<%BANNER%>

Time-Resolved Laser Spectroscopic Studies of Rate-Limiting Events in Protein Folding and Binding

Permanent Link: http://ufdc.ufl.edu/UFE0024151/00001

Material Information

Title: Time-Resolved Laser Spectroscopic Studies of Rate-Limiting Events in Protein Folding and Binding
Physical Description: 1 online resource (137 p.)
Language: english
Creator: Narayanan, Ranjani
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: binding, disordered, fluorescence, folding, fret, ia3, intrinsically, kinetics, lasers, nanosecond, protein, spectroscopy, tryptophan, tz2, viscosity
Physics -- Dissertations, Academic -- UF
Genre: Physics thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Proteins fold from simple poly-amino acid chains to compact three-dimensional structures capable of performing diverse functions in our cells. This disorder to order transition is both swift (~microseconds) and specific. While energy landscape theory and kinetic theories of diffusion have enriched our understanding of the mechanism by which proteins fold, we are still searching for the answers to key questions. What is the limiting speed for the folding process? What events drive the folding process at these limiting speeds? Protein folding begins with the formation of a contact between any two points of the chain diffusing towards each other. The rate of this event sets an upper limit to the overall folding rate. We use laser-triggered nanosecond-resolved multi-wavelength transient absorbance spectroscopy to study contact formation in simple amino acid chains in aqueous solvents. Studying variation of diffusion rates with solvent viscosity and temperature identifies the events limiting the folding rate. Small fast folding proteins are model systems to conduct such experiments, as their folding mimics the initial events in the protein folding process. Our laser-induced temperature-jump nanosecond- resolved fluorescence studies of the tryptophan zipper folding investigates the events that limit beta-hairpin formation in tryptophan zipper at various temperatures and solvent viscosities. We observe a fast (~100 ns) relaxation and a slower (~ ms) relaxation for the tryptophan zipper in our kinetic studies of TZ2 folding with varying time dependence at different temperatures and viscosities. The presence of more than one relaxation confirms the existence of multiple folding pathways for TZ2. They also open up the possibility of exploring these pathways in different regimes of solvent viscosity. We study the folding kinetics of the natively unfolded IA3 peptide coupled to its binding to the YPrA enzyme. This interaction results in specific and potent inhibition of YPrA by IA3. Our laser-triggered temperature jump studies enable a better understanding of the mechanism by which IA3 inhibits YPrA. The different folding kinetics of IA3 in the absence and presence of YPrA suggests a mechanism where IA3 first binds to YPrA, and then uses YPrA as a template to stabilize its folded state.
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 Ranjani Narayanan.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Hagen, Stephen J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-11-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024151:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024151/00001

Material Information

Title: Time-Resolved Laser Spectroscopic Studies of Rate-Limiting Events in Protein Folding and Binding
Physical Description: 1 online resource (137 p.)
Language: english
Creator: Narayanan, Ranjani
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: binding, disordered, fluorescence, folding, fret, ia3, intrinsically, kinetics, lasers, nanosecond, protein, spectroscopy, tryptophan, tz2, viscosity
Physics -- Dissertations, Academic -- UF
Genre: Physics thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Proteins fold from simple poly-amino acid chains to compact three-dimensional structures capable of performing diverse functions in our cells. This disorder to order transition is both swift (~microseconds) and specific. While energy landscape theory and kinetic theories of diffusion have enriched our understanding of the mechanism by which proteins fold, we are still searching for the answers to key questions. What is the limiting speed for the folding process? What events drive the folding process at these limiting speeds? Protein folding begins with the formation of a contact between any two points of the chain diffusing towards each other. The rate of this event sets an upper limit to the overall folding rate. We use laser-triggered nanosecond-resolved multi-wavelength transient absorbance spectroscopy to study contact formation in simple amino acid chains in aqueous solvents. Studying variation of diffusion rates with solvent viscosity and temperature identifies the events limiting the folding rate. Small fast folding proteins are model systems to conduct such experiments, as their folding mimics the initial events in the protein folding process. Our laser-induced temperature-jump nanosecond- resolved fluorescence studies of the tryptophan zipper folding investigates the events that limit beta-hairpin formation in tryptophan zipper at various temperatures and solvent viscosities. We observe a fast (~100 ns) relaxation and a slower (~ ms) relaxation for the tryptophan zipper in our kinetic studies of TZ2 folding with varying time dependence at different temperatures and viscosities. The presence of more than one relaxation confirms the existence of multiple folding pathways for TZ2. They also open up the possibility of exploring these pathways in different regimes of solvent viscosity. We study the folding kinetics of the natively unfolded IA3 peptide coupled to its binding to the YPrA enzyme. This interaction results in specific and potent inhibition of YPrA by IA3. Our laser-triggered temperature jump studies enable a better understanding of the mechanism by which IA3 inhibits YPrA. The different folding kinetics of IA3 in the absence and presence of YPrA suggests a mechanism where IA3 first binds to YPrA, and then uses YPrA as a template to stabilize its folded state.
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 Ranjani Narayanan.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Hagen, Stephen J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-11-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024151:00001


This item has the following downloads:


Full Text

PAGE 1

1 TIME RESOLVED LASER SPECTRO S COPIC STUDIES OF RATE LIMITING EVENTS IN PROTEIN FOLDING AND BINDING By RANJANI NARAYANAN 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 2009

PAGE 2

2 2009 Ranjani Narayanan

PAGE 3

3 ACKNOWLEDGMENTS Graduate study in a new country is both an exciting and occasionally terrifying prospect. I am indebted to the Gator community for making my experience in this wonderful University a memorable one. I would first like to acknowledge my parents, Mrs. And Mr. Narayanan for constantly encouraging me to aim high and achieve the best in all my endeavors. I also wish to thank my brother Sudharsan Narayanan for doing his bit to cheer me up during trying times. During my time as a graduate student, I met my husband, Balaji Krishnaprasad, whose constant support has sustained me through the past two years. Dr. S tephen Hagen my mentor, advisor and guide has instilled in me the zeal to achieve perfection in research. He has been a patient teacher who has taught me to deal with problems innovatively, communicate ideas effectively, and perform independent research e fficiently. I would also like to thank Dr. Arthur Edison, Dr. Adrian Roitberg, Dr. David Reitze, and Dr. Sergei Obukhov for serving on my supervisory committee. I have benefited from their suggestions and comments at various times. I consider myself f ortunate to have had a very talented committee to guide me. I am grateful to the faculty and staff of Physics department at the University of Florida, Gainesville. I am thankful to Marc Link, Edward Storch, Raymond Frommeyer, and Bill Malphurs at the Mach ine Shop for providing me with parts of exceptional quality for my research. I wish to thank the members of the Electronics Shop and Jay Horton for providing me with technical support and suggestions at odd times of the day and week. The administrative s taffs have also been very helpful with the smooth management of paperwork and other formalities.

PAGE 4

4 I would like to thank my colleagues Pablo Perez, Suzette Pabit, and Omjoy Ganesh for their companionship and useful suggestions (academic and otherwise). I have also benefited from collaborative associations with Dr. Cherian Zachariah and other members of Dr. Edisons lab. I would like to thank Dr. Alfred Chung for providing me with peptides for my research. Leslie Pelakhs help in solvent viscosity measurem ents for my project on TZ2 is also kindly acknowledged. I am indebted to the University of Florida for opening the doors to a rich educational experience. The staff at the International Center in particular Debra Anderson and Maud Fraser have been almost maternal in their concern for all students including myself. My stint as a graduate student also enabled me to befriend exceptional students and researchers from various disciplines of study at the University of Florida. I shall forever cherish the camar aderie and wonderful times spent in the company of my various Gator friends.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 3 LIST OF FIGURES .............................................................................................................................. 8 ABSTRACT ........................................................................................................................................ 11 CHAPTER 1 INTRODUCTION ....................................................................................................................... 13 Introduction ................................................................................................................................. 13 Protein Structure .......................................................................................................................... 13 Finding the Right Fold in Time .................................................................................................. 14 Rate Limit for Protein Foldin g ................................................................................................... 18 Diffusion Limits Rate of Protein Folding .......................................................................... 18 Polymer physics and diffusion -limited contact formation ......................................... 18 Gaussian statistics and the dynamics of the unfolded protein ................................... 19 Comparison of dynamics of internal and external loops ........................................... 22 Solvent Friction Effects on Protein Folding ...................................................................... 23 Internal Friction ................................................................................................................... 25 Intrinsically Disordere d Proteins ................................................................................................ 27 Definition of Disorder ....................................................................................................... 28 Characteristics of Intrinsically Disordered Proteins .......................................................... 29 Functions of Intrinsically Disordered Proteins .................................................................. 30 Theoretical Studies of Coupled Folding and Binding ............................................................... 31 Experiments on Coupled Folding and Binding in Natively Disordered Proteins ................... 32 Intrinsically Disordered Peptide IA3 and the Yeast Aspartic Proteinase A (YPrA) ............... 33 Intrinsically Disordered Peptide IA3 ........................................................................... 34 Yeast Aspartic Proteinase A (YPrA) .................................................................................. 35 Scope o f This Dissertation .......................................................................................................... 36 2 EXPERIMENTAL STUDIES OF FAST FOLDING KINETICS ........................................... 38 Introduction ................................................................................................................................. 38 Probes of Conformational Change in Proteins .......................................................................... 40 Fluorescence and Triplet Absorption Spectroscopy .......................................................... 40 Col lection of Fluorescence Spectra ............................................................................. 46 Circular Dichroism Spectroscopy ....................................................................................... 47 Collection of CD spectra .............................................................................................. 50 Tryptophan Photo physics .......................................................................................................... 50 Tryptophan Fluorescence .................................................................................................... 51 Tryptophan Triplet Relaxation ............................................................................................ 52 Techniques Used to Study Protein Folding Kinetics ................................................................ 55 Mixing .................................................................................................................................. 55

PAGE 6

6 Flash P hotolysis ................................................................................................................... 55 Excitation System: ....................................................................................................... 57 Probe System ................................................................................................................ 57 Sample Prepar ation ...................................................................................................... 58 Spectral Data Acquisition ............................................................................................ 59 Data Analysis ................................................................................................................ 60 Temperatu re jump Spectroscopy ....................................................................................... 60 Temperature -jump Apparatus ..................................................................................................... 62 Infrared (IR) Excitation ....................................................................................................... 62 Ultra violet (UV) Probe: ..................................................................................................... 63 Fiber and Sample Block ...................................................................................................... 65 Sample Handling: ................................................................................................................ 66 Sample Preparation: ............................................................................................................. 66 Temperature -jump Calibration ............................................................................................ 67 Kinetic Data Acquisition: .................................................................................................... 68 Spectral Data Acquisition: .................................................................................................. 69 3 KINETICS OF FOLDING AND BINDING OF THE INTRINSICALLY DISORDERED PEPTIDE IA3 WITH YPRA ........................................................................... 71 Introduction ................................................................................................................................. 71 Results .......................................................................................................................................... 73 Peptide Characterization ..................................................................................................... 73 Equilibrium Fluorescence Studies ...................................................................................... 73 Characterization of IA3 Folding Behavior at Equilibrium ................................................ 75 Kinetics of IA3 Folding in the Presence of TFE ................................................................ 78 Kinetics of IA3 Folding and Binding to YPrA ................................................................... 81 Discussion .................................................................................................................................... 82 Alpha helix Formation in Proteins ..................................................................................... 83 Scheme for IA3YPrA Interaction ...................................................................................... 84 Origin of Fast ~ 90 ns Relaxation ....................................................................................... 85 Future Research ................................................................................................................... 86 Summary ...................................................................................................................................... 87 4 HETEROGENEOUS FOLDING KINETICS OF TRYPTOPHAN ZIPPER ......................... 89 Introduction ................................................................................................................................. 89 Heterogeneous Folding of TZ2: Background .................................................................... 91 Results .......................................................................................................................................... 94 Circular Dichroism Spectroscopy ....................................................................................... 94 Kinetics of TZ2 Relaxation after Te mperature jump ........................................................ 95 Discussion .................................................................................................................................. 100 Conclusions ............................................................................................................................... 102 5 CONTACT FORMA TION IN POLYPEPTIDES .................................................................. 104 Introduction ............................................................................................................................... 104

PAGE 7

7 Background: Experiments on Contact Formation ........................................................... 104 Rationale for our Experiments .......................................................................................... 105 Results ........................................................................................................................................ 106 Characterization of Tryptophan Photochemistry ............................................................. 107 Determination of Power Law Dependence for Tryptophan Triplet Relaxation ............ 108 Tryptophan Triplet Lifetimes and Oxygen Quenching ................................................... 110 Challenges and Bottlenecks ...................................................................................................... 111 Discussion .................................................................................................................................. 113 Conclusions ............................................................................................................................... 114 6 FUTURE DIRECTIONS AND CONCLUSIONS .................................................................. 116 APPENDIX : NUMERICAL METHODS ....................................................................................... 118 LIST OF REFERENCES ................................................................................................................. 122 BIOGRAPHICAL SKETCH ........................................................................................................... 137

PAGE 8

8 LIST OF FIGURES Figure page 1 1 Re presentation of an amino acid. .......................................................................................... 14 1.2 Representation of the hierarchical nature of protein structure. ......................................... 15 1 3 Representation of energy landscape of protein folding. ...................................................... 16 1 4 F astest folding times for proteins .......................................................................................... 17 1 5 Internal loops and external loops .......................................................................................... 19 1 6 Statistical distribution of the conformations sampled by ends of a freely jointed polymer chain ........................................................................................................................ 21 1 7 Representation of barri er crossing events in Kramers model ............................................. 24 1 8 F ly -casting model. .................................................................................................................. 31 1 9 Complex of IA3 (red alpha helix) with YPrA (blue) ............................................................ 35 2 1 T wo -state folding in a protein ............................................................................................... 38 2 2 Jablonski diagram showing the interaction of light with matter ........................................ 41 2 3 Effects of solvent exposure on fluorescence of free tryptophan ......................................... 42 2 4 Energy transfer between two fluorophores tryptophan and dansyl attached to N2W K16C dansyl IA3 peptide ....................................................................................................... 43 2 5 Representation of design of IA3 mutants with FRET donor tryptopha n and acceptor dansyl ...................................................................................................................................... 45 2 6 Vie w of the inhibition complex of YPrA (blue) with IA3 (gray) based on crystal structure data (76) ................................................................................................................... 46 2 7 CD spectrum of thermal denaturation of alpha helical peptide IA3 and beta hairpin forming TZ2 peptide. ............................................................................................................. 49 2 8 Tryptophan structure ............................................................................................................. 51 2 9 M odel for diffusion controlled loop formation in a poly amino acid chain label ed with donor Trp and acceptor cysteine (in red). .................................................................... 52 2 10 Time -resolved transient spectroscopy apparatus ................................................................. 58 2 11 Response of fr ee tryptophan and the peptide TC5b to a temperatur e jump of 9.8oC. ...... 61

PAGE 9

9 2 12 T emperature jump set up with following elements ............................................................. 63 2 13 UV probe system ................................................................................................................... 64 2 14 Sample holder and fused silica fiber ..................................................................................... 65 2 15 Representation of thermal diffusion in sample after th e IR pulse. ...................................... 66 2 16 Temperature calibration of T jump system with fluorescence of free tryptophan in solvent used for experiments. .............................................................................................. 67 2 17 Saturation of aqueous NATA solution with UV photons. ................................................... 68 2.18 Spectral data acquisition system ........................................................................................... 69 3 1 Two potential schemes of interaction for YPrA with the intrinsically disordered peptide IA3 .............................................................................................................................. 72 3 2 Equilibrium fluorescence measurements with N2W -K16C dansyl IA3 ............................. 74 3 3 Equilibrium far -UV circular dichroism data for mutant N2W -K16C dansyl IA3 ............ 76 3 4 Analysis of thermodynamic parameters for folding transition of IA3 as a function of temperature and TFE.. ............................................................................................................ 78 3 5 T emperature -calibration of Temperaturejump system. ..................................................... 79 3 6 Temperature -jump data for IA3 in TFE ................................................................................. 81 3 7 Kinetics of coupled folding and binding interaction of IA3 with YPrA ............................. 82 3 8 P roposed model for interaction between IA3 (I) and YPrA (Y) .......................................... 84 4 1 Structure of tryptophan zipper TZ2. ..................................................................................... 90 4 2 CD signal for TZ2 in pH 7.0 phosphate buffer + 2M GdnHCl ........................................... 94 4 3 Comparison of fluorescent response of free tryptophan in N acetyl tryptophan amide (NATA) and tryptophan residues in TZ2 in phosphate buffer at pH 7.0 with 2M GdnHCl after a temperature jump o f 9.7oC ......................................................................... 95 4 4 Fluorescent response of 120 M TZ2 in 50 mM phosphate buffer at pH 7.0 and 2M GdnHCl to a temperature jump of 8 10oC. .......................................................................... 96 4 5 Shift in fluorescence intensity and red shift in peak wavelength of fluorescence at low viscosities (blue) and high viscosities (orange) ............................................................ 97 4 6 Viscosity dependence of relaxation of spectral shift at 35oC and 52oC .............................. 98

PAGE 10

10 4 7 Kinetic response of TZ2 in solvents of varying viscosities at different temperatures. ..... 99 5 1 Flash photolysis scheme for monitoring intra -chain diffusion via energy transfer between contact forming monomers. .................................................................................. 105 5 2 Optimisation of the transient spectroscopy apparatus ....................................................... 107 5 3 Absorption spectra for free tryptophan in pH 7.0 phosphate buffer ................................. 108 5 4 T ryptophan triplet relaxation (absorption maximum at 4 50 nm) fit to a single exponential and power law decay ....................................................................................... 109 5 5 Triplet relaxation as a function of triplet and oxygen concentration ................................ 111 5 6 Photo -damage of tryptophan by repeated UV irradiation in an experiment ..................... 112 A 1 T ime -resolved fluorescence spectra of double mutant N2W -K16C IA3 peptide ........... 118 A 2 Results of singular value decomposition of time resolved fluorescence spectra for IA3. ........................................................................................................................................ 120 A 3 Single exponential fit for both SV2 and SV3 in time .......................................................... 121

PAGE 11

11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy TIME RESOLVED LASER SPECTRO S COPIC STUDIES OF RATE LIMITING EVENTS IN PROTEIN FOLDING AND BINDING By Ranjani Narayanan May 2009 Chair: Stephen James Hagen Major: Physics Proteins fold from simple polyamino acid chains to compact three -dimensional structures capable of performing diverse functions in our cells. This disorder to order transition is both swift (~ microseconds ) and specific. While energy landscape theory and kinetic theories of diffusion have enriched our understanding of the mechanism by which proteins fold, we are sti ll searching for the answers to key questions. What is the limitin g speed for the folding process ? What events drive the folding process at these limiting speeds? Protein folding begins with the formation of a contact between any two points of the chain diffusing towards each other. The rate of this event sets an upper limit to the overall folding rate. We use laser triggered nanosecond resolved multi -wavelength transient absorbance spectroscopy to study contact formation in simple amino acid chains in aqueous solvents Studying variation of diffusion rates with solvent viscosity and temperature identifies the e ven ts limiting the folding rate. Small fast folding proteins are model systems to conduct such experiments, as their folding mimics the initia l events in the protein folding process. Our laser induced temperaturejump nanosecond resolved fluorescence studies of the tryptophan zipper folding investigates the events that limit beta -hairpin formation in tryptophan zipper at various temperatures a nd solvent viscosities We observe a fast (~100 ns) relaxation and a slower (~ s)

PAGE 12

12 relaxation for the tryptophan zipper in our kinetic studies of TZ2 folding with varying time dependence at different temperatures and viscosities. The presence of more tha n one relaxation confirms the existence of multiple folding pathways for TZ2. They also open up the possibility of exploring these pathways in different regimes of solvent viscosity. We study the folding kinetics of the natively unfolded IA3 peptide coupl ed to i ts binding to the YPrA enzyme. This interaction results in specific and potent inhibition of YPrA by IA3. Our laser triggered te mperature jump studies enable a better understanding of the mechanism by which IA3 inhibit s YPrA The different foldi ng kinetics of IA3 in the absence and presence of YPrA suggests a mechanism where IA3 first binds to YPrA, and then uses YPrA as a template to stabilize its folded state.

