Conjugated Polyelectrolytes Aggregation and Sensor Applications

MISSING IMAGE

Material Information

Title:
Conjugated Polyelectrolytes Aggregation and Sensor Applications
Physical Description:
1 online resource (152 p.)
Language:
english
Creator:
Yang, Jie
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, Yunwei Charles
Powell, David Hinton
Douglas, Elliot Paul

Subjects

Subjects / Keywords:
aggregation -- polyelectrolyte -- sensor
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:
Over the past several years, significant effort has beendevoted to explore the application of conjugated polyelectrolytes (CPE) aschemical and biosensors for the detection and analysis of a variety ofmolecules of environmental and biological interests. Well established sensingmechanisms have been reported and described to provide guidelines forunderlying concepts of CPE-based sensors. In this dissertation, we focus on thedesign and development of fluorescent biosensors based on functionalizedpoly(para-phenylene ethynylene)s (PPEs) mainly by taking advantage of theiraggregation-induced quenching mechanism.  A cationic polyamine-substitutedpoly(phenylene-ethynylene) (PPE-dNH3) is used to developa real-time fluorescent assay for acetate kinase (ACK) based on theaggregation-induced fluorescence quenching by PPi. The production of PPi fromdephosphorylation of acetyl-Pi which is catalyzed by ACK will cause asignificant spectroscopic change; therefore it allows us direct quantificationdetection of PPi as well as of rapid detection of ACK enzyme activity with adetection limit of 5 nM using ratiometric signals. Kinetic parameters relatedto this reaction are also derived. Then, we focus on theaggregation studies between different metal ions with another PPE derivative, abranched carboxylate-functionalized poly(phenylene-ethynylene) (PPE-dCO2)by FCS. Correlation curves for PPE-dCO2- inwater exhibit a pronounced increase in diffusion time when multivalent metal cationsare added to the solution. It is concluded that the diffusion time increases inthe order K+ »Na+ 2+ 2+ 2+3+, and large diffusion timeusually indicates the formation of large aggregates. Comparison of Stern-Volmerplot with diffusion time ratio plot shows the importance of aggregation in theamplified quenching effect. Last, we design anotherfluorescence “turn-on” sensor for PPi based on a homo poly(phenyleneethynylene) carboxylate (PPE-h-CO2). The “on-off-on” fluorescencechange is achieved on the basis of conformational change of PPE in the processof “deaggregate-aggregate-deaggregate” by adding triamines and PPi. Thenprinciple component analysis (PCA) method is utilized to convert a large set ofcomplicated fluorescence spectra intotwo linearly uncorrelated eigenvectors. And regressionanalysis is performed to predict the unknown concentration of PPi with ~ 95%accuracy.
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 Jie Yang.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Schanze, Kirk S.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-05-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045292:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 CONJUGAT ED POLYELECTROLYTES AGGREGATION AND SENSOR APPLICATIONS By JIE YANG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 201 3

PAGE 2

2 201 3 J ie Yang

PAGE 3

3 To my family

PAGE 4

4 ACKNOWLEDGEMENTS First of all, I would like to acknowledge my research advisor, Dr. Kirk Schanze, for his support, advice and encouragement He guided me into the amazing area of optical sensor s based on conjugated polyelectrolytes and encouraged me to think independe ntly and creatively. He is such a great supervisor with enthusiasm for science and kindness to help people and I sincerely apprecia ted all his advice and help for the past five years. My deepest gratitude also goes to all the former and current members from D r Schanze s group. Many of my projects couldn t have been completed without the guidance and help of Dr Fude Feng, who has broad knowledge and extensive experience about Polymer Chemistry and most importantly, he shared a lot with me and helped me to overcome several difficulties in my projects. Dr. Abby Shelton and Dr. Eunkyung Ji kindly taught me how to use the almost all the instruments in my lab and helped me out whenever I had a technical problem. I d like to thanks to Dr. Liao Chen for his patience and kindness to answering all kinds of my questions when we shared the same office. Dr. Jonathan Sommer gave me a lot of good advice on my first project and I still remembered his humor and his passion for surfing. I am also thankful to Dr. Aand Parthasarathy, Dr. Dongping Xie, Dr. Seoung Ho Lee Dr. Julia Keller, Zhuo Chen all the nice suggestion and discus sion about my research It is also dedicated to Xuzh Zhu for his help and advice on principal component analysis and matlab Of course, I d like to

PAGE 5

5 thank other group members, Dr. Galyna Dubinina, Dr. Jan Moritz Koene, Dr. Danlu Wu, Randi Price, Zhengxing Pan, Subhadip Goswami for their group work to make my research easier. I wo uld like to extend my appreciation to all my committee members: Dr. Ronald Castellano, Dr. Charles Cao, Dr. David Powell and Dr. Elliot Douglas for their time, encouragement, insig ht ful comment and suggestion. I d also like to thank Dr. Erkan Kose from North Dakota State University for his nice advice on my last project. Finally, I d like to express my gratitude to my family and my boyfriend Dou for their love and understanding for my work. Without their support and love, I will not make my dream come true.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGEMENTS ................................ ................................ ............................... 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGURES ................................ ................................ ................................ ........ 11 LIST OF ABBREVIATIONS ................................ ................................ ........................... 15 ABSTRACT ................................ ................................ ................................ ................... 19 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 21 Conjugated Polyelectrolytes ................................ ................................ ................... 21 Aggregation of Conjugated Polyelectrolytes ................................ ........................... 23 Factors that Influence the Aggregation of Conjugated Polyelectrolytes ........... 24 Solvent dependence ................................ ................................ .................. 24 Branched side groups and pH effects ................................ ........................ 25 Addition of oppositely charged ions ................................ ........................... 28 Interaction of surfactants with CPEs ................................ .......................... 30 Effect of Aggregation on Amplified Quenching ................................ ................. 32 Fluorescence Correlation Spectroscopy (FCS) ................................ ....................... 33 Principles of FCS ................................ ................................ .............................. 34 Typical FCS Setup ................................ ................................ ........................... 35 Applications of FCS ................................ ................................ .......................... 37 Aggregation behavior ................................ ................................ ................. 37 Size measurement ................................ ................................ ..................... 39 Sensor Applications ................................ ................................ ................................ 40 Small Io n Sensing ................................ ................................ ............................ 41 Protein Sensing ................................ ................................ ................................ 43 DNA Sensing ................................ ................................ ................................ .... 45 Challenges of CPE based Optical Sensors ................................ ...................... 47 Overview of This Dissertation ................................ ................................ ................. 48

PAGE 7

7 2 ION INDUCED AGGREGATION OF CONJUGATED POLYELECTROLYTES STUDIED BY FLUORESCENCE CORRELATION SPECTROSCOPY .................. 50 Background ................................ ................................ ................................ ............. 50 Results and Discussion ................................ ................................ ........................... 52 Investigation of I nteraction of A nionic PPE d CO 2 with M etal C ations ............... 52 Fluorescence q uenching of PPE d CO 2 i nduced by m etal c ations .............. 52 FCS measurements ................................ ................................ ................... 54 PPi Induced Aggregation of Cationic PPE d NH 3 ................................ ............... 61 Mechanisms ................................ ................................ ................................ ..... 65 Summary ................................ ................................ ................................ ................ 68 Experimental ................................ ................................ ................................ ........... 70 Materials ................................ ................................ ................................ ........... 70 Instrumentation ................................ ................................ ................................ 70 Theory of FCS ................................ ................................ ................................ .. 71 3 DIFFERENT AGGREGATIVE RESPONSES FOR NONQUENCHING MULTICATIONIC AMINES USING CONJUGATED POLYELECTROLYTES ......... 74 Background ................................ ................................ ................................ ............. 74 Results and Discussion ................................ ................................ ........................... 76 Photophysical Properties of P3 ................................ ................................ ........ 76 Fluorescence Quenching by Different Amines ................................ ................. 78 Effect on Amplified Quenching of P3 by MV 2+ ................................ .................. 80 Proposed Mechanism ................................ ................................ ....................... 84 Summary ................................ ................................ ................................ ................ 85 Experimental ................................ ................................ ................................ ........... 87 Materials ................................ ................................ ................................ ........... 87 Instrumentation ................................ ................................ ................................ 87 4 PRINCIPAL COMPONENT ANALYSIS FOR PYROPHOSPHATE SENSORS USING CONJUGATED POLYELECTROLYTES ................................ .................... 89 Backgr ound ................................ ................................ ................................ ............. 89 Principal Component Analysis ................................ ................................ ................ 91 Basic Methods and Procedures for PCA ................................ .......................... 91 Target Transformation ................................ ................................ ...................... 94 Results and Discussion ................................ ................................ ........................... 94 Overvi ew of PPi T urn on S ensor ................................ ................................ ...... 94 Fluorescence q uenching of P 3 by N4 ................................ ......................... 96

PAGE 8

8 Fluorescence recovery of P 3 / N4 by P P i ................................ .................... 97 F luorescence correlation spectroscopy measurement ............................... 98 PCA C alibration R esult for S pectroscopic D ata S et ................................ ......... 99 Loading data for each eigenvectors at different [PPi] ............................... 102 Regression analysis ................................ ................................ ................. 105 Summary ................................ ................................ ................................ .............. 109 Experimental ................................ ................................ ................................ ......... 110 Materials ................................ ................................ ................................ ......... 110 Instrumentation ................................ ................................ ............................... 111 5 ACETATE KINASE ASSAY USING POLY (PHENYLENE ETHYNYLENE) WITH POLYAMINE SIDE CHAINS ................................ ................................ ................. 112 Background ................................ ................................ ................................ ........... 112 Results and Discussion ................................ ................................ ......................... 115 Over view of ACK T urn off A ssay ................................ ................................ .... 115 Fluorescence Q uenching of PPE d NH 3 by P yrophosphate (PPi) ................... 117 ACK T urn off A ssay ................................ ................................ ........................ 119 Determination of ACK catalyzed kinetic parameters ................................ 123 Effect of Mg 2+ on ACK activity in turn off assay ................................ ....... 124 Specificity of ACK t urn off a ssay ................................ .............................. 126 Discussion ................................ ................................ ................................ ...... 1 27 Summary ................................ ................................ ................................ .............. 129 Experimental ................................ ................................ ................................ ......... 129 Materials ................................ ................................ ................................ ......... 129 General M ethods ................................ ................................ ............................ 130 Fluorescence turn off assay procedure ................................ .................... 130 Calculation of initial reaction rate ( 0 ) ................................ ....................... 131 Calculation of kinetic parameters ................................ ............................. 132 6 CONCLUSION ................................ ................................ ................................ ...... 134 Ion Induced Aggregation of Conjugated Polyelectrolytes Studied by Fluorescence Correlation Spectroscopy ................................ ............................ 134 Principal Component Analysis for Pyrophosphate Sensors Using Conjugated Polyelectrolytes ................................ ................................ ................................ 135 Acetate Kinase Assay Using Poly (phenylene ethynylene) with Polyamine Side Chains ................................ ................................ ................................ ............... 136

PAGE 9

9 APPENDIX MATLAB CODING FOR PCA ................................ ................................ ...................... 138 LIST OF REFERENCES ................................ ................................ ............................. 141 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 152

PAGE 10

10 LIST OF TABLES Table page 2 1 Diffusion time ratio of PPE d CO 2 ([PPE d CO 2 ] = 780 nM) with different metal ions at different concentrations. ................................ ................................ .......... 58 2 2 Hydrodynamic radius ( R H /nm) of PPE d CO 2 ([PPE d CO 2 ] = 780 nM) with different metal ions at different concentrations. ................................ .................. 60 3 1 Photophysical properties of P 3 in different solvents. ................................ .......... 77 4 1 Accuracy of average [PPi] obtained with PCA calibration method. ................... 108

PAGE 11

11 LIST OF FIGURES Figure page 1 1 Molecular structures of commonly used: (A) conjugated polymers; (B) conjugated polyelectrolytes. ................................ ................................ ............... 21 1 2 Molecular structures of PPEs with different backbone configurations. ............... 23 1 3 Absorption (left) and fluorescence (right) spectra of PPE SO 3 in MeOH (solid line), H 2 O (dashed line), (1:1) H 2 O/MeOH (dash dot line). ................................ 25 1 4 (A) Structure of PPEs with branched side chains; (B) Absorption and emission spectra of PPE d NH 3 in methanol and water. [PPE d NH 3 ...................... 26 1 5 Absorption and fluorescence spectra of PPE d CO 2 (A, B) and PPE d NH 3 (C, D) ................................ ...................... 28 1 6 (A) Structure of PPE CO 2 CO 2 in dot dash 2+ in methanol ................... 29 1 7 Structures of MPS PPV and DTA. ................................ ................................ ...... 31 1 8 PPV in water: (A) MPS PPV alon e and spectra of MPS PPV in water. ................................ ................................ ............ 31 1 9 Il l ustration of the molecular wire effect ................................ ............................... 33 1 10 Comparison of Stern Volmer response for various polymer/quencher ( MV 2+ ) systems ................................ ................................ ................................ .............. 33 1 11 Working principles of FCS. ................................ ................................ ................. 35 1 12 A typical FCS setup ................................ ................................ ............................ 36 1 13 (A) Structures of P1 and 2 (B) Mechanism of protein induced aggr egation ....... 38 1 14 Normalized correlation functions of 2 P1 / 2 2 /avidin, and P1 / 2 avidin with [avidin]/[ 2 ] = 0.25 in phosphate buffer solution (10 mM, pH 7.4). ....................... 39 1 15 Normalized fluorescence correlation curves for MEDs of different W 0 .............. 40

PAGE 12

12 1 16 Structures of PPE y and spermidine. ................................ ................................ ... 42 1 17 Schematic illustration of spermine induced aggregation of the anionic conjugated polyelectrolyte and the accompanying blue to green fluorescence color ch ange ................................ ................................ ................................ ....... 43 1 18 Structure of P2 ................................ ................................ ................................ ... 44 1 19 Photographs of fluorescence array of P2 presence of proteins (100 nM) ................................ ................................ ............ 45 1 20 Molecular structure of a PFP, c PFB15, and TR. ................................ ............... 46 1 21 Illustration of aggregation mediated fluorescence energy transfer to dye labeled DNA ................................ ................................ ................................ 47 2 1 St ructures of dentritic polymers. ................................ ................................ ......... 51 2 2 Stern Volmer Plot for PPE d CO 2 (780 nM) in water with different metal ions ..... 53 2 3 ( A) Normalized correlation curve for PPE d CO 2 (780 nM) in water with different ions (15 ( B) Count rates for PPE d CO 2 (780 nM) in water with ................................ ................................ ......................... 56 2 4 Normalized correlation curves of PPE d CO 2 (780 nM) in water with different [Fe 2+ ] (5 25 M) ................................ ................................ ................................ 57 2 5 Diffusion time ratio for PPE d CO 2 (780 nM) in water with different metal ions measured by FCS ................................ ................................ ............................... 59 2 6 AFM images of PPE d CO 2 ( 1 M) in water with different [ Fe 3 + ] ( 0 30 M). (A) PPE is well dispersed when no Fe 3 + is added. (B) Small aggregates formed when [ Fe 3 + ] = 5 M (C) Large aggregated formed when [ Fe 3 + ] = 30 M .......... 61 2 7 Fluorescence spectra of PPE d NH 3 solution (1 M) titrated with PPi in MES buffer (10 mM, pH 6.5). ................................ ................................ ...................... 62 2 8 (A) Normalized correlation curve for P PE d NH 3 (1 M) with different [PPi] in MES buffer (10 mM, pH 6.5). (B) Count Rates for PPE d NH 3 (1 M) with different [PPi] in MES buffer (10 mM, pH 6.5). ................................ .................... 64 2 9 Mechanism for PPE quenched by metal ions ................................ .................... 67

PAGE 13

13 2 10 Setup of FCS system in our lab. ................................ ................................ ......... 71 3 1 Chemical structures of P3 ethylenediamine dihydrochloride ( C2N2 ) cadaverine dihydrochloride ( C5N2 ) and tris(2 aminoethyl)amine ( N4 ) .............. 76 3 2 Normalized absorption and fluorescence spectra of P 3 in MeOH (solid line) and H 2 O, pH = 6.5 (dashed line) ................................ ................................ ......... 77 3 3 ( A) (C): Fluorescence spectroscopic changes of 2.0 M P3 solution observed upon titration of different amines in water : (A) N4 (B) C2N2 (C) C5N2 ; (D): S tern Volmer plots of I 0 / I as a function of amine concentration in water ............. 79 3 4 (A): Stern Volmer plots of I 0 / I as a function of MV 2+ concentration for d ifferent P3 /amine systems in water (B): Stern Volmer plots of I 0 / I as a function of MV 2+ concentration for P3 / C2N2 mixtures in w at different concentrations. ....... 82 3 5 Normalized correlation curves for different systems in water ............................. 83 3 6 Schematic representation of analyte induced aggregation between polymer and different amines. ................................ ................................ .......................... 85 4 1 Structures of polymers and substrates ................................ .............................. 91 4 2 ................................ ................................ .. 95 4 3 Fluorescence spectra changes upon titration of polyamine (0 10 P 3 in HEPES buffer solution ( 10 mM, pH = 7. 4 ). ................................ .......... 96 4 4 Fluorescence spectra changes upon titration of PPi and Pi (0 3 mM) into P 3 and 4 N4 in HEPES buffer solution ( 10 mM, pH = 7. 4 ) ; (A). PPi; (B). Pi. ................................ ................................ ................................ .. 98 4 5 Normalized correlation curves and diffusion time for different systems. Black: [ P 3 ] = 2 P 3 ] = 2 N4 ] = [ P 3 ] = 2 N4 ] = ................................ ................................ ................................ ... 99 4 6 ( A ) Original emission spectra of changes upon titration of PPi (0 3 mM) into P 3 and 4 N4 in HEPES buffer solution ( 10 mM, pH = 7. 4 ), ( B ) Fundamental spectra (matrices, R ) for two largest eigenvalues. ................ 100 4 7 Fundamental spectra for abstract factors ................................ ........................ 102

PAGE 14

14 4 8 ( A ) Contribution of the first fundamental spectrum ( R 1 ) to total emission at various [PPi]. (B ) Contribution of the second fundamental spectrum ( R 2 ) to total emission at various [PPi]. ( C ) Reconstructed spectra D = R 1 C 1 + R 2 C 2 .... 104 4 9 The contribution of two eigenvectors at different [PPi] ................................ ..... 105 4 10 [PPi] calibration curve obtained according to E quation 4 1 6 Standard calibration points from raw data are respented as black squares. Unknown points are resprented as blue triangles ................................ ............................ 108 5 1 ( A ). Structure of PPE d NH 3 and reaction scheme for acetate kinase. ( B ). Mechanism of ACK turn off assay. ................................ ................................ ... 115 5 2 Stern Volmer plot for PPE d NH 3 upon titration with Pi, acetate, PPi and AP, respectively. Solution conditions: 2 M PPE d NH 3 and 1 mM MgCl 2 in 25 mM MES (pH = 6.5) buffer, ex = 404 nm. ................................ ............................... 118 5 3 Fluorescence spectroscopic change upon the addition of ACK enzyme over a concentration range from 0 to 20 mL. Solution conditions: 2 M PPE d NH 3 and 1 mM MgCl 2 in 25 mM MES (pH = 6.5) buffer, ex = 404 nm. .................... 119 5 4 ( A ) Fluorescence spectroscopic changes as a function of time after addition ACK ( B ) Change of fluorescence intensity at 434 nm recorded every 30 s during the real time ACK turn off assay with various ACK concentrations ....... 120 5 5 Time based fluorescence intensity recorded every 30 s during the real time measurement of ACK turn off assay in a 1 0 min period at various [ACK] with different substrate concentrations ................................ ................... 122 5 6 Dependence of initial rate of reaction ( 0 ) on ACK concentrations. .................. 122 5 7 Enzyme kinetics parameters measurement by using AC K turn off assay. ....... 124 5 8 Effect of confactor MgCl 2 on the enzymatic activit y of ACK [Polymer] = 2 Red curve : [MgCl 2 ] = 1 mM No ACK added. B lack curve : [ACK] = 5 g/mL no MgCl 2 added. Blue curve : [ACK] = 5 g/mL and [MgCl 2 ] = 1 mM .................... 125 5 9 Specificity of ACK turn off assay. ( A ) Changes in fluorescence intensity at 434 nm after 60 min of incubation with six different proteins at 30 o C. ( B ) Time based fluorescence measurements for these proteins. ........................... 127

PAGE 15

15 LIST OF ABBREVIATIONS 3D Three dimension ACK Acetate kinase A DP Adenosine diphosphate A LP Alkaline phosphate A MP Adenosine monophosphate A P Acetyl phosphate A PD Avalanche photodiode A TP Adenosine triphosphate BSA Bovine serum albumin BT Benzothiadiazole C2N2 Ethylenediamine C5N2 Cavaderine CO 2 Carboxylate CPE Conjugated polyelectrolyte D Diffusion coefficient D A Donor acceptor DN A Deoxyribonucleic acid DTA D odecyltrimethyl ammonium bromide FCS Fluorescence correlation spectroscopy FET Field effect transistor FRET Fluorescence resonance energy transfer G ( ) Autocorrelation function

PAGE 16

16 GOX Glucose oxidase from Aspergillus niger HEPES 4 (2 Hydroxyethyl) 1 piperazineethanesulfonic acid H O K Hexokinase from saccharomyces cerevisiae, type III HTS High throughput screening I Fluorescence intensity LED Light emitting device MED Microemulsion droplet MES 2 (N morpholino) ethanesulfonic acid MIP M olecularly imprinted polymer MPS PPV Sulfonated poly(phenylene vinylene) MV 2+ Methyl viologen MW Molecular weight N4 Tris(2 aminoethyl)amine NADH Nicotine adenine dinucleotide, reduced NR 3 + Quarternary ammonium P1 Meta linked sulfonated poly(phenylene ethynylene) P2 Sulfonated poly(phenylene co bisthienylbenzothiadiazole ) P3 Poly(phenylene ethynylene) with bis carboxylate methylene side groups Pi Phosphate PCA Principal component analysis PFB Poly( fluor e ne co benzothiadiazole ) with alkyl ammonium side groups PFP Sulfonated poly( fluo e ine co phenylene )

