68 2559 2562. 165. M. Zhao, L. Jin, B. Chen, Y. Ding, H. Ma and D. Y. Chen, Appl. Opt. 2003, 42 4031 4036.

PAGE 175

175 166. S. Wilhelm, Confocal Laser Scanning Microscopy, http://www.zeiss.de/C1256D18002CC306/0/F99A 7F3E8944EEE3C1256E5C0045 F68B/$file/60 1 0030_confocal principles.pdf 167. Specific gravity, also called relative density, is the ratio of the substance to the density of water at 4 C 168. Free Space Fiber Coupler for Single Mode Fiber, http://www.thorlabs .com/catalogpages/V21/198.PDF Accessed Sep. 29th, 2012. 169. K. R. Spring, T. J. Fellers and M. W. Davidson, Confocal Microscope Objectives, http://www.olympusmicro.com/primer/techniques/confocal/confocalobjectives.html Accessed Sep. 15th, 2012. 170. S. Wilhelm, Carl Zeiss MicroImaging GmbH, Jena, Germany. 171. P. Sengupta, J. Balaji and S. Maiti, Methods 2002, 27 374 387. 172. M. Abramowitz and M. W. Davidson, Immersion Media, http://www.olympusmicro.com/primer/anatomy/immersion.html Accessed Sep. 15t h, 2012. 173. K. R. Spring, H. E. Keller and M. W. Davidson, Microscope Objectives Introduction, http://www.olympusmicro.com/primer/anatomy/objectives.html Accessed Sep. 15th, 2012. 174. D. E. Wolf What is the Confocal Volume?, http://www.fcsx pert.com/classroom/theory/pdfs/fcs_confocal.pdf Accessed Sep. 15th, 2012. 175. J. D. Muller, Y. Chen and E. Gratton, Biophotonics, Pt B 2003, 361 69 92. 176. D. L. Wu, F. D. Feng, D. P. Xie, Y. Chen, W. H. Tan and K. S. Schanze, J. Phys. Chem. Lett. 20 12, 3 1711 1715. 177. Although avidin only has 30% sequence identity to streptavidin, they have almost identical secondary, tertiary and quaternary structure 178. L. Haussling, H. Ringsdorf, F. J. Schmitt and W. Knoll, Langmuir 1991, 7 1837 1840. 179. G Balgi, D. E. Leckband and J. M. Nitsche, Biophysical Journal 1995, 68 2251 2260. 180. O. Livnah, E. A. Bayer, M. Wilchek and J. L. Sussman, P Natl Acad Sci USA 1993, 90 5076 5080. 181. Fude Feng, Kirk S. Schanze, paper in preparation.