PAGE 13

13 CHAPTER 1 INTRODUCTION Introduction Proteins are vital in biological p rocesses. Protein function is derived from structure, which in turn is derived from sequence encoded by the cellular DNA. The protein folding problem (1, 2) is the process by which the freshly synthesized polymer chain of amino acids assembles itself into a compact functional form on biologically relevant timescales. Current understanding of protein folding has benefited from advances in experimental methods for probing the protein folding reactio n and theoretical approaches that simulate folding using simple physical models. This has led to a giant leap in understanding of the proteins with a well -defined fold. The progress of similar research on intrinsically disordered proteins has been more r ecent. This dissertation focuses on the kinetics of sub-millisecond folding events in proteins, which includes binding induced folding of the intrinsically disordered peptide IA3. We use time resolved spectroscopic techniques to monitor the rate limiting events of the folding process such as intra -chain contact formation in simple polypeptide chains and turn formation in mini proteins such as tryptophan zipper. Our studies of the intrinsically disordered peptide IA3 follow its extremely rapid concerted fo lding and binding interaction with the protease YPrA. Before plunging headlong into a discussion of the details of the individual projects, this introductory chapter briefly reviews the major concepts in the protein folding literature relevant to our studi es. Protein Structure Proteins are polymer chains of amino acids, which are characterized by a carboxylic acid group ( -COOH), an amine group ( NH2) and another organic group (R) attached to the central carbon atom as shown in figure 11. R could represent an aliphatic or aromatic carbon group. The L and D -forms of the same amino acid are mirror symmetric: the hydrogen and R residue

PAGE 14

14 on the alpha carbon exchange places in the two forms. Genes can code for L amino acids only, hence protein chains constitut e L amino acids (3). Figure 1 1: Representation of an amino acid with the L -steric form having the side residue R and hydrogen at the alpha carbon as shown above. The peptide bond between two amino aci ds is formed when the carboxyl group ( COOH) of one amino acid reacts with the amine group ( NH2) of the other by condensation The primary structure of proteins is the polymer chain of amino acid residues held together by peptide bonds. This chain can rotate about the C C and N C bond angles respectively) to form helices, loops and sheets. These loops, turns and helices are held together by hydrogen bonds and constitute the secondary structure of proteins. Further linking of these secondary structural el ements by hydrophobic interactions between hydrocarbon side chains ionic salt bridges and disulphide linkages which extend the hierarchy of protein structure to tertiary and quaternary structure as shown in figure 1 2. Finding the Right Fold in Time The field of protein folding opened up with C. Anfinsens study on bovine ribonuclease (4, 5) demonstrating the ability of proteins to fold spontaneously and reproducibly. He showed that the fo lding process was a conformational search for the peptide configuration with lowest free energy. The sub-second timescales on which proteins complete this search suggests that this is not a random conformational search for the most stable conformation of the peptide. Levinthal estimated that a protein with ~ 100 residues would take close to 1010 years to sample all

PAGE 15

15 conformations available to it (Levinthals paradox) To fold on biologically relevant timescales, the protein must then access a fraction of all available conformations by choosing a particular folding pathway. Recent progress in strategies for simulation of protein folding by theoretical means has enabled a richer understanding of the mechanisms of protein folding. Figure 1.2: Schematic r epresentation of the hierarchical nature of protein structure. (Freely available illustrations from the public domain of the National Human Genome Research Institute.). The primary structure is the string of amino acids held together by peptide bonds. T his structure is retained when a protein is unfolded. Hydrogen bonds between amino acids lead to the formation of the elem ents of secondary structure such as helices, sheets and turns. Interactions between side chains of residues (ionic linkages, hydroph obic interactions) contribute to the tertiary and quaternary structure of the protein.

PAGE 16

16 Protein fo lding can be visualized as a chemical reaction with the unfolded and folded protein being initial and final states of the system. The protein can thus be de scribed as traversing a trajectory on a surface, which relates its energy to the co -ordinates of its constituent atoms in a particular conformation. In recent times, it has been possible to explore this energy surface by simulating the trajectories of the atoms of a protein while folding in solvent. Equations of motion have been written for the atoms constituting a protein, and a variety of potentials used to describe the electrostatic, van der Waals and covalent interactions in a protein. Unlike simple chemical reactions where transition state energies differ by large enthalpies of binding interactions, the complex energy landscape of protein folding is made up of states that differ slightly in configurational entropy. The initial state from which foldi ng is initiated is heterogeneous and involves several structures that are accessible to a polypeptide chain. Figure 1 3 : Schematic representation of energy landscape of protein folding, starting out with a broad range of poss ible initial folds that narrows down with the formation of stabilizing contacts between different parts of the polypeptide chain. The depth of this funnel like landscape (6, 7) is the free energy difference between unfolded and folded states. Energy Reaction Coordinates

PAGE 17

17 The folding process then proceeds as a downhill search in which interactions that are native like are more stabilizing than non -native like interactions. A given molecule samples only a subset of a ll possible conformations accessible to it, owing to the statistical nature of the process, thus allowing for a more realistic folding time. This picture of a folding funnel (6, 7, 8) has enabled us to understand several features of the folding reaction (figure 1 3 ). The concept of a funnel has been extended to protein binding, which is vital to the function ing of proteins. The stable complex of two proteins lies at the bottom of the funnel. It has been argued (9) that the more rugged th e landscape near the bottom of the funnel, the more likely it is for the protein to be flexible and make non -specific contacts during complex formatio n. If the funnel has a narrow minimum, it would imply the higher likelihood of a rigidly structured protein that would interact with other proteins in a lock and key scheme. This suggests that structural rigidity is not a pre -requisite for specificity in bim olecular recognition, and that protein -protein associations can be possible with flexible proteins too. Figure 1 4 : Schematic of fastest folding times for proteins measured in experiments based on a survey of literature (10, 11) for past decade

PAGE 18

18 Rate Limit for Protein Folding Experiments on small proteins have shown folding times on the scale of microseconds (10) to milliseconds (figure 1 4 ). Since fol ding involves compaction and contact formation events, the folding time is limited by the slowest of these events. These events include the collapse of the polypeptide chain to form the first contact to form loops and turns. T herefore, on e can expect min i -proteins that fold into loops and turns to fold at this limiting speed. Our studies of contact formation in small peptides and hairpin formation in the tryptophan zipper protein measure this limiting rate under different experimental conditions of tempe rature and solvent viscosity. We study fast folding proteins, so that we may understand what physically limits the rates of their folding. This implies that we have to implement strategies to acquire kinetic data for protein folding with sub-microsecond resolution and then make a connection between the observed kinetics and simple physical models for protein folding. Diffusion Limits Rate of Protein Folding The unfolded state is the starting point of the protein folding process. It is comparable to a random -coil like polymer chain that can sample a host of dynamic conformations. The formation of numerous contacts between segments of this chain facilitates the compaction and eventual folding of the protein. Intra -chain diffusion of these chain segments towards each other can thus be regarded as an elementary step in the folding reaction (12, 13) Polymer physics and diffusion -limited contact formation Theoretical studies of diffusion limited contact formation in proteins have gained from an understanding of cyclization reactions (14) in flexible polymers. The unfolded state of a protein (15, 16) can be modeled by statistical mechanical models for polymers. We have a simple picture of a multi unit chain whose monomers are brought in proximity to each other by

PAGE 19

19 diffusive motions in solvent. Sometimes, these monomers are the termini of the chain a nd they form an external loop. The other possibility is where contact -forming monomers form an internal loop with tails of monomers external to them (figure 1 5). The lengths of the loops and tails can be specified in terms of the number of monomers cons tituting them, if all monomers have similar dimensions. The space occupied by one monomer in a real chain is not accessible to another monomer, this volume being termed as excluded volume. Excluded volume is not just a measure of the geometrical volume occupied by a monomer; it also includes volume taken up by interaction with solvent molecules or other monomers. Excluded volume causes the ends of a chain to be farther away (on average) than they would be, in the absence of excluded volume. The dynamic s of a real polymer chain diffusing in solvent depend on the mechanical properties of the polymer chain such as loop lengths, tail lengths and excluded volume. Figure 1 5: A chain of linked spheres in white with two spheres s hown in black forming a loop. If these black spheres are at the ends, an external loop is formed. Loops between non terminal residues are termed internal loops Gaussian statistics and the dynamics of the unfolded protein A very simple model for the unfold ed protein (17) is the freely jointed chain of amino acid monomers with flexible linkers, making all bond and torsion angles about the links equally probable. This chain has no excluded volume and can execute a random walk in conformational TAILS Internal loops External loops LOO P

PAGE 20

20 space. The one-dimensional end -to -end distance vector R is calculated for a chain of n segments with uniform average bond length L (R max = n L) The mean value of this vector is zero when we average out equally probable bond angles. The root mean square value however is a non zero. Assuming each chain conformation to be equally probable, the probability of the ends being at a distance between R and R + dR is given by a Gaussian distri bution of the form shown in Eq.1 1 ) 2 / 3 exp( ) 3 / 2 ( 4 ) (2 2 2 / 3 2 2R R R dR R dR R P [1 1 ] The equation holds true for a phantom chain where two units could occupy the same space. Incorporation of excluded volume effects would reduce the conformations available to the polymer, reducing the random walk to a self avo iding walk. 2 2) ( ) ( r r P D t r P [1 2 ] Diffusion equations can be used to study the time -evolution of the probability distribution of r the distance between ends of a polymer and calculate the time taken by the termini of the polypeptide to diffuse towards each other, initiating the folding process (14, 16, 17) Peq(r) is the equilibrium distribution of the fluctuating distances between points in the chain. Schulten, Schulten and Szabo (18) use a one -dimensional model for diffusion of polymer to estimate time of first contact, obtaining a result that has been successful in explaining the dynamics of several peptides with reasonably good accuracy. This approach looks at the fluctuating end to -end distance between the ends of the loop and tries to estimate the time taken for the ends to diffuse towards each other under the action of an entropic potential and make the first contact.

PAGE 21

21 0 2 4 6 8 10 0 2 4 6 8 Distance r End-to-end distance p(r) 0 2 4 6 8 10 -2 0 2 4 6 Distance r Interaction potential U(r) U(r) = kB*T*log(p(r)) Probability distribution of end-to-end distance for freely jointed chain Figure 1 6 : Statistical distribution of the conformations sampled by ends of a freely jointed polymer chain interacting via a simple entropic potential U(r) The model, referred to as SSS theory from now, as sumes that the effective force field controlling intra -chain diffusion dynamics is given by the Eq.3 4. ) ( ln ) ( r P T k r UB [1 3 ] The average time of loop formation estimated in this fashion is shown in Eq.1 5 (18) 2) ( ) ( ) ( L x L ady y P x P x D dx [1 4 ] Here the contact radius a denotes separation between monomers forming contact and L is the contour length of the loop. We make a connection between kinetics of contac t formation and the equilibrium probability distribution P(r) In our studies, the assumption that D is the diffusion constant of a monomer in solvent is valid as long as we measure the relative rates of contact formation for different loop lengths, rather than abs olute rates. For a Gaussian n -segment

PAGE 22

22 chain without any excluded volume interactions, the theory predicts that the rate of contact formation scales as n3/2. Comparison of dynamics of internal and external loops We have looked so far at theoretical estimat es of loop rates by the ends of a chain. The more biologically relevant scenario is one where two points in the chain form an internal loop. Intuitively one would expect these loops to form slower than external loops due to the effect of dragging tails. The additional residues to the loop also have an excluded volume that reduces the probability of loop formation between interior residues. One study (19) on loop formation in three different scenarios; end -end, endi nterior and interior loops reported distinct differences in statistical behavior, which was manifest by the different scaling dependence of probability distributions for the loops. Doucet et al (20) have used a freely jointed chain model with hard sphere excluded volume interactions and applied SSS theory to the probability distribution for inter residue distance. Increased suppression of contact formation rates are reported for loops of increasing length on adding tai ls. This has been attributed to the increased interaction potential U(r) in loops with enhanced contour length and tails. The study of end interior loops (figure 3 3) seems to indicate that tails on one end slow the rate of contact formation, but not to t he same extent as a slowing down of contact formation rate by a tail on both ends. This study enables predictions for experimental determination of contact formation rates in polypeptide chains. The rate of diffusion limited contact formation estimated i n this fashion serves as an upper limit for the rate (18) at which a real polypeptide chain assembles itself (via multiple native like contacts) into a structured protein. Nanosecond-pulsed laser spectroscopy has ena bled measurement of intra -chain diffusion in the protein cytochrome c (21) This yields a time of 40 s for diffusion -limited contact between the heme group, and a residue 50 60 amino

PAGE 23

23 acids further along the polypept ide chain. This rate has been extrapolated to 1 s for contact formation between residues separated by 10 amino acids which represents the fastest intra -chain diffusion in short polypeptide chain segments. More recent experimental studies of contact form ation between diffusing segments of the polyamino acid chain have studied energy transfer between photo -excited triplet states of residues (22, 23, 24) in peptide chain. We us e this method of triplet triplet energy transfer to estimate the rate of diffusion limited contact formation in polypeptide chains (chapter 5 ). Solvent Friction Effects on Protein Folding The folding reaction of most fast -folding proteins is well described by a two -state transition between the unfolded state and the folded state. Since folding relies on the formation of multiple native like contacts between segments of the polypeptide chain by diffusion, the rate of folding is controlled by the rate of diffusion across the energy barrier separating the two states. Kramers theory of reaction rates (25) for a particle diffusing across a one -dimensional double well potential U(x) can be applied to this scenario (26) to estimate the escape of the particle. According to this theory, the diffusive dynamics of the particle is coupled to thermal noise inherent in the system, making the particles reaction co ordinate x and velocity stochastic in nature. The Langevin equation of motion for the particle of mass M can be written as shown in Eq.1 1 ) ( ) ( t x M x U x M [1 5 ] The term is the friction term that incorporates the damping effect of solvent interactions. The fluctuating force obeys the fluctuation-dissipation theorem. For a strong friction term, the x M can be neglected. The potential U can be a ssumed as shown in figure 1 -4 with an

PAGE 24

24 angular frequency 0 at initial state and b at top of barrier. Figure 1 7 : Schematic representation of barrier crossing events in Kramers model shows a particle of mass M in a double wel led potential described by U(x) with minima at x =A and x = C respectively. The a ngular frequencies at initial state x=A and barrier x=B are 0 and b respectively The escape rate for strong friction (large b) is determined from the probability density of particle crossing the barrier as (26) ) / exp( 20T k E kB b b [1 6 ] The energy of activation Eb = U(x = B) U(x=A) is expressed in kJ /mole The damping term could be friction due to solvent vis cosity, which limits the bulk motion of the polypeptide chain in solvent The rate of protein folding would then scale inversely with solvent viscosity Experimental measurements of protein folding kinetics have attempted to verify this dependence of f olding r ates on solvent viscosity ( / 1 Fk ) by adding viscous co -solutes to buffers. One problem with these studies is that the added co-solutes shift the stability of the folded state of the protein. In studies where additional measures were taken to U(x) x x = A x = B k + k x = C 0 b b M

PAGE 25

25 reverse this shift in stability, the inverse of the folding rate was observed to scale with solvent viscosity in a linear fashion (27, 28, 29, 30) Since experimental studies cannot probe conditions of very low solvent viscosity ( ), the kinetic results (1 Fk) can be extrapolated to zero solvent viscosity, to estimate the limiting rate of protein folding. As the fol ding kinetics are dominated more by internal reconfiguration of polypeptide chain than by bulk diffusion of chain in solvent. These interactions are weakly coupled to solvent and limit the folding rates. Internal friction is used to describe the contribu tion of these interactions to the overall friction affecting the protein. Early studies on polymer chains have indicated mechanisms by which a polymer chain can experience drag forces that do not arise from changes in solvent viscosity. These include the potential energy barriers to backbone rotations, longrange interactions between residues and the accessibility of free volume in a non -continuum solvent. Internal Friction The internal friction of the protein retards the motion of protein segments (27) relative to each other, and plays a key role in protein dynamics at low solvent viscosities. The consequences of internal friction have been interpreted in different ways by different models. Ansaris study of foldin g kinetics of myoglobin (27) observed a finite folding rate at low solvent viscosity. They used a modified Kramers model with an additional viscosity term in the reaction friction, as shown in equation 1 3. ) / exp( T k E A kB a F [1 7 ] Applying this model to the data for myoglobin, they obtained ~ 4.1 1.3 mPa s, 4 larger than the viscosity of water. More recent attempts to fit experimental data for protein

PAGE 26

26 folding to equation 1 3 (31, 32, 33, 34, 35) have not always provided conclusive evidence for a positive This has led to concerns that the interpretation of internal fr iction as a viscosity is flawed (28, 36) Instead, the protein folding time can be viewed as the sum of two timescales, one of which is solvent -controlled while the other is dominated by the effects of internal friction (28, 29, 36, 37) In equation 14, S reflects the timescale of protein conformational dynamics affected by changes in solvent viscosity, while int reflects the timescale of events that are insensitive to changes in solvent viscosity. int 1 S Fk [1 8 ] The timescale of internal friction (~ 106/s) has been seen to be slower (28, 29) than that of simple diffusion of an ide al polypeptide chain (~ 100 ns) It thus sets an upper limit to protein foldi ng that differs from diffusion limited rates for contact formation, hydrophobic collapse and similar bulk motions of the chain. Internal friction has a notable influence on folding dynamics at low solvent viscosities for proteins folding at relatively hi gh rates. This also explains the apparent absence of internal friction effects in proteins that fold on slower millisecond timescales (37, 38) Theoretical studies can offer an insight into the events underlying the folding process at different conditions of solvent viscosity. Simulations with lattice models (39) have shown the existence of multiple regimes for diffusional motion of peptides at different s olvent viscosities. It has been shown that the folding times scale linearly with solvent viscosity in the region of high solvent viscosities, but at low viscosities the scale factor is not unity (40) This suggests the weak coupl ing of folding dynamics to the solvent friction A recent simulation of folding dynamics for the protein BBA5 (41) has indicated a subtle shift in folding mechanism adopted by the protein at different c onditions of solvent viscosity. This study suggests that at low solvent

PAGE 27

27 friction, the protein adopts a folding pathway characterized by fast long range interactions. When solvent friction is large, short -range interactions are formed easily. Both the solvent and the protein determine the actual folding mechanism, as protein folding is a polymeric diffusive reaction affected by extrinsic solvent diffusion and intrinsic chain diffusion. This opens up the possibilities of observation of multiple pathways of folding for a protein by twe aking the solvent viscosity. I study the effects of solven t friction on the multi -state folding of the tryptophan zipper TZ2 (discussed in chapter 4 ). Intrinsically Disordered P roteins In the past few decades, protein -foldi ng research has largely focused on proteins with a well -defined structure. An awareness of the existence of functional proteins that lack a well defined three -dimensional structure has come about only in the last two decades (42, 43, 44, 45) (46, 47, 48) These proteins are termed as natively unfolded, natively unstruct ured, intrinsically disordered and intrinsically unstructured in the literature. The presence of a dynamic ensemble of conformations for the natively unfolded peptides poses a challenge to the sequence -structure -function paradigm (49) (50) in protein folding that asserts that a proteins amino acid sequence codes for one specific conformation that determines the proteins specific function. Disorder seems to confer several benefits to these proteins. The flexibility of disordered proteins enables them to interact with multiple molecules at different sites with high specificity and low affinity (42, 44, 45, 48, 49, 51, 52) because of their ability to rearrange their conformation to interact with a specific target molecule. Reduced steric constraints also en able faster association rates and easier dissociation by unzipping mechanisms (51) They have larger surface areas of interaction than conventionally folded proteins (53) due to their ability to wrap up or surround their partner molecule. They are also more tolerant of changes in pH and