PAGE 17

17 PLD Phospholipase D from peanut t ype II PO 3 2 Phosphonate POD P eroxidase from horseradish type I PPi Pyrophosphate PPE P oly(phenylene ethynylene) PPE CO 2 Carboxylated poly(phenylene ethynylene) PPE d CO 2 Anionic poly(phenylene ethynylene) with dendric carboxylate side chains PPE d NH 3 Cationic poly(phenylene ethynylene) with dendric ammonium side chains PPE SO 3 Sulfonated poly(phenylene ethynylene) PP P Poly( phenylene phenylene ) PPV P oly(phenylene vinylene) PT P olythiophene PTD Peptidase from porcine intestinal mucosa R H Hydrodynamic radius RN A Ribonucleic acid SS DN A Single strand DNA SV Stern Volmer TBT Bisthienylbenzothiadiazole THF Tetrahydrofuran TNT Trinitrotoluene TMR Tetramethylrhodamine TR Texas red

PAGE 18

18 V ef f Effective detection volume F Fluorescence quantum yield ma x Wavelength of maximum emission peak Viscosity of the solvent Diffusion time k Boltzmann s constant Structure parameter r Transversal or waist radius of confocal volume z Longitudinal radius of confocal volume

PAGE 19

19 Abstract of Dissertation Presented to the Graduate School of University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CONJUGATED POLYELECTROLYTES AGGREGATION AND SENSOR APPLICATIONS By Jie Yang May 201 3 Chair: Kirk S. Schanze Major: Chemistry Over the past several years, significant effort has been devoted to explore the application of conjugated polyelectrolytes (CPE s ) as chemical and biosensors for the detection and analysis of a variety of molecules of environmental and biological interests. Well established sensing mechanisms have been reported to provide guidelines for underlying concepts of CPE based sensors. In this d issertation, we focus on the design and development of fluorescent biosensors based on functionalized poly(phenylene ethynylene)s (PPEs) by taking advantage of th e aggregation induced quenching mechanism. The aggregation behavior of an anionic poly(phenyle ne ethynylene) (PPE d CO 2 ) upon the addition of different metal ions was first investigated by fluorescence correlation

PAGE 20

20 spectroscopy ( FCS ) FCS results show the diffusion time increases in the order K + Na + < Ca 2+ < Cu 2+ < Fe 2+ < Fe 3+ and large diffusion time usually indicates the formation of large aggregates. Comparison of the diffusion time ratio plot to Stern Volmer plot shows that most efficient Stern Volmer quenching is seen for ions that give rise to a large increase in FCS diffusion time, and it underscores the importance of aggregation o n the amplified quenching effect. Then, a fluorescen t carboxylat e substituted CPE was developed off uorescence change is based on the aggregate conformational change of PPE in the process of adding a tetraamine and PPi respectively Then principle c omponent analysis (PCA) method wa s utilized to convert a large set of complicate d fluorescence spect ra into two linearly uncorrelated eigenvectors R egression analysis wa s performed to predict the concentration of unknown PPi with ~ 95% accuracy. Last, a cationic polyamine substituted PPE (PPE d NH 3 ) wa s used to develop a real time fl uorescent assay for acetate kinase (ACK) based on the fluorescence quenching of PPE d NH 3 by PPi The introduction of ACK is able to catalyze the phosphate transfer from acetyl Pi to Pi and produce PPi, leading to significant spectroscopic change s. T herefore it allows a direct detection of PPi and measurement of ACK activity

PAGE 21

21 CHAPTER 1 INT RODUCTION Conjugated Polyelectrolytes c onjugated polymers such as poly(phenylene ethynylene) (PPE), poly(phenylene vinylene) (PPV), polythiophene (PT) and poly(para phenylene) (PPP) (structures shown in Figure 1 1 A ) have attracted considerable attention for their important roles in a variety of applications, including solar cells, 1 light emi tting diodes (LEDs), 2 3 field effect transistors (FETs), 4 5 and chemo and biosensors. 6 7 The delocalized electronic structures, variations in extent of conjugation and rapid excition transport along the backbone enable conjugated polymers to show strong optical absorption and emission, reversible electrochemical switching and increa sed sensitivity to target molecules 7 9 Figure 1 1 Molecular structures of commonly used: ( A ) conjugated polymers ; ( B ) conjugated polyelectrolytes

PAGE 22

22 Conjugated polyelectrolytes (CPEs) are water soluble conjugated polymers having ionic functional side groups. Commonly used charged side groups include sulfonate (SO 3 ), carboxylate (CO 2 ), phosphonate (PO 3 2 ) and quarternary ammonium (NR 3 + ) and some examp les are presented in Figure 1 1 B 10 CPEs combine the intrinsic electronic and optical properties which arise from the organic conjugated ba ckbone 11 and good solubility in water or polar organic solvents 12 13 The unique structures provide CPEs with additional advantages. For example CPEs are able to bind strongly with other i onic species via electrostatic interaction s with their charged side groups. Many CPE based sensors have been established and investigated by taking advantage of this property to detect target ions of environmental and biological interest such as toxic met al ions, proteins and oligo and polynucleic acids. 7 14 T he ionic and amphi phi lic nature of CPEs also mak es them capable of self assembl y into films, colloids or other supermolecular structures. 15 Poly(phenylene ethynylene) s (PPEs) comprise one of the most important classes of CPEs where the backbones are composed of conjugated ethynyl ene linked with phenylene rings. Traditionally, PPEs have been used as electroluminescent devices since they are electron system is either oxidized or reduced. 16 17 More recently PPEs have been of increasing interest in areas like chemo and biosensors. M any sensory systems, including TNT sensors, 12 13 metal ion sensors, 18 glucose sensors 14 and avidin sensors, 19

PAGE 23

23 have been built based on PPEs. Based on the main chain conformation, PPEs can exist as three isomers: para meta and ortho poly(phenylene ethynylene)s (structures shown in Figure 1 2). Compared to other CPEs, PPEs have sev eral superior properties. First because of their versatile chemistry small modifications to the structure can lead to very different electronic and photophysical properties. Second, they can adopt various conformations in different solvents. For example, in polar solvents meta link ed PPEs can self assemble into a helical conformation which is stabilized by stacking interactions between the phenyl rings. 20 21 Third, they usually feature high photostability. All of these properties enable them to be good candidates for use in sensors Figure 1 2 Molecular structures of PPEs with different backbone configurations Aggregation of Conjugated Polyelectrolytes In general, CPEs exhibit photophysical properties similar to these conjugated polymers having the same backbones because the optical properties are primarily conjugated backbone. However, because of the inherent amphi p hilic structures of the CPEs (hydrophobic backbones and hydrophilic side chains) they tend to self assemble into sup ra molecular aggregates in aqueous solutions or polar

PAGE 24

24 organic solvents, and the aggregation proc ess induce s significant change s i n both absorption and fluorescence spectr a 22 Factors that Influence the Aggregation of Conjugated Polyelectrolytes Solvent dependence The existence of CPEs i n different conformational and aggregation states in different solvents was explored in 2002 by Tan and Schanze 22 23 who investigated the absorption and fluorescence p roperties of an anionic CPE, PPE SO 3 ( Figure 1 1 B) in water, methanol and a 1:1 water : methanol mixture. As shown in Figure 1 3 the absorption and fluorescence spectra both undergo red shift s with an increase in the water:methanol ratio with the effect most pronounced in the fluorescence spectrum. In methanol, the fluorescence appears as a narrow band with a sharp peak at 450 nm corresponding to a small S toke s shift relative to the absorption maximum at 430 nm. U pon the introduction of water, the fluorescence decrease s significant in intensity at 450 nm and a weaker and broad er band appears at longer wavelength ~ 550 nm. Schanze et al suggested that in a n organic solvent like methanol, PPE SO 3 is with photophysical properties similar to those of neutral structurally related conjugated polymers dissolved in good solvents like CHCl 3 or THF. 24 As a result the emission is dominat ed by excitions that are restricted to single chains. In comparison, in a queous solution, it is believed the PPE SO 3 chains form supramolecular aggregates through

PAGE 25

25 adjacent polymer chains. 25 27 The dec r eased fluorescence intensity and the shift to longer emission wavelength are due to emission from i nter chain excimer like states which have lower energ ies and longer radiative lifetime s than the intra chain exciton state. It is also important to note that the quantum yield of CPEs decrease s with increas ing water concentration providing additional evi dence of formation of low emissive interchain excited states that arise polymer chains. 28 Figure 1 3. Absorption (left) and fluorescence (right) spectra of PPE SO 3 in MeOH (solid line), H 2 O (dashed line), (1:1) H 2 O/MeOH (dash dot line). Fluorescence spectra are area normalized to reflect relative quantum yields Reprinted with permission from Tan et al 29 Branched side groups and pH effects Considerable effort ha s been devoted to decreas ing the degree of aggregation in aqueous solution by incorporati ng with different functional groups. Zhao and Lee 30 31 introduced a series of conjugated polyelectrolytes modified with two branched polyionic

PAGE 26

26 side chains (Figure 1 4 A ). T hese bulky and highly charged functional groups are capable of increasing the electrostatic repulsion between adjacent polymer chains effectively decreasing the hydrophobic stacking interaction T he presence of these large ionic groups also significantly enhances the solubility of CPE s in aqueous solution. Figure 1 4 (A ) Structure of PPEs with branched side chains ; (B ) Absorption and emission spectra of PPE d NH 3 in methanol and water. [PPE d NH 3 ] = Reprinted with permission from Lee et al 30 Figure 1 4 B shows the absorption and emission spectra for one of these CPEs (PPE d NH 3 ) in both methanol and water. As described above, PPE SO 3 exhibits structured emission in methanol wit h a small Stokes shift while in aqueous solution it appears as a broad and red shift ed band 22 In contrast, PPE NH 3 shows negligible change in its absorpti on spectrum with a subtle red shift of about 3 nm in water. In water, the emission spectrum is similar to that in methanol with em at the same peak wavelength (~ 440 nm ), al though the fluorescence intensity is slightly reduced in water

PAGE 27

27 This comparison ind icates that PPE d NH 3 is much less aggregated in water and compared to PPE SO 3 most likely du e to the high ionic charge densit y and the large size of the branched side groups which are assumed to interrupt the coplanarity of the adjacent conjugated backbones. Th e quantum yield is moderately retained in water to The strong electrostatic repulsion among the multiple NH 3 + groups effectively r educe s CPE aggregation in w ater T herefore it is believed the modification to the pH conditions can significantly change the optical properties. Lee studied the pH dependence of their branched polymers in the range of 4.5 to 10.5 and the results are presented in Figure 1 5 The PPE d CO 2 absorption spectrum exhibits a strong peak with a max of 404 nm in the pH range of 8.5 to 10.5. As pH decreases, the absorption red shifts and a shoulder at 435 nm increases. The change in its flu orescence is more significant. At high pH, there is a strong peak for structured emission at 444 nm. As the pH is lowered, another wide band appears at ~ 515 nm, indicating tha t PPE d CO 2 ha s a tendency to aggregate at lower pH where the anionic carboxylate groups are gradually protonated T he charge neutralization process lead s to reduction of the electrostatic repulsion and the more hydrophobic environment induce s aggregation of the polymer chains through hydrophobic and stacking interactions. The conformational change of the PPE backbones finally contributes to the spectra l change Similar to PPE d CO 2 the optical properties of c ationic PPE d NH 3 are also pH dependent At higher pH, t he

PAGE 28

28 weakly basic groups are deprotonated resulting in a decrease of the high energy fluorescence, a low quantum yield emission and a red shif t in both the absorption and emission spectra Figure 1 5 Absorption and fluorescence spectra of PPE d CO 2 ( A B ) and PPE d NH 3 ( C D ) as a functional of pH in water. [PPE] = is 1.0 unit. Reprinted with permission from Lee et al 30 Addi tion of oppositely charged ions T he addition of opposite ly charged ions can also induce the inter chain aggregation of CPEs. 22 32 In 2006, Jiang and Schanze found that the spectral changes observed when calcium (II) ions a re added to a nionic PPE CO 2 (Figure 1 6 A ) in methanol, 33 are similar to those seen when PPE SO 3 is aggregated in water. A broad band appear s at longer wavelength with low fluorescence intensity. In a good solvent like methanol,

PAGE 29

29 PP E CO 2 show s structur ed emission with a strong and narrow band. However, the addition of Ca 2+ results in a pronounced decrease in fluorescence intensity, emission red shift and band broadening (Figure 1 6 B ) Since Ca 2+ has a closed shell structure and is therefore unable to act as an electron or energy acceptor, so the observed spectra l changes are probably due to the formation of CPE aggregates. Since PPE CO 2 has two negatively charge d side groups on each repeat unit, the positive ly charged Ca 2+ is assume d to induce the aggregation of PPE CO 2 in methanol by acting as an ionic bridge between the polymer chains through electrostatic interaction s and complexation with the carboxyl functional groups. The aggregation behavior of other CPEs systems have a lso bee n investigated by adding different additives ligands such as oxalic acid and diamines. 34 35 Figure 1 6. ( A ) Structure of PPE CO 2 ; ( B CO 2 in dotted line dot dot dash line dot dash line dashed line solid line ) Ca 2+ in methanol Reprinted with permission from Jiang et al 33

PAGE 30

30 Interaction of s urfactants with CPEs Due to the amphiphilic nature of CPEs, it is suggested they can form stable complexes with oppositely charged molecules, such as surfactants. 36 37 In 2000, Whitten and co workers first reported that the geometric conform ation of CPEs can be adjusted by adding appropriate su rfactants, lead ing to dr amatic change s in their photophysical properties. 38 In their work, Whitten et al synthesized an anionic poly(phenylene vinylene) CPE (MPS PPV Figure 1 7 ) whose optical proper ties are strongly dependent on solvent Upon adding the cationic surfactants dodecyltrimethyl ammonium bromide (DTA, Figure 1 7 ) to the aqueous solution of MPS PPV, the emission spectrum of the CPE/surfactant complex changed dramat ically. As shown in Figure 1 8 the fluorescence spectrum after addition of DTA has well defined vibrational structure with a noticeable blue shift and dramatic intensity enhancement; while the absorption spectrum is red shifted and narrowed. MPS PPV tends to aggregate in po or solvents like water and its fluorescence spectrum shows a broad emission However, the enhancement of the fluorescence intensity and the shape of the emission spectrum for the CPE/surfactant complex are very similar to the properties of the structur ed emission exhibit ed by single po lymer chains in good solvents. Whitten et al proposed t hat CPE and D TA form a polymer surfactant complex due to a Columbic interaction between the anionic polymer side chains and the cationic surfactant headgroups as well a s hydrophobic interacti ons between the surfactant tail and the conjugated polymer backbone successfully

PAGE 31

31 disrupting polymer polymer interaction (aggregation). Additionally, they suggested that this is an entropically favored process due to the release of interfacial water mol e cules (hydrophobic effect). Both factors reduce polymer interchains interactions and reduce the conformational disorder. Therefore, the polymer backbone becomes more extended and ordered and the fluorescence quantum efficiency is inc reased. Figure 1 7 Structures of MPS PPV and DTA. Figure 1 8 PPV in water: (A) MPS PPV alone and spectra of MPS PPV in water. Reprinted with permission from Chen et al 38

PAGE 32

32 Effect of A g g regation on Am plified Q uenching Amplified quenching i n conjugated polymers was first reported by Swager and co workers 39 who showed that th e quenching response of a fluorescent PPE functionalized on each repeat unit with a cyclophane receptor to methyl viologen (MV 2+ structure in Figure 1 10 ) is enhanced 50 to 100 fold compared to that of a small fluorophore. As illustrated in Figure 1 9 t hey attributed the amplified quenching of PPE to a the polymer side chains to complex with the quenchers leading to extended exciton transport along the conjugated polymer chains. Several studies found that multiple factors including molecular weight, 40 polymer aggregation, 22 4 1 chain length 42 and quenche r binding affinity 43 44 c an influence the amplified quenching effect. The effect of aggregation on amplified quenching has been widely discussed. T he comparison shown in Figure 1 10 indicates that the quenching efficiency has increased from ~ 10 3 M 1 to ~10 6 M 1 using a negatively charged conjugated polymer compared to its oligomer with the same MV 2+ quencher. This behavior was used to design a CPE based sensor with sig nificant quenching amplification and a detection limit for MV 2+ is in the submicromolar concentration range. Most importantly, even higher Stern Volmer constants in excess of 10 7 M 1 have been reported for the aggregated CPE systems, further improv ing the sensing response to the low nanomolar concentration range. It is believed th at formation of aggregates provides

PAGE 33

33 the opportunity for excitions to diffuse in a three dimension al pathway leading to a significant increase in their chance to encounter the quenchers. Figure 1 9 Il l ustration of the molecular wire effect Reprinted with permission from Zhou et al 39 Figure 1 1 0 Comparison of Stern Volmer response for various polymer/quencher ( MV 2+ ) systems. The numbers in boxes are the Stern Volmer quenching constants and the concentration units to the right give quencher ion d etection limit. ( A ) Uncharged small molecule chromophore s ; ( B ) anionic small molecular ion chromophore s ; ( C ) unaggregated anionic CPE s ; ( D ) aggregated anionic CPE s Fluoresce nce Correlation Spectroscopy (F C S) Fluorescence correlation spectroscopy (FCS) is a spectroscopic technique based on the statistical analysis of spontaneous fluorescence fluctuations. It was first

PAGE 34

34 introduced by Madge, Elson and Webb in 1972 45 who use d the technique to study the kinetics of DNA drug intercalation. This method is a very sensitive tool for high resolution spatial and temporal analysis of fluorophores in the nanomolar concentration range as well as for single molecule detection. 46 In addition, FCS pr ovides parameters about dynamic events such as diffusion constants hydrodynamic radii, chemical reaction rate s and conformational change s, and it is an ideal approach to investigate the thermodynamic s and kinetic s of molecular in teractions. 47 49 Principles of FCS FCS analysis calculates a correlation function from the time dependent intensity fluctuations of fluorescent particles observed by confocal microscopy. As shown in Figure 1 11 FCS records the emission fluctuations from fluorescent particles moving in and out of a femtoliter confocal volume formed by a focused laser beam. These fluctuations are due to Brownian motion which can be affected by a number of factors including enzymatic activity, protein folding, phase fluctuations and conformational dynamics. 50 51 T he number of fluorescent particles in the observation volume is measured as a function of time. After an auto correlation function G( ) is applied the data are transformed in to a correlation curve. Two major results can be obtained from this correlation curve: the temporal autocorrelation defines the time scale of the diffu sion time ; and the variance of the intensity provides the average number of fluorescent particles in the detection volume. As illustrated in Figure 1 11 large molecule s usually

PAGE 35

35 feature a longer diffusion time as they diffuse slowly, and the correlation curve shift s to the right when the particle size increases, a nd the value of G ( ) is in versely proportional to the number of particles in the volume. In addition changes in particle fluorescence intensity can be recorded during observation. Hence, FCS can be used to d etermine sample concentrations dynamic information diffusion co efficient s and rate const ant s related to rotation and translation all important parameters in biochemical rese arch, biophysics and chemistry. 52 Figure 1 1 1. Working principles of FCS Typical FCS S etup A block diagram of a confocal microscope as the foundation for an FCS system is schematically depicted in Figure 1 12 A laser beam at the desired wavelength passes

PAGE 36

36 through a microscope objective via a dichroic mirror after being optimized with several optical elements Then the laser is focus ed on a small volume and excite s the fluorescen t particles in the sample The emitted fluorescence is collected and it pass es through the same dichroic mirror and an emissi on filter. There is a pinhole right below the emission filter to block any background fluorescen t light that is not from the sample volume. By properly adjusting the pinhole, the resolution and signal in the desired plane is greatly enhanced and optimized. The pinhole is f ollowed by a detector preferably an avalanche photodiode (APD) with single photon sensitivity, to convert all the photon signals to electric signal s and to amplif y the signals, which are passed to the correlator and software for calculat ion of the auto correlation function and display of the decay curve. Figure 1 1 2 A t ypical FCS setup. L, laser; M1, M2, mirrors; SF, spatial filter; SMF, single mode filter; FP, fiberport; DM, dichroic mirror; MO, microscope objective lens; S, sample; EF, emission filter; PH, pinhole; MMF, multi mode fiber; APD, Avalanche Photo Diode ; C, correlator; PC, personal computer. Black dash line represents the outline of fluorescence microscope.

PAGE 37

37 Applications of FCS Recently, FCS has been employed in various areas to study molecular concentrations and diffusion times, 49 size s and conformational change s 53 and molecular interactions. 54 Most importantly, FCS is also capable o f obtain ing kinetic parameters for bimolecular interactions if the y cause fluorescence fluctuations on a short er time scale than the diffusion time 54 55 These i nclude unimolecular reactions, intersystem crossing, triplet state dynamics, and reversible photobleaching. The following section describes the use of FCS to address problems with size change s of the fluorescent p articles Aggregation b ehavior In a recent paper Schanze and co workers applied FCS to explain the intercalative binding between a meta linked CPE ( P1 ) that adopt s a helical conformation in water and a biotin tetramethylrhodamine (TMR) conjugate ( 2 Figure 1 13A ). 19 The biotin can bind to P1 via intercalation between the stacked phenylene units in the helical assembly. A ddition of avidin, which is a protein well known to bind strongly to biotin, led to supramolecular polymer aggregates for mation by non covalent cross linking between the biotin unit of intercalated 2 and avidin (See Figure 1 13) Since the fluorescence change wa s not significant when avidin wa s added, FCS was used to detect the siz e change, leading to a detection limit for avidin of less than 100 pM

PAGE 38

38 Figure 1 1 3 (A) Structures of P1 and 2 ( B) Mechanism of protein induced aggregation. Reprinted with permission from Wu et al 19 The FCS results are displayed in Figure 1 14 The small dye molecule 2 has the lowest diffusion time of ~ 0.2 9 ms After the introduction of P1 the P1 / 2 complex exhibit ed a longe r diffusion time ~0.79 ms due to the intercalative binding of 2 with P1 The most significant increase of the diffusion time was observed when avid i n wa s added to the P1 / 2 mixture at a stoichiometric concentration, and the correlation curve reveal ed the highest diffusion time of 11.50 ms. The dramatic shift of the correlation c urve suggest s The avi din/ 2 complex (c ontrol experiment ) has a diffusion time of 1.21 ms slightly larger than that of the P1 / 2 complex, but considerably smaller than the time for the supramolecular P1 / 2 avidin assembly.