PAGE 176

176 182. K. Lee, J. C. Cho, J. DeHeck and J. Kim, Chemical Communications 2006, 1983 1985. 183. D. P. Xie, in Department of Chemistry University of Florida, Gainesville, FL, 2012, p. 152. 184. S. Taira and K. Yokoyama, Biotechnol. Bioeng. 2004, 88 35 41. 185. M. X. You, R. W. Wang, X. B. Zhang, Y. Chen, K. L. Wang, L. Peng and W. H. Tan, Acs Nano 2011, 5 10090 10095. 186. Y. Kawabe, L. Wang, S. Horinouchi and N. Ogata, Adv. Mater. 2000, 12 1281 1283. 187. A. E. Friedman, J. C. Chambron, J. P. Sauvage, N. J. Turro and J. K. Barton, J. Am. Chem. Soc. 1990, 112 4960 4962. 188. R. M. Hartshorn and J. K. Barton, J. Am. Chem. Soc. 1992, 114 5919 5925. 189. D. M. Mock, N. I. Mock, C. W. Stewart, J. B. LaBorde and D. K. Hansen, J. Nutr. 2003, 133 2519 2525. 190. H. P. Le sch, M. U. Kaikkonen, J. T. Pikkarainen and S. Yla Herttuala, Expert Opinion on Drug Delivery 2010, 7 551 564. 191. M. Wilchek and E. A. Bayer, Biomol. Eng 1999, 16 1 4. 192. J. T. He, Z. H. Shi, J. Yan, M. P. Zhao, Z. Q. Guo and W. B. Chang, Talanta 2005, 65 621 626. 193. M. Jeon, J. Kim, K. J. Paeng, S. W. Park and I. R. Paeng, Microchem. J. 2008, 88 26 31. 194. K. Tanaka, S. Yokoi, K. Morimoto, T. Iwata, Y. Nakamoto, K. Nakayama, K. Koyama, T. Fujiwara and K. Fukase, Biorg. Med. Chem. 2012, 20 1865 1868. 195. M. Q. Lu, W. Ngo, Y. Mei, V. Munshi, C. Burlein, M. H. Loughran, P. D. Williams, D. J. Hazuda, M. D. Miller, J. A. Grobler, T. L. Diamond and M. T. Lai, Protein Expres. Purif. 2010, 71 231 239. 196. E. Ji, in Department of Chemistry Univ ersity of Florida, Gainesville, FL, 2009, p. 167. 197. M. Slim, N. Durisic, P. Grutter and H. F. Sleiman, Chembiochem 2007, 8 804 812. 198. C. M. Ruan, K. F. Zeng, O. K. Varghese and C. A. Grimes, Biosens. Bioelectron. 2004, 19 1695 1701.

PAGE 177

177 199. G. S. Ho ng, C. Li and L. M. Qi, Adv. Funct. Mater. 2010, 20 3774 3783. 200. S. Otsuki and M. Ishikawa, Biosens. Bioelectron. 2010, 26 202 206. 201. X. D. Song, H. L. Wang, J. Shi, J. W. Park and B. I. Swanson, Chem. Mater. 2002, 14 2342 2347. 202. P. Schwille, S. Kummer, A. A. Heikal, W. E. Moerner and W. W. Webb, Proc. Natl. Acad. Sci. U. S. A. 2000, 97 151 156. 203. R. Paolesse, D. Monti, F. Dini and C. Di Natale, in Luminescence Applied in Sensor Science Springer Verlag Berlin, Berlin, Heidel berg, 2011, pp. 139 174. 204. A. P. Demchenko, Trends Biotechnol. 2005, 23 456 460. 205. D. Perahia, R. Traiphol and U. H. F. Bunz, J. Chem. Phys. 2002, 117 1827 1832. 206. A. Ajayaghosh and S. J. George, J. Am. Chem. Soc. 2001, 123 5148 5149. 207. A s shown in the webpage of Sigma company, HK3 has two isoforms, PI and PII (A and B), pI = 5.25 and 4, respectively. 208. HRP has at least seven isozymes with pI values range from 3.0 9.0 as descriped in the webpage of Sigma company 209. M. M. Bradford, Anal. Biochem. 1976, 72 248 254. 210. Y. Wang, E. Y. Chi, K. S. Schanze and D. G. Whitten, Soft Matter 2012, 8 8547 8558. 211. M. Loog, R. P. W. Duin and R. Haeb Umbach, IEEE Transactions on Pattern Analysis and Machine Intelligence 2001, 23 762 766. 212. M. Uray, P. M. Roth and H. Bischof, in Proc.Int'l Conf. on Computer Vision Theory and Applications 2009. 213. Bradford Protein Assay, http://www.science.smith.edu/departments/Biochem/Biochem_353/Bradford.html Accessed Mar. 15th, 2012. 214. Extincti on Coefficients, < http://www.piercenet.com/files/TR0006 Extinction coefficients.pdf> Accessed Dec. 22nd, 2012, 2012. 215. Y. Chen, in Department of Biophysics & Computational Biology University of Illinois at Urbana Champaign, Urbana, Illinois, 1999, p. 140. 216. I. B. Kim, A. Dunkhorst, J. Gilbert and U. H. F. Bunz, Macromolecules 2005, 38 4560 4562.