PAGE 28

28 temperature, which might cause destabilization of well -folded proteins (45) Disordered regions are more amenable to regulation by molecules or proteins, which bind to them at specific sites. This is used for regulation of several signaling and degradation pathways (42, 49) Natively disordere d proteins are involved in vital biological functions that include molecular recognition (51, 54) regulation of transcription and translation (42, 55, 56) assembly and proteolytic regulation (42, 52) To perform these diverse functions, most disordered proteins have to bind to other proteins (50, 55, 57, 58) metal ions (59) (60) and radicals (61) They also undergo a disorder to order transition on binding to their targets (62) The exact mechanism of this concerted folding and binding is not very well understood. While the structural features of the se peptides prior to and after the coupled folding and binding interaction have been studied exhaustively (63) (58) few theoretical studies (64, 65) (66, 67) and fewer experimental strategies (68, 69) have focused on the hierarchy of events in this int eraction. This chapter reviews the literature on intrinsically disordered proteins and the handful of experiments focused on the coupled folding and binding interaction of intrinsically disordered proteins with their target molecules. Finally, we shall i ntroduce the intrinsically disordered peptide IA3 and its target molecule, the enzyme YPrA. We have used temperature -jump triggered time resolved fluorescence to monitor the binding induced helix folding in the natively unfolded IA3. We present the resul ts and conclusions of our experimental studies in detail in the next chapter. Definition of Disorder Order or structure in a protein is defined by the degrees of freedom available to the residues of the protein. These degrees of freedom are characteriz ed by the bond angles for the C C and C N bonds and the atom positions that limit the conformations available to each residue. Intrinsically disordered proteins exhibit a dynamic ensemble of conformations (44) (46) available to each constituent residue under physiological conditions of pH and temperature. The existence

PAGE 29

29 of disorder is thus determined by the dynamical properties of the protein and not necessarily by the presence or absence of local secondary structure. Disorder could be local e.g. a short region in a protein or global (full -length protein) (42) Characteristics of I ntrinsically D isordered P roteins Intrinsically disordered p roteins (70) (71) have an amino acid composition that is different from that of their folded counterparts. They are depleted in hydrophobic and aromatic residues (ordering re sidues), and enriched in charged hydrophilic (disorder promoting) residues such as glutamine, glycine etc (71) (72) A high (usually negative) charge and low hydrophobicity is a typical attribute of an intrinsically disordered protein (46), (44) The solvent exposed surface area for each residue in an intrinsically disordered protein is much l arger than the solvent exposed area for residues in structured proteins. This enables a larger surface area of interaction per residue (73) The ratio of interacting surface area per residue to total surface are a per residue is as high as 50% in intrinsically unstructured proteins (70) compared to only 5 15% in well -structured proteins. This implies that intrinsically disordered proteins use a large portion of their surfa ce area to interact with other proteins or molecules (74) In comparison to ordered proteins, intrinsically unstructured proteins have more hydrophobic residues exposed to surface than buried within. Their presence in exposed surface area of intrinsically disordered proteins indicates their role in interactions with their target proteins that ultimately stabilize the fold of the intrinsically disordered peptide (74) (73) To summarize, intrinsically disordered proteins can be distinguished from their structured counterparts by their larger interacting surfaces and interactions mediated by hydrophobic residues in interacting regions. This s uggests a unique mode of interaction where hydrophobic residues make contacts during binding that could promote stabilization of the protein core

PAGE 30

30 Functions of Intrinsically Disordered Proteins Intrinsically disordered proteins are involved in several v ital processes like bio -molecular recognition, regulation of transcription, translation, signaling pathways, and proteolysis (42, 54, 57) The absence of a rigid struct ure enables them to associate with multiple target molecules with high specificity and low affinity. Their ease of interaction with their targets also leads to faster rates of interaction. Intrinsically disordered proteins are involved in nucleic acid r ecognition (48, 51, 52, 57) (75) and unwinding and bending of DNA. Several transcriptional regulators like p3 00 and p27 (42, 55) have domains that are completely unfolded and on binding to their targets, the binding stabilizes their tertiary structures. The relative instability of intrinsically disordered pr oteins involved in transcriptional regulation and signaling enables a higher level of control through proteolytic degradation (42, 51) Post -translational modification of complexes of proteins i s also easier with the flexible disordered peptides (45, 54) Simple biological switching is possible by covalent modification of binding of intrinsically disordered proteins with their tar gets. Chemical modification of side chains requires a close association of target protein and modifying enzyme. Steric hindrances affect the rate of this association if the side chain lies within a structured region. A disordered region facilitates subs trate binding with the side chain, as seen in several proteins with disordered regions undergoing acetylation, methylation, phosphorylation and glycosylation (51) The binding of intrinsically disordered regions or peptides with specific biological molecules regulates inhibition or activation of processes like degradation. We study the inhibition of a protein degrading enzyme in yeast, YPrA (76) by an intrinsically disordered IA3 pe ptide in the same yeast cell. Structural studies of the IA3YPrA complex show that the N terminus of the IA3 peptide is folded into an alpha helix that occupies the active site of the

PAGE 31

31 enzyme. Since access to the enzymes active site is blocked for all po tential substrate molecules, the enzyme activity is inhibited. Theoretical Studies of Coupled Folding and Binding The functionality of intrinsically disordered proteins is linked to their coupled folding and binding interaction with their target protei ns. Some theoretical studies have employed energy landscape theory to model the association of a natively unfolded protein with another protein. One of the most cited models is this regard is the fly -casting model (65) proposed by Shoemaker et al which suggests that an unstructured protein can make non -specific contacts with its target, which help stabilize the folded state of the unstructured peptide. The unstructured protein is visualized as a fishing line ca st out to explore its (figure 1 6) environment. Figure 1 8 : Schematic illustration of fly -casting model, which shows the extended disordered peptide (in blue) with a larger reach and a better chance of initiating contact with the target biological molecule shown in brown. The more compact peptide (in black) cannot explore its environment as fast as the disordered peptide can. The model suggests that the disordered peptide binds to its target via long range non-specific conta cts that help stabilize its folded state. Target Bimolecule

PAGE 32

3 2 It can diffuse through this space in a shorter time than a compact structured protein, enabling faster target recognition. The initiation of binding by the formation of non-specific contacts could be an enhancemen t over binding in well -folded proteins by formation of specific contacts. Simulations of the coupled folding and binding interaction of p27 with cdk42 (64) have revealed the presence of transient bound intermed iates that precede the folding of the natively unfolded p27. A recent experiment on coupled folding and binding kinetics (68) in the intrinsically unstructured CREB peptide also suggests a fly -casting -like mode of inte raction. Another model for coupled folding and binding is the model of conformational selection. It suggests that an unstructured peptide folds first (77 78) and then binds to its target by making specific contacts. An interesting comparison of binding in natively unfolded proteins with binding in well folded proteins suggests that the roughness of the folding landscape of the unstructured peptide may determine the binding mech anism adopted by it (79) The free energy landscape for structured proteins showed two narrow minima in folding reaction co -ordinate with one entropic barrier between unbound and bound states. This would imply the f ormation of specific contacts by a peptide with a well -defined fold, to stabilize its complex with its target (80) (conformational selection). Unstructured proteins had an energy landscape characterized by a broad val ley with several minima implying transient intermediates. Promiscuous contacts made by the unstructured peptide with its target protein (flycasting model) in these intermediates stabilize the fold of the unstructured peptide. Experiments on Coupled F old ing and Binding in N atively Disordered P roteins Experimental studies of coupled folding and binding protein associations have usually focused on structural features of the interacting proteins prior to and after the reaction. We are

PAGE 33

33 aware of a few experiments that characterize the mechanism of the interaction. Cofactor assisted folding of flavodoxin from Desulphuris sulphuricans in the presence of the FMN cofactor has been studied (81) in mixing experiments. The folding of the flavodoxin protein proceeds via the fast formation of a bound intermediate followed by the slower conversion to the folded state. The authors of this study also reported a ten -fold acceleration in the folding rate due the binding interactio n. Recent studies (68) of the interaction of the unstructured pKID domain of the CREB protein with the KIX domain of CBP have shown that an encounter complex with non -specific contacts is formed prior to the folding of the unstructured domain. Simulations in a later study (82) have reiterated these findings. Studies of the disordered TC 1 peptide with Chibby protein (83) have revealed regions of high helical propensity in the TC 1 peptide that conformationally rearrange themselves prior to binding. Another study on folding in intrinsically unstructured mutants of the SNase proteins (69) has revealed the preference of some mutants to fold prior to binding, and others to bind before folding. Intrinsically Disordered P eptide IA3 and the Yeast Aspartic Proteinase A (YPrA) We study the coupled folding and binding interaction of the intrinsically disordered p eptide IA3 (PDB ID 1dpj) with the proteolytic enzyme, aspartic proteinase A or YPrA (PDB ID 2jxr) in yeast. The aspartic proteinase cleaves proteins and is localized to the cellular vesicles. The IA3 peptide is found in the cytoplasm of the same cell suggesting that it does not encounter the proteinase unless the vesicles containing YPrA are ruptured. In the event of a vesicular rupture, the IA3 peptide binds to the proteinase at its active site. This blocks access by any other substrate to the active si te of the proteinase, thereby inhibiting the protease. Thus, the disordered peptide IA3 effectively protects the cellular machinery from possible damage by the protease. It is unique as an endogenous inhibitor (84) o f an enzyme, i.e. the same cell that expresses the enzyme also expresses its inhibitor. Pure IA3 is unstructured (85) in solution. When bound to

PAGE 34

34 YPrA, it is observed that the N terminal residues of IA3 (76) are folded into an alpha helix. One of the simplest questions one can ask about the YPrA IA3 interaction is whether binding precedes folding or folding precedes binding. In one simple scheme, a small subpopulation of IA3 that is already folded to a helix can bind to YPrA. This drives the equilibrium towards the folded, bound state. The other possibility is that free IA3 remains unfolded, but is able to make non -specific contacts with YPrA that facilitate the folding transition. Our study of the folding kinetics of the intrinsically unstructured IA3 in the absence and presence of YPrA enables us to test these two scenarios. Intrinsically D isordered P eptide IA3 IA3 is a 7.7 kDa peptide of 68 amino acids found in the cytoplasm of the yeast cell Saccharomyces cerevisiae (84) The amino acid sequence of wild type IA3 shown in the figure 4.2 (86 87) shows 47 polar residues and 6 aromatic acids. Pure IA3 is unstructured as evidenced by circular dichroism and nuclear magnetic resonance studies (85) Crystallography studies of the complex of IA3 bound to YPrA (76) indicate that the N -terminus of IA3 (residues 2 32) is folded into an alpha helix, with the hydrophobic residues facing into the active site of the enzyme. This alpha helix occupies the space in the active cleft of the protease. IA3 is very specific to YPrA (88, 89) to which it binds with sub nanomolar binding affinity. It does not inhibit other proteases that are structurally similar to YPrA. In fact IA3 is cleaved by these other similar proteases. Experiments with truncated versions of the wild-type IA3 show that the N terminus alone is capable of inhibiting the protease with the same affinity as the wild -type IA3 peptide (89, 90) While the N terminus is vital to the inhibitory action, it is believed that the C terminus also plays a role in the binding interaction of YPrA and IA3.

PAGE 35

35 Yeast Aspartic Proteinase A (YPrA) Aspartic proteinases are enz ymes characterized by a dyad of aspartic acid groups (91) that is necessary for protein digesting action or proteolysis. Structural studies of these proteinases (87, 91, 92) show a molecule with predominant beta sheet content and right handed alpha helical segments that form two lobes, with the active site in the interface of the two lobes. Each lobe contains an aspartic acid residue po sitioned close to the other aspartic acid in the other lobe and together they cleave substrate that is positioned inside the active cleft (92) Most aspartic proteinases have a unique structural motif, termed the flap and consisting of a beta hairpin loop extending over the active cleft and implicated in catalytic action. Figure 1 9: Complex of IA3 (red alpha helix) with YPrA (blue), with polysaccharide chain attached to Asn 266 and aspartic groups of Asp32 and As p215 near the active site. Reproduced with permission from Yeast (2007), 24, 467480 The flap region (91) is the most flexible element of the structure and has been reported to move by as much as 8.7 among differe nt crystals structures of the enzyme. Superimposition

PAGE 36

36 of crystal structures of free enzyme onto a crystal structure of enzyme inhibitor complex shows important differences in the conformation of the flap, particularly the orientation of a tyrosine residue (Tyr75) that is conserved in all proteinases. YPrA is an aspartic proteinase that is synthesized as a 405 amino acid peptide in the ribosome of the yeast cell ( Sachharomyces cerevisiae ), as an inactive enzyme or zymogen. This peptide is then shuttled to the acidic vesicles where it is activated after cleavage of a segment of the peptide. The final mature peptide (42 kilo Dalton) has 329 amino acids (43% polar residues and 12% aromatic re sidues) and is localized to acidic vesicle s Mature YPrA is glycos ylated with sugars attached to certain amino acids (Asn67 and Asn266) (93 94) as in figure 1 7. We have surveyed the literature on intrinsically disordered peptides and seen several examples where biological function hinges on the association of proteins via a disorder to order transition. This concerted folding and binding mechanism is observed for the IA3-YPrA binding interaction. Prior studies have focused on the structur e of IA3 in its free and folded forms. The exact sequence of events that is responsible for YPrA inhibition is still not clearly understood. Our study of the kinetics of the IA3YPrA interaction in chapter 3 addresses the simple question of whether bind ing precedes folding, or folding precedes binding. We monitor the folding transition of IA3 in the absence and presence of YPrA, so that we may compare possible schemes for the IA3YPrA interaction. Scope of T his D issertation I study fast dynamics of protein folding and binding using nanosecond time resolved spectroscopic methods such as absorption and fluorescence spectroscopy. Three projects are presented in this dissertation

PAGE 37

37 (1 ) Studies of coupled folding and binding interaction between the intrinsi cally disordered peptide IA3 and its binding partner YPrA : These studies combine equilibrium circular dichroism and fluorescence spectroscopy with kinetic measurements using a temperature jump system. (2 ) Studies of solvent viscosity dependence of foldi ng kine tics of tryptophan zipper TZ2: These studies combine equilibrium circular dichroism and kinetic studies with temperature jump triggered time resolved fluorescence spectroscopy. (3) Time resolved transient absorption studies of contact formation in polypeptide chains to estimate the upper limit to protein folding Following this introductory chapter, I shall review experimental techniques used to monitor the protein folding reaction in chapter 2 with a focus on techniques capable of probing the sub-m illisecond response of a protein to a thermal, chemical or optical perturbation. Subsequent chapters provide a detailed description of the different projects mentioned above. Chapter 3 provides a detailed description of our kinetic studies of the coupled folding and binding interaction between the protease YPrA and its intrinsically disordered inhibitor IA3. I study solvent viscosity effects on beta hairpin formation in the tryptophan zipper, a mini -protein of twelve residues designed to form a beta hair pin, in chapter 4. S tudies of contact formation in polypeptide chains are discussed in detail in chapter 5 The concluding chapter, Chapter 6 is a general summary of results and conclusions. An appendix on methods of numerical analysis is provided to su pplement the information on data analysis in all the projects.

PAGE 38

38 CHAPTER 2 EXPERIMENTAL STUDIES OF FAST FOLDING KINETICS Introduction Experiments on fast folding of proteins have contributed greatly to our knowledge of the early events (2, 10, 95, 96) in this process. Advances in genetic engineering (10, 97, 98, 99, 100) have enabled expression of site -directed mutants with fluorescent probes of the folding transition. The adven t of pulsed lasers (101, 102, 103) has o pened up the opportunity of probing events in the sub-microsecond timescale. We study folding on sub-millisecond timescales (11, 95, 101, 104, 105, 106) because the earliest events in the folding of small proteins, which set the pace for the entire folding process occur on these very timescales. These processes could be the elementary steps (12, 23, 107) in the folding reaction such as loop formation (12) turn formation (108, 109, 110) a nd helix nucleation (107, 111) Figure 2 1: Schematic of two -state folding in a protein with a free energy difference G between folded state F and unfolded state U Protein folding can be treated as a chemical reaction that proceeds from unfolded state to the folded state, after crossing an ener gy barrier as shown in figure 2 1 (1 12) The folding of U F G F U Free energy Reaction Coordinate k F k U

PAGE 39

39 many proteins can be treated as a two -state transition where the Gibbs free energy difference between unfolded state U and folded state F is GF U In addition, the rates of folding and unfolding are kF and kU respectively (113) [2 1 ] The equilibrium constant for the folding reaction KUF is given by ) / exp( ] [ ] [ T k G k k U F KB U F U F m equilibriu F U [2 2 ] Here kB is the Boltzmann c onstant and T is the temperature in Kelvin. At any time t the rate of change of population of folded molecules [F] is given by ] [ ] [ ] [ F k U k dt F dU F [2 3] Since the total population of protein is fixed to C, the Eq. 2 3 can be written as )) exp( 1 ))( /( ( ] [ ][ ) exp( )} exp( ] {[ ; )] 0 ( [ ); exp( )} exp( ] {[ ) exp( )exp( ] }[ { ] [ ] [ ) ( ] )[ ( ] [ ]} [ { ] [ ][ ] [ 0 0 0t k k C k F F dt t C k t F d F tF t C k t F dt d t C k t F dt d C k F dt F dk k k F k k C k F k F C k dt F d F U C U F F t F F F F F F observed U F U F F U Fo [2 4 ] U F k F k U

PAGE 40

40 The solution for this differential equation is a single exponential with relaxation rate For a known free energy of unfolding and unfolding rates ( kF, kU) can be determined. Experiments designed to determine usually employ chemical, optical or thermal means to destabilize the protein from its folded state, and then probe its relaxation back to equilibrium. We shall first rev iew the commonly used probes of conformational change in proteins and then review commonly used techniques to trigger conformational change of the protein. Probes of Conformational Change i n Proteins When electromagnetic radiation is incident on a sample, the properties of the radiation that emerge from the sample can provide structural information about the sample. The simplest measurable property is the fraction of excitation absorbed (optical absorption). Radiation emitted by a sample at wavelengths o ther than the excitation wavelength can also be monitored (i.e. fluorescence, phosphorescence and Raman scattering). Circular dichroism is the phenomenon of differential absorption of left and right circularly polarized light by chiral molecules. We use circular dichroism spectroscopy to determine the secondary structure content of the IA3 peptide and tryptophan zipper in different solvent conditions, and time resolved fluorescence spectroscopy to measure the folding kinetics of these proteins. Fluoresce nce and Triplet Absorption Spectroscopy A qualitative study of these techniques would require a quantum mechanical description of the electronic states of the molecules in the sample before and after their interaction with the radiation (114) If we assume two states a and b described by the wave -functions a b, respectively, the intensity of emission or absorption associated with a transition between these two states depends on the transition dipole moment defined by a b ~ and the magnitude of

PAGE 41

41 the incident electric field that induces an electric dipole moment in the sample. The energy gap and subsequently the wavelength of maximal absorption between these two states is a consequence (114) of this interaction between electric field of inc ident light and sample. Different electronic transitions in proteins and polypeptides (115) give rise to their unique absorption spectra. A transition from the ground state to the excited state ( *) occurs i n the long-wave region while the presence of heteroatoms such as nitrogen, oxygen or sulphur enables n and n transitions. The delocalization of pi electrons also results in a unique spectral response. Figure 2 2: Jablonski diagram showing the int eraction of light with a sample and the different possible pathways for de -excitation. Absorption of light (red arrow) causes excitation of molecule (optical probe) to excited electronic singlet state S2. It could relax back to electronic ground state S0 by fluorescent emission (blue arrow at right). An intersystem crossing (pink dashed arrow) causes excitation of triplet state, which could relax back to ground state by phosphorescent (blue arrow at left) emission. Figure 2 2 illustrates the several poss ible de -excitation pathways available to a molecule after photo-excitation. The excited state can relax back to the ground state by fluorescence on nanosecond timescales, or it can be excited to a different spin state, the triplet state from which it can relax back by phosphorescence on s -ms timescales. The fraction of excited singlet states that relax by fluorescence (116) is termed as fluorescence yield F. Similarly, the fraction of

PAGE 42

42 excited singlet states that are converted to excited t riplet states is termed triplet yield T. The violation of spin conservation principles in the singlet -triplet conversion results in a low probability of phosphorescent emission, and thus low phosphorescence intensity and a long triplet relaxation lifetim e (115) De excitation of these excited states occurs by an electron transfer mechanism after diffusion limited collision with oxygen triplet and other quenchers in solvent. The triplet state lifetime of a mole cule can be employed as a sensitive and specific probe of collisional quenching. We employ this strategy in our measurements of contact formation in polypeptides (chapter 3). Figure 2 3: Effects of solvent exposure on fluorescence of free tryptophan a re illustrated for various solvent conditions. TFE or 2, 2 ,2, trifluoroethanol as a solvent is less polar than water. Increased amounts of TFE in the environment of tryptophan cause an increase in fluorescence intensity and a small blue shift in the peak wavelength of tryptophan emission. The emission spectrum of these excited states is also affected by exposure to solvent. Electric dipole interaction with solvent molecules affects the dipole strength and thus the energy difference between excited and g round states. The fluorescent emission spectrum intensity and peak wavelength are altered, making fluorescence a good probe of solvent exposure (116) during folding of proteins (figure 2 3).