PAGE 39

39 Figure 1 1 4 Normalized correlation functions of 2 P1 / 2 2 /avidin, and P1 / 2 avidin with [avidin]/[ 2 ] = 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 et al 19 Size m easurement FCS is also an ideal tool to measure diffusion and size distribution for single molecul e detection. In 2011, Pal and co workers demonstrated the use of FCS to measure the size, size distribution and polydispersity of a supramolecular nanostructure (microemulsion droplets, MEDs). 56 A group of MEDs with 13 different core sizes ( W 0 ) rang ing from 2 to 50 nm were prepared with sulforhodamine B as a fluorescent marker. A maximum entropy based FCS data fitting model was used to analyze the correlation curves for these MEDs in solution. As shown in Figure 1 15 a s the size of the droplet increase d the correlation curve shift ed to a longer diffusion time. It is clea r that FCS is an ultra sensitive anal ytical tool that it can detect very smaller size difference s From the

PAGE 40

40 relationship between the diffusion time and the hydrodynamic radius of their nan oparticles the authors could predict the size of an unknown nanopart icle. Figure 1 1 5 Normalized fluorescence correlation curves for MEDs of different W 0 As W 0 increases, the curves shift to the longer lag times, indicating the increase in size of the droplets. The correlation curve for SRhB (control) in water (dotted line) is also included for comparison. Reprinted with permission from Pal et al 56 Sensor Applications During the past few decades, CPE systems have been broadly investigated as o ptical chemical and bio sensors and researchers have design ed different CPEs systems for application to various analytes, including small ions, biomolecules, proteins and nucleic acids. In contra st to conventional sensing methods, fluorescence based CPE systems provide several advantages. First and most essentially, CPEs are water soluble and are more adaptable and compatible to biological systems. Second, these CPE based sensors usually utilize f luorescence detection which is rapid and easy, and it affords great sensitivity and real time measurements. Third, CPE based sensor

PAGE 41

41 systems can be easily adapted to high throughput screening (HTS) formats. As a consequence, many sensory systems based on C PEs have been studied and reported. In general, most CPE based sensors utilize one of the following mechanisms: amplified quenching, fluorescence resonance energy transfer (FRET) and conformational change (including aggregation induce d quenching). I n mos t cases, sensors utilize more than one mechanism in a given assay since these three mechanisms are not independent. In the following, we will introduce some examples primarily based on the aggregation induced quenching mechanism. Small I on S ensing Many f luorescence based sensing systems have been developed for s mall species such as heavy metal ions and small organic ions, which can be toxic hazards or threats. In 2007, Swager and co workers demonstrated the detection of nonquenching, multicationic small molecules including spermine, spermidine and neomycin by using a blue emitting polyanionic PPE doped with green emitting exci ton traps sites (anthryl units, Figure 1 16 ). 57 In their investigation, they found that these multicationic amines were able to effectively induce the formation of strongly associated aggregates between the polymer chains in solution. As illustrated in Figure 1 17 anionic PPE y was dissolved as isolated single chains because of the presence of bulky side groups. Therefore most of the fluorescence observed from the polymer arise s from the inhere nt blue emission band at around 430 nm. Spermidine, which predominantly has four positive charges

PAGE 42

42 under acidic pH conditions, was then introduced to the polymer solution. The small and multicationic spermidine effectively attract s and complex es with the po lyanionic PPE to form large aggregates, bringing the polymer chains into close proximity This leads to expansion of excitions migration to a three dimensional space greatly enhancing exciton transport. As long as these excitions were trapped by the lower energy anthryl sites, a visually noticeable blue to green fluorescence emission wa s observed As little as ~ 1.6 spermidine was sufficient to produce a fluorescence color change from blue to green Among all the ana lytes they have tested, only th ose w ith more than two positive charges were capable of ind uc ing efficient inter chain aggregation of PPE thus giving this system selectiv ity towards different small molecule amines. This color change sensor method is expected to be useful for detect ing and mo nitor ing of biological ly relevant and nonquenching multicationic species. Figure 1 1 6 Structures of PPE y and spermidin e

PAGE 43

43 Figure 1 1 7 Schematic illustration of spermine induced aggregation of the anionic conjugated polyelectrolyte and the accompanying blue to green fluorescence color change. Reprinted with permission from Swager et al 57 Protein S ensing Proteins play an essential role in facilitating physiological an d biochemical functions, cataly sis of tens of thousands of reactions as enzymes, cell signaling, molecular recognition, cellular communication, and gene expression. 58 59 Many methods and strategies have been developed for protein detection and analysis for medical diagnostics and clinical research. In 2011, Lee et al synthesized an anionic conjugated polyelectrolyte which can exhibit blue to green or blue to red change in fluorescence upon exposure to different charged proteins. 60 The conjugated polyelectrolyte ( P2 Figure 1 18 ) wa s functionalized with sulfonate groups and a certain percentage of bisthienylbenzothiadiazole groups (TBT) which serve d as energy acceptors T he aggregation induced intermolecular excition can migrate from the energy donor ( phenylene groups ) to these TBT units to exhibit different emissions. T he authors

PAGE 44

44 conducted a screening assay array for different biologically relevant proteins and the results are presented in Figure 1 19 The proteins examined in their project and their isoelectric points were lysozyme ( pI = 11.0), Bovine serum albumin ( BSA pI = 4.9), Lecin ( pI = 5.47) and cytochrome C ( pI = 10.7). The sensing mechanism wa s mainly attributed to electrostatic and hydrophobic interactions between the polymers and proteins. As a result of these factors, different degree s of aggregation occu rred, lead ing to development of different emission color change s Over the entire pH range, proteins may carry one positive/negative charge or remain neutral in aqueous solution depend ing on their pI value. Therefore, the pI for each specific protein is ve ry important in this sensing strategy. This rapid and visually noticeable (blue to red emission) CPE based method will be useful in sensory arrays for detecting various cationic proteins under pH control. Figure 1 1 8 Structure of P2

PAGE 45

45 Figure 1 19 Photographs of fluorescence array of P2 s in the presence of proteins (100 nM). From top to bottom: polymer; Lysozyme (pI = 11.0); BSA ( pI = 4.9); Lecin ( pI = 5.47) and Cytochrome C ( pI = 10.7). Reprinted with permission from Lee et al 60 DNA S ensing DNA has be en recognized as a key target for cancer diagnosis and detection of bacteria and viruses 61 62 and significant developments have been achieved to make DNA sensing more accessible. 63 64 In 2011, Woo and co workers reported a FRET based DNA biosensor which can be mediated by the aggregation of a complex between an anionic CPE (a PFP, Figure 1 2 0 ) and a cationic CPE (c PFB15, Figure 1 2 0 ). 65 In their approach (see Figure 1 2 1) electrostatic complexation induced by the oppositely charged polyelectrolytes a PFP and c PFB15 led to the aggregation of

PAGE 46

46 polymer chains with a concomitant decrease in intersegment distances. Thus the excited energy state was transfer red to the BT (benzothiadiazole, 15% in c PFB15) segments and a green emission was observed. W hen a single stranded DNA labeled with Texas Red (ssDNA TR, Figure 1 2 0 ) was introduced t he negatively charged DNA bound to the polymer complex to give rise to further energy transfer fr om the BT segment to TR and the red emission w as used to signal the presence of target DNA. O verall, this method involves a two step FRET process : t he energy of the excited fluorine phenylene on a PFP (initially blue fluorescence ) is delivered to the BT segment on c PF B15 due to the formation of the CPE complex leading to green fluorescence. Then the green emission disappear s with a red emission appearing via a subsequent FRET after complexation with ssDNA TR. The optical signal is also significantly amplified, resulting in increased sensitivity in CP E based F RET DNA detection assays. It should be pointed out that several parameters are importan t in this sensory system, including t he D A distance, spectra overlap, DNA:CPE concentration ratio, and dipole orientation. Figure 1 2 0 Molecular structure of a PFP, c PFB15, and TR

PAGE 47

47 Figure 1 2 1 Illustration of aggregation mediated fluorescence energy transfer to dye labeled DNA. Reprinted with permission from Jin et al 65 C hallenges of CPE based O p tical S ensors Most CPE based sensory systems provide several advanta ges including superior sensitivity, a high degree of selectivity and simplicity. While highly useful, there are some shortcomings associated with the methods. First photobleaching which is inherent to most CPEs can lead to false signals and cause erro rs in analytical studies where quantitative analysis is required. 7 66 Second the selectivity of the sensors to the specific target molecules remains a problem, as most of the sens ing mechanisms involve relatively weak electrostatic and/or hydrophobic interactions Thus, i nterference from unwanted nonspecific interaction s between CPEs and non target species must be carefully examined. 67 68 Third, subtle change s in the experimental conditions (ionic strength, temperature, pH, and additives) may induce significant change s in the sensor systems A s a result all experimental conditions must be controlled and optimized to achieve reproducibility.

PAGE 48

48 Overview of This Dissertation The primary goal of the present study is to investigate the aggregation behavior of functionalized poly(phenylene ethynylene) (PPE) upon the addition of different a nalytes and to explore their optical sensor applications towards various targets including metal ions and pyrophosphate (PPi). Chapter 2 reports a study of the ion induced quenching of an anionic PPE ( PPE d CO 2 ) using FCS. The interaction of the C PE with a variety of metal ions including Na + K + Ca 2+ Cu 2+ Fe 2+ and Fe 3+ was investigated. FCS results showed that the diffusion time increases in the order K + Na + < Ca 2+ < Cu 2+ < Fe 2+ < Fe 3+ As discussed above, long diffusion times usually indicate large CPE aggregates. Comparison of the diffusion time ratio plot from FCS to the Stern Volmer plot from fluorescence spectroscopy showed different response s of the two techniques, most likely due to an amplified quenching effect Chapter s 3 and 4 illustrate the aggregation based detection of struc turally similar diamines and t etra amines using a carboxyl ate functionalized poly (phenylene ethynylene). Upon the addition of amines, the PPE can cross linked via electrostatic i nteractions between the polymer carboxylic acid binding sites and the amino groups of the multiamines to form interpolymer aggregat es leading to fluorescence self quenching of the polymer FCS was also utilized to study the size change of the polymer with different amines. Only tetraamines was able to bind strongly to PPE to form tight

PAGE 49

49 aggregates In addition, a turn on sensor for pyrophosphate ( PPi ) was designed based on the strong association of PPi with tetraamine. Application of p rincip al component a nalysis (PCA) to this system yielded t wo princip al components The PCA method was able to determine the concentration of PPi with an approximately 95% accuracy when PPi is within concentration range from 100 M t o 3 mM Chapter 5 describe s a real time fluorescence assay for acetate kinase (ACK) using a PPE with branched ammonium side chains ( PPE d NH 3 ). The cationic PPE can be quenched very efficiently by PPi due to the formation of large aggregates, while other substrates includ ing Pi, a ce tyl phosphate and acetate fail to induce any significant fluorescence change. Therefore, a real time fluorescence turn off assay for the enzyme acetate kinase (ACK) using PPE d NH 3 as the signal transducer wa s developed. The assay operates with substrates in the micromolar range, and it offers a straightforward and rapid detection of ACK activity for enzyme present in the nanomolar concentration range.

PAGE 50

50 CHAPTER 2 ION INDUCED AGGREGATION OF CONJUGATED POLYELECTROLYTES STUD IED BY FLUORESCENCE CORRELATION SPECTROSCOPY Background As discussed in Chapter 1, t he addition of oppositely charged small ions into a CPE solution lead s to significant changes in the photophysical properties. 39 69 Early studies found that some metal ions, such as Pd 2+ and C a 2+ exhibit a superquenching ability towards carb oxylated poly(phenylene ethynylene)s (PPE) in aqueous solution. 18 22 32 33 The sensitive photophysical response of conjugated polymer has been ascribed to either the electron transfer bound by more than one carboxylate belonging either to a single polymer chain or to adjacent poly mer chains to form aggregates. Several ion sensors have been developed based on cation i nduced CPE fluorescence change. 18 70 Although fluorescence spectroscopic studies on CPE aggregation have been fully elucidated in the literature, 18 22 33 direct evidence of polymer aggrega tion upon addition of various types and concentration of ions is still insufficient. In order to better interpret the spectral change of CPE and to clarify potential ambiguities, this work made use of fluorescenc e correlation spectroscopy (FCS ). FCS has been primarily employed for the analysis of biological systems, 46 52 but applicat ion to polymers, in particular CPEs, has gained increasing interest throughout the past decade due to the high sensitivity and the

PAGE 51

51 capability of single molecular analysis. Because of their inherent fluorescence, CPEs can be observed directly, avoiding the tedious dye labeling process and simplifying studies of conformational or diffusional changes. In the current w ork we introduce an FCS system coupled with a 405 nm diode laser and it was applied to investigate the aggregation behavior of two CPEs induc ed by small molecules. Figure 2 1.Structures of dentritic polymers Th e two C PEs (PPE d CO 2 and PPE d NH 3 Figure 2 1) studied here have poly(phenylene ethynylene) backbones substituted with bulky, highly charged ionic side chains. 30 31 Anionic PPE d CO 2 contains carboxylate groups as the side chain while PPE d NH 3 is cationic with a branched polyamine as the functional group. As discussed previously both polymers exhibit similar optical properties in methanol and water, since they exist as free single polymer chain in these solvents. 71 In this ch ap t er, the interactions of the CPEs with six metal cations (PPE d CO 2 ) or pyrophosphate anions

PAGE 52

52 ( PPE d NH 3 ) were investigated by FCS to observe the relationship between the chain aggregation and fluorescence quenching. The FCS results indicate that addition of Fe 2+ Fe 3+ and Cu 2+ all transition metal ions with high charge density, can readily induce aggregation of PPE d CO 2 chains. Similarly, the addition of negatively charged PPi to the cationic PPE d NH 3 system also induces aggregate formation. The se findings agree well with the results obtained by fluorescence titration experiments. We explain the fluorescence change by different quenching mechanisms and also explore the amplified quenching mechanism in terms of aggregation/physical interactions of CPE. The results demonstrate that FCS with 405 nm excitation has great potential for studies of the molecular conformational and diffusional behavior of CPEs. Results and Discussion Investigation of I nteraction of A nionic PPE d CO 2 with M etal C ations Fluorescence q uenching of PPE d CO 2 i nduced by m etal c ations The objective for developing the FCS system was to utilize it to probe the interactions of CPE chains with ions in aqueous solution. We also have the goal to correlate the metal ion induced change s in CPE solutions with the changes in fluorescence. Six different metal cations (Na + K + Ca 2+ Cu 2+ Fe 2+ and Fe 3+ ) with a concentration range of 0 25 M were added to 780 nM aqueous PPE d CO 2 and the fluorescence spectra were recorded. Figure 2 2 illustrates the Stern Volmer plots of I 0 /I as a function of the concentration for different metal ions at the maximum emission

PAGE 53

53 wavelength of 436 nm for PPE d CO 2 where I 0 and I are the emission intensities of polymer before and after the addition of metal ions. Little quenching is observed for PPE d CO 2 for [M n+ ] < 2 M. However, when the concentration increases above 3 M, a dramatic increase in the slope of the plots is observed for Fe 2+ Fe 3+ and Cu 2+ signaling the onset of a highly efficient quenching process. The fluorescence intensity for PPE d CO 2 begins to decrease very sharply from this point for Fe 2+ and Fe 3+ until the SV plot levels off. At the highest concentration level, 25 M, the fluorescence intensity is quenched 97%, 96% and 95% relative to fluorescence of the unquenched PPE d CO 2 for Fe 2+ Fe 3+ Cu 2+ respectively. In comparison, all the other ions Na + K + and Ca 2+ were unable to induce any significant quenching. Figure 2 2. Stern Volmer Plot for PPE d CO 2 (780 nM) in water with different metal ions, I 0 and I are the fluorescence intensities of PPE d CO 2 before and after the addition of metal ions.

PAGE 54

54 FCS m easurements The diffusion behavior of 780 nM aqueous PPE d CO 2 with the same six metal ions was investigated via FCS. The normalized FCS correlation curves for PPE d CO 2 with six metal ions at 15 M are shown in Figure 2 3A Although the sizes of the polymer and the polymer/metal complexes are not absolutely uniform, the single species fitting equation ( see E quation 2 2 in experimental section ) was nonetheless used to fit the FCS curves, so that the average diffusion time for each CPE sample could be obtained. As can be seen, the polymer alone has the shortest diffusion time at 60.2 1.8 s. The addition of Na + K + and Ca 2+ ions induces very little change in the correlation curves, and the diffusion time increases only slightly to 69.2 3.1 s for Na + 66.8 2.7 s for K + and 2+ indicating that the size of the polymer increases very little in the presence of these metal ions. However, when the same concentration of Cu 2+ is added to the polymer solution, the curve shifts to a much longer diffusion time of 0.406 ms, which is almost 7 times that of the pure polymer. The curve shifts even further with a diffusion time of 1.06 ms upon the addition of Fe 2+ suggesting the formation of large aggregates in the presence of these metal ions. The largest change is observed after addition of Fe 3+ which features the longest diffusion time of 4.38 ms. From the comparison, we can conclude that the ability of metal ions to induce aggregation increases in the order of K + + < Ca 2+ < Cu 2+ < Fe 2+ < Fe 3+ which is consistent with the increase of the positive charge density on the metal ions as well as their bin ding

PAGE 55

55 affinities to the carboxylate groups on the polymer side chains. In addition, Fe 2+ and Fe 3+ and Cu 2+ tend to bind to multiple donor atoms, and thus can bridge two of the tri acid units on two adjacent PPE d CO 2 chains. Note there are some unusual spike s showing up in the diffusion curve with Fe 3+ and this is due to the formation of very large aggregates. 53 The count rates (corresponding to the fluorescence intensity) versus observation time for polymer/metal ion mixtures at [ metal io 2 3B Most of the fluctuation profiles have a narrow distribution of fluorescence events, and therefore the fluctuation profiles often have a stable baseline. 62 However, in some cases, 51 53 a few peaks are observed in the fluctuation profiles of some polymer metal mixtures (such as PPE d CO 2 with Fe 2+ ). Such peaks have usually been attributed to the existence of some relatively large particles passing through the excitation volume, 72 73 indicating the formation of aggregates. Note that the count rate, which reflects the corresponding to the fluorescence intensity, decreases in the order of Na + > K + > Ca 2+ > Fe 2+ > Cu 2+ > Fe 3+ According to previous reports 29 and the results described above, the formation of aggregates can efficiently quench the CPE fluorescen ce, because the emission is dominated by excitons, which are trapped in the aggregated state. Thus, the above intensity order is consistent with the observed quenching ability of these metal ions.

PAGE 56

56 Figure 2 3. ( A) Normalized correlation curve for PPE d CO 2 (780 nM) in water with different ions (15 ( B) Count rates for PPE d CO 2 (780 nM) in water with The black solid lines are single specific fitting curves. The effect of metal ion concentration on the size of the PPE aggregate s was also investigated. The correlation curves for PPE d CO 2 / Fe 2+ mixtures at [Fe 2+ ] = 5 25 M are specifically selected and presented in Figure 2 4 In general, the diffusion time increases gradually with increasing [Fe 2+ ], reflecting an increase in agg regate size. In particular,

PAGE 57

57 this is a considerable increase in size from 5 M to 7.5 M, indicating that the formation of very large aggregates is triggered at this point, with [Fe 2+ M as the onset concentration for formation of large CPE aggregate s. A small spike is observed near the end of the curve when [Fe 2+ ] = 25 M. This is presumably due to the formation of larger particles. 53 Figure 2 4 Normalized correlation curves of PPE d CO 2 (780 nM) in water with different [Fe 2+ ] (5 25 M) The black so lid lines are single specific fitting curves. In order to gain a complete picture of the effect of metal ions on the diffusion behavior of PPE d CO 2 the d iffusion time ratios / for PPE d CO 2 in water before and after the addition of metal ions with varying concentrations from 5 M to 25 M were calculated and are plotted in Figure 2 5 Detailed diffusion time ratio data for each ion is summarized in Table 2 1.