PAGE 178

178 217. B. Liu, H. G. Dai, Y. Y. Bao, F. F. Du, J. Tian and R. K. Bai, Polymer Chemistry 2011, 2 1699 1705. 218. B. Liu, Y. Y. Bao, F. F. Du, H. Wang, J. A Tian and R. K. Bai, Chem. Commun. 2011, 47 1731 1733. 219. B. Liu, Y. Y. Bao, H. Wang, F. F. Du, J. Tian, Q. B. Li, T. S. Wang and R. K. Bai, J. Mater. Chem. 2012, 22 3555 3561. 220. D. Winter and C. D. Eisenbach, J. Polym. Sci., Part A: Polym. Chem. 2004, 42 1919 1933. 221. C. H. Li, Y. B. Guo, J. Lv, J. L. Xu, Y. L. Li, S. Wang, H. B. Liu and D. B. Zhu, J. Polym. Sci., Part A: Polym. Chem. 2007, 45 1403 1412. 222. Y. Q. Hu, Y. Xiao, H. M. Huang, D. L. Yin, X. M. Xiao and W. H. Tan, Chem Asian J 2011, 6 1500 1504. 223. J. C. Wataha and C. T. Hanks, J. Oral. Rehabil. 1996, 23 309 320. 224. J. Kielhorn, C. Melber, D. Keller and I. Mangelsdorf, Int. J. Hyg. Envir. Heal. 2002, 205 417 432. 225. B. Dimitrova, K. Benkhedda, E. Ivanova and F. Adams J. Anal. At. Spectrom. 2004, 19 1394 1396. 226. C. Locatelli, D. Melucci and G. Torsi, Anal. Bioanal. Chem. 2005, 382 1567 1573. 227. K. Van Meel, A. Smekens, M. Behets, P. Kazandjian and R. Van Grieken, Anal. Chem. 2007, 79 6383 6389. 2 28. R. N. Rao and M. V. N. K. Talluri, J. Pharm. Biomed. Anal. 2007, 43 1 13. 229. E. Unterreitmaier and M. Schuster, Anal. Chim. Acta 1995, 309 339 344. 230. H. M. Huang, K. Wang, W. H. Tan, D. An, X. H. Yang, S. S. Huang, Q. Zhai, L. Zhou and Y. Jin, Angew. Chem. Int. Ed. 2004, 43 5635 5638. 231. T. Schwarze, H. Muller, C. Dosche, T. Klamroth, W. Mickler, A. Kelling, H. G. Lohmannsroben, P. Saalfrank and H. J. Holdt, Angew. Chem. Int. Ed. 2007, 46 1671 1674. 232. F. L. Song, A. L. Garner and K. Ko ide, J. Am. Chem. Soc. 2007, 129 12354 12355. 233. C. Y. Tan, in Department of Chemistry University of Florida, Gainesville, FL, 2004, p. 163.

PAGE 179

179 234. C. H. Chen, W. H. Chen, Y. H. Liu, T. S. Lim and T. Y. Luh, Chem. Eur. J. 2012, 18 347 354. 235. B. Wan g and M. R. Wasielewski, J. Am. Chem. Soc. 1997, 119 12 21.

PAGE 180

180 BIOGRAPHICAL SKETCH Danlu Wu was born in the city of Longyan, Fujian province, China. In 2003, s he graduated from No. 1 Middle School of Longyan and got the admission of University of Science and Technology of China (USTC ) S he started her college education i n September and obtained her polymer chemistry four years later In A ugust of 2007 she came to United States and enrolled in Department of Materials Science and Engineering, Universi ty of Florida for graduate study One year later, she got the master degree in materials science and then she transferred to Department of Chemistry in the same university. Danlu joined the group of Dr. Kirk S. Schanze to pursue her Ph.D. in analytical che mistry. In the past four and half years, she researched in the amazing area of development of conjugated polyelectrolytes based fluorescent sens ing methods She received her Ph.D. in the fall of 2012.