PAGE 43

43 300 350 400 450 500 550 0 0.002 0.004 0.006 0.008 0.01 Wavelength (nm) Fluorescence (a.u.) 0%TFE 8% TFE 12% TFE A r D D Energy Transfe r b etween F luorescent M olecules A pair of natural or synthetic fluorophores tagged to a protein can exchange energy through a coupled electric dipole electric dipole interaction (117) This transfer is termed as fluo rescent resonant energy transfer or FRET and is a non -radiative transfe r of energy. Figure 2 4: Energy transfer between two fluorophores tryptophan and dansyl attached to N2W K16C dansyl IA3 peptide increases as distance between them decre ases due to folding of peptide. This is evident from reducing fluorescence emission at 350 nm (donor) and increasing fluorescence emission at 495 nm (acceptor). The spectra are acquired at room temperature for peptide in different concentrations of helix -promoting co solvent TFE. The inset shows the orientation of acceptor and donor dipoles A, D respectively with respect to distance vector between them ( D, A) and the planes containing the dipoles ( ) The induced electric dipole moments in donor molecule D and acceptor molecule A are D and A respectively. V gives the dipole -dipole interaction between donor and acceptor and the rate of energy transfer k FRET is proportional to the expression in Eq 2 5 A The factor 2 contains

PAGE 44

44 all the effects of orien tation of donor and acceptor molecules on dipole strengths and cannot be measured directly. Figure 2 4 (inset) illustrates the acceptor dipole and donor dipole oriented along planes, which have an angular separation The donor makes an angle D with th e distance vector r between donor and acceptor, while acceptor makes an angle A with the same. From equation 2 5 A we can see that 2 can take a value from zero to four. Uncertainties in the value of 2 are a major source of error in R0 and the distance s estimated from FRET. A D A D A D A D A D A D A D A DR R R R R R V cos cos 3 cos sin sin / ~ ~ } cos cos 3 cos ) {cos( / ~ ~ / ) ~ )( ~ ( 3 / ~ ~ ~3 3 5 3 [2 5 A ] 2 2 6 2 2~ ~ ) / ( ~Ab A Aa Db D Da FRET Aa Db Ab Da FRETR k V k [2 5B] The efficiency of this energy transfer E varies inversely with the sixth power of distance R between the fluorophores (116, 117) as shown in Eq 2 5B and Eq 2 6 6 / 1 0 4 4 2 5 0 1 6 0) ( ) ( 10 8 8 ) ) / ( 1 ( / 1 d F n R R R k k EA D D D FRET FRET [2 6 ] Donor relaxation in the absence of acceptor is given by D. R0 is the distance between the fluorophores that corresponds to an energy transfer efficiency of 50%. It is defined in terms of the fluorescence yield of donor D refractive index n integral of overlap of donor absorption spectr um and acceptor emission spectrum and the orientation fa ctor 2. The efficiency of energy transfer can also be estimated from a measurement of donor fluorescence intensity at peak emission wavelength in the absence and presence of the acceptor

PAGE 45

45 species. As donor and acceptor come closer and the efficiency of energy transfer increases, the intensity of donor fluorescence falls and acceptor fluorescence rises. Conversely, an increase in the distance between them will be evident by an increase in donor emis sion and a decrease in acceptor emission intensities (Figure 2 4). I use tryptophan as a FRET donor and (1, 5 IAEDANS) dansyl as FRET acceptor in our studies of folding kinetics of the intrinsically unstructured peptide IA3. The Forster constant R0 const ant for the tryptophan -dansyl pair is 2.2 nm (116) while the CC separation between the fluorophores in the mutant IA3 peptide (N2WK16C -dansyl IA3) is 2.03 nm. Figure 2 5: Schematic representation of d esign of IA3 mutants with FRET donor tryptophan and acceptor dansyl in the N terminus at positions 2 and 16. As the N -terminus folds, the FRET labels come closer, enhancing the efficiency of energy transfer. Conversely, the labels move apart as the pepti de unfolds, say after a temperature jump. This causes energy transfer efficiency to drop. Fluorescence spectroscopy can be used to monitor the variation in donor and acceptor fluorescence intensities during the folding transition. Tryptophan has a fluore scence emission peak at 350 nm, while dansyl emits at a peak wavelength of 500 nm. As they come closer, the increased energy transfer causes a drop in tryptophan emission at 350 nm and a rise in the dansyl emission at 500 nm as seen in Figure 2 4. Conver sely, as the peptide unfolds, the distance between the residues is increased, decreasing the efficiency of energy transfer. This is visible in the increased emission of the donor molecule (tryptophan) and reduced emission of acceptor (dansyl).

PAGE 46

46 To study the binding interaction of IA3 with YPrA, we used an acceptor only tagged IA3 mutant peptide (K16C -dansyl IA3), as the YPrA has four naturally occurring tryptophan residues near the IA3 binding site (118) In the crys tal structure of the IA3YPrA complex, three of the tryptophan residues (W39, W190 and W241) are within the Frster distance of residue 16 of IA3 peptide. Residue W181 lies at a di stance of 2.8 nm away (Figure 2 6 ). Figure 2 6: View of the inhibition c omplex of YPrA (blue) with IA3 (gray) based on crystal structure data (76) The N terminal residues form a helix, with cysteine of K16C mutant shown in orange. The tryptophan residues of YPrA near the active site are sho wn in red. Collection of Fluorescence S pectra All equilibrium fluorescence data for our experiments are acquired on a JASCO FP 750 spectropolarimeter. Lyophilized protein is dissolved in 50 mM sodium phosphate (pH 7.0) to make a buffered solution with ~ 1520 M protein. This solution is taken in a cell with an optical path length of 10 mm Fluorescence spectra are acquired over wavelengths 260500 nm after excitation by light of wavelength of 266 nm. Sample temperature can be controlled to a

PAGE 47

47 precision of 0.1oC with the use of a circulating water bath (Neslab). All samples are sealed to eliminate evaporation loss and stirred continuously during data acquisition. Circular Dichroism S pectroscopy We understand fluorescence in terms of the electric eff ects of light incident on the sample. The effects of the magnetic field of the incident light wave are negligibly small in such cases. There are molecules in which magnetic effects due to the oscillating magnetic field of the incident radiation (114) induces a substantial magnetic dipole m Both the electric dipole and magnetic dipole m act on the wave functions representing the electronic states giving rise to the phenomenon of optical activity. Circular dichroism is one manifestation of optical activity. Circular dichroism (CD) is the phenomenon of differential absorbance of left and right circularly polarized light by an optically active sample. If a beam of light is incident on a sample that absorbs right and left circularly polarized light equally, there is no effect on the polarization state of the transmitted light. When the left and right circularly polarized light waves are differentially absorbed however, an elliptical polarized light wave is obtained. Therefore, CD measurements are measurements of the ellipticity of resultant polarized light. The minor and major axes of the ellipse traced by the resultant light wave (electric field vector E with subscripts L and R for le ft and right circular polarization) depend on the extinction coefficients for left and right circularly polarized light L and R respectively, as in Eq uation 2 7 The intensity of the left and right polarized light is proportional to the magnitude of the corresponding electric field vector as shown in equation 2 7. The intensity of signal after differential absorption ca n be estimated from the extinction coefficients for the left and right circularly polarized light protein concentration C and optical path length l respectively. Values

PAGE 48

48 for CD for peptides are also reported in terms of molar ellipticity and molar ellip ticity per residue. 4 / 180 )( 303 2 ) 2 / ) exp(( 1 ) 2 / ) exp(( 1R L R L R L R L R L R L RLA A Cl Cl I I I I E E E E [2 7 ] The circular dichroism signal of an isolated optically active molecule is not usually very in formative about structure The CD signal of a group of such molecules in a macromolecule is en riched by the contributions of electric dipole -magnetic dipole coupling and an exciton coupling. The exciton coupling ( *) dominates the CD spectrum and has a value that depends on distance between the monomers and their respective orientation. Prote ins typically absorb light only in the ultra -violet wavelength range ( < 300 nm). The chemical groups that dominate the observed spectra for proteins are peptide bonds, amino acid side chains and groups such as hem e (114, 115) etc. The peptide group contains a delocalized electron cloud that extends over the peptide nitrogen, carbon and oxygen. The transition is the most easily observable transi tion at 190nm, followed by an n transition at 210 220 nm. The amino acids that contribute maximally to absorption spectra of peptides are tryptophan, tyrosine and phenylalanine (115) The handedness of alpha helices and betaturns ena bles circular dichroism studies of protein secondary structure in the wavelength 170 260 nm and tertiary structure near -UV circular dichroism signal (260 330 nm) (112) Alpha helices have a characteristic signal at 208 nm and 220 nm, while beta turns have a characteristic CD signal at 212 nm. Changes in temperature or chemical environment alter the CD spectrum, making it an excellent probe of changes in secondary structure.

PAGE 49

49 200 210 220 230 240 -500 0 500 1000 Wavelength (nm)CD (mdeg) cm2/dmol 210 215 220 225 230 235 240 245 250 -10 0 10 Wavelength (nm)CD (mdeg/cm/M) Temperature Isodichroic point A B Temperature Temperature Isodichroic point TZ2 in 50 mM pH 7.0 phosphate + 2M GdnHCl IA3 in 50 mM pH 7.0 phosphate + 25% TFE v/v Figure 2 7 : CD spectrum of alpha he lical peptide IA3 thermally denatured to random -coil in subplot A shows isodichroic point at 202 nm, implying a two state folding transition. Panel B is the CD spectrum monitoring the thermal denaturation of beta -hairpin forming mini -protein tryptophan zi pper. A large CD signal is seen at 227 nm due to exciton splitting. As temperature increases, the exciton splitting is reduced as are the intensities of peaks at 215 nm and 227 nm. I study the helix-coil transition in the intrinsically unstructured IA3 peptide at various temperatures and concentrations of 2, 2, 2 trifluoroethanol or TFE (a helix promoter discussed in chapter 3). Figure 2 7 A shows the loss of helical structure content with increased temperature. I also use circular dichroism to investigate the effects of ethylene glycol and temperature on the beta hairpin folding transition of the tryptophan zipper protein TZ2. Figure 2 7 B shows an exciton peak at 227 nm and minimum at 215 nm, which indicates a splitting of exciton levels due to interac tion between the pi -electron clouds of the tryptophan residues holding together the tryptophan zipper. The intensity of the exciton peaks is reduced with increased temperature, as

PAGE 50

50 TZ2 unfolds. The CD curves for the thermal transition of IA3 and TZ2 inter sect at a single wavelength, termed as isodichroic wavelength. The presence of an isodichroic point is indicative of a two -state folding transition in proteins. Collection of CD spectra Spectra were collected on an AVIV 202 CD spectropolarimeter at wavel engths 195240 nm in a 1cm path length cell over temperatures 5 85oC (5oC increments). A reference spectrum (solvent) was subtracted from each sample spectrum as a background correction. Equilibrium far -UV circular dichroism (CD) spectra were collected f or all IA3 peptides at ~ 15 M concentration in 50 mM pH 7.0 phosphate buffer and 0 25% TFE v /v (increments of 5% TFE v/v). I acquir e circular dichroism data for 32 M TZ2 in 50 mM pH 7.0 phosphate buffer with varying concentrations of ethylene glycol (percentage by weight) from 0 5 0% in increments of 12.5% by weight over temperatures 5 95oC and wavelengths 200260 nm. Tryptophan P hoto -p hysics The projects discussed in this dissertation focus on studies of peptides that have tryptophan as a probe of conformational change during the folding transition of the peptide under different experimental conditions. We briefly review the properties of tryptophan that make it an excellent spectroscopic probe for our experiments. The side chain of the amino acid tryptophan is an indole molec ule with a substitution at the third carbon atom as shown in figure 2 8 It has ten electrons delocalized about the aromatic groups, and a nitrogen atom that contribute to a complex absorption spectrum. A peak at 280 nm with a wing at 271273 nm (115) characterizes its absorption spectrum The fluorescence spectrum is broad, featureless and centered around 348 350 nm. The fluorescence quantum yield

PAGE 51

51 is only 0.20 and is affected by intra -molecular and inter -mol ecular quenching by other side groups in the tryptophan molecule. Figure 2 8: Tryptophan structure shows the indole ring with substitution at ca rbon 3, with carbon atoms in gree n, nitrogen atoms in blue, oxygen atoms in red and hydrogen atoms in white. Tryptophan F luorescence T ryptophan fluorescence is very sensitive to the environment of its side chain (114, 115, 116) The delocalized pi -electron clou d of tryptophan is rich in electrons and forms hydrogen bonds with solvent molecules that can donate a proton. A change of solvent polarity alters the strength of these bonds, leading to a blue shift of peak wavelength in less polar solvents (figure 2 3) and a red shift in more polar solvents. Fluorescence intensity of tryptophan is also affected by quenching due to solvent exposure. These specific features of tryptophan fluorescence are used to probe conformational changes in proteins, e.g. unfolding of protein can expose tryptophan residues in the interior to solvent, reducing their emission intensity as well as causing a shift in peak wavelength (119) The tryptophan molecule can exchange energy with other

PAGE 52

52 fluoroph ores such as dansyl and fluorescein by a distance -dependent dipole -dipole interaction. The efficiency of this energy transfer can be measured by measurements of donor and acceptor fluorescence intensity over time. We use it as a FRET donor with dansyl as the acceptor in our kinetic studies of the coupled folding and binding interaction between IA3 and YPrA. Tryptophan Triplet R elaxation Tryptophan can be excited to a triplet state ( T = 0.27), which has an absorption maximum at 456 nm (120, 121, 122) This transition from excited singlet to excited triplet state is characterized by a relaxation ~ 4 40 s (122, 123) Quenching of the triplet state can occur by electron transfer after diffusion -limited collision with other groups such as the oxygen triplet (121, 124) and the sulphydryl (125) group in the amino acid cysteine. This quenching reduces the tryptophan -triplet relaxation lifetime making it a useful probe of diffusion limited contact formation in polypeptide chains with cysteine and tryptophan monomers (figure 2 9) Figure 2 9: A simplified model for diffusion controlled loop formation in a poly amino acid chain labeled with donor Trp and acceptor cysteine (in red). The UV excitation excites the donor (color change from white to blue). The excited donor diffuses towards acceptor, making a contact, and then transferring energy (color change from blue to white again). Our kinetic studies of contact formatio n in peptides labeled with tryptophan and cysteine employ flash photolysis to excite tryptophan to triplet state. Cysteine is the most efficient T rp* Trp* Trp h Trp k D k D Q Cys Cys Cys Cys*

PAGE 53

53 quencher of the tryptophan triplet state among all amino acids (123, 125) Its quenching rate ~ 5 108/M/s is 400 fold larger than that of other amino acids including tryptophan. Contact with the sulphydryl group of cysteine is a short range interaction involving an electron transfer (123, 124, 125) from the triplet state of tryptophan to the cysteine. Therefore, the rate of the tryptophan triplet decay in the presence of cysteine is a direct probe of the rate of contact formation (22) in a chain labeled with tryptophan and cysteine. This rate can be determined from the relaxation of tryp tophan triplet, as the rate of a tryptophan-cysteine contact is coupled to the dynamics of the chain diff using in solvent. I assume a simple model for the interaction (21) between excited triplet of tryptophan (Trp) and cysteine (Cys). Contact formation is facilitated by diffusion occurring at a rate k D, while k D i s the rate at which the ends diffuse aw ay without forming a contact. It is assumed that once the contact is formed, energy transfer with a quenching rate Q is imminent. ] /[ *] [ D Q Q D observed observedk k k k k dt Trp d k (22) [2 8 ] When the rate of energy transfer (Q) is much faster than the rate of disassociation ( k D -), the observed rate in Eq. 2 8 is the rate of diffusion k D, in Eq.2 9 given in terms of end-to -end distance distributi on for a polymer by SSS theory. 2 / 3 2 / 3 2) ) ( 2 /( 4 ~ ) ( ~ n k nl Da k k k k k kobserved D observed D Q D observed [2 9 ] Trp* + Cys k D Trp*.Cys k D Trp.Cys Q

PAGE 54

54 The formation of a loop is said to be diffusion limited in such a scenario. The rate of diffusion is given by the SSS result in Eq. 1 2 for a Gaussian chain. The contour l ength L can be ascertained from the number of peptides separating the tryptophan and cysteine. Loop formation by contact between these two amino acids occurs at a distance equaling the sum of their van der Waals radii, which replaces a in Eq.1 2 T he eff ective diffusion coefficient of tryptophan in water 7 106cm2/s is used The simple estimate of contact formation time using SSS theory enables us to relate the rate of loop formation with the length of the loop as in Eq.2 9 A number of transients are c reated on photolysis of tryptophan (126, 127, 127) I focus on those species that have a significant contribution to the absorbance signal in the wavelength ra nge of our interest. The triplet yield of tryptophan in water is 0.27 (120, 122) (127) 1 Tryptophan triplet: This species has a peak absorbance at 460 nm (122, 128) and its relaxation to the ground state, which is typically ~ 4 40 s, can occur via several pathways. Trp T Trp (relaxation at 104/s) (122, 126) (121) [2 10] Trp T + Trp T 2 Trp (t riplet annihilation at 1010/M/s) (126) Trp T + Trp quenching (ground state quenching at 107/M/s) (126) Trp T + O2 quenching (@ 5.3 109/M/s) (121, 129) 2 Cation radical: This species is created during photo ionization of tryptophan. The absorption spectrum of this species has a peak at 550 nm. In neutral pH solutions, this decays rapidly within a microsecond. (130) 3 Neutral radical: The cation species deprotonates, giving a neutral radical that exhibits a peak in the absorption spectrum at 510 nm, and decays slowly in a few hundred microseconds. (130, 131) 2 Trp0 2 Trp (radical annihilation @ 7.9 108/M/s) (126, 130, 132) [2 11] 4 Solvated electrons: The photo ion ization of tryptophan also yields solvated electrons, which have a peak in absorption at 720 nm. Degassing samples with nitrous oxide enables efficient scavenging of this species. (122, 133) Th e following equations summarize the photophysics of tryptophan

PAGE 55

55 Trp Trp* Trp+ + e(photo-ionization ) (126, 130, 133) [2 1 2 ] (Excitation) Cation radical) (Solvated electrons) Trp+ Trp0 + H+ (Deprotonation) [2 13] (Neutral radical) (Protons) (122, 131, 132) Trp Trp T (Inter -system crossing after excitation ) [2 14] In experiments, [Trp] ~ 100 M, the ground state quenching mechanism occurs on a timescale of milliseconds, when the processes of interest to us occur in the microsecond scale. The kinetics of relaxation of the tryptophan triplet and the neutral radical are discussed in detail in chapter 5. Techniques Used to Study Protein Folding K inetics Mixing This is one of the oldest techniques used to study folding of proteins. A protein in highly destabilizing solvent conditions is mixed with a buffer, favoring refolding of protein. The kine tics of relaxation of protein to the new equilibrium is probed by an optical probe, such as fluorescence, circular dichroism etc (134) The time resolution of this technique is limited by the time of mixing, which is r arely faster than a few tens of microseconds (135) Continuous flow mixing has enabled the detection and characterization of intermediates along the folding pathway of proteins such as cytochrome c. Flash P hotolysis A laser pulse can be used to trigger a photochemical change, with the wavelength chosen to specifically excite molecules in the protein that serve as optical probes of the folding transition. This photochemical change could be a rupture of specific bonds or fast electron transfer that leads protein to the native state. While flash photolysis has been around for a long while, nanosecond and femtosecond pulsed lasers allow studies of sub microsecond events in