PAGE 58

58 Table 2 1 Diffusion time ratio ( / ) of PPE d CO 2 ([PPE d CO 2 ] = 780 nM) with different metal ions at different concentrations. Concentration of ions ( ) Na + K + Ca 2+ Fe 2+ Fe 3 + Cu 2 + 5 1.04 1.07 1.05 1.38 2.97 1.3 7.5 7.52 11.84 2.4 10 8.97 18.18 3.5 12.5 12.10 41.61 4.5 15 1.15 1.11 1.51 17.66 72.72 6.75 20 43.53 >100 7.24 25 1.16 1.21 1.90 77.34 >100 8.59 Based on E quation s 2 3 and 2 5 (see experimental section) the diffusion time is proportional to R the hydrodynamic radius of the particle. Therefore, any changes in the hydrodynamic radius will affect the mobility of the molecules; a large particle would usually have a longer diffusion time than small particles in a certain fixed volume. 74 75 As a result, each of the plots provides insight as to how the size of the polymer changes in the presence of different metal ions at varying concentrations. As illustrated in Figure 2 5 the / is very close to 1 for Na + and K + at all concentrations, suggesting that monovalent metal ions fail to induce the aggregation of PPE d CO 2 The ratio increases to 2 for Ca 2+ at 25 M and for Cu 2+ the ratio is about 8 at this concentration, indicating that divalent metal ions intend to bind with polymers and induce the formation of small aggregates. However, for Fe 2+ and Fe 3+ / increases greatly with the increasing of

PAGE 59

59 concentrati on and reaches 70 when [ Fe 2+ ] = 25 M and [Fe 3+ ] = 15 M Very large CPE aggregates are believed to be formed in these systems. Figure 2 5 Diffusion time ratio for PPE d CO 2 (780 nM) in water with different metal ions measured by FCS, and are the diffusion time of PPE d CO 2 before and after the addition of metal ions. By comparing Figures 2 2 and 2 5 a direct relationship between fluorescence quenching and polymer aggregation is observed. It is evident that metal ions display similar trend s in polymer fluorescence quenching and aggregation. The most efficient Stern Volmer quenching is observed for the metal ions that give rise to the longest diffusion times, i.e., largest polymer aggregates, verifying previous reports that polymer aggregati on plays an important role in the amplified quenching effect. 18 22 23 33 Moreover, these results also shed light on the origin of the super linear Stern Volmer correlations that are frequently observed for CPE quencher systems. From the FCS results, it is

PAGE 60

60 evident that there is a minimum concentration of metal ions needed to induce aggregation (7.5 M for Fe 2+ in this study). Size c alculation and AFM s tudies Although it is d ifficult to know the exact shapes of the particles in different systems, we assume that they are approximately spherical particles to calculate approximate hydrodynamic radius, R H By using E quation 2 3 and 2 5 (see experimental section) the R H of PPE d CO 2 is calculated to be ~1.9 nm. After the addition of 15 M metal ions, the values of R H are 2.2, 2.1 and 2.9 nm for PPE d CO 2 with Na + K + and Ca 2+ respectively, while for Fe 2+ Fe 3+ and Cu 2+ the calculated R H values are 34, 140 and 13 nm, respectively. Additional information about the size of the mixture is summarized in Table 2 2 Table 2 2. Hydrodynamic radius ( R H /nm) of PPE d CO 2 ([PPE d CO 2 ] = 780 nM) with different metal ions at different concentrations. Concentration of ions ( ) Na + K + Ca 2+ Fe 2+ Fe 3 + Cu 2 + 5 2.02 2.08 2.04 2.68 5.76 2.52 7.5 14.6 23.0 4.66 10 17.4 35.3 6.79 12.5 23.5 80.7 8.55 15 2.23 2.15 2.93 34.3 141 13.1 20 84.4 >194 14.0 25 2.25 2.35 3.69 150 >194 16.7

PAGE 61

61 AFM was also used to study the size change of the polymer in the presence of Fe 3+ at different concentrations. The solutions were deposited on the surfaces of mica substrates and dried overnight before undergoing AFM measurement. As shown in Figure 2 6, pure PPE d CO 2 without any metal ions is well dispersed and no large 3+ to the polymer solution, several clusters of the polymer are observed with fairly large sizes. As [Fe 3+ a much larger aggregate is observed with a significant height of 200 nm. It was not easy to find this aggregate during sample scanning, indicatin g that the aggregates formed in dilute solution are few in number with many polymer strands per particle. The AFM image supports the FCS results that Fe 3+ can successfully induce strong inter chain aggregation of PPE d CO 2 Figure 2 6. AFM images of PPE d CO 2 ( 1 M) with different [ Fe 3 + ] ( 0 30 M). ( A) PPE is well dispersed when no Fe 3 + is added. ( B) Small aggregates formed when [ Fe 3 + ] = 5 M ( C) Large aggregated formed when [ Fe 3 + ] = 30 M PPi Induced Aggregation of Cationic PPE d NH 3 In an earlier study, 30 31 we reported that PPi (structures in Figure 2 7) can efficiently quench the fluorescence of cationic PPE d NH 3 (structure in Figure 2 1 ). Due to the fact

PAGE 62

62 that PPi cannot directly participate in any electron or energy transfer process with the polymer, the m ost likely reason for this quenching capability is aggregation of the polymer through electrostatic or hydrophobic interactions. In our previous work, di rect eviden ce for aggregation induced quenching was insufficient, and in the present report we utilize FCS to prove the hypothesis. Initially, fluorescence spectra of 1 M PPE d NH 3 in MES buffer in the presence of PPi from 0.5 to 10 M were recorded (Figure 2 7). When the PPi concentration increases, the strong blue emission at = 430 nm decreases, accompa nied by appearance of a broad and structureless green band at around = 520 nm. The large red shift (~90 nm) of the fluorescence spectrum suggests that the photoluminescence emanates from a lower energy state 29 and that strong inter may enhance the conjugation effect and lower t he overall energy level. Figure 2 7 Fluorescence spectra of PPE d NH 3 solution (1 M) titrated with PPi in MES buffer (10 mM, pH 6.5).

PAGE 63

63 To further investigate the reason for the PPi induced fluorescence quenching of PPE d NH 3 FCS measurements are carried out on 1 M PPE d NH 3 solutions with varying PPi concentrations of 1, 10 and 25 M in MES buffer solution. The normalized correlation functions obtained for each mixture are displayed in Figure 2 8 A The initiate diffusion time of PPE d NH 3 is 58.0 s, however, after the addition of 1 M PPi, the correlation curve does not change much ( = 58.5 s). When [PPi] increases to 10 M, a shift in the FCS curve is observed, indicating the formation of large aggregates with an estimated diffusion time of 2.03 ms. After 25 M PPi is added to the polymer solution, a more pronounced shift is observed and the diffusion time is ~ 11.8 ms. The result demonstrates that even larger aggregates are formed with increasing [PPi]. The calculated R H ar e 1.88 nm for PPE d NH 3 itself; and 65.7 nm and 382 nm when [PPi] = 10 M and 25 M, respectively. Figure 2 8B illustrates the count rate of PPE d NH 3 (1 M) /PPi mixtures with different [PPi] in MES buffer In the control experiment of PPE d NH 3 without PPi, the baseline is quite stable with no obvious peaks and it has relatively high fluorescence intensity. A large peak appears after the addition of 10 M PPi to the polymer solution, indicating that large aggregates are passing through the illumi nation volume. The fluorescence intensity largely decreases when PPi is added. When the [PPi] increases to 25 M, a higher peak appears, attributed to the formation of even larger aggregates in

PAGE 64

64 the solution. The fluorescence intensity of this system is the lowest, and the intensity trend is consistent with the spectra shown in the Figure 2 7. Figure 2 8. ( A) Normalized correlation curve for PPE d NH 3 (1 M) with different [PPi] in MES buffer (10 mM, pH 6.5). The black solid lines are single specific fitting curves. ( B) Count Rates for PPE d NH 3 (1 M) with different [PPi] in MES buffer (10 mM, pH 6.5).

PAGE 65

65 Mechanisms In the first system with PPE d CO 2 the cationic ions supposedly form ion pair complexes with the anionic polymer. A photon induced electron transfer mechanism is responsible for the significant quenching of PPE by electron deficient ions Cu 2+ Fe 2+ and Fe 3+ where the excitons will be trapped and deactivated. 76 78 Other ions, like Ca 2+ have close shell structures and are incapable of attracting electrons from the polymer chains. It has been reported that the quenching efficiency of oppositely charged quenchers on CPEs is more eff icient when the polymer chains are highly aggregated, due to the interchain exciton migration. 23 76 79 80 Larger aggregation is induced by those ions (Fe 3+ Fe 2+ and Cu 2+ ) with higher positive charge density compared to Na + and K + Moreover, the reason Fe 2+ and Fe 3+ have higher quenching efficiencies than Cu 2+ is likely due to a higher association constant for binding of iron ion to the anionic side chains in contrast to Cu 2+ which primarily forms complexes containing four ligands (tetrahe dral or distorted octahe dra l with water in axial position). Fe 2+ and Fe 3+ form octahe dral complexes with six donor atoms, resulting in a greater tendency to crosslink two sets of tri CO 2 pendants on different polymers and form aggregates. Iron ( III ) in particular, forms strong compl exes with oxygen donors. Consequently, the amplified quenching is the result of the combined effects of both ion pairing charge transfer and multivalent related aggregation.

PAGE 66

66 The mechanism is illustrated in Figure 2 9 When a small amount of Cu 2+ Fe 2+ or Fe 3+ ([metal ions] < 3 M ) is added to the PPE d CO 2 aqueous solution, the metal cation will bind with the anionic carboxylate groups on the polymer due to electrostatic interactions. At this point, the negative charge density of the polymer is partially d ecreased, but the polymer still retains the characteristics of molecularly dissolved chains, and photo induced electron transfer is not pronounced for the electron deficient metal ions. Thus, only slight quenching of the polymer is observed and the small s ize change is not detectable by FCS. When the concentration of metal ions exceeds 3 M inter chain crosslinking of the polymer is induced, and small aggregates of PPE d CO 2 begin to appear in the solution. As a result, an exciton generated upon photo excitation can migrate along different polymer chains until it meets a metal ion, where the exciton is quenched. Since any electron receptor, i. e., a metal ion, attached to a single polymer chain is capable of quenching a number of polymer chains in the aggregate, amplified quenching occurs, resulting in the onset of a large decrease of fluorescence in th e Stern Volmer plot in Figure 2 2 However, the molecular weight becomes only three or four times that of the single chain, as only small aggregate s form at this point. Based on E quation s 2 3 and 2 8 (see experimental section) the change of the molecular weight is the polymer changes only slightly, and only small increase in the diffusion time is observed (Figure 2 5 ). After a large excess of metal ions is added, PPE d CO 2 is highly

PAGE 67

67 crosslinked by the metal cat ions to form large aggregates. A significant extension of the diffusion time of polymer is observed, corresponding to the large increase of molecular weight and a large rise of the diffusion time ratio is observed (Figur e 2 5 ). This model explains the response lag between the Stern Volmer plot and the diffusion time ratio plot in Figures 2 2 and 2 5 The diffusion time ratio increases remarkably at 10 M Fe 2+ and at 7.5 M Fe 3+ while in the Stern Volmer plot, the fluorescence decreases significantly after the addition of only 3 M Fe 2+ or Fe 3+ Figure 2 9. Mechanism for PPE quenched by metal ions A similar mechanism can be applied to the other system with PPE d NH 3 PPE d NH 3 exists as a single chain without PPi in MES buffer solution and e mits high intensity blue light. After PPi is attached to the NH 3 + groups, the intense negative charge density on

PAGE 68

68 P Pi neutralizes the polyamine groups and also induces inter chain interaction of PPE d NH 3 to form aggregates. Fluorescence of PPE d NH 3 is highly quenched and shifts to longer wavelength. The FCS result gives solid proof of formation of large aggregates, esp ecially after adding 10 M PPi. By introducing FCS, the amplified quenching mechanism has been investigated from a new aspect. More importantly, the comparison of different responses by two techniques based on different mechanisms (fluorescence quenching a nd FCS), provides insight into the details of molecular interactions during the quenching process. Summary We have successfully constructed an FCS system with blue laser as light source and provided the details regarding the alignment, optimization, and calibration of the setup. Two examples were carefully investigated. In the study of PPE d CO 2 with metal ions, Fe 3+ Fe 2+ and Cu 2+ which have much higher polymer quenching efficiencies compared to other ions, were shown to induce significant polymer aggre gation based on the FCS results. The electron densities of the metal ions and their binding affinities to the carboxylate groups on the functional chains play an essential role in aggregate formation. The correlation between the Stern Volmer and diffusion time ratio plots demonstrates that aggregation of the CPE contributes significantly to the amplified quenching phenomenon. Furthermore, the large increase in diffusion time observed by FCS clearly proved that the quenching of PPE d NH 3 is due to a conformat ional or

PAGE 69

69 aggregation induced mechanism. Both of the results are consistent with the data from fluorescence measurement. The changes in photophysical property of CPE resulted from the combined effects of electron transfer and conformation change of polymer s. It is difficult to distinguish the effects from the two mechanisms based solely on the fluorescence spectroscopy measurements. FCS is capable of measuring polymer size in solution, providing insight and direct information about physical state of the pol ymer. With the help of FCS, we were able to explain the fluorescence change caused by different quenching mechanisms in greater detail and also to prove the amplified quenching mechanism via a different experimental method. Ions with high charge density ca n induce formation of larger aggregates, consistent with their higher quenching efficiency as shown in the fluorescence spectra. Thus ion induced aggregation significantly contributes to the fluorescence quenching of CPE. In this work, use of a 405 nm las er has successfully expanded application of FCS to CPE area and established a platform for in depth study of conformational changes and the diffusion behavior of CPE. FCS is clearly a promising and powerful tool for investigations of the interactions betwe en CPE and other ions or molecules, as well as resulting conformational changes.

PAGE 70

70 Experimental Materials PPE d CO 2 and PPE d NH 3 were synthesized and prepared according to the reported procedures. 30 31 Molecular weights of the PPE d CO 2 and P PE d NH 3 were ca. 27,300 and 16,600 Da, respectively. Fluorescein was purchased from Fisher. 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 gr ade materials (Fisher). All polymer concentrations are reported as the polymer repeat unit concentration. Fluorescein was prepared in 10 mM phosphate buffer (pH = 8). MES buffer solution (10 mM, pH = 6.5) was prepared from 2 (N morpholino) ethanesulfonic a cid and sodium hydroxide. All metal ions were purchased from Sigma Aldrich Company and used as received. Sodium pyrophosphate was obtained from J. T. Baker Chemical Co. Instrumentation FCS measurements were taken on a homemade setup using a 405 nm diode l aser (Coherent, CUBE) as the excitation light (Setup in Figure 2 10) Fluorescein (30 nM in 10 mM phosphate buffer, pH = 8) was used as the calibration standard for the system. Fluorescence spectra were recorded on a Photon Technology International fluorom eter and corrected by using correction factors generated with a primary standard lamp.

PAGE 71

71 Figure 2 10. Setup of FCS system in our lab Theory of FCS In principle, f luctuations in the fluorescence signal are quantified by temporally autocorrelating the recorded emission signal s collected within the confocal volume. The normalized autocorrelation function, defined as 52 ( 2 1) is used to characterize the temporal fluctuations In E quation 2 1, describes the fluctuation of the fluorescence signal as deviations from the temporal average of the signal at time t A three dimensional fitting model representing a single component system is written as: ( 2 2)

PAGE 72

72 w here is the longitudinal radius and is the transversal or waist radius of the confocal volume ; and the structure parameter, equates to N is the average number of fluorescent molecules in the confocal volume ; is the average time of fluorescent molecules diffusing in the de tection volume, which i s characteristic for a specific molecule. The relationship of to the molecular diffusion coefficient D (m 2 s 1 ) is given by: ( 2 3 ) The waist radius is obtained from its conversion equation : ( 2 4 ) where is the diffusion coefficient of the standard calibration dye. The translational diffusion coefficient D of a molecule is related to its s ize by the Stokes Einstein equation : ( 2 5 ) where k ; T is the temperature ; is the viscosity of the solvent ; and R is the hydrodynamic radius. E quation 2 5 can be used to estimate the size of diffusing particles by assuming the particles has a spherical sha pe with radius R which is related to the molecular weight MW of the molecule with a specific gravity by ( 2 6 ) where V is the volume of molecule. Thus we have

PAGE 73

73 ( 2 7 ) These equation s show that the radius R and diffusion coefficient D are weakly dependent on the molecular weight. By combining E quation 2 5 and 2 7 we have: ( 2 8 ) This relationship is useful for estimation of the molecular weight of a spherical particle from its diffusion coefficient.

PAGE 74

74 CHAPTER 3 DIFFERENT AGGREGATIVE RESPONSES FOR NONQUENCHING MULTICATIONIC AMINES USING CONJUGATED P OLYELECTROLYTES Background Amines are a class of small organic molecules that are used extensively in industries for colorant manufacture, fertilizer production and pharmaceutical preparation. 81 85 Biogenic polyamines (e.g. cadaverine, spermine, spermidine, and putrescine) are present in living organisms and are known to be involved in regulation of gene expression, cell proliferation and differentiation. 34 86 In additi on, some amines have been identified as potential indicators for various health risks, including cancer, bacterial infection and food poisoning. 87 89 T he role amines play in human health and food quality is so important that it is imperative to design and develo p an easy, rapid and effective method to detect the target amines. In recent decades, some methods have been developed to sense different amines including using antibodies, 90 molecularly imprinted polymers (MIPs) 91 92 enzymes, 93 single molecule 94 95 array sensors, 96 97 electrophoretic analysis 98 and the most commonly used method high performance liquid chromatograph y. 99 T hese traditional approaches have several limitations. For example, existing amine assays are usually laborious, requir ing expensive equipment, and are rel atively slow and therefore unsuitable for rapid screening applications. As a result, a rapid and sensitive detection method for amines would be very promising to screen a large number of amines samples, especially those with similar structures and properti es.

PAGE 75

75 While many highly sensitive CPE based sensors have been reported, most signal transduction mechanisms for these CPEs reply on the electron or energy transfer between the polymer and a quencher. 71 100 101 Other analytes such as amines cannot participate in these direct quenching processes due to their incompatible redox and spectral properties. In this situation, an alternative approach driven by analyte induced aggregation has been employed, le ading to self quenching processes. This strategy takes advantage of the interchain conformational change of the polymer induced by electrostatic and/or hydrophobic interactions between these nonquenching analytes and polymers. Moreover, the assembly of pol ymers by different amines leads to different aggregative responses. In this chapter, we utilize P 3 (an anionic PPE functionalized with carboxylate groups Figure 3 1) to investigate the different aggregative responses by a group of three structurally similar amines : ethylenediamine ( C2N 2 ), cavaderine ( C5N2 ) and t ris(3 aminoethyl)amine ( N4 ) (structures shown in Figure 3 1 ) Upon the addition of amines, only N4 was found to effectively induce the formation of tightly associated aggregates be tween the polymer chains in solution, while C2N 2 and C5N2 showed negligible effect on the polymer aggregate formation. The results are consistent with similar work reported by Swager. 57 Diffe rent fluorescence responses to different amines we re also observed, indicating that the interaction between polymer chains and amines varie d depend ing on the shape, valence and length of the amines. This work provides an

PAGE 76

76 easy way to discriminate these simi lar amines and a better idea of how different analytes interact with polymers. Figure 3 1 Chemical structures of P 3 ethylenediamine dihydrochloride ( C2N2 ) cadaverine dihydrochloride ( C5N2 ) and tris (2 aminoethyl)amine ( N4 ) Results and Discussion Photophysical Properties of P 3 A conjugated polyelectrolyte substituted with carboxylate groups ( P 3 ) was designed and utilized in our experiments. 102 The normalized absorption and fluorescence spectra for P 3 in various solvents are shown in Figure 3 2. In methanol, P 3 exhibits an absorption wavelength maximum at 391 nm with a structured emission of 425 nm. In aqueous solutions P 3 shows s imilar spectro scopic profiles with almost the same wavelength maxima or spectral band shape for the absorption and fluorescence spectra

PAGE 77

77 (Table 3 1 ) This suggests P 3 is not aggregated in water, existing as molecularly dissolved chains. 103 It is likely the presence of the h igh negative charge density on each repeat u nit of polymer minimizes the electrostatic interaction and interrupts th e interchain aggregation leading to enhancement in their fluorescence properties. For example, P 3 maintains a relatively high quantum yield in water ( F = 0.42, using quinine sulfate in 0.1 M H 2 SO 4 as standard), as it decreases only ~20 % compared with that in methanol ( F = 0.52). Table 3 1. Photophysical properties of P 3 in different solvents MeOH H 2 O max max abs max em F max abs max em F P 3 391 425 0.52 391 426 0.42 Quinine sulfate in 0.1 M H 2 SO 4 is used as actinometer, F =0.545. Figure 3 2. Normalized absorption and fluorescence spectra of P 3 in MeOH (solid line ) and H 2 O pH = 6.5 (dash ed line ), ex = 360 nm

PAGE 78

78 Fluorescence Quenching by Different Amines Since anionic P 3 the aggregation behavior of this polymer to oppositely charged, small molecule amines, was investigated. The aggregative responses of P 3 to different amines were carried out i n water at a polymer concentration of 2.0 M. In this experiment, three amines we re chosen: eth ylene diamine ( C2N2 ) cadaverine ( C 5 N 2 ) and tris(2 aminopropyl)amine ( N4 ) (see their chemical structures in Figure 3 1 ) A mong these amines the diamines vary by the number of methylene unit separat ing the diamino groups The fluorescence titration spectr um for each amine at varying concentrations is displayed in Figure 3 3. As the concentration of N4 increased, the fluorescence intensity significantly decreases a nd a red shifted band show s up at ~ 5 3 0 nm consistent with the aggregation induced planarization of the polymer chains. In comparison, t he addition of C2N 2 and C5N2 were able to promote the planarization and a small degree of aggregation (possibly dimerization) between the polymer chains, resulting in fluorescence self quenching. However, neither of them was capable of inducing the polymer aggregate formation. T he de gree of aggregation or possibl y dimerization depend ed on the charge density and the structure of the amines. 104 105 Figure 3 3D shows the Stern Volmer plot of t he fluorescence spectra of P3 with different amines. Overal l, N4 ha s the highest positive charge density (carrying three positive charges in pH 6.5 water based on pKa values 106 ) and therefore features the most effective quenching ability compared to the other

PAGE 79

79 diamines. D iamines we re able to quench the polymer; however, they demonstrate d different quenching efficiencies. The quenching efficiency decreases as the chain length of diamine d increases, indica ting that different structure of the diamines led to the stacking of polymer chains to a different degree. 107 C5N2 wa s the least effective quencher because its long and flexible structure that cannot bring two polymers in close distance to form tightly associated aggregates. Therefore, this CPE based chemical sensor still exhibits some selectivity towards different amines. Figure 3 3 ( A ) ( C ): Fluorescence spectroscopic changes of 2.0 M P3 solution observed upon titration of different amines in water : (A) N4 (B) C2N2 (C) C5N2 ; ( D ): Stern Volmer plots of I 0 / I as a function of amine concentration in water pH = 6.5, ex = 360 nm.