PAGE 56

56 folding. The earliest such study was conducted on a complex of cytochrome c protein (11) and carbon monoxide, which was photo -disassociated by a laser pulse. The release of the carbon monoxide that binds preferentially to the heme group allows the heme group to bind to other ligands in the protein. This enables observation and measurement of rates of initial formation of contacts as protein samples its conformational space. Recent studies of contact formation in polypeptide chains have employed flash photolysis methods to excite triplet states whose relaxation (22, 24, 136, 137, 138) enables an observation of the dynamics of a disordered polypeptide ch ain. We use flash photolysis triggered energy transfer between tryptophan and cysteine in a polypeptide chain to monitor intra -chain diffusion, discussed in chapter 5 A flash of UV light excites the tryptophan molecule to the triplet state, which can b e detected through its visible absorption. Quenching of the triplet state by collisional contact with the sulphydryl group of cysteine enables an actual measurement of contact formation in the chain. The acquisition of time resolved absorption spectra pro vides information about the relaxation of the excited triplet state, which is linked to conformational dynamics of the polypeptide chain. Our transient absorption spectroscopy system (figure 2 10) monitors tryptophan triplet relaxation over wavelengths 400700 nm with nanosecond time resolution on sub-millisecond timescales. Sample is excited by a laser pulse ( = 289 nm, pulse width 7 ns) that is generated by Stimulated Raman emission in hydrogen gas at 650 psig pumped by the fourth harmonic of Nd: YAG laser at = 266 nm. A xenon flash lamp ( = 400700 nm) probes the excitation volume at a programmable time d elay D after the excitation. A kinetic profile of triplet relaxation can be obtained for values of D from a few nanoseconds to a millisecond. The system consists of the following elements

PAGE 57

57 Excitation and probe optics Sample preparation Spectral data acq uisition Excitation S ystem : The 266 nm fourth harmonic of Nd: YAG laser (Spectra Physics INDI 50 10, 5Hz, 57ns) induces stimulated Raman emission in deuterium gas at a pressure of 650 psig in Raman cell (Light Age PAL 101 RC). The first Stokes shift at this wavelength for deuterium is 2991 cm1, providing an output at wavelength of 289 nm. The two wavelengths are separated using two Pellin Broca prisms, following which the 289 nm beam is focused onto the sample cell using a long focus( f = +50 cm) fuse d silica lens. The size of the UV spot on sample at focus is 2.12 mm. Typical energy of this excitation beam is 1 1.6 mJ/pulse. Probe S ystem A xenon flash lamp (EG &G) provides a flash of visible light (400700 nm) at time D after the pump beam. The tim e delay between probe pulse and pump pulse is varied from 10 ns 10 ms by the use of a Lab-view program that controls a SRS DG535 delay generator whose output signals externally trigger the flash lamp trigger and laser Q switch trigger. A combination of beam -splitters and mirrors yields two probe beams, which are focused onto sample. One of these beams is focused onto the excitation volume of sample, while the other beam is focused to a spot located 5 mm below the other. The control track for unphotolyz ed sample provides the reference absorbance intensity I0, with respect to which the transient absorbance I for photolyzed sample is calculated using Beer -Lambert law } 10 {) ( 0ClI I

PAGE 58

58 Figure 2 10: System triggers photolysis in sample with UV pulse at time t, and probes sample response at time delay after the UV flash. The timing and trigger of UV flash and probe xenon flash is controlled remotely. The UV pulse is produced by Raman shifting the 266 nm harmonic of an Nd: YAG laser in a cell filled with deuterium gas at 650 psig. Pellin Broca prism s (PBP) separate the fundamental from Raman -shifted 289 nm light. A fused silica lens (f = +50 cm) focuses the UV light onto sample. The xenon lamp light is focused onto two spots in sample, one of which coincides with the UV focus. The light from both photolysed and unphotolysed sample is focused onto the CCD camera after dispersion by a diffraction grating. The time varying absorbance of sample is calculated at each time delay t from the ratio of the light intensities emerging from the two spots. Here C is concentration of transient (tryptophan triplet in our experiment), l the optical path length of the sample and the extinction coefficient of our transient. The triplet state of tryptophan has = 5000/M/cm at the peak absorption wavelength of 460 nm. We optimize and characterize our system with the absorbance of the organic dye 9anthracene carboxylic acid. Sample P reparation We typically work with concentrations of peptides or NATA that would give 100 M of tryptophan in the ground state and ~ 2 5 M triplet tryptophan, so that we can observe the best

PAGE 59

59 signal for tryptophan triplet in a 1 cm optical path length cell. Samples are prepared in 50 mM pH 7.0 phosphate buffer and purged with nitrous oxide gas for 15 20 minutes prior to data acquisition. The nitrous oxide scavenges electrons produced (122, 129) during experiment as their large absorption signal at 720 nm can interfere with the signal under consideration. It also reduces the concentration of ambient oxygen (129) in sample from 250 M to ~ 20 M. Oxygen is a very efficient quencher (121, 123, 124, 126) of the triplet state of tryptophan. We also use a glucose oxidasecatalase enzyme system (139) to lower ambient oxygen levels in the sample to a few M. Glucose oxidase uses oxygen to break down glucose to give hydrogen peroxide, which is broken down to water and oxygen by the c atalase enzyme. Addition of the enzymes (~20 nM) a few minutes before the actual experiment bring down oxygen to the levels where it does not affect the kinetics of tryptophan triplet relaxation. We also add a layer of mineral oil on top of sample to pre vent any sample interaction with oxygen in atmospheric air during experiment. Samples are stirred constantly with a magnetic stir bar to prevent photo damage of sample in the excitation volume over the time -course of experiment. A thermo electric chip con nected to a temperature controller (MPT 5000, Wavelength Electronics), which can set sample temperature from 5 100 oC with 0.1 oC precision, controls the sample temperature. Spectral Data A cquisition The xenon spectra (control and signal tracks) are focu sed onto the entrance slit of a monochromator (Princeton Instruments Acton) with a diffraction grating, which disperses a spectrum from 400700 nm. This spectrum is now focused onto the CCD array (Roper Scientific Instruments PI -MAX) operating in nanosecond -gated mode. A GG400 filter cuts out light below 340 nm wavelength, to filter out tryptophan fluorescence signal at 350 nm. We calibrate the

PAGE 60

60 ICCD camera for the wavelengths mentioned using the standard lines of the mercury argon penlamp (Oriel 6033). Optical elements focusing light from sample onto spectrograph must be optimally aligned so that the light coming from the signal track and control track have close to equal intensities in the wavelength region of interest. The programmable timing generato r (PTG) triggers camera ON at delays ~ D after excitation and controls timing of spectral acquisition by camera. These delays are programmed to be logarithmically spaced between 10 ns 10ms using Labview. For each delay, 20 40 spectra are acquired to enh ance signal to noise ratio. We use Labview software to trigger the pump and probe beams, set parameters for instruments acquiring data and acquire data. Data Analysis The time delays from the HP counter and the spectra from the CCD are combined by the M atlab program tspec_data_3.m (Hagen SJ) to yield a matrix with data as a function of wavelength along rows and time delays (along columns). Singular value decomposition (explained in appendix A) or SVD resolves the data matrix into independently evolving spectral and temporal eigenvectors. The temporal eigenvectors can be fit to simple exponential relaxations. Temperature jump S pectroscopy Changes in temperature can destabilize the folded state of the protein, allowing for a method of probing their fold ing mechanism. Early T jump methods used a capacitive discharge to heat the protein solvent quickly. This provided a time resolution (140) of 10 s, varying with the electrical conductivity of the solvent. Pulsed IR -laser temperature jump has now become the method of choice to (141) trigger sub-microsecond events in protein folding (102, 142)

PAGE 61

61 Figure 2 1 1 : Response of free tryptophan and the peptide TC5b to a temperature jump of 9.8oC. We do not observe any relaxation in the fluorescence of free tryptophan on the timescales of a few microseconds (panel A). Panel B however shows a clean relaxation of TC5b peptide on these timescales. The rates of formation of secondary structural elements such as alpha helices (143) and beta hairpins (144) have been determined with nanosecond time resolution using the laser triggered temperature jump system. In a typical temperature -jump (T jump) experiment, an IR laser pulse deposits a pulse of heat energy into the solvent. Fluorescence, absorption, Raman spectroscopy, or infrared spectroscopy is then used to probe the relaxation of the protein to a new equilibrium set by the elevated temperature. Our studies of fast folding kinetics in tryptophan zipper and IA3 systems rely on a tempe rature jump to trigger a decrease in stability of the folded state of the protein. The solvent equilibrates rapidly ~ 20 30 ns to the raised temperature, while the protein relaxes to equilibrium on a slower timescale (figure 2 11). We detect fluorescence of tryptophan in IA3 mutant and tryptophan zipper. We discuss the temperature jump apparatus in the next section.

PAGE 62

62 Temperature-jump A pparatus The laser temperature jump apparatus used in the present work (figure 2 12) consists of the following: Infra r ed excitation Ultra violet probe Sample block Sample handling and preparation Kinetic data acquisition Spectral data acquisition Infrared (IR) E xcitation We use an Nd: YAG laser (Continuum Surelite I 10) whose flash lamp and Q -switch are triggered externally by the falling edge of electrical pulses (10Hz, 10 s wide, 5V 0V). The laser output at the fundamental wavelength 1064 nm (2Hz, 5 7 ns) is used to pump a Raman cell (Light Age 101 PAL RC) filled with hydrogen gas. The first Stokes line for stimulated Raman scattering is at 4155 cm1, hence the 1064 nm fundamental output of the Nd:YAG laser is shifted to 1890 nm. Light at the shifted wavelength is separated from the fundamental wavelength by dispersion through a Pellin Broca prism split into two beams that are incident on the sample from opposit e side s (hot spot diameter 1 mm) as shown in figure 2 12 ensuring a homogenous heating of the sample. At 1890 nm, the decay length of the beam in water is 0.03 cm. Temperature jumps of 10 15 oC can be achieved for aqueous solutions by varying the Q -switch de lay setting of the IR laser in the range of 116 120 s.

PAGE 63

63 F1 IR laser Raman Shifter BS UV 266 nm PMT CCD Sample + holder TFP /4 plate PBP VC G M1M2M3M4M5F2L1L2L3L4BS I M6L5M7M8 Figure 2 1 2 : Schematic of temperature jump set up with following elements TFP thin film polarizer BS beam splitter PBP Pellin Broca prism to disperse remaining 532 nm light present in the 266 nm beam M1, M2, M3, M4, M5 IR reflecting mirrors to steer IR beams F1, F2 Schott glass filters 59880 to ensure equal IR energy on both sides of sample L1, L2 IR focusing lens with f = +50cm L3 Objective lens, f = +20 cm to focus sample emission onto PMT L4 UV focusing lens, f = +17.5 cm, L5 Fused silica collimating lens, f = +10 cm M6 flip mirror M7 -M8 Aluminum protected mirrors PMT Hamamatsu photomultiplier R1166 CCD Princeton Instruments CCD camera with f = +2.5 cm lens focusing light onto CCD chip G Diffraction grating 600 grooves/mm, 400nm BLZ VC video camera for alignment Ultra -violet (UV) Probe: The Continuum Minilite I 1 Nd: YAG laser contains second and fourth harmonic crystals to produce nanosec ond laser pulses of wavelength 266 nm (2 Hz, 5 ns wide). The laser is externally triggered by the rising edge of electrical pulses (10 Hz, 10 s wide, 0V 5V). This light is focused onto the sample at the center of the region heated by the infrared beams. The correction for fluctuations in laser intensity can be made, if necessary (Figure 2 1 3 ).

PAGE 64

64 Figure 2 1 3 : Schematic of optics for ultra v iolet probe beam P = equilateral prism to disperse 532 nm and 266 nm light M = UV reflecting mirror L1 = PCX converging lens (f = + 50 mm) I = pinhole made with punctured aluminum foil W = silica wedge to deflect light towards photodiode for detection of UV intensity F = Silica filter 0.5 OD to cut laser intensity and protect silica fiber L2 = PCX converging lens (f = + 100 mm) The size of the focused UV spot is 75 100 m. UV -excited sample fluorescence is collected by an f/1 fused silica lens and split by a silica wedge. The front surface reflection is directed toward a photomultiplier (Hamamatsu R1166). The time delay between the thermal trigger and the ultra -violet probe is varied so that we may sample the fluorescent response from pre trigger to 500 ms after the thermal trigger. We obtain a rough alignment of the laser beam focus by the use of an aluminum block with a 75 m diameter clearance. This alignment is f ine tuned by adjusting the beam position to maximize the temperature perturbation induced by the IR laser pulse in the fluorescence of a solution of free tryptophan. Labview programs are used to trigger the IR pump and UV probe laser pulses, define param eters for instruments and acquire kinetic and spectral data. Matlab (Mathworks) can then combine the time and wavelength data for analysis.

PAGE 65

65 Figure 2 1 4 : Sample holder and fused silica fiber Fiber and Sample B lock : The sample flows through a fused sili ca fiber of rectangular cross -section (0.1 mm X 1 mm), which is mounted on an aluminum block as shown in figure 2 1 4 Conducting silver paint applied to the ends of silica fiber enhances thermal contact between silica capillary and block. A thermo -electr ic stage and controller maintains sample and holder at the desired temperature. A thermistor monitors temperature of sample during data acquisition. The temperature of the sample can be varied from 5 100 oC to 0.1 oC accuracy using a thermo electric chi p connected to a PID temperature controller (MPT 5000, Wavelength Electronics). The time resolution of acquired experimental data depends on the IR laser pulse width of 5 7 ns and the 20 30 ns thermal equilibration time for solvent following T jump. The s ize of the largest time window for the experiment depends on the duration the final temperature Tf is stable. The duration of this final temperature is determined by thermal diffusion in capillary, which is constrained by the geometry of the sample holder Our aqueous sample flows through a capillary of rectangular cross -section 100 m x 1mm with the pump and probe laser beams incident on the 1 mm wide face of the capillary. We can consider thermal diffusion in a cylinder of radius r and height 100 m.

PAGE 66

66 Figure 2 1 5 : Schematic representation of thermal diffusion in sample after the IR pulse (red arrow) is incident on sample in silica capillary (optical path length 0.1mm). If this diffusion occurs on a timescale t, (figure 2 1 5 ) t r2 ~2 [2 14] The factor denotes the thermal diffusivity of the medium, which is 1.44 x103 mm2/s for water at 293K. The radius r is the radius of the heated region of our sample. This corresponds to IR beam size of 1 mm. We get thermal diffusion on timescales of 500 ms in wa ter, which encases the flowing sample. We set our laser repetition rate to be 500 ms for this very reason. Sample H andling: A syringe pump pushes the aqueous protein sample through the capillary at a flow rate of typically 0.2 ml/hr. For sample concentrations of ~ 1 mg/ml, this allows ~1 day of data collection wit h 1 mg of protein. W e are thus able to probe of the folding transition in a protein over a range of temperatures while using barel y a milligram of protein. S amples are degassed and filtered with a 0.22 m inline filter to minimize cavitation problems during data acquisition. Sample P reparation: IA3 in its pure lyophilized form was dissolved in buffer solutions of 50 mM pH 7.0 sodium phosphate or pH 4.5 sodium acetate respectively. The aspa rtic proteinase YPrA was obtained in lyophilized form from Sigma (P 8892) and then hydrated in pH 4.5 sodium acetate buffer. 0.1 mm 1 mm

PAGE 67

67 For experiments with TZ2, I prepared solutions of 100 120 M TZ2 in 50 mM pH 7.0 phosphate buffer with 2M GdnHCl and ethylene gl ycol. T he viscosity of solvents is enhanced to about 3.6 mPas (3 4 water) at 25oC by adding 0 50 percent by weight of ethylene glycol to buffer containing 2M GdnHCl. The kinematic viscosity () of all solvents (density ) is directly measured with a calibrated Cannon -Fenske viscometer immersed in a water bath to a n accuracy of better than 1% by this method. Temperature -jump C alibration Figure 2 1 6 : Temperature calibration of T jump system with fluorescence of free tryptophan in solvent used for experiments. Panel A shows the equilibrium fluorescence data of free tryptophan in pH 7.0 phosphate buffer with 17% v/v tri -fluoroethanol for temperatures 5 85oC in increments of 5oC. The integrated fluorescence is fit to an exponential function of tem perature, show in panel B. The difference in tryptophan fluorescence intensity before and after the temperature jump is correlated to fit parameters obtained for equilibrium data in panel B and calibrated in terms of change in temperature (panel C)

PAGE 68

68 To de termine the magnitude of the temperature jump prior to each experiment we measure the fluorescence of free tryptophan solution in the T jump apparatus. Comparison of the change in fluorescence after T jump to an equilibrium calibration measurement of try ptophan fluorescence in same buffer over temperatures 5 85oC enables us to determine the magnitude of the temperature-jump. The logarithm of the equilibrium fluorescence intensity is fit to a second order polynomial function of temperature as shown in pan el B of figure 2 16. A typical calibration dataset for NATA is acquired prior to experiment with peptide and is shown in panel C of figure 2 16. The difference in intensity of fluorescence of pre trigger (panel C) and post trigger fluorescence signal is measured and then, using the fit parameters obtained from equilibrium fluorescence data we can estimate the size of the temperature jump in Celsius. Kinetic Data A cquisition: 0 100 200 300 400 500 600 700 800 0 100 200 300 400 500 UV intensity (mV) Emission intensity on PMT (mV) Saturation of tryptophan fluorescence for optimal UV intensity y = 451.03 405.37*exp(-0.0028*x) Figure 2 1 7 : Saturation of aqueous NATA solution with UV photons. We wish to operate in the region where the slope of the exponential decay of PMT signal with changing UV intensity is close to zero, implying negligible dependence of PMT signal on UV intensity, and thus reduced noise due to shot to -shot variation in UV intensity.

PAGE 69

69 Emission from the sample volume probed by the focused ultra -violet laser pulse is collected and sampled by a photomultiplier (Hamamatsu R1166 D s well as a CCD camera. The primary sources of noise in data collection arise from shot to -shot variations in UV and IR laser intensity as well as a small amount of electrical noise due to RF interference from the IR laser Q -switch. Noise due to fluctuations in UV laser intensity can be reduced by optimizing PMT signal while varying the UV intensity and determining the region of operation where PMT signal is insensitive to fluctuations in UV intensity (Figure 2 1 7 ). Spectral D ata A cquisition: Figure 2.18 : Sample fluorescence is collected by a lens that focuses light at the focal plane of a collim ating lens. Collimated light is dispersed by a diffraction grating to yield a spectrum spanning wavelengths 250 700 nm. This spectrum is focused onto a CCD array that acquires spectra synchronously with the photomultiplier. We can acquire multi -wavelen gth time resolved fluorescence spectra over ns s timescales with nanosecond resolution. Sample fluorescence is collimated and directed onto a diffraction grating (150 grooves/mm, 400 nm blaze wavelength), which disperses light to provide spectrum over

PAGE 70

70 wavelengths 290725 nm. The spectra are centered at 400 nm and focused onto a CCD camera array with a focusing lens (f = 2.5 cm). Wavelength calibration of this CCD array is performed with standard mercuryargon pen lamp (Oriel 6033). A video camera with a focusing lens is also used to optimize alignm ent of emitted light through the optical system (figure 2 1 8 ). We have a laser triggered temperature jump system that can probe the kinetics of relaxation of proteins that can absorb UV light of wavelength 266 nm and have a fluorescent response in the wave length region 250 700 nm. We can simultaneously acquire spectral and kinetic data with nanosecond time resolution on timescales of ns -ms. We employ temperature jumps of 6 8oC to destabilize the folded state of the tryptophan zipper protein, so that we may probe its folding by observing the fluorescence of tryptophan (emission maximum at 350 nm) at time delays extending out to a few tens of microseconds. We have used this system to monitor the folding kinetics of tryptophan zipper in solvents of varying viscosity. Our temperature jump studies of the folding kinetics of the intrinsically unstructured peptide IA3 coupled to its binding with YPrA enable us to identify the sequence of events that lead to the effective inhibition of the YPrA.