PAGE 80

80 Effect on Amplified Quenching of P 3 by MV 2+ P reviously it is has been noted that aggregation has a significant effect on amplified quenching, 101 therefore we examined the effect of amine induced aggregation on the am plifi ed quenching of the polymer by methyl viologen (MV 2+ an electron transfer quencher ). Figure 3 4 A illustrates a series of Stern Volmer plots for MV 2+ quenching of P 3 fluorescence in the presence of various amines at a concentration of 25.0 M in water (pH = 6.5 ) It is observed MV 2+ quenches P 3 fluorescence with very different efficiency in different P 3 /amine systems. Interestingly, c ompared to th e control experiment without amine N4 significantly enhances the quenching by MV 2+ ; whereas the other amines ( C2N2 and C5N2 ) decrease the amplified quenching effect. O verall, the quenching efficiency decreases in the order: N4 > c ontrol (no amine) > C2N2 > C5N2 corresponding to the decreasing positive charges and increasing chain length of the diamines. I n the P 3 / N4 complex, less than 2 00 nM MV 2+ is required to quench 90% of the polymer fluorescence while only ~65% of the fluorescence is quenched for the control experiment at the same level of [ MV 2+ ] More importantly, the Stern Volmer plot in P 3 / N4 syste m is superlinear even at low quencher concentrations. S uperlinear quenching is observed immediately upon addition of a very small amount of MV 2+ (less than 300 nM) in the presence of P 3 and N4 s ignaling the presence of a ggregates. The enhanced quenching efficiency is ascribed from the increased exciton diffusion pathway in the PPE aggregates. In contrast, in the other two P 3 /diamine mixtures where the

PAGE 81

81 polymer is much less effectively aggregated, the SV plot features a nea rly linear plot and the slope ( K sv constant) is significantly less than that of the control at the same concentration level of MV 2+ We assume the reduced quenching efficiency is probably due to the fact that longer diamines like C2N5 and C 5 N5 are unable t o induce efficient interchain aggregation at lower concentrations; therefore less quenching sites are available for MV 2+ due to the excess presence of positively charged diamines. These diamines will preferably occupy one or two side groups within one poly mer chain, which will screen the binding between P 3 and MV 2+ Therefore, it is suggested that the effect of adding amines on the amplified quenching is strongl y depend ent on the association constant between the polymer and diamine, which is related to the charge density as well as the rigidity of the amine structures. Only amines featuring high charge density lead to efficient aggregation of polymer. This finding illustrat es the correlation between the quencher induced aggregations of CPEs and amplified que nching effect s To get a detailed perspective of how the presence of diamines affect amplified quenching, the amplified quenching effect of P3 by MV 2+ in the P 3 / C2N2 complex solutions was investigated with varying [ C2N2 ] from 0, 10, 25 to 50 M. The Stern Volmer results are illustrated in Figure 3 4B The addition of C2N2 decreases the amplified quenching efficiency and increasing [ C2N2 ] will even lower the amplified quenching effect For example, at the highest [MV 2+ ] l evel, the quenching efficiency is ~ 97 % for the control without addition of C2N2 ; but it decreases to ~ 90 %, ~85 % and

PAGE 82

82 ~ 78 % when [ C2N2 ] is increased to 10, 25 and 50 M. From the comparison, we can conclude the more and more binding sites on the polymer chain will become unavailable for MV 2+ as [ C2N2 ] is increased, leading to a higher competition and a decreased amplified quenching efficiency. The result is basically consistent with what we have discussed above Figure 3 4 ( A ): Stern Volmer plots of I 0 / I as a function of MV 2+ concentration f or d ifferent P3 /amin e systems in water [ P3 ] = 2.0 M, [amine] = 25.0 M. ( B ): Stern Volmer plots of I 0 / I as a function of MV 2+ concentration for P3 / C2N2 mixtures in w at different concentrations. ([ C2N2 ] = 0, 10, 25 and 50 M). [ P3 ] = 2.0 M pH = 6.5, ex = 360 nm. Quantitation of Aggregation by Fluorescence Correlation Spectroscopy Fluorescence correlation spectroscopy ( FCS ) was carried out to investigate the aggregation behavior of P 3 upon addition of different amines. Figure 3 5 summarize s the FCS experimental results for sample solutions containing 2.0 M P 3 and 20 M amine of N4 C2N2 and C2N5 respectively. The result show s that P 3 / N4 complex exhibits a significantly longer diffusi on time (4.21 ms) compared with that of P 3 (0.0625 ms),

PAGE 83

83 indicating the formation of large aggregates. In contrast, the correlation curve for P 3 / C2N2 complex reveal s a diffusion time of 0.0832 m s, which is only slightly larger than that of the free polymer, suggesting the size of the polymer increases a little but not too much. The increase is even less for P 3 / C5N2 com plex with a diffusion time 0.0741 m s, indicating longer amines are unable to induce effective aggregation of polymers. Therefore, it is expect ed both diamines were only able to promote a small degree of aggregation (most likely dimerization) between polymer chains. The FCS result demonstrated the quenching efficiency of amines decreases with decreasing positive charge and also decreases with inc reasing of chain length, which is consistent with the results from the fluorescence quenching studies. Figure 3 5 Normalized correlation curves for different systems in water Black: [ P 3 ] = 2 P 3 N4 P 3 C2N2 ] [ P 3 C5N2 curves pH = 6.5

PAGE 84

84 Proposed Mechanism The quenching response s from P 3 toward s structurally similar amines result from the multiple contacts between the carboxylate side chains on the polymer and the amino groups from amines through electrostatic and/or hydrophobic interactions. However, the quenching efficiency is highly dependent on the charge density as well as the shape of the amines. For example, tight and more packed interpolymer will form when adding a highly charged amine such as tetra amine. 35 108 109 In this situation, large PPE aggregates are observed and the polymer fluorescence is more quenched ( Figure 3 6 path A). F or diamines like C2N5 and C5N2 they are unable to cross link the polymer to form interpolymer stacking aggregates due to their lower binding affinity (low positive charges) In addition, the chain leng th of these diamines is relatively long and ther ef ore their structure is highly flexible and rotated, consequently most of them will preferentially bind to one polymer chain instead multiple polymer chains (Figure 3 6 path B). In this situation, no large PPE aggregates formed The quenching is resulted from the reduced electrostatic repulsion on the side chains, leading to the decrease of the fluorescence by increasing the free rotation of the polymer chains

PAGE 85

85 Figure 3 6. Schematic representation of analyte induced aggregation between polymer and different amines. Summary In this chapter, we have investigated the interactions between a carboxyl ate functionalized poly (phenylene ethynylene) ( P3 ) and different amines. P3 i s d emission in aqueous solution. Both fluorescence quenching measurements and FCS measurements have been carried out. The results demonstrated that the positive charge density and the length of the amines have significant effect s on the aggregated states of P 3 and also amplified quenching effect. Based on the fluorescence quenching studies, it is suggested the quenching efficiency decreases in the order of: N4 > C2N2 > C5N2 which correspond s to the decreasing positive charge density and the increasing chain length of the diamines. The

PAGE 86

86 effect of the presence of a mines on amplified quenching has also been studied. It is found that different amines exhibit different effect s on amplified quenching of P 3 by MV 2+ S ignificantly enhance d amplified quenching was observed in the presence of N4 resulting from its strong complexing ability with P 3 to form large and tight aggregates The formation of aggregates increase s the three dimensional diffusion pa thway for excitons While for the diamines incl uding C2N2 and C5N2 they greatly weaken th e amplified quenching effect. It is assume d that these diamines attach to the multiple binding sites of one polymer chain instead of adjacent polymer chains due to the flexibility of their chemical structure as well as the low positive charge density. Therefore they are not able to induce efficient interchain aggregation of the polymers. Instead, the occupancy of the diamines will reduce the available bindin g sites for the incoming MV 2+ leading to decreased quenching efficiency. The amplified quenching efficiency of MV 2+ decreases with the increasing [ C2N2 ]. At the same time, the FCS measurements demonstrate that only N4 is capable of forming large aggregate s with P 3 while the size changes for the other two P 3 /amines mixtures are negligible. Possible aggregation/quenching mechanisms for P 3 with different amines are carefully explained Upon the addition of different amines, P 3 can be either cross linked to fo rm interpolymer aggregates or remain as free single chain or possibly dimers The former case leads to the fluorescence quenching of P 3 while the latter also results in the fluorescence

PAGE 87

87 quenching owing to the free rotation of the polymer chains after the s ide groups have been neutralized. From previous reports by Lavigne et al 104 they showed that diamines form either tight or loose aggregates with polymers While in our study, we proposed that longer diamines will preferably bind to one polymer chain instead of multiple polymer chains based on both amplified quenching studies and FCS measurements which is consistent with the results reported by Swager 57 However, direct evidence of how the aggregates look like is still insufficient. But our CPE based sensor study provides a simple and rapid way to discriminate different amines Experimental Materials The synthesis of P3 is described in the literature. 110 All s ample solutions were prepared using water which is distilled and purified by using a Millipore purification system (Millipore Simplicity Ultrapure Water System). Methyl viologen dichloride hydrate, ethyl ene diamine dichloride cavaderine dichloride and t ris(3 aminoethyl)amine were purchased from Sigma Aldrich All chemicals were used as rec eived, unless otherwise noted. I nstrumentation UV vis absorption spectra were recorded on a Shimadzu 1800 photospectrometer. Steady state emission spectra were measured and obtained from a spectrofluorometer

PAGE 88

88 from Photon Technology International. All the spectra have been corrected by using correction factors generated with a primary standard lamp. FCS measurements are taken on a homemade setup using a 405 nm diode laser (Coherent, CUBE) as the excitation light 30 nM fluorescein in 10 mM phosphate buffer (pH = 8) is used as the calibration standard for the system.

PAGE 89

89 CHAPTER 4 PRINCIPAL COMPONENT ANALYSIS FOR PYROPHOSPHATE SENSORS USING CONJUGATED POLYELECTROLYTES Background Pyrophosphate (P 2 O 7 4 PPi and Figure 4 1 ) is a biologically significant anion that plays an essential role in bioenergetics and metabolic processes including energy transduction, extracellular signal mediations and protein synthesis. 111 114 In particular, PPi is known to be involved in several biochemical reactions, such as DNA and RNA polymerization, 115 hydrolysis of ATP, 116 cyclic AMP synthesis and many other enzymatic reaction s. 61 117 It also has been reported that the abnormal level of PPi is related to various diseases including cancer, 118 120 arthritis, 121 and vascular calcification 122 Therefore the selective detection and sensing of PPi is importan t to understand its role in these biological processes. Among all the techniques, fluorescence detection methods received most attention due to their advantages such as superior sensitivity and selectivity, low cost, easy detection, versatility and wide dynamic range. 7 In the past few years, the development of fl uorescence sensors for PPi has become a major research focus and many PPi fluorescence probes have been reported m ost ly based on small molecule sensing methods. 123 125 In their approach es small chromophores incorporated wi th metal cations including zinc, 126 copper, 127 cadmium, 120 palladium, were employed as PPi rece ptors. PPi will coordinate with the metal ions to f orm host guest complex resulting in fluorometric (and/or colorimetric) signals changes

PAGE 90

90 The high binding affinity of PPi towards metal ions can be attributed to its good chelation and bridging capabilit ie s Even though PPi sensors based on small molecules have been broadly studied, there is increasing incentive to find alternative methods for PPi sensing owing to low PPi/Pi selectivity, considerable synthetic effort, involvement of heavy metal ions and lim ited applications in fabrication devices. 128 130 Principal component analysis (PCA) is a standard mathematical method for data analysis that is generally applie d to multivariate analytical system 131 133 It was first established in 1901 by Karl Pearson and it has become one of the most used tools in exploratory data analysis and making predictive models since the n. 134 Many sensory systems based on PCA methods have been reported 135 138 and it has been proved that PCA provides a simple and robust way to reduce data dimensionality and reveal useful information behind the complex data set with minimal effort. H owever, only a few papers described the use of PCA for the analysis of spectroscopic data set, 139 141 which is probably due to the complication of the data set, interference of background spectra and problems regarding uncalibrated s pectral features. In this chapter, we report a sensitive and selective PPi based on conjugated polyelectrolyt es. The earlier discussions in C hapter 3 provide the basis for the study presented here, where we find the addition of tris(3 aminoethyl)amine ( N4 ) is capable of inducing inter chain aggregation of P 3 in aqueous solutions. The N4 quenched fluorescence of P3 can be turned on selectively by

PAGE 91

91 PPi. T he fluorescence recovery is achieved by breaking the P3 / N4 complex, due to the strong association between PPi and N4 Particularly, the enhanced fluorescence of the CPE is blue shifted upon complexation with PPi. PCA was used as the calibration method to q uantitatively measure the concentration of PPi in the sensory system. The results demonstrate d that the PCA calibration method provide d a fast and convenient way for PPi concentration measurement with a high accuracy ~ 95% when PPi is within concentration range from 100 M t o 3 mM Figure 4 1 Structures of polymers and substrates Principal Component Analysis Basic Methods and Procedures for PCA Principal component analysis is an analytical technique to identify the similar patterns in data by reducing dimensions without loss of information, which in turn results in the predictions for some recognizable factors. In order to perform a principal com ponent analysis, we need to first establish a data matrix D 0 from the raw fluorescence data which are presented below as E quation 4 1, D 0 = [ X i, j ] ( 4 1)

PAGE 92

92 where X i, j is the jth factor associated with row i. In our study, we refer X i, j as the fluorescence intensity at wavelength i in the jth group from the raw data set. In order to avoid any inconsistency during experiment operation, each fluorescence spectrum is area normalized using E quation 4 2 D = D 0 10 9 / A ( 4 2 ) where A is the integration area for the fluorescence spectrum of P3 ( 2 M ) in HEPES buffer ( 10mM, pH 7.4 ) and 10 9 is an arbitrary number for the normalization. To make this data matrix to be factor analyzable, it can be written as a linear sum of product term s in the form of: D = RC ( 4 3 ) Here, R and C are referred as scores and loading matrices for D which can be used to reproduce the original matrix D Consequently, the next step is to find out R and C matrices from a kn own matrix D First, we need to find out the covariance matrix of D which is inherently a square matrix: Z = D T D ( 4 4 ) This ma trix is then diagonalized into E quation 4 5 : Q 1 ZQ = [ j jk ] ( 4 5 ) Since the covariance matrix Z is square, we can calculate the eigenvectors and the corresponding eigenvalues for the matrix based on E quation 4 6 Z q j = j q j ( 4 6 )

PAGE 93

93 where q j is the j th column of Q Therefore matrix Q can be easily achieved by using the eigenvectors from Z Since these columns (or eigenvectors) constitute a mutually orthonormal set, one can show that Q 1 = Q T ( 4 7 ) By exploitin g Equation 4 7 two equation s are easily obtained and R and C matrices are consequently calculated based on the relationship below, R = DQ ( 4 8 ) C = Q T ( 4 9 ) The procedure explained so far is a ge neral scheme for abstract factor analysis. 141 In this approach, the eigenvectors are consecutively calculated in order to minimize the residual error in each step. As a result, ea ch successive eigenvector accounts for the largest variation in the data. When all of the eigen values are calculated by using Equation 4 5 the variation corresponding to the largest eigenvalue and eigenvector is subtracted from the covariance matrix as sh own in E quation 4 10: R 1 = Z 1 q 1 q 1 T ( 4 1 0 ) From this residual matrix, the second principal eigenvector and its associated eigenvalue are calculated. R 1 q 2 = 2 q 2 ( 4 1 1 ) To obtain the third eigenvector, we define R 2 as R 2 = Z 1 q 1 q 1 T 2 q 2 q 2 T ( 4 1 2 )

PAGE 94

94 When one continues in this fashion, the remaining eigenvectors and eigenvalues are extracted in succession. 141 Target Transformation As previously mentioned, the R and C matrices do not have any physical or chemical meanings as they constitute an abstract solution. Therefore a target transformation is necessary to change them into meaning factors, and this is achieved by applying a transformat ion matrix and combining with Equation 4 3 D = RC = ( RT ) ( T 1 C ) ( 4 1 3 ) In E quation 4 13 transformation matrix T is a square matrix of dimension n, where n is the number of significant factors determined by PCA. T has the following form for a data matrix that can be described with two principal factors: ( 4 1 4 ) If the transformation is orthogonal (i.e., it preserves the angles between the factor axes), then a b c and d are unity. However, if the transformation is nonorthogonal, then th ese constants should be determined by taking into account prior information about the real factors. 141 Results and Discussion Overview of PPi T urn on S ensor An a nionic conjugated polyelectrolyte featuring carboxylate functional groups is used in our study ( P 3 Figure 4 1 ). The cartoon in Figure 4 2 illustrates the PPi turn on

PAGE 95

95 quench polymer dissolves as free single chains in aqueous solution and it displays strong fluorescence emission with a maximum peak at ~434 nm. The high q uantum yield (0.42 in water) also indicates this polymer does not aggregate in water. Then the fluorescence of P 3 is quenched by the addition of tris(3 aminoethyl)amine ( N4 Figure 4 1 ) which is attributed to the inter chain aggregation of P 3 A blue to g reen emission change is observed and the red shift is caused by the aggregated polymer. 22 At last, the quenched fluorescence is recovered a fter PPi is introduced to the P 3 / N4 solution. Fluorescence recovery occurs because PPi chelates N4 with a higher binding affinity leading to breakup of the P 3 / N4 cluster. Therefore, N4 can no longer complex with the polymer and liberate d the polymer from aggregated clusters. The addition of P Pi is signaled by a blue off process affords a convenient way to determine the presence of PPi. Figure 4 r

PAGE 96

96 Fluorescence q uenching of P 3 by N4 The interaction of P 3 with N4 was investigated first As discussed in Chapter 3 P 3 remains as single chains in water and it is probably due to the high negative charge density on the side groups. In HEPES buffer solutions (10 mM, pH = 7.4) N4 c an induce efficient aggregation of P 3 indicating that three positive side groups of N4 bridge the single polymer chains to form aggregates, contributing to the most significant quenching effect. The detailed titration spe ctra have been represented in Figure 4 3 W hen N4 (c = 0 10 M) wa s titrated into 2 M of P 3 in the HEPES buffer solution ( 10 mM, pH = 7.4), a strong peak at wavelength ~ 424 nm gradually decrease s which is originally from the blue emission of single C PE chains. At the same time, another broad band shows up at longer wavelength ~ 540 nm and this is attributed to the green emission of the aggregated C PE. The Stern Volmer constant ( K sv ) was calculated to be ~ 3.5 10 4 M 1 Figure 4 3. Fluorescence spect ra changes upon titration of polyamine (0 10 P 3 in HEPES buffer solution ( 10 mM, pH = 7. 4 ).

PAGE 97

97 Fluorescence recovery of P 3 / N4 by P P i In earlier literatures, it was reported PPi could complex with polyamine or similar structures through coordination very effectively 142 143 In order to investigate the i nteraction between PPi and P 3 / N4 mixtures a series of PPi titration with concentration conducted in HEPES ( 10 mM, pH 7.4) buffer solutions P 3 N4 As shown by Figure 4 4 A the fluorescence intensity of P 3 at 424 nm is quenched almost 90 % (red) relative to the initial fluorescence of pure P 3 (black) N4 is introduced Upon the ad dition of PPi, the fluorescence intensity at 424 nm increase s while the green band at 540 nm decreases After 3 mM PPi is titrated, the fluorescence at 424 nm is recovered to ~91% of the initial intensity for t he pure polymer and meanwhile ~ 70% of the fluorescence intensity at 540 nm is decreased compared to mixture solution in the presence N4 The recovered fluorescence intensity can be explained by the reduced amou nt of aggregated polymer chains Note that every spectrum wa s obtained after a complicat ed kinetic equilibrium has been reached, where PPi compete d with the polymer to bind with N4 In contrast, the effect of Pi on the fluorescence recovery of P 3 / N4 complex was investigated (Figure 4 4B Pi failed to induce any significant fluorescence recovery due to its lower binding affinity to N4 suggesting that this fluorescent sensor has a high selectivity towards PPi over Pi.

PAGE 98

98 Figure 4 4. Fluorescence spectra changes upon titration of PPi and Pi (0 3 mM) into P 3 and 4 N4 in HEPES buffer solution ( 10 mM, pH = 7. 4 ) ; (A). PPi; (B). Pi. F luorescence correlation spectroscopy measurement off change of P 3 by adding different substrates, fluorescence correlation spectroscopy (FCS) technique wa s carried out to investigate the size change s of the polymer. The correlation curves as well as the nonlinear fittings for different systems are described in Figure 4 5 where [ P 3 ] is fixed at 2 P 3 alone, it has a diffusion time ~ 6.3 10 5 s, which is quite close to the normal value for a well dispersed conjugated polyelectrolyte in aqueous solution. 144 T he correlation curve for the P 3 / N4 complex ([ N4 ] = shifts to the right with a more than 10 fold in crease of the diffusion time ( ~ 65 10 5 s ) This significant increase clearly indicated the addition of N4 induces the aggregation of the polymer. 19 As expected, the correlation curve shifts back to a smalle r diffusion time (~ 6. 4 10 5 s ) wa s applied to the P 3 / N4 mixture where it almost

PAGE 99

99 overlap with the original correlation curve for the polymer alone Overall, the introduction of PPi successfully brok e the polymer aggregates and released t he free polymer chains, which has a small diffusion time and increased fluoresce nce intensity Figure 4 5 Normalized correlation curves and diffusion time for different systems. Black: [ P 3 ] = 2 P 3 ] = 2 N4 ] = [ P 3 ] = 2 N4 ] = PCA C alibration R esult for S pectroscopic D ata S et In order to apply PCA in our study a titration experiment with PPi was initially carried out to obtain a calibration standard and the spectroscopic data set was collected and investigated by PCA method. The titration was performed in the P 3 / N4 mixture solution ([ P 3 N4 10 mM, pH = 7.4) and the emission spectra were obtai ned by a fluorescence spectrometer at thirteen different PPi concentrations. This set of 13 emission data w as normalized by adjusting the integration areas for the pure polymer emission data to 10 9 and displayed at Figure 4 6 A Then the

PAGE 100

100 spectroscopic data was transformed into a 31613 data matrix, D 31613 (where 316 row s correspond to emission wavelength covering spectral range from 384 to 700 nm and 13 column s correspond to the PPi concentrations ). Figure 4 6 ( A ) Original emission spectra of changes upon titration of PPi (0 3 mM) into P 3 and 4 N4 in HEPES buffer solution ( 10 mM, pH = 7. 4 ), ( B ) Fundamental spectra ( matrices, R ) for two largest eigenvalues. Target transformation This data matrix D 31613 was applied and calculated by PCA algorithm. Two significant eigenv ectors of covariance matrices Z (out of a total of 13) were obtained and

PAGE 101

101 they account for 99.99 % of the total v ariation. As previously discussed, the initial eigenvectors generated by PCA do not have any physical meanings, for example, one of the eigenvectors features a negative amplitude (Figure 4 7 ). Hence, an appropriate target transformation is necessary to rotate these initial eigenvectors into physically meaningful factors. Two guide lines must be followed before the rotation: 141 (1) Negative emission intensity is not allowed; and (2) both eigenvecto rs should be related to the emission spectra that are attributed from either fluorescent state of the polymer. Consequently, a new transformation matrix was produced and shown as E quation 4 1 6 : ( 4 1 5 ) where the coefficients and rotation angles were determined semiempirically. Followed by the multiplication of the initial R matrix by this transformation matrix T two new eigenvectors were obtai ned and displayed in Figure 4 6 B Red lin e ( R 1 ) represents the fundamental spectrum for the first eigenvector, whereas blue line ( R 2 ) represents the fundamental spectrum for the sec ond eigenvector. As displayed in Figure 4 6 B R 1 exhibits a strong structur e d emission spectrum with a maximum peak ~430 nm, very similar to the fluorescence emission of P 3 at its single chain state. The results suggest that the first eigenvector is mo stly single polymer dependent, representing the single chain polymer emission. R 2 shows a broad emission ban d at longer wavelength ~540 nm which i s dominated by the aggregated polymer state. Note that there are also two other small bands showing up at ~ 430 nm and ~ 450 nm which is still slightly affected

PAGE 102

102 by the presence of single state polymer. Even though the separation is not perfect for R 2 however, it accounts for the contribution mostly from aggregated polymer emission to the total emission spectrum. Hence, the PCA method successfully resolves most of the emission variations from single chain state to aggregated state polymers by providing two eigenvectors after target transformation, leading to the improved accuracy and estimation for PPi concentrations Figure 4 7 Fundamental spectra for abstract factors Loading data for each eigenvectors at different [PPi] Based on E quation 4 9 the loading percentage for each eigenvector at different PPi conditions ([PPi] = 0 3 mM) can be calculated and therefore they can be replotted to get a reconstructed data matrix. Figure 4 8 A and B illustrated the fractional contributions of both eigenvectors to the total reconstructed emission spectra at varying [PPi]. By combining them together, a reconstructed fluorescence emission spectrum is finally achieved shown in Figure 4 8 C which displays a similar pattern compared to the

PAGE 103

103 original emission spectrum (Figure 4 6 A ). The result clear shows that PCA analysis successfully reduces a l arge set of data into two principal components that can explain most of the spectroscopic change at different conditions without loss of much information. The plot in Figure 4 9 gives a detailed loading percentage graph for both eigenvectors at varying [P Pi]. With increasing amount of PPi, the effect of the first eigenvector (red line) increases while the second eigenvector (black line) decreases. The blue line is actually the total amount loading of first and second component as a sum, which appears to be a horizontal line. As noted above, the first eigenvector is mainly single polymer dependent and the second one is primarily related to the effect of fluorescence from polymers at aggregated states; therefore the blue line should be expected to be a horizo ntal line if we assume the polymer can only stay as either state (single chain or aggregate). In addition, it is of note that the value of the blue line is close to a constant (in this case, it equals to 0.25) at the standard experimental condition, sugges ting the total contribution from both eigenvectors at any [PPi] condition is a constant. This can be probably attributed to t he fact that the amount of polymers and amine in the standard conditions are fixed. Since the total emission spectrum is the combin ation of fluorescence spectrum from single polymers as well as the aggregated polymers, the total effect of both eigenvectors under the same experimental condition should be constant as long as the concentrations of polymer and amine are fixed.