PAGE 71

71 CHAPTER 3 KINETICS OF FOLDING AND BINDING OF THE INTRINSICALLY DISORDERED PEPTIDE IA3 WITH YPRA Introduction We have seen the diverse roles played by intrinsically disordered proteins in initiation and regulation of vital biological processes (42, 46, 51) The activity of these proteins arises through their folding to a structure that is conducive to their function (42, 45, 52, 58, 62) This ordering transition occurs during or prior to their association with other proteins or cofactor s (58) The inhibition of the prot eolytic enzyme YPrA by the intrinsically unstructured IA3 is a case in point. Free IA3 has no residual secondary structure, as indicated by CD (85) and NMR studies (145) X ra y crystallographic studies of YPrA and IA3 (76) show the N terminal residues of IA3 bound to the active site of YPrA and folded into an alpha helix. This folded segment of IA3 blocks substrate access to YPrA inhibiting th e enzyme. Despite overwhelming evidence of a unique mechanism of interaction that couples helix formation and binding (76, 84, 88, 89, 145) no study so far has investigated the sequence of events that lead to YPrA inhibition by IA3. We study the kinetics of this interaction to gain an insight into the function of IA3. The simplest mechanisms that can be envisaged for the YPrA IA3 interaction are that (Scheme I) folding precedes binding to YPrA or that (Scheme II) a transient complex is formed prior to folding of the intrinsically disordered peptide (Figure 5.1). Scheme I would suggest that the folding kinetics of IA3 would be unchanged by the binding interaction with YPrA. If scheme II were to be valid, however, the folding kinetics of IA3 should be significantly different in the presence of YPrA. Our approach is to do a comparative study of the kinetics of IA3 folding in the abs ence and presence of YPrA, or effectively IA3 folding independent of binding and IA3 folding coupled to binding. We use the helix promoting co -solvent 2, 2, 2 trifluoroethanol (146, 147) (TFE) to

PAGE 72

72 in duce helix formation in free IA3. Equilibrium circular dichroism measurements characterize the stability of the helix coil transition in IA3 as a function of TFE and temperature. We then measure folding rates of IA3 in the presence of varying concentrati ons of TFE (and no YPrA) by measuring time resolved fluorescence signal following a nanosecond temperature jump. Figure 3 1: Two potential schemes of interaction for YPrA with the intrinsically disordered IA3. Scheme I suggests that folding precedes bin ding to YPrA, while scheme II suggests that binding to YPrA precedes folding of IA3. Our studies employ peptides with the fluorescent labels tryptophan and dansyl incorporated in the N terminus, so that we may use distance dependent (FRET) energy transfer to monitor the folding of the N terminus of the IA3 peptide. The folding kinetics of IA3 in water are estimated by an extrapolation of the rates in various TFE concentrations to zero TFE. We measure folding kinetics of IA3 coupled to binding to YPrA using laser triggered temperature jump fluorescence spectroscopy. The differences in the observed kinetics of IA3 folding enable us to draw some conclusions as to the merits of the two differ ent schemes proposed in figure 3 1.

PAGE 73

73 We observe that in the absence of YPrA, IA3 is a slow folder, with a folding time of 5 s at room temperature. Comparisons to literature values for alpha helix folding (143, 148, 149) times of 200300 ns suggest that IA3 is unable to nucleate sufficiently fast, leading to inefficient and sluggish folding. In the presence of the protease YPrA, IA3 exhibits a fast ~ 90 ns relaxation that we believe is the folding of IA3. This result and other observations in our s tudies convince us that IA3 first makes a series of contacts with YPrA, and then uses YPrA as a template to assist its folding. Results Peptide C haracterization All peptides (wild type IA3 and mutants N2W IA3, K16C IA3, N2W -K16C IA3) were expressed in E coli. Synthesis, purification and characterization (with standard biochemical techniques) of all peptides was completed at Dr. Edisons laboratory at the University of Florida. Inhibition assays were used to confirm that mutations and dansyl labeling d id not affect the potency of inhibition of the IA3 peptides. The dissociation constant for binding of enzyme and inhibitor is a measure of the potency of the inhibitor. It is termed inhibition constant or KI and has units of concentration. Inhibition co nstants for all IA3 mutants were at or below the limits of instrument detection at 25oC, implying an inhibition constant KI < 1nM at 25oC. Equilibrium F luorescence S tudies We verified the occurrence of FRET between the fluorophores in the double mutant IA3 by monitoring the equilibrium fluorescence of IA3 in different concentrations of TFE. Previous NMR studies of IA3 in TFE have indicated an increase in helical content of IA3 on addition of TFE, with a helix folding midpoint concentration of 18% v/v TF E in buffer. A representative spectrum of the double mutant IA3 is shown in Figure 3 2 (A). We select tryptophan in N2W and N2WK16C dansyl IA3 as a molecular probe of the folding transition.

PAGE 74

74 Figure 3 2 : The emission of N2W -K16C -dansyl IA3 after excitat ion by 266 nm light shows donor (tryptophan) emission at 350 nm and acceptor (dansyl) emission at 500 nm (panel A). Equilibrium fluorescence of ~ 20 M N2W IA3 (B), tryptophan of N2W K16C dansyl IA3 (C) in various TFE concentrations from 0 25% TFE v/v shows a decrease in tryptophan emission in the dansylated peptide with increased TFE. In the absence of dansyl, the emission of tryptophan increases sli ghtly with increasing TFE concentration. We compared the emission of tryptophan in the presence of dansyl in the double mutant N2WK16C -dansyl IA3 with tryptophan emission in donor only labeled N2W IA3 peptide and observed that on adding TFE, the emission of tryptophan is progressively reduced in the double mutant, while in the donor only peptide, the emission is pro gressively increased.(Figure 3 2 B and C). This is expected as the peptide is induced to fold in the presence of increasing amounts of TFE, bringing the tryptophan and dansyl closer and increasing energy transfer from tryptophan to dansyl.

PAGE 75

75 Characterization of IA3 F olding Behavior at E quilibrium Previous NMR studies have revealed that free IA3 forms a random -coil (no residual secondary struct ure) (145) in the absence of TFE. In the presence of TFE, the N terminus of IA3 acquires alpha helix structure. The use of TFE to induce helix folding in peptides with little or no helical secondary structure in th e absence of TFE has been well studied (146, 147, 150) It is possible to determine the equilibrium thermodynamic parameters (free energy of folding /unfolding, associated enthalpy and entropy) for the helix to coil transition in water by an extrapolation of the folding parameters measured at various TFE concentrations (150, 151) Equilibrium far -UV circular dichroism probes the secondary structure content of IA3 at various temperatures and TFE concentrations. We observe that at low TFE concentrations, the CD spectrum is typical of a random coil with a dominant minimum at 200 nm. As the concentration of TFE increases, the CD spectrum begins to exhibit the signature of an alpha helical structure with a minimum near 208 nm and 220 nm. All the spectra for all IA3 mutants also exhibit an isodichroic point near 202 nm, which is suggestive of the existence of two populations of IA3, one that is folded into an alpha helix and one that is unfolded. This observation is consistent with previous NMR studies and enables us to fit all CD data over all wavelengths, temperatures and TFE concentrations to a two state model for IA3 folding. Our model assumes that the free energy of unfolding of IA3, G varies linearly with TFE concentration (Eq 3 1). Our unfolding free energy for all IA3 mutants is then well defined by two enthalpy parameters mH and H0, and two entropy parameters S0 and mS. ] [ ] )[ ( ) ( ] [ ] [0 0 0 0 0TFE m G TFE Tm m S T H TFE T G TFE m S S TFE m H H S T H GS H S H [3 1]

PAGE 76

76 The parameters H0 and S0 represent the enthalpy and entropy of IA3 helix -coil transition in water, while mH and mS represent the energetic contributions of IA3 interactions with solvent (TFE) that affect the helix coil equilibrium. The population of IA3 in the folded (F) or unf olded state (U) is related to the free energy of unfolding G and temperature T as shown in Eq. 3 2. Figure 3 3 : Equilibrium far -UV circular dichroism data for mutant N2W -K16C -dansyl IA3 (~ 15 M ) in pH 7.0 phosphate buffer with varying concentrations of TFE from 0 25% v/v in increments of 5% v/v (each panel shows one concentration). For each concentration, scans were taken over temperatures 5 85oC, represented by the multiple spectra that intersect at 202 nm. CD spectra are typical of a random -coil peptide at low TFE concentrations (0 5% v/v) while the spec tral characteristics of an alpha helix structure are seen at higher TFE concentrations. The data in black is fit well by a two state folding model for the helix -coil transition in IA3 as a function of temperature and TFE. Our fits are plotted in red. )) / ) ( exp( 1 /( ) / ) ( exp( 1 ( )) / ) ( exp( 1 ( ) (1RT TFE T G RT TFE T G f T T f RT TFE T G TFE T fU F U [3 2]

PAGE 77

77 We can then represent our CD signal at each wavelength, temperature and TFE concentration as arising from a sum of two signals, each the product of fraction of molecules in the folded (or unfolded state) i.e. fF and fU and the temperature independent basis CD spectrum for the alpha helix (or random -coil) i.e. F and U. (Eq 3 3 ) ) ( ) ( ) ( ) ( ) ( TFE T f TFE T f TFE T CDU U F F [3 3 ] We estimate the fraction of IA3 in the folded and unfolded states for each condition of temperature and TFE concentration and use the actual observed CD spectrum for that experimental condition to estimate our basis spectra. Our global fit method uses an initial guess for H0, S0, mH and mS for estimation of fF and fU which are then used to estimate the basis spectra. The predicted CD spectra as a function of tempe rature and TFE are calculated from above, and deviation from experimental spectra is calculated. The initial guess parameters are varied until the best match between predicted and actual CD spectra is obtained (least squares minimization of deviation of f it from actual CD spectra). We find a very good fit between predicted and actual CD spectra for IA3 mutants at all temperatures, TFE concentrat ions and wavelengths (Figure 3 3 ). This vindicates our choice of the two state folding model, for which our es timated basis spectra are seen in figure 3 4 We observe that all IA3 mutants have equal sub -populations of folded and unfolded states at a TFE concentration of 1819% v/v (the folding mid-point) at room (85, 145) temperature as observed in previous NMR studies of IA3 folding in the presence of TFE. The thermodynamic parameters for the helix -coil transition in water for all IA3 mutants is very similar with a free energy of unfolding in water at 25oC being ~ 6.8 kJ/mol. Contour plots of G and fU as a function of temperature T and TFE concentration are very similar for all IA3 peptides. The contour of population of unfolded IA3 at low TFE concentrations (figure 3 4C) is

PAGE 78

78 Figure 3 4 : Analysis of thermodynamic parameters for folding transition of IA3 as a function of temperature and TFE. Panel A shows the basis spectra for the two states of IA3 which are comparable to CD spectra for random coil and alpha helix in the CD literature. Panel B an d C represent contour plots of free energy and unfolded subpopulation as functions of temperature T in Celsius and TFE concentration in v/v. almost parallel to the temperature axis. This implies that a temperature jump at very low TFE concentrations does not significantly change the fraction of IA3 in the folded state. This means that we would not trigger significant amounts of folding or unfolding by applying a thermal perturbation to IA3 in the absence of TFE. Therefore, we measure the folding kinetics of IA3 at various TFE concentrations above 8% v/v and then extrapolate the folding /unfolding rates to zero TFE to determine the folding of free IA3 in aqueous conditions. Kinetics of IA3 F olding in the P resence of TFE We monitor the folding transition of free N2W -K16C dansyl IA3 in pH 7.0 phosphate buffer with varying concentrations of TFE by volume by acquiring time resolved fluorescence data after a thermal trigger. A typical dataset is shown in figure 3 5B

PAGE 79

79 Figure 3 5 : Time resolved fluorescence data of N2W -K16C dansyl IA3 in phosphate buffer with 17% TFE v/v at a final temperature 24.8oC. Raw data in panel (B) is analyzed by singular value decomposition, which separates spectral response from kinetic response. The spectral vectors U1 to U3 are shown in panel A, while the weighted kinetic eigenvectors SV2 and SV3 are shown in panel C with best fits for a monoexponential relaxation. The thermal trigger causes a rapid (~ 20 30 ns) decrease in overall fluorescence emission, due to the intrins ic negative temperature dependence of the fluorescence quantum yield. This fast 2030 ns relaxation is observed in free tryptophan as well and is not related to the protein folding dynamics. The interesting relaxation is a slower microsecond relaxation, w hich is studied by a singular value decomposition (see Appendix A) analysis of data. Singular value decomposition resolves the data into independent spectra Ui () whose evolution in time is described by Vi(t) The weight of each component of the dataset is given by an S -value. In the present case, U1() resembles the average spectrum of the peptide and shows the fluorescence peaks corresponding to emission from tryptophan at 350 nm and dansyl at 500 nm. The time -dependence of the average fluorescence i s defined by S V1 (t) which shows a step

PAGE 80

80 response to the temperature -jump. The relaxation in SV2 (t) and S V3 (t) describes the time course of the spectral components U2 () and U3 () respectively. SV2 and SV3 show a rise on the same timescale indicating that this relaxation corresponds to the rise in tryptophan emission at 350 nm and fall in dansyl emission at 500 nm with a wavelength shift. The rise in donor emission concurrent with fall in acceptor emission is consistent with the idea of reduced energ y transfer due to helix melting at higher temperatures. Both SV2 and SV3 show a relaxation that can be fit to a single exponential with a timescale of 500 ns ( = 1/krelax). Since we have observed a very good agreement of our helix coil equilibrium (CD) to a two -state thermodynamic model, we use a two -state model to relate kre lax to the folding rate kF and unfolding rate kU as in Eq. 3 4. ) / ) ( exp( / 1 RT TFE T G k k k k kU F U F relax relax [3 4 ] The ratio of the folding and unfolding rates is related to the free energy of unfolding G which has been evaluated for each temperature and TFE concentration from our analysis of the CD data. We can estimate the values of kF and kU from the experimentally measured relax and the estimated G for each solvent condition. We thus estimate the folding and unfolding rates of free IA3 for various TFE concentrations and temperatures. We observe that the overall relaxation rate is independent of variations in TFE (Figure 36A) concentration, while the folding and unfolding rates vary linearly with TFE concentration on a semi logarithmic scale. This is to be expected as our free energy of folding shifts linearly on addition of TFE a s in equation 31. The extrapolation of the folding and unfolding rates of IA3 at each temperature to zero TFE yields the folding and unfolding rates of free IA3 in water at that temperature (Figure 3 6B).

PAGE 81

81 Figure 3 6 : Fitted results of IA3 kinetic data w ith panel A showing the observed relaxation time for IA3 in different TFE concentrations and different temperatures. The relaxation time is dependent on temperature, but not on TFE concentration. Panel B focuses on the rates of relaxation for IA3 in diff erent TFE concentrations at 25oC, and the extracted folding and unfolding rates ( kF, kU) from equation 3 4. These rates are extrapolated to zero TFE to estimate the folding and unfolding rates of IA3 in water at 25oC. We can see that the folding rate of IA3 in water at room temperature is ~ 3.3 s. This is extremely slow in comparison with other folding rates observed for alpha helices in the protein folding literature where rates typically exceed 2 106/s. We can conclude that in the absence of YPrA, IA3 is a very sluggish folder in comparison with other alpha helical peptides. At 25oC Folding rate kF = 0.3 0.04 / s Unfolding rate kU = 3.3 0.5/ s Observed relaxation time krelax = 3.6 0.5/ s Kinetics of IA3 F olding and B inding to YPrA We measure the relaxation of the complex formed by single mutant K16C dansyl IA3 and the aspartic proteinase YPrA after a laser triggered temperature jump. We use pH 4.5 acetate buffer for our experiments with YPrA as it mimics the acidic environment of the vesicle s in which YPrA is located in yeast cells. Since IA3 has a very strong binding affinity with YPrA, a temperature jump cannot radically alter the population of IA3 bound to the protease. This gives us a very weak signal and prevents us from measuring the dependence of the relaxation rates on

PAGE 82

82 IA3 concentrations. We use a slight excess of YPrA in our experiments, typically 50 M YPrA and 40 M IA3. This ensures that all IA3 is bound to YPrA, and largely eliminates the fluorescence background from dansyl in free IA3. We observe a fast ~ 95 20 ns relaxation for the complex in our CCD data (figure 3 7 B) with a wavelength shif t at the tryptophan and dansyl wavelengths after a temperature jump from 17.3 to 23.8oC. We also observe a fast 80 12 ns relaxation in our photomultiplier data sensitive to tryptophan emission only. This relaxation is faster than the relaxation that we observe for unbound IA3. We also see a shift in wavelength for both dansyl in IA3 and tryptophan in YPrA (Figure 3 7 A ). There are several such indications that this relaxation is the signature of an intermediate binding step in the coupled folding and binding interaction of IA3 with YPrA. Figure 3 7 : Kinetics of coupled folding and binding interaction of IA3 with YPrA: Panel A shows the spectral response of the system with U1 (blue curve) resembling the average fluorescence of the system. U2 and U3 (i n green and red respectively) represent the wavelength shift of the tryptophan and dansyl fluorescence during the binding reaction. Panel B shows t he time relaxation corresponding to the wavelength shift in the tryptophan and dansyl, SV2 and SV3 are well fit by a single exponential relaxation 95 20 ns. Discussion IA3 has the distinction of being the first specific inhibitor of an aspartic proteinase to be discovered. We study the kinetics of the mechanism by which IA3 binds to YPrA and folds into

PAGE 83

83 an al pha helix that blocks access to the enzyme active site. Our approach first decouples the folding and binding aspects of the interaction, and monitors helix formation in free IA3. The kinetics of the coupled folding and binding IA3YPrA interaction are compared against this control scenario. Alpha helix F ormation in P roteins Helix formation as one of the elementary events in the protein folding process has been the focus of several theoretical (152, 153, 154) and experimental studies (143, 148, 149, 154) Theoretical studies have recognized the role of two events in the helix f ormation process: (155, 156) helix initiation or nucleation and helix propagation. Helix initiation involves hydrogen bond formation between two residues on a chain separated by three pepti de bonds and is entropicall y expensive. Once the helix is initiated, the bond energy compensates for the loss of entropy due to fixation of each additional residue. For the first turn in a peptide, this compensation is not available leading to a free ene rgy barrier to the helix formation process. Helix initiation (155) can hold up the overall process of helix formation in short peptides, while helix propagation limits the rate of helix formation in long peptide s. It is estimated that peptides of 20 30 amino acids take 0.3 0.5 s to fold into an alpha helix (152, 153) This is also observed in experiments (100, 102, 111, 148, 149, 157, 158) The N terminus of IA3 is 32 residues long, and predicted to fold in 200 300 ns. Our experiments with IA3 in TFE indicate a relatively sluggish folder that takes close to 3 5 s to fold. This would imply inefficient nucleation in the absence of stabilizing contacts with YPrA. In the context of binding induced folding, a previous theoretical study of dimeric coiled coil GCN4 peptide folding has indicated that folding is faster with the collision of two unstructured chains than collision of two preformed helices. This implies that the interaction of

PAGE 84

84 unstructured chains precedes their folding. A very similar scheme can also be envisaged for the interaction of the unstructured IA3 peptide w ith YPrA. We discussed two simple schemes (Figure 3 1) in our introductory section. The first scheme proposed a collision of helically folded IA3 with YPrA, leading to formation of specific contacts. The second scheme proposed the formation of non-speci fic contacts prior to folding of IA3. We can compare the folding kinetics of IA3 in water in the absence of YPrA (extrapolation of TFE results) and presence of YPrA to assess the virtues of above schemes. Figure 3 8 : In the proposed model for interacti on between IA3 (I) and YPrA (Y), binding of the unstructured IA3 precedes its folding into an alpha helix. The formation of the bound intermediate (E) precedes the formation of the complex (C) of folded IA3 bound to YPrA. Helix nucleation and folding are assisted by the binding interaction with YPrA. Scheme for IA3-YPrA I nteraction We have observed a fast ~ 90 ns relaxation that is three times faster than the estimated folding rate (3.3 s) or the unfolding rate ( 300 ns) of free IA3 in water at 25oC. We do not see such a fast relaxation in control experiments with just K16C dansyl IA3 or YPrA in isolation. Thus, the rates of the observed relaxation cannot correspond to the folding of f ree IA3 in isolation or a relaxation in the YPrA alone. The spectral response also shows a change in both tryptophan wavelengths (belonging to protease) and the dansyl wavelength (corresponding to IA3). This must signify a change in spatial separation or relative orientation of the tryptophan

PAGE 85

85 residues of the protease and the dansyl in K16C -dansyl IA3. We can conclude that the 90 ns relaxation cannot arise from the folding or unfolding step of the IA3 alone, but that it corresponds to an interaction betwe en IA3 and YPrA. Since the folding kinetics of free IA3 are significantly different from those of folding coupled to binding, and the spectral response shows a change in the fluorescence of both donor and acceptor, scheme I can be eliminated. Origin of F a s t ~ 90 ns R elaxation We now need to identify the source of the fast 90 ns relaxation. For KI = 1nM, [IA3] = 40 M, [YPrA] = 50 M and a relaxation of 90 ns, the simplest two state association model would have rates kON, kOFF that follow kON/kOFF = KI and kON [extra YPrA] + kOFF = 1/krelax. We have 10 M extra YPrA in our experiments, which suggests two possibiliti es. One possibility is that the relaxation is large because the rate of association is large, i.e. the order of 1012 (Ms)1. However, this result exceeds the diffusionlimited rate of 109 (Ms)1 by several orders of magnitude and is therefore unphysical. The other possibility is that the rate of disassociation is large i.e. of the order of 107/s. This suggests a very weak association of IA3 and YPrA, which could lead to tight inhibition only if kON is also very fast. Even for diffusion limited associati on kON ~ 109(Ms)1 the disassociation rate kOFF ~ 107(Ms)1 gives [I] [Y]/ [E] = kOFF/kON ~ 0.01 M. This makes the binding affinity of IA3 for YPrA close to 1M. In the light of the high binding affinity of IA3 for YPrA (KI ~ 1 nM), it is then difficult to explain the combination of weak association with very fast kinetics. Even for diffusion limited association kON ~ 109(Ms)1 the disassociation rate kOFF ~ 107(Ms)1 gives [I] [Y]/ [E] = kOFF/kON ~ 0.01 M. A subnanomolar binding affinity is possible in this scenario only if C is favored over E (figure 3 8 ) by a factor of 107. Since only a small population ~ 10% of IA3 is helically folded in water, this would necessitate an association that enhances the folded sub -population of IA3 by eight orders of

PAGE 86

86 magnitude. These estimates are hard to believe especially since the literature on folding and binding in proteins suggests typical association rates of 106(Ms)1 and disassociation rates of 10 100 /s (68 69, 81) Based on the above arguments, we conclude that the fast relaxation observed does not correspond to the association of IA3 with YPrA, but rather the folding/unfolding step in scheme II. We also see an increase in donor signal, signifying an enhanced separation between tryptophan in YPrA and dansyl in IA3, due to unfolding of the IA3 helix at the active site. We see a very fast 80 90 ns relaxation that is nearly threefold faster than the expecte d rate of IA3 unfolding in water (~300 ns), implying that the binding interaction accelerates the folding/unfolding kinetics of IA3. To achieve sub nanomolar inhibition, the complex C should be favored over E, implying that the 80 90 ns relaxation is real ly dominated by the folding rate of IA3 (kF). So, kF >> kU, with kF + kU ~ (90ns)1. This implies that the relaxation we observe is the folding rate of IA3 in contact with YPrA. Thus interaction with YPrA accelerates the folding process of IA3 by three orders of magnitude to a folding rate that is comparable to the rates of the fastest folding helices in the protein folding (100, 143, 148, 157) literature. Future R esearch Our studies point towards a mechanism of interaction where IA3 forms an encounter complex, and uses YPrA as a template to assist its folding. Our studies have focused on the interaction of the N terminus of IA3 with YPrA, as it is crucial to inhibitory action. Studies indicate that N -terminal extension of IA3 peptides can relax the selectivity of IA3 towards YPrA, but the deletion or mutation of the C terminus does not have such effects. NMR studies show interactions between YPrA with C terminal residues of IA3. The 15N HSQC peaks for three residues in the C terminus (G40, G62 and G64) are seen to broaden and then disappear with addition of sub -stoichiometric quantities of YPrA. It is intriguing that the residues of the C -

PAGE 87

87 terminus inter act with the protease, but they do not affect the equilibrium constant for inhibition. This suggests the interesting possibility of the C terminus binding to YPrA by long range contacts and forming the intermediate complex. We could speculate that this s teers the peptide so that the N -terminus can fold and bind to the active site. This is similar to the fly -casting scheme (65) which suggests that long range contacts initiate the formation of a bound intermediat e that promotes folding and binding. Future investigations of IA3YPrA interaction kinetics with mutants would be useful in determining the nature of these interactions that stabilize the bound and folded state of IA3. Summary Our kinetic study of IA3 folding coupled to its binding interaction with YPrA is possibly the very first investigation of the events underlying the inhibition of YPrA by IA3. We have characterized the equilibrium thermodynamics of the helix to coil transition in free IA3 as a fun ction of the helix promoting co-solvent TFE and temperature. Our estimate of the free energy of unfolding G for each solvent condition from the aforementioned analysis enables us to extract the folding and unfolding rates of free IA3 from its relaxation after a nanosecond temperature jump. These rates can be extrapolated to water to estimate the folding rates of free IA3 in water. The folding rates of free IA3 in water can then be compared to the relaxation observed following a thermal perturbation of a complex of IA3 and YPrA. The fast relaxation in that system seems to indicate that IA3 and YPrA form a transient unfolded intermediate prior to forming the bound complex Scheme II in figure 3 1. Our observations point towards a fly-casting mode of int eraction where the intrinsically disordered peptide interacts with YPrA forming an intermediate and then uses the protease as a template to facilitate its helical folding.

PAGE 88

88 We have a better understanding of the events leading to inhibition of YPrA by IA3. Our study does not directly identify the interactions at the residue level that stabilize the folded state of the peptide. Simulation or experimental studies of mutants without key residues can enrich our understanding of the IA3YPrA system at the mole cular level. Experiments suggest that the C-terminus of IA3 far from the binding domain interacts with the protease and yet does not affect the inhibition constant of the reaction. Future experiments that identify the role of the C terminus would greatly enhance our current understanding of the specific details of the inhibition reaction.

PAGE 89

89 CHAPTER 4 HETEROGENEOUS FOLDING KINETICS OF TRYPTOPHAN ZIPPER Introduction The free energy surface of protein folding is visualized as a funnel (6, 7, 8, 159, 159, 160) where the protein moves from regions of high conformational energy (the unfolded state) to regions of l ow conformational energy (the folded state). The topology of this folding funnel (160, 8) influences the folding route taken by the protein molecule. If the shape of the free e nergy is such that the trajectories of all protein molecules are confined to a narrow region of conformational space, a single folding pathway can describe folding satisfactorily. There are proteins whose energy terrain is such that the trajectories of di fferent protein molecules pass through broad regions of conformational space. The presence of multiple folding routes in such proteins is observed by different folding behavior in a variety of experimental studies (8, 119, 161, 162, 163) Kinetic studies of multi -state folding thus offer a unique insight into the events that drive folding as the protein explores its c onformation and energy options by diffusion. Kramers theory of reaction rates (25) can effectively describe the effects of friction on the dynamics of diffusion for the protein. The theory proposes that in the limit of strong fri ctional damping, the protein folding rates scale inversely with the frictional drag coefficient that appears in the Langevin equations of motion for the protein molecul e in the medium. This frictional drag is often attributed to the dynamic viscosity of t he solvent s. Experiments on protein folding in different regimes of solvent viscosity (27, 28, 29, 34, 36, 37, 164) indicate the existence of two timescales One is the timescale determined by the friction external to the protein and varies with solvent viscosity (28, 37) The second timescale is set b y internal friction effects which do not vary with solvent viscosity. Studies have focused so far on the influence of solvent viscosity on the rate of folding along one pathway (28, 29, 32, 33, 34, 36)

PAGE 90

90 Figure 4 1: Structure of tryptophan zipper TZ2 shows the side chains of four tryptophan residues Trp2, Trp4, Trp9 and Trp11 respectively. The stacking interactions be tween these indole side -chains hold the tryptophan zipper structure together. The above figure is generated from the PDB file 1le1 using PYMOL software. A recent study on effects of solvent viscosity on protein folding dynamics has suggested that under different conditions of solvent viscosity (41) the folding pathway itself could change. This study predicts that at low solvent viscosities, internal friction hampers the formation of short range contacts. So the fol ding route is chosen such that longrange contacts are formed swiftly. Conversely, at high solvent viscosities, short range contacts form earlier as the bulk diffusion of peptide in solvent is greatly restricted. Our kinetic studies of multi -state foldin g in tryptophan zipper TZ2 enable us to test this prediction. The tryptophan zipper is a 12 residue polypeptide chain designed by Cochran (165) et al that forms a beta -hairpin held together by stacking interactions between four tryptophan residues in the peptide. We focus on the tryptophan zipper TZ2 with sequence SWTWENGKWTWK (figure 4 1). Tryptophan zippers are model systems (30, 119, 166) for the study of beta hairpin formation in proteins. Circular dichroism studies of TZ2 and other tryptophan zippers exhibit a spectrum characteristic of exciton splitting due to the interactions (figure 4 -2A) between the -

PAGE 91

91 electron clouds of the tryp tophan residues (167) A couplet is observed with a maximum near 227 nm and a minimum near 215 nm, resulting from the dipole -dipole interaction between tryptophan residue pairs in close proximity (165) Besides a distinctive CD signal, tryptophan fluorescence and infra red absorption can also be used as a probe of the folding transition (30, 119, 165, 166) Heterogeneous Folding of TZ2: Background TZ2 is a very stable peptide that does not unfold completely even at temperatures of 85oC (119, 165, 166) Studies of TZ2 folding at different concentrations of denaturant and temperature (119) have indicated that at high denaturant concentrations ~ 6M GdnHCl, CD and fluorescence indicate an apparent two state foldi ng with the same midpoint temperature Tm. At lower denaturant concentrations of 0 5M GdnHCl, different midpoint temperatures are indicated by CD and fluorescence studies with a disparity of larger than 20oC in the midpoint temperatures. MD simulations (119) of the thermal transition for parameters like fraction of backbone hydrogen bonds, radius of gyration etc also showed three clusters of melting temperatures ( Tm ~ 30oC, 65 75oC and 160oC). The distribution of free energies for the different parameters sampled by simulations, indicated a bumpy energy landscape with several local minima separated by significant energy barriers for TZ2 at low temperatures. T -jump measurements of TZ2 folding kinetics using a two-state folding model have determined the folding time of 1.8 s and unfolding time of 18 s (30) These rates agreed well with accompanying simulations that predict a folding time of 3 6 s and unfolding time of 1420 s. Heterogeneous folding kinetics have been observed in te mperature -jump triggered fluorescence emission and infra red absorption studies (166, 168) of TZ2 folding at different temperatures. Wavelength -dependent fluorescence kinetics were reported in T ju mp

PAGE 92

92 fluorescence studies (168) The TZ2 relaxation was observed to occur by a fast ~100 ns relaxation attributed to enhanced tryptophan mobility and a slower ~ s relaxation attributed to a breaking of the TZ2 core and solvation of tryptophan residues in TZ2 The study also reported an intriguing temperature dependence of the wavelength-dependent kinetics. At low temperatures ( T < 45oC) TZ2 would exhibit one re laxation for probed wavelengths, and different relaxations for probed wavelengths at higher temperatures. The more recent study of folding in TZ2 variants focused on relaxation of amide I 12C=O vibrational modes corresponding to loss in beta -strand stru cture and gain of disordered structure (166) Different relaxations were observed for these different mode s at low temperatures ( T < 40oC) with the loss in beta structure relaxing faster than the gain in disordered st ructure after a temperaturejump of 10oC. Labels with 13 C on selected amide C=O positions on opposite strands of TZ2 were added to probe the effects of altered cross -strand interactions on the folding dynamics. The different bands probed (including the 13C amide band) relaxed with different rates at temperatures below 40oC. The relaxation of the 13C amide band in mutants with 13C labels in the middle of the hairpin resembled the relaxation of the 12C amide band that corresponds to loss of beta strand s tructure. In mutants with labels near the turn, the 13C band relaxation coincides with 12C band relaxation corresponding to rise of disordered structure. In mutants with labels near the ends, the same relaxation was slower than the relaxation of the ban d corresponding to rise in disordered structure and faster than the 12C=O band corresponding to loss of beta -strand structure suggesting a partially folded intermediate. The observation of a fairly stable core in TZ2 mutants was interpreted to suggest an unfolding mechanism, which proceeded from the terminals to the center.

PAGE 93

93 We are motivated to understand this heterogeneity in folding behavior in TZ2 by studying the kinetic response of TZ2 fluorescence after a temperature -jump at different temperatures an d in solvents of varying viscosities We need to add guanidine hydrochloride to enhance the temperaturejump signal as the fraction of unfolded TZ2 molecules is very small (~10%) at room temperature in plain buffer. Addition of guanidine hydrochloride in creases the fraction of unfolded TZ2 by reducing the denaturation (melting) temperature of the peptide. We vary the viscosity of solvents by adding ethylene glycol (0 50% by weight) (33) to 50 mM phosphate buffe r maintained at pH 7.0 The kinemat ic viscosity of solvents with ethylene glycol (maximum ~ 4 water) wa s measured by Leslie Pelakh in our laboratory with a CannonFenske viscometer immersed in a temperature-controlled water bath. It is important in such studies (165) to verify that the stability of the protein is not affected by the change in solvent composition. Therefore, we employ equilibrium circular dichroism (CD) to characterize the folding transition of TZ2 in ethylene glycol at various temperatures. We observe that although ethylene glycol enhance s the viscosity of the protein solvents, it does not stabilize or destabilize the folded state of TZ2. We then study the folding kinetics of TZ2 in five different concentrations of ethylene glycol from 0 50% by weight in buffered solutions of pH 7.0 with 2M guanidine hydrochloride. We observe two relaxations in our temperature jump experiments; one is fast (~100 ns) while the other is slow (~ s) and well described by a single exponential. These two relaxations correspond to changes in tryptophan fluor escence intensity and a red shift in wavelength of peak emission of tryptophan in TZ2 after the temperature -jump. As we alter the conditions of temperature and viscosity, we observe different kinetic responses for the change in fluorescence

PAGE 94

94 intensity and the wavelength shift in the fluorescence of TZ2. These responses are the kinetic signa tures of different events in the folding process for TZ2. Results Circular Dichroism Spectroscopy CD spectra for TZ2 (~ 3 2 M) in 50 mM phosphate buffer (pH 7.0 ) with 2M GdnHCl and varying concentrations of ethylene glycol show the characteristic exciton splitting at 215 nm and 227 nm (167) The intensity of these peaks is reduced as the temperature is raised, melting the peptide. Figure 4 2: CD signal for TZ2 in pH 7.0 phosphate buffer + 2M GdnHCl shows a peak at 227 nm and a dip in ellipticity at 215 nm. The intensity of signal at these wavelengths drops with increasing temperatur e as seen in panel A. There is good agreement bet ween the data ( black ) and the fit to a two -state model (red ). The stability of TZ2 is unaltered by addition of ethylene glycol, as shown by horizontal contours ( panel B ). The CD signal at the highest tempe rature of 94.7oC is not however the spectrum of a fully unfolded peptide. All spectra are fit to a two -state model to estimate the thermodynamic parameters ( G) of the thermal melting of TZ2. The stability of TZ2 folding ( G) as a function of temperature and ethylene glycol concentration is seen to be unaffected by the presence of ethylene glycol (horizontal isostability contours in figure 4 2B). Our CD results verify the

PAGE 95

95 negligible stabilizing or destabilizing effect of ethylene glycol on the stability of the folded state of TZ2 (31, 33) Kinetics of TZ2 R elaxation after Temperature-jump The laser temperature jump instrument, which uses a CCD camera to record a time resolved fluorescence emissio n spectrum of the peptide after an IR pulse is described in detail in chapter 2. The data analysis is based on singular va lue decomposition ( appendix A -1 ). We observe two contributions to our acqui red fluorescence data The spectra corresponding to th ese compo nents of our signal, (figure 44A and B) U1 and U2 represent the average fluorescence emission of the TZ2 peptide with a peak at 350 nm and the red shift in the peak wavelength of TZ2 emission, respectively. The relaxations that correspond to the evolution in time of spectra described by U1 and U2 are distinctly different (168) Figure 4 3: Comparison of fluorescent response of free tryptophan in N acetyl tryptophan amide (NATA) and tryptophan residues in TZ2 in phosphate buffer at pH 7.0 with 2M GdnHCl after a temperature jump of 9.7oC is shown above. The post trigger fluorescence data for NATA (open circles) shows no relaxation, while the post trigger data for TZ2 (closed circles) shows a fast ~ 100 ns relaxation. The fast relaxation ~ 100 ns in the fluorescence intensity (corresponding to V1) has been previously observed. It has been attributed to (168) a weakening of the stacking interactions

PAGE 96

96 between the indole s ide chains of TZ2. The shift in intensity stems from the lesser degree of quenching of tryptophan emissi on by contact with other tryptophan residues by contact with adjacent tryptophan residues as the tryptophan -tryptophan interactions break apart during unfolding. This relaxation is not observed in a control experiment with free tryptophan in same solvent as TZ2 at same temperatures (figure 4 3). Figure 4 4: Fluorescent response of 120 M TZ2 in 50 mM phosphate buffer at pH 7.0 and 2M GdnHCl to a temp erature jump of 8 10oC is shown in panels A -C. The mean fluorescent spectrum and its first derivative (spectral shift in wavelength) in panel A can be compared to the spectral eigenvectors U1 and U2 obtained after singular value decomposition of the fluore scence data for TZ2. The kinetic relaxation of the spectral shift corresponding to U2 is shown in panel C for different temperatures 25oC, 35oC and 52oC. A vertical offset is added to the data (full circles) at different temperatures to make the distinct ion between the kinetics at different temperatures clearer. The kinetic data is fit to single exponentials (fits indicated by dotted lines), and error estimates calculated using a bootstrap analysis.

PAGE 97

97 The second relaxation corresponds to U2, which resemble s the first derivative (dU1/d of the mean spectrum given by U1. This derivative is the red shift in the peak wavelength of tryptophan emission and relaxes on slower, microsecond timescales (168) This relaxation has also been observed previous ly and has be en attributed to a solvent exposure of tryptophan side chains as the hairpin unfolds. We are able to fit the relaxation in V2 to a single exponential. Figure 4 4C shows the relaxation in V2 for TZ2 at different temperatures (25 55 oC). We see a progressive speeding of the relaxation as temperature is raised from 25oC ( s, blue circles) to 35oC ( s, green circles) to 55oC ( s, red circles). Intensity Shift in peak Intensity Shift in peak Intensity Shift in peak Figure 4 5: Shift in fluorescence intensity and red shift in peak wavelength of fluorescence at low viscosities (blue) and high viscosities (orange) The relaxati ons in V1 and V2 occur independently (blue arrows in figure 4 5 ), as there is no evidence of a fast relaxation in the spectral shift on those timescales As viscosity increases, a fast relaxation is also observed in the spectral shift (orange arrows in f igure 4 5 ). This suggests a scheme of events after the temperature jump where TZ2 unfolds by a fast weakening of the cross -strand interactions holding together its hairpin structure and a simultaneous solvent exposure of the tryptophan residues. This is followed by another slower solvent exposure of the core residues in TZ2. So a change in the unfolding mechanism is clearly visible at different regimes of solvent viscosity (figure 4 6 ).