PAGE 104

104 Figure 4 8 ( A ) Contribution of the first fundamental spectrum ( R 1 ) to total emiss ion at various [PPi]. (B ) Contribution of the second fundamental spectrum ( R 2 ) to tot al emission at various [PPi]. ( C ) Reconstructed spectra D = R 1 C 1 + R 2 C 2 obtained by combining the total contributions from the two largest eigenvalues.

PAGE 105

105 Figure 4 9 The contribution of two eigenvectors at different [PPi] Regression analysis One of the main objectives in our study is to allow the prediction of PPi concentr ation from an unknown sample by using PCA calibration method. In order to apply the data from standard experiment to other unknown system, regression analysis was carried out to determine the values of the coefficients for a model function that best fits t he set of data from standard experiment. 145 146 For example, the C 1 and C 2 values obtained from PCA were fitted to the PPi concentrations using several nonlinear model functions. The best fit function is displayed as E quation 4 1 6 with confidence level of fits at 95%, ( 4 1 6 )

PAGE 106

106 where y is the PPi concentration and x is the contribution percentage of C 1 ( C 1 %). The correlation coefficients for the empirical fits were 0.999 or better. Note that earlier we have explained that the sum of C 1 % and C 2 % equals to 1 since the total of C 1 and C 2 is a constant when the experimental condition is fixed; therefore in this situation we use C 1 % C 1 C 2 %) instead of using C 1 and C 2 directly for the regression analysis. By reducing the variables from 2 to 1, the complexity of the system and the order of the polynomial equation s are reduced A pplications to unknowns Figure 4 1 0 provides the calibration plot for the standard data that can be used to predict the concentration of PPi from unknown sample. In this two dimensional graph, the black points represent the values of C 1 % calculated from PCA analysis for the standard data se t, whereas the red nonlinear plot represents the function in E quation 4 1 6 The regression function provides the fundamentals for the unknown prediction. In our system, the largest two eigenvectors from PCA are corresponding to the two states of the polyme r, and there is only variable (PPi concentration) that would affect the fluorescence emission spectrum of the polymer as long as the standard experimental conditions are fixed. Most importantly, the variance of [PPi] will only balance the equilibrium betwe en single state polymer and aggregated state polymer. Therefore they will not change the eigenvectors as well as their fundamental matrices ( R 1 and R 2 ) for the sample system but only their loading percentage ( C 1 % and C 2 %); therefore, the

PAGE 107

107 eigenvectors obtai ned from standard sample should be able to be directly applied to the unknown samples under the same experimental condition. To calculate the C 1 and C 2 values for an unknown sample, the fluorescence data matrix for the unknown should be firstly obtained af ter normalization, which is to avoid any inconsistency during experimental operation. Then we apply the fundamental matrices ( R 1 R 2 ) for the eigenvectors from standard data matrix into our unknown matrix according to E quation 4 3 the loading value ( C u,1 and C u,2 ) for each eigenvector can be easily achieved. There are 17 groups of unknown samples that have been tested and their loading percentage for C u,1 is calculated. All the values from unknown samples are put into the calibration plot in Figure 4 1 0 s hown as blue triangles. From the comparison it suggested most of the data from unknown samples locate very close to the calibration plot and consequently the concentration of PPi for each unknown sample can be estimated. Table 4 1 summaries the predicted average [PPi] as well as standard errors for 17 grou ps of unknown samples by using E quation 4 1 6 From the results it reveals that the estimated [PPi] stays close to the real concentrations that most of the errors are ~ 5%. It is be lieved some of the errors are system errors, which are resulted from solution preparation, instrument errors and operation errors. Regarding to so many possible interferences, in overall, PCA analysis provides good precision and accuracy when predicting th e concentrations of PPi.

PAGE 108

108 Figure 4 1 0 [PPi] calibration curve obta ined according to E quation 4 1 6 Standard calibration points from raw data are respented as black squares. Unknown points are resprented as blue triangles Table 4 1. Accuracy of average [PPi] obtained with PCA calibration method. [PPi] real ( M) [PPi] est ( M) % error in [PPi] 80 85 6.25 150 149 0.67 320 291 9.06 360 363 0.83 480 462 3.75 650 609 6.31 675 624 7.56 780 754 3.33 825 780 5.45 850 866 1.88 900 849 5.67 950 916 3.58 1200 1101 8.25 1250 1162 7.04 1400 1333 4.79 1500 1388 7.47 1600 1484 7.25

PAGE 109

109 Summary A CPE on was designed and developed The sensory system is composed of a highly fluorescent conjugated polyelectrolyte ( P3 ) featuring carboxylate side chains and a positively charged polyamine ( N4 ) and the recovery of N4 quenched fluorescence of P3 is highly selective for PPi over other in organic anions such as Pi. The presence of N4 can effectively cross polymers to form large aggregates, leading to fluorescence decrease. F ollow ed by the addition of PPi which exhibits a higher binding affinity to N4 the fl uorescence will be recovered eventually after N4 is released from the polymer aggregates The size change is also monitored by FCS technique, indicating this is an analyte induced aggregation/deaggregation mechanism. The green to blue fluorescence change i s readily visible to the naked eye This approach avoids the use of heavy metal ions which are typically involved in traditional PPi sensing methods, significantly eliminating any risks or pollutions to the environment. Principal component analysis (PCA) has been used to the analysis of fluorescence spectroscopic data of the sensory system as a calibration method to measure the concentration of PPi. Two principal components were obtained by PCA analysis and they have been applied to find out their relation ship with [PPi] by establishing a nonlinear equation via regression analysis. The unknown investigation reveals that PCA calibration affords high accuracy and precision (~5% error) in the predication of [PPi] from unknown

PAGE 110

110 samples. Additionally, PCA calibra tion method can be applied for unknown [PPi] predication across a wide concentration range from lower micromolar to up to several millimolar (10 M 3 mM) significantly broadening its application in different areas with various needs. Most importantly, e xpertise is less required for the quantitative measurement of PPi with the help of computational calculation via Matlab. In overall, this method provides an easy and fast way for PPi detection without requirement of any controls. Even this work shows a lo t of advantages as an efficient qualitative and quantitative tool for PPi detection, it has some limitations. First of all, the PCA analysis is computationally demanding and it is relative time consuming when there is a large set of data involved. Second, the regression equation obtain in our study is only working properly under a specific experimental condition and any changes to the experimental parameters (including substrate concentration, buffers, temperature, etc.) is subject to a consequently change to the regression equation Last, more efforts are needed to improve the detection limit of the system. Experimental Materials All sample solutions were prepared using water which is distilled and purified by using a Millipore purification system (Millipo re Simplicity Ultrapure Water System). HEPES buffer solution ( 10 mM, pH = 7.4) was provided with

PAGE 111

111 4 (2 hydroxyethyl) 1 piperazineethanesulfonic acid and sodium hydroxide. Tris (3 aminoethyl)amine and potassium pyrophosphate were purchased from Sigma Aldrich All chemicals were used as received, unless otherwise noted. Instrumentation Fluorescence spectra were measured and obtained from a spectrofluorometer from Photon Technology International. All the spectra have been corrected by using correction factors generated with a primary standard lamp. FCS measurements are taken on a homemade setup using a 405 nm diode laser (Coherent, CUBE) as the excitation light. 30 nM fluorescein in 10 mM phosphate buffer (pH = 8) is used as the calibration standard for the sys tem.

PAGE 112

112 CHAPTER 5 ACETATE KINASE ASSAY USING POLY (PHENYLENE ETHYNYLENE) WITH POLYAMINE SIDE CHAINS Background Since its discovery by Lipmann in 1944 and isolation from Escherichia coli by Ochoa in 1954, 147 148 acetate kinase ( ACK, EC 2.7.2.1) has been considered as an important enzyme for energy production and one of the earliest phosphoryl transfer enzymes. 149 151 In nature, acetate kinase is widely distributed in both anaerobic and aerobic microbes of the Bacteria and Archaea domains. 152 The enzyme plays an essential role in carbon metabolism as it is ab le to decompose complex organic materials in anaerobic conditions to m ethane by microbial food chains 153 154 Early reports demonstrated that acetate kinase can catalyze the reversible transfer of a phosphate group from acetyl phosphate to ADP and then produce acetate and ATP 155 156 Other substrates including phosphates (Pi) can also function as phosphoryl acceptors to form pyrophosphate (PPi) which is a biologically significant anion involved in many cellular processes. Note that the acetate kinase which particularly uses Pi and acetyl phosphate as substrates are recognized as acetate kinase (diphosphate) ( EC 2.7.2.12 ) in enzymology since A D P or other common nucleoside diphosphat es ca nnot replace Pi as phosphoryl acceptor in the direction of acetate formation. 157 It is well known that acetyl phosphate serves not only as a precursor for many important intermediates in metabolism including acetyl coenzyme A, 149 158 but also

PAGE 113

113 as a potential response regulator in bacte rial signal transduction pathways 159 160 As a result, the phosphorylation reactions by ACK contribute to the regulati on of central reactions in metabolism via acetyl phosphate. Moreover, acetate kinase is reported to be involved i n the process of glycolysis in Aerobacter aerogenes 161 162 T hese bacteria can use acetate as their sole source of carbon and also excrete acetate when grow ing on glucose. The growth of the bacteria l mutant lacking acetate kinase is confirmed to be inhibited by glucose, indicating that the nature enzyme is able to excrete the excess carbohydrate. 163 165 The amphibolic role of acetate kinase in acetate excretion and activation makes this enzyme significantly necessary in microorganism s Several ACK assays that use different substrates in the direction of acetate formation based on biolu minescent and enzyme couple d method s have been developed. 166 168 For the ACK catalyzed system involved with acetyl phosphate and ADP, the most commonly used assay i s to couple ATP formation to the reduction of NADH through hexokinase and glucose 6 phosphate dehydrogenase. 168 T he reaction i s monitored by measuring the changes of NADH emission at 340 nm by exciting at 290 nm. However, f ew papers have reported assa ys f or the other ACK system that generates acetate and P Pi from acetyl phosphate and Pi Although PPi sensing methods have been widely studied, there is only one paper published by Reeves and Guthrie in 1975 that discusses the kinetic investigation f or ACK (diphosphate) activity. 157 The authors used four standard methods for different substrates coupled with several other kinases

PAGE 114

114 and substrates D ue to the complexity of the system this assay suffers several disadvantages including interference and contamination from other substrates as well as issues concern ed wi t h sensiti vity and selectivity. Additionally, it is not possible to perform a real time measurement for ACK using either ADP or Pi as substrates on all the methods discussed above and they are both labor ious and time consuming. We have recently developed cationic conjugated polyelectrolyte (CPE) having bulky branched side chains. 31 This CPE (PPE d NH 3 Figure 5 1 A ) features six ammoni um fun ctional groups on each rep eat unit electrostatic repulsion. 30 The fluorescence intensity of this unaggregated CPE can be efficiently quenched by the addition of PPi, due to the aggregate formation in duced by PPi. The blue to green fluorescence change is readily observed by naked eye and the fluorescence based sensing strategy offers good sensitivity and real time measurement. In addition, the effects of other inorganic anions including Pi on this polymer were also i nvestigated only negligible changes were observed. Th is high selectivity for PPi over Pi indicates that this CPE can be used in biological assays involv ing these two anions. T h is chapter described t he develo pment of the CPE assay for ACK (diphosphate) that require s acetyl phosphate and Pi as substrates. With the introduction of ACK to a solution containing PPE, acetyl phosphate, Pi and Mg 2+ ACK catalyze s the phosph ate transfer from acetyl phosphate to Pi produc ing PPi and a cetate.

PAGE 115

115 The decrease of fluorescence intensity is quantitatively related to the production of PPi concentration as a function of incubation time. The real time measurement allows the determination of kinetic parameters for ACK and the assay exhibits high s ensitiv ity and selectivity comparable to that of other enzyme assays The assay is conducted at physiological pH and offers a straightforward and rapid detection method for ACK activity with the enzyme present in the nanomolar concentration range. Figur e 5 1. ( A ). Structure of PPE d NH 3 and reaction scheme for acetate kinase. ( B ). Mechanism of ACK turn off assay. Results and Discussion Overview of ACK T urn off A ssay A cationic conjugated polyelectrolyte carrying dendritic ammonium side groups, PPE d NH 3 is the signaling element for ACK activ ity As described previously, the introduction of the bulky charged side groups significantly increases electrostatic repulsion between adjacent polymer chains; there by circumventing the aggregation of

PAGE 116

116 PPE d NH 3 in aq ueous solution 31 The cartoon shown in Figure 5 1 B illustrates the turn off mech anism for the ACK turnoff assay, which is based on aggregation of the polymer. In 25 mM 2 (N morpholino) ethanesulfonic acid (MES) bu ffer solution (pH 6.5), PPE d NH 3 exhibits a strong and sharp fluorescence peak at 43 4 nm with a quantum yield o f ~ 0.23 31 After substrates acetyl phosphate (AP) and Pi are added, the fluorescence intensity decrease s only slightly, although the cationic side groups on the polymer are partially neutralized by the anionic substrates. Therefore at this time PPE d NH 3 remains molecularly dissolved in the presence of Pi and Acetyl Pi. In a turn off approach, ACK is ad d ed to the polymer/substrate solution and initiates phosphate transfer from acetyl Pi to Pi and produce s PPi and acetate 157 169 The product PPi is capable of efficiently binding and cross linking adjacent polymers, leading to formation of inter chain polymer aggregates. As a result, the structure d emissio n of the polymer decrease s and shift s to longer wavelength. As the reaction proceeds, the amount of PPi available to cross link the polymer increases, and the fluorescence is further quenched ( the polymer ) Another product, acetate, similar to Pi and acetyle Pi, having only one negative charge is unable to induce interpolymer interactions and th e production of PPi is the only reason for the quenched fluorescence Therefore, the catalyti c activity of ACK is signaled by the fluorescence of PPE d NH 3 state. The decreased fluorescence intensity is associated with the product concentration, [PPi], which increases as a function of time in the enzymatic reaction thereby allowing

PAGE 117

117 quantitative investigatio n o f ACK activity. A divalent metal ion Mg 2+ is required for the ACK activit y because it is reported to be an important cofactor for ACK. 170 172 Fluorescence Q uenching of PPE d NH 3 by P yrophosphate (PPi) The f luorescence response of PPE d NH 3 to Pi, AP, PPi and acetate provides the basis for development of an ACK assay. A series of titrations were carried out using Pi, acetyl Pi, PPi or acetate res pectively ove r a concentration range from 0 to 1 0 M into a MES buffer ( 25 mM, pH = 6.5) solution conta ining 2 M PPE d NH 3 Figure 5 2 is the S tern V olmer plot I 0 /I q of PPE d NH 3 where I 0 and I q are the fluorescence intensit ies of PPE d NH 3 before and after the addition of each component as a function of the concentration of Pi, AP, PPi and acetate, giving a clear demonstration of the fluorescence quench ing efficiency of each species. From Figure 5 2 we can see that the quenching efficienc y decreases in the order of PPi > > AP > Pi = a cetate at the same concentration level, consistent with the decreasing negative charge density on these species. PPi has the largest negative charge density and thus features the strongest electrostatic interaction with the dendritic ammonium groups on the polymer T he rigid structure of PPi enables it to cross link two polymer chains, leading to the formation of inter chain aggregat es Both characteristics contribute to the greatest quenching efficiency of PPi The Stern Volmer constant ( K sv ) for PPi is calculated to be ~1.3 10 5 M 1 The aggregation mechanism wa s also proved by fluorescence correlation spectr oscopy (FCS). 173 As described above, AP, Pi and acetate exhibit very weak

PAGE 118

118 quenching effect s on PPE d NH 3 even t hough the negative charges can partially neutralize the positive charges on the polymer, they are unable to induce efficient interpolym er interactions. Therefore the fluorescence decrease of the experiment is associated with the production of PPi B y observing the fluorescence change we can monitor the enzyme activity in the presence of low substrate concentrations. Note this assay is n ot appropriate for high substrate concentrations (>1 mM), since the large concentration of negatively charged substrates will have a considerable effect on the aggregation of PPE d NH 3 Figure 5 2. Stern Volmer plot for PPE d NH 3 upon titration with Pi, acetate, PPi and AP, respectively. Solution conditions: 2 M PPE d NH 3 and 1 mM MgCl 2 in 25 mM MES (pH = 6.5) buffer, ex = 404 nm.

PAGE 119

119 ACK T urn off A ssay In order to eliminate possible interference s, the effect of addition of ACK on the fluorescence of PPE d NH 3 wa s examined. Figure 5 3 demonstrate s that there is negligible change in polymer fluorescence intensity with the addition of [ACK] up to 20 g/ mL, thereby rul ing out the possibility that the fluorescence change of the assay is caused by the added enzyme. Figure 5 3. Fluorescence spectroscopic change upon the addition of ACK enzyme over a concentration range from 0 to 20 g/ mL. Solution conditions: 2 M PPE d NH 3 and 1 mM MgCl 2 in 25 mM MES (pH = 6.5) buffer, ex = 404 nm. In a mixture solution containing PPE d NH 3 AP, Pi and the cofactor MgCl 2 the introduction of ACK initiate s phosphate transfer from AP to Pi to generate PPi and acetate as the products (see chemical reaction in Figure 5 1A) The enzymatic activity is monit ored by the continuous decrease of the polymer fluorescence intensity. Figure 5 4 A presents the fluores cence spectroscopic changes observed for the standard assay

PAGE 120

120 solution at 30 o C every 1 min for 10 min. The initial polymer fluorescence spectrum is well shaped with a strong structure d emission at ~434 nm. Following addition of 5 g/m L ACK the enzymatic reaction is activated and the product PPi act s as a cross linker to form inter chain aggregates. Therefore as the incubation time incre ases, the structure d band decrease s significantly as the broad and structureless band at longer wavelength ( ~520 nm ) becomes more prominent consistent with PPi induced aggregation. 22 71 33 Figure 5 4 ( A ) Fluorescence spectroscopic changes as a function of time after addition of 5 g/mL ACK in a 10 min period. ( B ) Change of fluorescence intensity at 434 nm recorded every 30 s during the real time ACK turn off assay with various ACK concentrations (0 10 g/mL) Solution conditions: 2 PPE d NH 3 1 mM MgCl 2 1 00 Pi and 1 0 AP in 25 mM MES (pH = 6.5) buffer at 30 C, ex = 404 nm In order to test the feasibility of using this PPE d NH 3 / AP/Pi system as a real time turn off assay for ACK activit y a series of ACK catalyzed assays w as carried out as a function of time at varying ACK concentration s at standard assay conditions ( 10 M AP, 100 M Pi and 1 mM Mg 2+ at 30 o C ) Figu re 5 4 B shows plots of the continuous decrease of fluorescence intensity at 434 nm for solutions with five different

PAGE 121

121 concentrations of ACK from 0 to 10 g/m L where the fluorescence intensity was measured at 30 s intervals. I n a 10 min measurement period, it is clearly noted that the initial reaction rate increases with [ACK] The fluorescence intensity decreases linearly during the initial period (first 2 min) and then it slows down as the reaction co ntinues. This effect is more significant at higher [ACK] and is believed to be caused by the reaction s coming to equilibrium. Note that the fluorescence decrease of the control experiment (black curve) is caused by photobleaching. Under standard experimental conditions, the reaction favors the forward reaction which produces PPi, but formation of PPi also initialize s the reverse react ion and slows the rate of PPi production. In a series of investigations, we examine d the effect of substrate concentration (AP and Pi) on the reaction rate. Figure 5 5 demonstrates the fluorescence decrease at varying substrate concentrations with Pi from 50 to 300 and AP from 5 to 30 T he reaction rate increases with increasing [substrate]. At higher substrate concentration ( Figure 5 6 C & D), the reaction reached equilibrium in less tha n 2 min and more than 90% of the initial fluorescence intensity was quench ed indicating that more PPi was produced at higher substrate concentrations. Figure 5 6 illust rates a plot of initial reaction rate ( 0 ) as a function of ACK concentrations. Within 1 min after the reaction begins, 0 is directly proportional to [ACK] in the range of 0 0.8 g/mL, suggesting that the enzymatic reaction is kinetically controlled by ACK in the initial stage of reaction. It is expected fro m the plot that the detectio n limit of this assay is ~ 0 .05

PAGE 122

122 g/mL using the following equation : D.L. = 3 bk /m ( bk standard deviation of the blank; m, slope of the curve ) Figure 5 5 Time based fluorescence intensity recorded every 30 s during the real time measurement of ACK turn off assay in a 1 0 min period at various [ACK] with dif ferent substrate concentrations ; (A). [Pi] = 50 (B). [Pi] = 100 (C). [Pi] = 200 (D). [Pi] = 300 Different colors indicate different [ACK], 0 (black), 1 (red), 3 (blue), 5 (pink) and 10 (green) g/m L F igure 5 6. Dependence of initial rate of reaction ( 0 ) on ACK concentrations. Solution conditions: 2 PPE d NH 3 1 mM MgCl 2 1 00 Pi and 1 0 AP in 25 mM MES (pH = 6.5) buffer at 30 C, ex = 404 nm, em = 434 nm.

PAGE 123

123 Determination of ACK c atalyzed k inetic p arameters The kinetic parameters ( K m and V max ) for the ACK catalyzed reaction can be obtained from th e CPE based fluorescence turn off assay. As described in the experimental section, in a bi substrate enzymatic reaction, the standard approach to study the kinetic behavior is to measure the rate of product formation whil e varying the concentration s of both substrates. For example, we first varied the AP concentration from 2.5 to 25 at different level of Pi (100, 150, 200 and 250 respectively ) and all the assays were conducted with 0.2 g/mL ACK in 25 mM MES ( pH 6.5 ) at 30 C The initial rate ( 0 ) for each experiment was calculated. N ext, a plot of 1/ 0 vs 1/[Pi] was prepared and fitted to the modified Michaelis Menten equation (see E quation 5 5 in the experimental section) as illustrated in Figure 5 7 A The derived intercepts for each linear fitting were plotted vs 1/[Pi] and another linear plot was obtained as shown in Figure 5 7 B According to E quation 5 6 (see experimental section) K m for AP were calculated with values of 135 In a similar way, Figure 5 7 C & D are used to calculate the K m for Pi with values of 625 And V max was calculated to be 0. 617 0.089 M/sec Compared to report ed apparent K m values for AP (~60 ) and Pi ( ~ 2.2 mM), 157 the values obtained from our approach are in good agreement with the reported data.

PAGE 124

124 Figure 5 7. Enzyme kinetics parameters measurement by using ACK turn off assay. ( A ): Double reciprocal plot of 1/initial rate vs. 1/[AP] at several different concentrations of Pi. The [AP] is varied from 2.5 to 25 ( B ): Secondary replot of intercept vs. 1/[Pi] ( C ): Double reciprocal plot of 1/initial rate vs. 1/[Pi] at several different concentrations of AP. The [Pi] is varied from 25 to 200 ( D ): Secondary replot of intercept vs. 1/[A P] Effect of Mg 2+ on ACK a ctivity in t urn off a ssay It has been reported that Mg 2+ is an important cofactor in the catalyzed ACK transfer of a phosphate group from AP to Pi in the acetate kinase assay. 172 174 Therefore the enzyme activities in both presence and absence of Mg 2+ are presented in Figure 5 8 T he red curve is the control experiment without adding kinase It clearly shows that there is almost no change of the polymer fluorescence for the control experiment The slight decrease of the fluorescence in the time period is probably caused by photobleaching of

PAGE 125

125 PPE under constant illumination 175 176 The bl ue curve illustrate s the fluorescence change in a complete assay system with both [Mg 2+ ] at 1 mM and [ACK] at 5 g/mL. As previous ly described the addition of ACK activate s the phosphate transfer from AP to produce PPi resulting in a decrea se in fluorescence intensity, with n early 90% of the fluorescence quenched in 10 min In comparison, the black curve shows how the PPE d NH 3 fluorescence behave s without adding Mg 2+ Even in the presence of 5 g/mL ACK, no efficient quenching process happen s, with results qu ite similar to the control experiment (< 25% decrease of the fluore scence in 10 minutes ) From these results, we ca n conclude that the presence of Mg 2+ play s an essential rol e in the acetate kinase assay. Figure 5 8 Effect of confactor MgCl 2 on the enzymatic activit y of ACK [Polymer] = 2 Red curve : [MgCl 2 ] = 1 mM No ACK added. B lack curve : [ACK] = 5 g/mL no MgCl 2 added. Blue curve : [ACK] = 5 g/mL and [MgCl 2 ] = 1 mM

PAGE 126

126 Specificity of ACK t urn off a ssay The specificity of the ACK turn off assay wa s also tested in order to prove that it is an effective and selective biosensor for the target enzym e, and a number of experiments we re conducted using other proteins including peptidase from porcine intestinal mucosa ( PTD ) hexokinase from saccharomyces cerevisiae ( H O K ) phospholipase D from peanut ( PLD ) glucose oxidase from Aspergillus niger ( GOX ) and peroxidase from horseradish ( POD ) as controls. None of these proteins ha s any specific interactions with PPi. ACK along with these control proteins were incubated with PPE d NH 3 und er the same standard conditions (1 00 Pi and 1 0 AP in the 25 mM MES buffer at 30 o C ) In each assay, the protein concentration wa s controlled at 0.2 g/m L. In a 60 min period of incubation, the fluorescence intensity wa s recorded every 30 second at 434 nm and the final fluorescence change is illustrated and compared in Figure 5 9 A O verall, the assay containing ACK show ed a ~ 90% decrease in fluorescence intensity in one hour, while all the control proteins showed no sign ificant change with fluorescence decrease in the range between 13% and 25% Figure 5 9B provides the detailed information of how the fluorescence intensity varied with time. Over the 1 hour period, the fluorescence intensity decreased very rapidly after A CK wa s added. In contrast, the addition of other proteins induced a n initial decrease in the first 5 min followed by a very slow decrease in the fluorescence intensity. We attribute this phenomenon to non specific binding between proteins and any species in the solution. 177 It is believed that the introduction of

PAGE 127

127 these proteins disturb the electrostatic balance which may affect the fluorescence intensity. The fluorescence decrease in the initial 5 min caused by non specific interaction s for the control proteins accounted for most of the overall fluorescence change. Regardless of the slight fluorescence decrease by non spe cific binding, it is clear that the fluorescence change for the other protein s is negligible when compared to that for ACK. As a result, we are confident to conclude that this turn off assay is specific for ACK and most of the fluorescence decrease is due to the ACK enzyme activi ty with only a small portion aris ing from non specific interactions. Figure 5 9 Specificity of ACK turn off assay. ( A ) Changes in fluorescence intensity at 434 nm after 60 min of incubation with six different proteins ( ACK, PTD, H O K, PLD, GOX and POD ) at 30 o C ( B ) Time based fluorescence measurements for these proteins. All protein concentrations are set at 0.2 g/ mL Standard assay condition: 2 PPE, 1 mM MgCl 2 200 Pi and 20 AP in 25 mM MES (pH = 6.5) buffer ex = 404 nm. Discussion T his chapter described a novel CPE based fluorescence turn off assay to monitor ACK catalyzed enzymatic activit y in the direction of PPi production To the best of our

PAGE 128

128 knowledge, this is the first report of using conjugated polyelectro lytes for ACK real time measurement s Compared to the traditional enzyme coupled sensing methods, the CPE method eliminates the need for additional enzymes and therefore reduces the possibility of contamination and complexity Mostly importantly, the assa y can be carried out with substrate concentration s as low as micromolar range which are much lower than those of other reported ACK assay conditions, and the K m values obtained for both substrates are close to the reported values. In addition the CPE method offers superior sensitivity in terms of enzyme concentrations and exhibits selectivity over other enzymes. Finally, this fluorescence based assay can be easily adapted to a high throughput screening format O verall this real time CPEs based sensing system affords an easy and rapid method for ACK detection. While this CPE based turn off approach provides many advantages, it also presents some challenges. First, photobleaching of the CPEs and non specific interaction s between the polymer an d other proteins may cause systematic errors Second, the assay is not ideal at high substrate concentrations. Even though low substrate concentration s ha ve little effect on fluorescence of the polymer ; the presence of high concentration s of negatively cha rged substrates will quench the fluorescence intensity, leading to systematic errors Last, the inhibition study is still inefficient and more inhibition stud ies needed to be tested in the future.

PAGE 129

129 Summary This chapter describe d the design and development of a novel fluorescence turn off assay for ACK enzym e activity which offers easy, rapid and real time measurement s with high sensitivity and selectivity. Based on the interchain aggregation of PPE induced by the product PPi, the CPE based system is utilized as a turn off sensor for ACK. Th e assay allows for determination of kinetic parameters for both substrates and it operates with substrates in the micromolar range with relatively go od selectivity. Since this work demonstrate s high specificity for PPi over Pi, this method can be extended to other substrate/enzyme systems that involve the discrimination of PPi and Pi. For example, in a recent work published by our group, an anionic PPE CO 2 coupled with Cu 2+ using PPi as the sub strates was applied to study the enzymatic activity of alkaline phosphatase (ALP). However, our PPE system does not require addition of a heavy metal ion ( Cu 2+ ) and affords similar speed, sensitivity and selectivity. We believe this system will be promisi ng and useful as a signal transducer for other enzymatic reactions. Experimental Materials The synthesis of PPE d NH 3 is described in the literature 31 The aqueous stock solution of PPE d NH 3 was diluted with M aqueous solution s w ere prepared with water that was pre distilled and then purified by

PAGE 130

130 using a Millipore purification system. MES buffer solution (25 mM, pH 6.5) was prepared with 2 (N morpholino) ethanesulfon ic acid and sodium hydroxide. Lithium potassium acetyl phosphate was obtained from Sigma Aldrich. Magnesium chloride, sodium phosphate tribasic and sodium acetate were purchased from Fisher Chemical. Sodium pyrophosphate was purchased from J.T.Baker Chemic al Co. Acetate kinase was prepared and provided by Dr. Cheryl Ingram Smith at Clemson University. The enzyme stock solution was prepared immediately before use in the fluorescence assays. General M ethods Fluorescence t urn off a ssay p rocedure The enzyme assays were conducted in 25 mM MES buffer ( pH 6.5) at 3 0 o C For real time assays, the fluorescence intensity was recorded with excitation and em ission wavelengths of 404 and 43 4 nm, respectively. A standard assay procedure was carried out as f ollows: a 2.0 mL aliquot of a solution containing PP E d NH 3 AP Pi, and Mg 2+ was mixed and allowed to reach equilibrium at 3 0 o C and t he initial polymer fluorescence intensity ( I b0 ) at 4 34 nm was recorded. In order to avoid interference caused by polymer photobleaching, the solution was placed in the spectrometer and the fluorescence intensity change as a function ( I bt ) of time was measured every 10 s as a blank control. At the sam e time, a second 2.0 mL aliquot of fresh PPE/AP/Pi/ Mg 2+ solution was prepared and incubated at 3 0 o C Then ACK was quickly added via a microliter pipet, and the fluorescence intensity ( I t ) of the solution was measured at 10 s increments. The

PAGE 131

131 fluorescence intensity for both blank and sample solutions was record ed under the same conditions. T he sample fluorescence intensity was corrected to I t c using blank solution according to Equation 5 1: 178 (5 1) To monitor the assay in an end point format, the fluorescence intensity versus waveleng th spectra were recorded with the excitation wavelength at 404 nm. Calculation of initial reaction rate ( 0 ) To obtain the relationship between fluorescence change and product PPi concentration, t he corrected fluorescence intensity was converted into [PPi] as function of time using E quation 5 2 which is derived from a calibration plot obtained from a standard sample solution with known concentration of PPi : (5 2) where [PPi] 0 is the known PPi concentration of the standard sample, [PPi] t is the product concentration at time t I p is the fluorescence intensit y of PP E d NH 3 before addition of substrate, I 0c is the initial corrected fluorescence intensity at t = 0, that is, the flu orescence intensity after addition of substrate but before the addition of enzyme and I tc is the corrected fluorescence intensity at time t after enzyme addition. A plot of [PPi] t vs t was then derived and 0 was calculated from the slope by using the data where [PPi] is a linear function of time.

PAGE 132

132 Calculation of k inetic p aram e ters The Michaelis Menten equation is generally applied to calculate the kinetic parameters and is expressed as (5 3) w here [S] 0 is the initial substrate concentration, V max is the maximum rate of the enzyme catalyzed reaction at saturation substrate concentration, and K m is the Michaelis constant However, our enzyme system involves two substrates Pi and AP and is more complicated than a single substrate system. Therefore a modified approach based on the Michaelis Menten equation for a bisubstrate enzyme reaction was used 179 (5 4) In Equation 5 4 K ia is the intrinsic dissociation constant for substrate A and K a and K b are the Michaelis constants for substrate A and B respectively. In double reciprocal form, the rate equation becomes Equation 5 5 which can be analyzed via suitable linear plots, (5 5) For example, when [ A ] is varied at constant [ B ] the equation becomes ( 5 6 ) w here a double reciprocal plot of 1/ 0 vs 1/[ A ] h as the form of y = m x + b in which the slope term [( K ia K b / V max [ B ]+ K b / V max ] and intercept term [ K a / V max [B]+1/ V max ] show a

PAGE 133

133 dependency on the concentration of substrate B T he slope and intercept values are both in the form of linear equation s and can thus be plotted to give values corresponding to V max and Michaelis constants To evaluate K a and V max a seri es of Equation s 2 5 plots are prepared for different constant [ B ], and this intercept b follow E quation 5 7 ( 5 7 ) Thus, by plotting b vs. 1/[ B ], we can obtain a secondary plot with intercept equals to 1/ V max and slope corresponding to to K a /V max Thus, two kinetic parameters of interest, V max and K a can be evaluated by implementing this approach. Defining A as Pi and B as AP, first vary Pi at different constant levels of AP to obtain a series of plots based on Equation 5 6. Then, the intercepts are plotted vs 1/[ B ] to obtain V max and K a for Pi. T h en the same procedure is preformed varying AP for different constant levels of Pi to obtain V max and K b for AP.

PAGE 134

134 CHAPTER 6 CONCLUSION In conclusion, the design and development of optical biosensors using functionalized water soluble conjugated polyelectrolytes (CPEs) have been presented and discussed By taking advantage of analyte induced aggregation mechanism, all three CPE based sensory systems exhibit ed good sensitivi ty and selectivit y. Both fluorescence spectroscopy and fluorescence correlation spectroscopy (FCS) were involved to investigate the aggregation behaviors of C PEs in the prese nces of different target molecules These sensory systems provide d rapid and convenient ways for detection of metal ions pyrophosphate ions (PPi) and enzyme activity Ion Induced Aggregation of Conjugated Polyelectrolytes Studied by Fluorescence Correlation Spectroscopy While the mechanisms for fluoresc ence quenching of CPEs are widely studied and elucidated in the literatures, detailed information of how different types of quenchers interact with CPEs is still insufficient. Therefore fluorescence correlation spectroscopy (FCS) wa s introduced to study the physical state change s of CPEs upon the addition of different ions Two examples were described. First, the interaction of an anionic CPE PPE d CO 2 with a variety of metal ions including Na + K + Ca 2+ Cu 2+ Fe 2+ and Fe 3+ was investigated. Correlation c urves for each CPE/metal ion mixtures at different [metal ion] were depicted and compared. It was found that the diffusion time increase d in the order

PAGE 135

135 of K + Na + < Ca 2+ < Cu 2+ < Fe 2+ < Fe 3+ ; and large diffusion times usually indicate large CPE aggregates. The results suggested that monovalent metal ions bond ed to the CPE chain but can not induce aggregation and divalent metal ions (Ca 2+ Cu 2+ ) were able to bind and induce th e formation of small aggregates ; while for iron catioins (Fe 2+ and Fe 3+ ) they can bind strongly and induce the fo rmation of large CPE aggregates The outstanding abilities of iron to induce aggregation was attributed to the propensity of these metals to bind to six ligands, and thus bridge two of the tri acid units on two PPE d CO 2 chains. Then the diffusion time ratio plot from FCS wa s compared with Stern Volmer plot from fluorescence spectroscopy, and the slow response of FCS plot explain ed the effect s of amplified quenching in a new aspect We also investigated the interact ion of PPE d NH 3 with PPi and the solid evident supported large aggregates formed when PPi was titrated into PPE d NH 3 solution. Principal Component Analysis for Pyrophosphate Sensors Using Conjugated Polyelectrolytes In t his chapter a novel CPE based turn on sensor for PPi was designed and applied to detect pyrophosphate concentration with the help of principal component analysis Polymer P3 exists as single chains in both methanol and water because of the novel carboxylate methylene side groups Upon the addition of tris(3 aminoethyl)amine, very strong and large aggregates of P3 were formed, driven by both electrostatic interaction and hydrophobic interaction. Then the a ddition of PPi will take the tetra amine

PAGE 136

136 away and release the polymer chains becaus e of the highly b ind affinity between PPi and t etra amine resulting in a fluorescence recover y Since the system wa s involv ed with several different substrates and the fluorescence spectroscopic spectrums were also relatively complex with one peak going up and the other band go down, principle component analysis (PCA) method wa s used to analyze this system. Two principle components were obtain ed as well as their loading percentage at each [PPi] concentration. The established equation obtained from regression analysis can be applied to calculate the unknown [PPi] in such a system without a reference. And it has been proved that the sensor together with PCA method is able to predict the unknown concentration for PPi with a ~ 95% accuracy when PPi is within concentration range from 100 M t o 3 mM Acetate Kinase Assay Using Poly (phenylene ethynylene) with Polyamine Side Chains As previous described in C hapter 2 t he fluorescence of a cationic poly(phenylene ethyn ylene) polymer ( PPE d NH 3 ) with branched side chains can be quenched very efficiently by pyrophosphate (PPi) The quenching occurs because PPi can effectively cross link PPE to form large aggregates O ther substrates includ ing Pi, Acetyl phosphate and acetate fail to induce any significant fluorescence change. Therefore, a real time fluorescence turn off assay for the enzyme acetate kinase (ACK) using PPE d NH 3 as the signal transducer wa s developed. ACK will initiate the phosph ate

PAGE 137

137 transfer from ac etyl Pi to Pi and produc e PPi and acetate T he fluorescence of the polymer PPE d NH 3 will decrease gradually after ACK is introduced into the Acetyl Pi/Pi system. The assay operates with substrates in the micromolar range, and it offers a straightforward and rapid detection of A CK activity with the enzyme present in th e nanomolar concentration range The effect of Mg 2+ on the ACK catalytic activity wa s also investigated and it was proved that Mg 2+ wa s an important activator for the ACK de phosphorylation A modified approach based on Michaelis Menten equation wa s used to study the kinetic parameters for this bi su bstrate system, from which the equilibrium constants K m for Pi and Acetyl Pi were obtained Besides, the highly selectivity of this system was proved by the comparison with other enzymes.

PAGE 138

138 APPENDIX MATLAB CODING FOR PCA load CheerSpectraAmineAll.txt; D1=CheerSpectraAmineAll(2:317,11:37); Concentration=CheerSpectraAmineAll(1,11:37); Wavelength=CheerSpectraAmineAll(2:317,1); [m,n]=size(D1); O=zeros(1,n); O=sum(D1); for g=1:n N(g)=(1/O(g)*O(g)); end for a=1:n for l=1:m D(l,a)=D1(l,a)*N(a); end end D; Lambda1=0; RR=zeros(n,n); Z=D'*D; CC=zeros(m,n); Rc=zeros(m,n) ; HH=zeros(m,n); e=1; Vec=0; p=1; while e > 0 if p > 1 Z=RR; end RR=Z Lambda1*Vec*Vec'; [Q,Lambda]=eig(RR); Lambda1=max(max(Lambda)); if p ==1 maxii=Lambda1; end for y=1:n for t=1:y if Lambda1 == Lambda(t,y); x=y;

PAGE 139

139 end end end Vec=Q(:,x); if Lambda1 > maxii*0.001 HH=D*Q; Rc(:,p)=HH(:,x); CC(p,:)=Vec'; else e=0; end p=p+1; end R=z eros(m,p 2); C=zeros(p 2,n); for aa=1:p 2 R(:,aa)=Rc(:,aa); C(aa,:)=CC(aa,:); end R; C; U=0; O=0; N=zeros(1,n); U=D.*D; O=sum(U); for g=1:n N(g)=sqrt(1/O(g)*O(g)); end for a=1:n for l=1:m Dn(l,a)=D(l,a)*N(a); end end Dn; R1=R; C1=C; for phii=295:295 phi=phii*pi/180; T=[1.1*cos(phi) 1.8*sin(phi); 1.9*sin(phi) 6.5*cos(phi)]; R=R1*T;

PAGE 140

140 C=inv(T)*C1; figure(1) plot(Wavelength,D) title('Original Spectra') xlabel('Wavelength (nm)') ylabel('Intensity (a.u.)') figure(2) clf plot (Wavelength,R*C) title('Reconstructed Spectra') xlabel('Wavelength (nm)') ylabel('Intensity (a.u.)') figure(3) plot(Wavelength,R) title('Fundamental Spectra') xlabel('Wavelength (nm)') ylabel('Intensity (a.u.)') grid on figure(4) plot(Concentration,C(1,:),'bo ',Concentration,C(2,:),'g+ ',Concentration,C(1,:)+C(2,:),'rx ') title('Loadings vs. actual') xlabel('Concentration') ylabel('Actual Loadings'); F=C'; if abs(F(1,1)) abs(F(2,1)) < 0.002 a=1; b=2; else a=2; b=1; end phii; pause(0.5) end

PAGE 141

141 LIST OF REFERENCE S (1) Gunes, S.; Neugebauer, H.; Sariciftci, N. S. Chem Rev 2007 107 1324. (2) Kraft, A.; Grimsdale, A. C.; Holmes, A. B. Angew Chem Int Edit 1998 37 402. (3) Montali, A.; Smith, P.; Weder, C. Synthetic Met 1998 97 123. (4) Torsi, L.; Dodabalapur, A.; Rothberg, L. J.; Fung, A. W. P.; Katz, H. E. Science 1996 272 1462. (5) Sirringhaus, H. Adv Mater 2005 17 2411. (6) McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem Rev 2000 100 2537. (7) Thomas, S. W.; Joly, G. D.; Swager, T. M. Chem Rev 2007 107 1339. (8) Gierschner, J.; Cornil, J.; Egelhaaf, H. J. Adv Mater 2007 19 173. (9) Nguyen, T. Q.; Wu, J. J.; Doan, V.; Schwartz, B. J.; Tolbert, S. H. Sci ence 2000 288 652. (10) Pinto, M. R.; Schanze, K. S. Synthesis Stuttgart 2002 1293. (11) Moroni, M.; Lemoigne, J.; Luzzati, S. Macromolecules 1994 27 562. (12) Yang, J. S.; Swager, T. M. J Am Chem Soc 1998 120 11864. (13) Yang, J. S.; Swager, T. M. J Am Chem Soc 1998 120 5321. (14) Schanze, K. S.; DiCesare, N.; Pinto, M. R.; Lakowicz, J. R. Langmuir 2002 18 7785. (15) Wilson, J. S.; Frampton, M. J.; Michels, J. J.; Sardone, L.; Marletta, G.; Friend, R. H.; Samori, P.; Anderson, H. L.; Cacialli, F. Adv Mater 2005 17 2659. (16) Barata, P. D.; Costa, A. I.; Ferreira, L. F. V.; Prata, J. V. J Polym Sci Pol Chem 2010 48 5040. (17) Tour, J. M. Chem Rev 1996 96 537. (18) Bunz, U. H. F.; Kim, I. B.; Dunkhorst, A.; Gil bert, J. Macromolecules 2005 38 4560.

PAGE 142

142 (19) Wu, D. L.; Feng, F. D.; Xie, D. P.; Chen, Y.; Tan, W. H.; Schanze, K. S. J Phys Chem Lett 2012 3 1711. (20) Nelson, J. C.; Saven, J. G.; Moore, J. S.; Wolynes, P. G. Science 1997 277 1793. (21) Prince, R. B.; Saven, J. G.; Wolynes, P. G.; Moore, J. S. J Am Chem Soc 1999 121 3114. (22) Tan, C. Y.; Pinto, M. R.; Schanze, K. S. Chem Commun 2002 446. (23) Tan, C. Y.; Alas, E.; Muller, J. G.; Pinto, M. R.; Kleiman, V. D.; Schanze, K. S. J Am Chem Soc 2004 126 13685. (24) Halkyard, C. E.; Rampey, M. E.; Kloppenburg, L.; Studer Martinez, S. L.; Bunz, U. H. F. Macromolecules 1998 31 8655. (25) Kim, J.; Swager, M. Nature 2001 413 548. (26) Levitus, M.; Zepeda, G.; Dang, H.; Godinez, C.; Khuong, T. A. V.; Schmieder, K.; Garcia Garibay, M. A. J Org Chem 2001 66 3188. (27) Walters, K. A.; Ley, K. D.; Schanze, K. S. Langmuir 1999 15 5676. (28) Rothberg, L. J.; Yan, M.; Papadimitrakopoulos, F.; Galvin, M. E.; Kwock, E. W.; Miller, T. M. Synthetic Met 1996 80 41. (29) Schanze, K. S.; Tan, C. Y.; Pinto, M. R. Chem Commun 2002 446. (30) Lee, S. H.; Komurlu, S.; Zhao, X. Y.; Jiang, H.; Moriena, G.; Kleiman, V. D.; Schanze, K. S. Macromolecules 2011 44 4742. (31) Zhao, X. Y.; Schanze, K. S. Chem Com mun 2010 46 6075. (32) Muller, J. G.; Atas, E.; Tan, C.; Schanze, K. S.; Kleiman, V. D. J Am Chem Soc 2006 128 4007. (33) Jiang, H.; Zhao, X. Y.; Schanze, K. S. Langmuir 2006 22 5541. (34) Reinert, S.; Mohr, G. J. Chem Commun 2008 2272. (35) Sun, H.; Feng, F.; Yu, M. H.; Wang, S. Macromol Rapid Comm 2007 28 1905.

PAGE 143

143 (36) MacKnight, W. J.; Ponomarenko, E. A.; Tirrell, D. A. Accounts Chem Res 1998 31 781. (37) Kabanov, A. V.; Bronich, T. K.; Kabanov, V. A.; Yu, K.; Eisenberg, A. J Am Chem Soc 1998 120 9941. (38) Chen, L. H.; McBranch, D.; Wang, R.; Whitten, D. Chem Phys Lett 2000 330 27. (39) Zhou, Q.; Swager, T. M. J Am Chem Soc 1995 117 12593. (40) Whitten, D. G.; Chen, L. H.; McBranch, D. W.; Wang, H. L.; Helgeson, R.; Wudl, F. P N atl Acad Sci USA 1999 96 12287. (41) Gaylord, B. S.; Wang, S. J.; Heeger, A. J.; Bazan, G. C. J Am Chem Soc 2001 123 6417. (42) Zhao, X. Y.; Pinto, M. R.; Hardison, L. M.; Mwaura, J.; Muller, J.; Jiang, H.; Witker, D.; Kleiman, V. D.; Reynolds, J. R. ; Schanze, K. S. Macromolecules 2006 39 6355. (43) Fan, C. H.; Hirasa, T.; Plaxco, K. W.; Heeger, A. J. Langmuir 2003 19 3554. (44) Jiang, H.; Zhao, X. Y.; Schanze, K. S. Langmuir 2007 23 9481. (45) Magde, D.; Webb, W. W.; Elson, E. Phys Rev Lett 1972 29 705. (46) Krichevsky, O.; Bonnet, G. Reports on Progress in Physics 2002 65 251. (47) Yin, Y. D.; Yuan, R. F.; Zhao, X. S. J Phys Chem Lett 2013 4 304. (48) Shin, H. S.; Okamoto, A.; Sako, Y.; Kim, S. W.; Kim, S. Y.; Pack, C. G. J Phys Che m A 2013 117 27. (49) Zettl, U.; Hoffmann, S. T.; Koberling, F.; Krausch, G.; Enderlein, J.; Harnau, L.; Ballauff, M. Macromolecules 2009 42 9537. (50) Fogarty, K.; Van Orden, A. Methods 2009 47 151. (51) Waldeck, D. H.; Kaur, P.; Yue, H. J.; Wu, M. Y.; Liu, M.; Treece, J.; Xue, C. H.; Liu, H. Y. J Phys Chem B 2007 111 8589. (52) Haustein, E.; Schwille, P. Annu Rev Bioph Biom 2007 36 151.

PAGE 144

144 (53) Waldeck, D. H.; Yue, H. J.; Wu, M. Y.; Xue, C. H.; Velayudham, S.; Liu, H. Y. J Phys Chem B 2008 11 2 8218. (54) Chen, Y.; Munteanu, A. C.; Huang, Y. F.; Phillips, J.; Zhu, Z.; Mavros, M.; Tan, W. H. Chem Eur J 2009 15 5327. (55) Humpolickova, J.; Benda, A.; Beranova, L.; Hof, M. Chem Listy 2009 103 911. (56) Pal, N.; Verma, S. D.; Singh, M. K.; Sen, S. Anal Chem 2011 83 7736. (57) Satrijo, A.; Swager, T. M. J Am Chem Soc 2007 129 16020. (58) Selkoe, D. J. Nature 2003 426 900. (59) Aebersold, R.; Mann, M. Nature 2003 422 198. (60) Lee, J. H.; Kim, D. G.; Kwon, N. Y.; Jang, G. S.; Son, J. H.; Lee, M.; Cho, H. J.; Kweon, H. S.; Lee, T. S. J Polym Sci Pol Chem 2011 49 138. (61) Nyren, P. Anal Biochem 1987 167 235. (62) Blagbrough, I. S.; Adjimatera, N.; Kral, T.; Hof, M. Pharm Re s 2006 23 1564. (63) Bazan, G. C.; Li, H.; Yang, R. Macromolecules 2008 41 1531. (64) Gaylord, B. S.; Heeger, A. J.; Liu, B.; Wang, S.; Fan, C. H.; Bazan, G. C. Abstr Pap Am Chem S 2003 225 U985. (65) Jin, Y.; Yang, R.; Suh, H.; Woo, H. Y. Macromo l Rapid Comm 2008 29 1398. (66) Andrew, T. L.; Swager, T. M. Macromolecules 2008 41 8306. (67) Wang, D. L.; Gong, X.; Heeger, P. S.; Rininsland, F.; Bazan, G. C.; Heeger, A. J. P Natl Acad Sci USA 2002 99 49. (68) Kim, I. B.; Dunkhorst, A.; Bunz, U. H. F. Langmuir 2005 21 7985. (69) Swager, T. M.; Thomas, S. W.; Joly, G. D. Chem Rev 2007 107 1339. (70) Kim, I. B.; Bunz, U. H. F. J Am Chem Soc 2006 128 2818.

PAGE 145

145 (71) Schanze, K. S.; Liu, Y.; Ogawa, K. J Photoch Photobio C 2009 10 173. (72) Holdcroft, S. J Polym Sci Pol Phys 1991 29 1585. (73) Balgi, G.; Leckband, D. E.; Nitsche, J. M. Biophys J 1995 68 2251. (74) Chiantia, S.; Kahya, N.; Schwille, P. Langmuir 2007 23 7659. (75) Sasmal, D. K.; Mondal, T.; Sen Mojumdar, S.; Choudhury, A.; Banerjee, R.; Bhattacharyya, K. J Phys Chem B 2011 115 13075. (76) Chen, L. H.; McBranch, D. W.; Wang, H. L.; Helgeson, R.; Wudl, F.; Whitten, D. G. Proceedings of the National Academy of Sciences of the United State s of America 1999 96 12287. (77) Zhao, X. Y.; Jiang, H.; Schanze, K. S. Abstr Pap Am Chem S 2007 233 (78) Smith, T. A.; Ramachandran, G.; Gomez, D.; Ghiggino, K. P. Synthetic Metals 2005 152 17. (79) Wang, D. L.; Wang, J.; Moses, D.; Bazan, G. C.; Heeger, A. J. Langmuir 2001 17 1262. (80) Wang, J.; Wang, D. L.; Miller, E. K.; Moses, D.; Bazan, G. C.; Heeger, A. J. Macromolecules 2000 33 5153. (81) Shalaby, A. R. Food Res Int 1996 29 675. (82) Bibillo, A.; Figlerowicz, M.; Kierzek, R. Nucleic Acids Res 1999 27 3931. (83) Kusano, T.; Yamaguchi, K.; Berberich, T.; Takahashi, Y. J Plant Res 2007 120 345. (84) Larque, E.; Sabater Molina, M.; Zamora, S. Nutrition 2007 23 87. (85) Groppa, M. D.; Benavides, M. P. Amino Acids 2008 34 35. (86) Onal, A. Food Chem 2007 103 1475. (87) Fujita, K.; Nagatsu, T.; Shinpo, K.; Maruta, K.; Teradaira, R.; Nakamura, M. Clin Chem 1980 26 1577. (88) Ma, Y. F.; Liu, G. S.; Du, M.; Stayton, I. Electrophoresis 2004 25 1473.

PAGE 146

146 (89) Casero, R. A.; Marton, L. J. Nat Rev Drug Discov 2007 6 373. (90) Yeh, C. Y.; Lin, S. J.; Hwang, D. F. J Food Drug Anal 2004 12 128. (91) Mertz, E.; Zimmerman, S. C. J Am Chem Soc 2003 125 3424. (92) Greene, N T.; Shimizu, K. D. J Am Chem Soc 2005 127 5695. (93) S. G. G. Cheng and Z. M. Merchant, Biosensors in Food Analysis,in Characterization of Foods: Emerging Methods, ed. A. G.Gaonkar, Elsevier Science, New York, 1995, ch. 14 (94) Jung, J. H.; Lee, S. J .; Kim, J. S.; Lee, W. S.; Sakata, Y.; Kaneda, T. Org Lett 2006 8 3009. (95) Secor, K.; Plante, J.; Avetta, C.; Glass, T. J Mater Chem 2005 15 4073. (96) Rakow, N. A.; Sen, A.; Janzen, M. C.; Ponder, J. B.; Suslick, K. S. Angew Chem Int Edit 2005 44 4528. (97) Zhou, H. C.; Baldini, L.; Hong, J.; Wilson, A. J.; Hamilton, A. D. J Am Chem Soc 2006 128 2421. (98) Teradaira, R.; Fujita, K.; Nagatsu, T.; Shinpo, K.; Maruta, K.; Nakamura, M. J Clin Chem Clin Bio 1980 18 698. (99) Rossi, S.; Lee, C.; Ellis, P. C.; Pivarnik, L. F. J Food Sci 2002 67 2056. (100) Liu, M.; Kaur, P.; Waldeck, D. H.; Xue, C. H.; Liu, H. Y. Langmuir 2005 21 1687. (101) Liu, Y.; Schanze, K. S. Anal Chem 2009 81 231. (102) Fude Feng, Kirk S. Schanze paper in preparation (103) Tan, C. Y.; Pinto, M. R.; Schanze, K. S. Abstr Pap Am Chem S 2002 223 D43. (104) Nelson, T. L.; O'Sullivan, C.; Greene, N. T.; Maynor, M. S.; Lavigne, J. J. J Am Chem Soc 2006 128 5640. (105) Maynor, M. S.; Nelson, T. L. ; O'Sullivan, C.; Lavigne, J. J. Org Lett 2007 9 3217.

PAGE 147

147 (106) Blackman, A. G. Polyhedron 2005 24 1. (107) Schanze, K. S.; Zhao, X. Y.; Jiang, H. Macromolecules 2008 41 3422. (108) Bao, B. Q.; Yuwen, L. H.; Zheng, X. N.; Weng, L. X.; Zhu, X. R.; Zha n, X. W.; Wang, L. H. J Mater Chem 2010 20 9628. (109) Nelson, T. L.; Tran, I.; Ingallinera, T. G.; Maynor, M. S.; Lavigne, J. J. Analyst 2007 132 1024. (110) Feng, F. D.; Yang, J.; Xie, D. P.; McCarley, T. D.; Schanze, K. S. J Phys Chem Lett 2013, submitted. (111) Chakrabarti, P. J Mol Biol 1993 234 463. (112) Bodin, P.; Burnstock, G. Neurochem Res 2001 26 959. (113) Zyryanov, A. B.; Tammenkoski, M.; Salminen, A.; Kolomiytseva, G. Y.; Fabrichniy, I. P.; Goldman, A.; Lahti, R.; Baykov, A. A. B iochemistry Us 2004 43 14395. (114) Pathak, R. K.; Tabbasum, K.; Rai, A.; Panda, D.; Rao, C. P. Anal Chem 2012 84 5117. (115) Lipscomb, W. N.; Strater, N. Chem Rev 1996 96 2375. (116) Mathews, C. K.; van Holde, K. E. Biochemistry; The Benjamin/ Cummings Publishing Company, Inc.: Redwood City, CA, 1990 (117) Tabary, T.; Ju, L. Y.; Cohen, J. H. M. J Immunol Methods 1992 156 55. (118) Satarug, S.; Baker, J. R.; Urbenjapol, S.; Haswell Elkins, M.; Reilly, P. E. B.; Williams, D. J.; Moore, M. R. T oxicol Lett 2003 137 65. (119) Jiang, G. F.; Xu, L.; Song, S. Z.; Zhu, C. C.; Wu, Q.; Zhang, L. Toxicology 2008 244 49. (120) Goyer, R. A.; Liu, J.; Waalkes, M. P. Biometals 2004 17 555. (121) Timms, A. E.; Zhang, Y.; Russell, R. G. G.; Brown, M. A. Rheumatology 2002 41 725.

PAGE 148

148 (122) Lomashvili, K. A.; Cobbs, S.; Hennigar, R. A.; Hardcastle, K. I.; O'Neill, W. C. J Am Soc Nephrol 2004 15 1392. (123) Gao, J.; Riis Johannessen, T.; Scopelliti, R.; Qian, X. H.; Severin, K. Dalton T 2010 39 7114. (124) Huang, X. M.; Guo, Z. Q.; Zhu, W. H.; Xie, Y. S.; Tian, H. Chem Commun 2008 5143. (125) Wu, C. Y.; Chen, M. S.; Lin, C. A.; Lin, S. C.; Sun, S. S. Chem Eur J 2006 12 2263. (126) Su, G. Y.; Liu, Z. P.; Xie, Z. J.; Qian, F.; He, W. J.; Guo, Z. J. Dalton T 2009 7888. (127) Schanze, K. S.; Zhao, X. Y.; Liu, Y. Chem Commun 2007 2914. (128) Cheng, T. Y.; Wang, T.; Zhu, W. P.; Chen, X. L.; Yang, Y. J.; Xu, Y. F.; Qin, X. H. Org Lett 2011 13 3656. (129) Park, C.; Hong, J. I. Tetrahedron Lett 2010 51 1960. (130) Lee, H. N.; Swamy, K. M. K.; Kim, S. K.; Kwon, J. Y.; Kim, Y.; Kim, S. J.; Yoon, Y. J.; Yoon, J. Org Lett 2007 9 243. (131) Lau, Y. T. R.; Weng, L. T.; Ng, K. M.; Chan, C. M. Anal Chem 2010 82 2661. (132) Zhang, Z. F.; Liu, Y.; Zha ng, H. Spectrosc Spect Anal 2009 29 3263. (133) Farcomeni, A. Computation Stat 2009 24 583. (134) Pearson, K. Philos Mag 1901 2 559. (135) Tan, Z. J. Opt Eng 2009 48 (136) Chen, C. C.; Shih, J. S. J Chin Chem Soc Taip 2008 55 979. (137) Lu, Y. J.; Partridge, C.; Meyyappan, M.; Li, J. J Electroanal Chem 2006 593 105. (138) Sasaki, I.; Tsuchiya, H.; Nishioka, M.; Sadakata, M.; Okubo, T. Sensor Actuat B Chem 2002 86 26.

PAGE 149

149 (139) Vogt, F.; Mizaikoff, B. Anal Chem 2003 75 3050. (140) Cui F. X.; Fang, Y. H.; Lan, T. G.; Xiong, W.; Yuan, Y. M. Spectrosc Spect Anal 2011 31 2794. (141) Kose, M. E.; Omar, A.; Virgin, C. A.; Carroll, B. F.; Schanze, K. S. Langmuir 2005 21 9110. (142) Watchasit, S.; Kaowliew, A.; Suksai, C.; Tuntulani, T. ; Ngeontae, W.; Pakawatchai, C. Tetrahedron Lett 2010 51 3398. (143) Avaeva, S. M.; Sklyanki.Va; Botvinik, M. M. J Gen Chem Ussr 1969 39 559. (144) ang, J. Wu, D.L.; Xie, D.P.; Schanze, K.S. Ion induced aggregation of CPEs studied by Fluorescence Correlation Spectroscopy. Submitted. (145) Keithley, R. B.; Heien, M. L.; Wightman, R. M. Trac Trend Anal Chem 2009 28 1127. (146) Leng, C. L.; Wang, H. S. J Comput Graph Stat 2009 18 201. (147) Lipmann, F. J Biol Chem 1944 155 55. (148) Rose, I. A.; Grunbergmanago, M.; Korey, S. R.; Ochoa, S. J Biol Chem 1954 211 737. (149) Schonheit, P. Hyperthermophilic Enzymes, Pt B 2001 331 168. (150) Spector, L. B. P Natl Acad Sci Biol 1980 77 2626. (151) Bowman, C. M.; Valdez, R. O.; Nishimura, J. S. J Biol Chem 1976 251 3117. (152) Ren, N. Q.; Lin, H. L.; Zhang, K.; Zheng, G. X.; Duan, Z. J.; Lin, M. Curr Microbiol 2007 55 167. (153) Mukhopadhyay, S.; Hasson, M. S.; Sanders, D. A. Bioorg Chem 2008 36 65. (154) Thauer, R. K.; Mollerzinkhan, D.; Spormann, A. M. Annu Rev Microbiol 1989 43 43. (155) Kahane, I.; Razin, S.; Muhlrad, A. Israel J Med Sci 1978 14 807.

PAGE 150

150 (156) Kahane, I.; Razin, S.; Muhlrad, A. Fems Microbiol Lett 1978 3 143. (157) Reeves, R. E.; Guthrie, J. D. Biochem Bioph Res Co 1975 66 1389. (158) Schafer, T.; Selig, M.; Schonheit, P. Arch Microbiol 1993 159 72. (159) Perraud, A. L.; Weiss, V.; Gross, R. Trends Microbiol 1999 7 115. (160) Buss, K. A.; Cooper, D. R.; Ingram Smith, C.; Ferry, J. G.; Sanders, D. A.; Hasson, M. S. J Bacteriol 2001 183 3536. (161) Brown, T. D. K.; Stormer, F. C.; Pereira, C. R. S. J Bacteriol 1972 112 1106. (162) Sanwal, B. D. Bacteriol Rev 1970 34 20. (163) Elmansi, E. M. T.; Holms, W. H. J Gen Microbiol 1989 135 2875. (164 ) Fox, D. K.; Meadow, N. D.; Roseman, S. J Biol Chem 1986 261 3498. (165) Koplove, H. M.; Cooney, C. L. J Bacteriol 1978 134 992. (166) Lipmann, F.; Tuttle, L. C. J Biol Chem 1945 159 21. (167) Rose, I. A. Methods in Enzymology 1955 1 591. (168) Gorrell, A.; Lawrence, S. H.; Ferry, J. G. J Biol Chem 2005 280 10731. (169) Fowler, M. L.; Ingram Smith, C.; Smith, K. S. Eukaryot Cell 2012 11 1249. (170) Ingram Smith, C.; Gorrell, A.; Lawrence, S. H.; Iyer, P.; Smith, K.; Ferry, J. G. J Bac teriol 2005 187 2386. (171) Miles, R. D.; Gorrell, A.; Ferry, J. G. J Biol Chem 2002 277 22547. (172) Oultram, J. D.; Burr, I. D.; Elmore, M. J.; Minton, N. P. Gene 1993 131 107. (173) Jie Yang, Kirk S. Schanze, paper in preparation. (174) Grundy, F. J.; Waters, D. A.; Allen, S. H. G.; Henkin, T. M. J Bacteriol 1993 175 7348. (175) De Jong, M. J. M.; Vissenberg, M. C. J. M. Philips J Res 1998 51 495.

PAGE 151

151 (176) Shibata, T.; Hattori, T.; Onodera, S.; Kaino, T. Nippon Kagaku Kaishi 1998 831. (177) Dwight, S. J.; Gaylord, B. S.; Hong, J. W.; Bazan, G. C. J Am Chem Soc 2004 126 16850. (178) Liu, Y.; Ogawa, K.; Schanze, K. S. Anal Chem 2008 80 150. (179) Smith, J. H., Simons, C., Enzymes and Their Inhibition: Drug Development; CRC Press 2004

PAGE 152

152 BIOGRAPHICAL SKETCH Jie (Cheer) Yang was born in Hubei Province, China. Jie started her undergraduate studies at Wuhan University in 2004. Four years later, she received her bachelor degree in chemistry. During that same year, she went to Unite d States to continue her graduate stu dies in D epartment of C hemistry at University of Florida, where she joined Dr. In the past five years, she focused her research on the design and development of analytical methods based on fluorescent c onjugated polyelectrolytes for sensing applications She received her Ph.D from the University of Florida in the spring of 2013.