PAGE 98

98 At high temperatures and low solvent viscosities, we see only one relaxation (figure 4 6B and 4 6D) We could interpret this as the kinetic signature of a more two -state like folding behavior of TZ2. In solvents of higher viscosity, relaxation in the wavelength shift of TZ2 at high temperatures still conta ins a fast component (figure 4 6 D and F). The fast component is significantly weaker than at low temperatures (compare figure 4 6E and 4 6 F). Figure 4 6 : Viscosity dependence of relaxation of spectral shift at 3 5oC and 52oC. At low temperatures and viscosities, a slow relaxation in the wavelength shift is observed. At higher viscosities, we see a fast relaxation in the wavelength shift as a shoulder on the slower s relaxation. The presence of a slow and fast relaxation is weakened by raising temperature to 52oC. The variation of observed relaxation in spectral shift with solvent viscosity is shown in figure 4 7. At high temperatures (45oC and 55oC), the observe d relaxation in V2 varies linearly

PAGE 99

99 with solvent viscosity (0.8 2.5 mPa s). At lower temperatures, the observed relaxation rates in V2 vary in a nonlinear fashion as viscosity increases from 1 3.6 mPa s). Figure 4 7 : Observed relaxation time of spec tral shift ( ) after a temperature jump to various final temperatures in solvents of varying viscosities is interpolated to temperatures 2552oC. These interpolated times are fit to the best linear function of the solvent viscosities at those temperatures. At high temperatures, there is a good agreement between a line fit and the actual data. The dotted lines represent the best linear fit, while the data are shown in circles. The observed relaxation for TZ2 at a particular temperature is the kinetic signature of i ts diffusion along its folding free energy landscape. A protein with one folding pathway would exhibit one single relaxation, which slows down with increased solvent friction. The plot of observed relaxation rate versus solvent viscosity would then be a straight line. When the protein can access multiple folding routes, it takes different folding routes at different conditions of solvent viscosity. The change in slope of the plot of observed relaxation rates with change in solvent viscosity would then b e an indicator of a change in the folding route adopted by TZ2. Thus, this result also serves as an experimental verification of Pande and Rhees prediction (41) that changes in solvent viscosity shift its folding path way. However, the exact interpretation of the changes in the folding pathway at different solvent viscosities differs subtly. We shall

PAGE 100

100 discuss the expected changes in the folding pathway as predicted by Rhees results and our observations of the changes in the folding pathway in TZ2 in the next section. The nonlinear response of TZ2 folding kinetics to increases in solvent viscosity is more pronounced at low ( T 35oC) temperatures This suggests that TZ2 folding tends to be multi -state at these temperatures. Discussion The rate at which a polypeptide chain diffuses through the solvent physically (28, 29, 36, 37) limits the overall rate of protein folding. Both solvent friction external to solvent, as well as internal friction due to intra -chain interactions influences the diffusional dynamics of the protein. Sim ilar experiments in the past have focused on simple proteins with one folding pathway (28, 29, 37, 157) where different sources of friction influence the rates o f folding along the folding route differently. Our studies of TZ2 folding in different regimes of solvent viscosity indicate that these different sources of friction can also influence the folding route adopted by the protein. TZ2 folding has been observe d to be heterogeneous at varying conditions of temperature and denaturant concentration. The interpretation of this heterogeneity has been different in different studies probed by UV -circular dichroism, fluorescence, and infrared absorption inTZ2 (30, 119, 166, 168) One previous study of TZ2 folding kinetics at different temperatures in 2M GdnHCl observed two different microsecond relaxations at temperatures exceeding 45oC. These relaxations corresponded to the fluorescence intensity of TZ2 at different wavelengths (168) and were attributed to sampling of different local minima on the free energy landscape. A recent stud y found this heterogeneity in folding kinetics to be temperature -dependent but with multi state fold ing at low temperatures and two -state like folding at high temperatures. This was explained in terms of the rougher folding energy landscape for TZ2 at low temperatures (166) Our studies of TZ2 folding in phosphate buffer with 2M GdnHCl definitely confirm the multi -state nature of folding at low temperatures. The fluorescence intensity and the peak

PAGE 101

101 wavelength of the f luorescence emission traverse different trajectories along the energy l andscape as shown (figure 4 3) by the different relaxations in the V1 and V2 kinetic vectors over all wavelengths. We observe a fast ~ 100 ns relaxation in V1 for the mean fluorescence intensity followed by a slower ~ s relaxation for the red shift in wavelength for TZ2 emission. Th e fast relaxation of the fluorescence intensity can be attributed to the weakening of stacking interactions holding the hairpin together. The slower wavelength shift is the effect of a slo wer so lvation of residues involved in the stacking interactions. This suggests an unfolding mechanism where the interactions holding the TZ2 hairpin structure together break apart first followed by slower solvation exposure. As solvent viscosity is en hanced, the relaxations progress from being independent (figure 4 4 and 4 5A) to the situation where both fast and slow relaxations are observed in the red shift in wavelength. The mechanism of unfolding the hairpin now progresses by a fast simultaneous rupture of the interactions holding the hairpin together and solvent exposure of the tryptophan residues in the core, followed by a slower solvent exposure of core residues This i s a change in the folding route adopted by the protein molecule. This cha nge in the folding route can be induced by a subtle shift in the balance of influence from friction external to the protein and internal friction internal to the protein. The influence of solvent viscosity on the folding routes is weakened at high tempera tures (figure 4 5B and 4 5F). In fact at low solvent visc osities and high temperatures, TZ2 folding resembles a two -state transition. Our experiments demonstrate the use of solvent tuning the folding pathways adopted by TZ2 at various temperatures (figur e 4.6) The rough free energy landscape accessible to TZ2 can be smoothed by increases in temperature (119, 166). The solvent friction on the other hand influe nces the roughness of the same landscape in a complex ma nner. So if increases in

PAGE 102

102 temperature can change the folding of TZ2 from multi -state to apparent two -state folding, increases in medium friction can potentially enable TZ2 to sample multiple folding routes ( figure 4 5 ) even at higher temperatures A recent study by Pande and Rhee predicted that at low solvent viscosities, the protein folding would occur by early formation of long -range contacts. At high viscosities, it would occur by early compaction followed by formation of longrange contacts. Our stu dies of TZ2 folding kinetics suggest a different change in folding mechanism At low solvent viscosity (~water), unfolding occurs by a fast rupture of the hairpin followed by solvation of the core. Reversibly folding would occur by solvent exclusion (compaction) followed by longrange contact formation. At high solvent viscosity (~ 3 4 water), we observe the kinetic signature of a folding mechanism that involves simultaneous compaction and interactions between the cross -strands of TZ2. The access to different folding routes can enable an enriched understanding of the energy landscape of folding for TZ2 and the different possible schemes of events underlying the formation of beta hairpin structure in TZ2. It also opens up the possibility of future experiments on TZ2 mutants with site -specific fluorescent labels, so that we may understand better the events that limit the rate of hairpin formation in TZ2. These events could include turn formation and the formation of contacts that stabilize the core of TZ2. Conclusions Kinetic studies of TZ2 folding in solvents with varying concentrations of ethylene glycol are conducted to study multi -state folding at different conditions of solvent viscosity and temperatures. The fluorescence intensity and the peak wavelength of the fluorescence emission traverse different trajectories along the energy landscape as shown (figure 4 -3) by the different relaxations in the V1 and V2 kinetic vectors over all wavelengths. This confirms the (30, 119)

PAGE 103

103 heterogeneous nature of TZ2 folding. We also investigate the temperature dependence of this heterogeneity in folding for different conditions of solvent viscosity and observe that folding is more multi -state at low tempe ratures and more like a two -state transition at high temperatures Increases in solvent friction enhance the multi -state fol ding behavior, even at high temperatures of 55oC. The non linear kinetic response at different experimental conditions of solvent viscosity experimentally verifies the (41) prediction that folding pathways are influenced by the diffusional properties of the protein and its environment.

PAGE 104

104 CHAPTER 5 CONTACT FORMATION IN PO LYPEPTIDES Introduction Protein folding begins from an ensemble of random coil like configurations for the unfolded polypeptide chain, which proceeds to the final folded state by formation of contacts between residues of the chain. The first contact (12) between any two regions of the chain is an important event (23) that initiates the compaction of the chain and the formation of secondary structural elements such as turns, hairpins a nd helices. This event is termed as contact formation or loop formation in the protein folding literature. It sets the limit for the protein folding process, as a protein can fold only as fast as its folding nucleus can form. Theoretical estimates of rat es of contact formation have used simple polymer models for the unfolded polypeptide chain (18, 169) to estimate the rates at which its ends diffuse towards each other in solvent. Experimental mea surements of intra -chain diffusion rates of the ends of a polypeptide chain (22, 23, 24) indicate a rate of contact formation ~ 107/s or (100 ns)1 for very short polypeptides an d slower rates for longer loops. This chapter focuses on studies of intra -chain diffusion in polypeptide chains in aqueous solvent and the effects of loop length, temperature and solvent viscosity on the rates of loop formation. Background: Experiments on C ontact F ormation Energy transfer methods such as FRET and triplet -triplet energy transfer (TTET) are used to estimate the rate of loop formation in peptides (170) and nucleic acids (171) Studies of end to -end loop formation kinetics in peptides with seven to twenty amino acids show that the time of formation of longer loops are in agreement with times predicted by SSS theory i.e. 2 / 3n For short loops with f our to six amino acids between terminal amino acids, however the time of loop formation saturate s at 100 ns.

PAGE 105

105 While contact formation between termini of the polypeptide chain has been subject to theoretical and experimental investigation, the more interest ing and biologically relevant case of internal loop formation where two nonterminal points in the chain make a contact has not been as well studied. The only study (172) that we are aware of, measures loop formation r ates for peptides labeled with synthetic dyes. The rates of loop formation decreased with increasing tail length in loops with tails added to one end and loops with tails added to both ends. In general, the presence of tails on both ends of the termini greatly lower the speed of contact formation as compared to the presence of a tail at one end. Figure 5 1: Flash photolysis scheme for monitoring intra -chain diffusion via energy transfer between contact forming monomers. The sample is excited by trigger at time t in stage (i), causing donor excitation to triplet state (color change from black to brown) in stage (ii). Contact formation in (iii) precedes energy transfer (iv) which can be measured by monitoring relaxation of don or triplet at variable time delays D Rationale for our E xperiments Our motivat ion is to study the process of contact formation in chains with and without tails (external and internal loops) so that we may truly understand what parameters exert a rate limiting influence on the rates of this process. I choose the triplet triplet energy transfer (TTET) O PTICAL TRIGGER O PTICAL PROBE t t + D 10 ns < D < 200 s (i) (ii) (iii) (iv) Time A

PAGE 106

106 method for our studies of contact formation in polypeptide chains, as it is a more sensitive and specific probe of contact formation, compared to FRET. The m echanisms for TTET and FRET differ in that actual overlap of electron clouds is essential for TTET. Therefore, it is more reliable as a reporter for loop formation than FRET. I use tryptophan and cysteine as triplet donor and acceptor respectively, in our experiments (figure 5 1). Quenching of the triplet state can occur by electron transfer after diffusionlimited collision with the sulphydryl (125) group in the amino acid cysteine. This quenching reduces the tryptophantriplet relaxation lifetime making it a useful probe of diffusionlimited contact formation in polypeptide chains with cysteine and tryptophan monomers (figure 2 9). Results Optimization of the Transient Spectroscopy System The transient spectrometry set up is capable of acquiring multi -wavelength spectra over wavelengths 400700 nm at sub-millisecond time delays with nanosecond time resolution. Energy transfer experiments reported in the contact formation literature typically probe sample respon se at a s ingle wavelength. M ulti -wavelength data provides additional information and thus a more global view of the system response. This system is designed to excite molecules which can absorb 260290 nm light, such as tryptophan ( max = 280 nm) in peptides. The s ystem performance is optimized with the complex of carbon monoxide and protein cytochrome c (11) which is photolyzed by the UV beam. Cytochrome c contains a heme group that binds to specific ligands His18 and Met80 in the natively folded protein in the absence of carbon monoxide. It can be unfolded under destabilizing conditions by the preferential binding of carbon monoxide to the haem group. This complex is photo-disassociated by a laser pulse after which the haem group rebinds to its ligands in the protein while sampling the

PAGE 107

107 conformational space available to it. Thus, the laser pulse initiates a conformational change (figure 5 2 ) in the protein. Time resolved difference absorpt ion spectroscopy (A( ,t) A( t = 0) in the region of the Soret band associated with the haem group (400 430 nm) at time delays from 10 ns to 100 ms can be used to monitor this conformational change. Figure 5 2 : Systems check with 3 M Cytochrome c + C O, whose differential absorbance (A) can be resolved into (B) spectra ui relaxing on timescales given by vi The huge change in absorbance is due to the re -binding of haem in protein with its ligands. I can observe the change in absorption of the cyt ochrome c after flash photolysis at ~ 417 nm over logarithmic timescales spanning nanoseconds to seconds. T he alignment of the transient spectrometry system can also be optimized by observing the magnitude of the change in absorbance after photolysis of t his sample. Characterization of Tryptophan P hotochemistry I have observed the s pectra of transients (figure 5 3 ) generated after laser photolysis of aqueous solutions of tryptophan and their relaxations on the microsecond timescale. Singular value decomp osition of our absorption spectra yields triplet spectrum at 460 nm and neutral radical at 510 nm. The time -evolution of these spectra can be fit to a single exponentia l on

PAGE 108

108 microsecond timescales. A 5 s relaxation is observed for the tryptophan triplet despite meticulous sample preparation and system optimization. This does not match literature values of 38 s for tryptophan triplet relaxation. I observe a relaxation of 53 70 s for the neutral radic al, which is comparable to a slow ~ 60 100 s relaxation attributed to it in the literature. Figure 5 3 : Absorption spectra for free tryptophan in pH 7.0 phosphate buffer shown in panel A is resolved into major spectral contributions to the signal (low er two panels) and their evolution in time (SV1 and SV2) shown in panel B. D etermination of Power Law D ependence for Tryptophan T riplet R elaxation The tryptophan triplet can relax by multiple pathways (chapter 2, Eq 2 10). Previous studies (22, 173, 174) of tryptophan triplet relaxation in the context of triplet triplet energy transfer experiments have assumed that the triplet state decays exponentially with time. T his is true only when the mechanism of triplet relaxation is not triplet triplet annihilation or triplet quenching by oxygen. If triplet concentration is relatively low (~1050 M), the triplet prefers to relax by an exponential decay or oxygen quenching. Under concentrations of high concentration (> 100 M), the triplet is de -excited pre-dominantly by a bimolecular annihilation mechanism. For the T T annihilation reaction,

PAGE 109

109 t Trp k Trp t Trp Trp k dtTrp d Trp TrpT TT T T T TT T k TTT)] 0 ( [ 2 1 )] 0( [ )] ( [ } ] [ 2 { ] [ ] [ ] [ 22 (126) [5 1 ] This functional form in Eq.5 1 demonstrates the power law dependence of triplet relaxation times. We verify this for different concentrations of tryptophan. We observe that the higher the tryptophan concentration, th e faster the triplet decay, implying an increased triplettriplet annihilation. This also explains our consistently measured relaxation times of 5 s for the triplet of tryptophan (TrpT). Figure 5 4 : The tryptophan triplet relaxation (absorption maximum at 450 nm) fit to a single exponential and power law decay Figure 5 4 plots the kinetic data at wavelength 450 nm corresponding to the peak in t ryptophan triplet absorption to a single exponential decay and decay described by above equation. The plot clearly demonstrates that the triplet relaxation is fit best by a power law and not by an exponential decay

PAGE 110

110 Tryptophan T riplet Lifetimes and O xyge n Q uenching The triplet relaxation in the presence of oxygen quenching and triplet triplet annihilation can be summed up by Eq 5 2 Trp Trp Trp O TrpTT oxk T k T2 22 (126) [5 2 ] The rate of oxygen quenching kox is an order slower than the rate of triplet-triplet annihilation kTT. T he functional form of triplet relaxation in Eq.5 3 is given by a combination of annihilation and oxygen quenching mechanisms. ) 2 exp( )) /( ( 1 ) 2 exp( )) /( ( ) ( 2 2 )) /( 1 / 1 ( ) ( ) 0 ( ]; [ 2 / ] [ ] ][ [ ] [ 2 ] [ln0 2 2 2t mk m C C t mk m C mC t y t mk dt k m y y dy m y y dy C t y Trp y k O k m Trp O k Trp k dt Trp dTT TT TT y C t TT y C y C T TT ox T ox T TT Tm y y [5 3 ] Figure 5 5 is a simulation of the change in tryptophan triplet concentration over time assuming a fixed oxygen concentration of 14 M and variable initial triplet concentration (from 3 300 M) The concentration of oxygen is calculated from the concentration of ambient oxygen in water (250 M) and deoxygenation parameters. The results are dominated by the triplettriplet annihilation reaction at triplet concentrations exceeding 30 M. At lower concentrations, the relaxation of the tryptophan

PAGE 111

111 triplet is markedly different. Thus, for the purpose of our experiments, we can say that the triplet relaxation occurs ~ kTT due to triplet triplet annihilation. Very few experimental studies on tryptophan photochemistry and loop formation using tryptophan photochemistry actually (126) address the issue of triplet triplet annihilation. The observ ed values of 40 s for tryptophan triplet relaxation reported in the literature are valid as they have been reported for experiments involving low concentrations of tryptophan triplet (~ 1030 M) after meticulous deoxygenation, so that the mono-exponential decay is the only major relaxation pathway. Figure 5 5 : Triplet relaxation as a function of triplet and oxygen concentration Challenges and B ottlenecks Progress in determining an accurate relaxation time for the tryptophan triplet has bee n hampered by the following factors: 10-8 10-7 10-6 10-5 10-4 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 tTrpT/TrpT(0) Calculated TrpT decay for [O2] = 0.014 mM 22-Nov-2006 TrpT 0 = 1 uM 3 uM 10 uM 30 uM 100 uM 300 uM

PAGE 112

112 The raw absorbance signal shows two signals at 460 nm and 510 nm decaying rapidly on different timescales. The neutral radical has a large broad peak at 510 nm and a very slow relaxation time of about a hundred microse conds. The decay of this species also occurs by an annihilation reaction on a slower timescale. This greatly affects our analysis of triplet state kinetics. Tryptophan is easily photo -bleached by UV light. This affects our signal levels over the time of an experiment. A typical experiment uses 1500 UV shots, for optimal signal to noise. This is also close to the number of UV shots after which tryptophan signal drops by 1/e (figure 56 ). This makes alignment with tryptophan cumbersome. Figure 5 6 : P hoto -damage of tryptophan by repeated UV irradiation in an experiment. A solution of free tryptophan (3 ml, 0.1mM) in phosphate buffer is sealed in a 1 cm path length absorption cell, and subject to flash photolysis repeatedly in ou r transient spectrometr y setup. Each experiment involves 1800 excitation events or UV laser shots. The fluorescen ce of the tryptophan sample is monitored after each successi ve irradiation o n a JASCO FP 750 fluorimeter. It shows the irreversible photo-da mage to tryptophan durin g experiments. We typically start with 0.35 micromoles and lose almost half of it after one experiment. We are unable to get a signal from our sample after three experiments, as indicated by the negligible fluorescence of the tryptophan sample in figure.

PAGE 113

113 Discussion The formation of a loop in a polymer chain of amino acids is a significant step (13, 22) towards the folded state of the protein. The accurate estimation of these rates is very e ssential for a better understanding of the mechanism of protein folding. Energy transfer methods have been used to probe the diffusion limited rates of loop formation in polymer chains tagged with optically excitable molecular probes. The observed kineti cs have been fit to simple models for polymer chains diffusing in solvent enabling an estimate of the speeds at which intra -chain diffusion of an unfolded polypeptide chain can initiate the creation of a loop. Experimental studies of loop formation have f ocused largely on the rates of formation of external loops and their dependence on factors such as loop length, temperature, amino acid composition and solvent viscosity. A very recent study has also investigated the slowing effect of tails of residues ex ternal to the loop, on rates of loop formation. None of the studies mentioned so far addresses the role of excluded volume interactions and their effect on loop formation kinetics in exterior and interior loops. The motivation in studying loop formation in polymer chains 5 20 amino acids long was to investigate the change in rates of loop formation with and without tails as a consequence of excluded volume interactions between segments of the loop. Towards this end, I constructed a transient spectroscopy system capable of probing the diffusion limited energy transfer between tryptophan and cysteine separated by glycine glutamine alanine repeats in amino acid chains of 5 20 residues. Cysteine is known to quench the triplet state of tryptophan most effecti vely among amino acids, by a mechanism that involves actual overlap of electron clouds. Our rationale for using this system is that the triplet state relaxation provides a sensitive and specific probe of actual contact preceding the formation of a loop.

PAGE 114

114 I have been able to characterize the triplet state of tryptophan in the wavelength and tim e basis. I am unable to determine the tryptophan triplet relaxation accurately due to the presence of other transients with a large optical signal in the wavelength region of our interest. The low triplet yield of tryptophan necessitates the use of high concentrations ~ 50100 M for our experiment. This results in relaxation of tryptophan by other mechanisms like triplet -triplet annihilation. The complex relaxation kinetics makes it harder to obtain a reliable estimate of the relaxation rate. The enhanced tryptophan concentrations also result in the production of other photoproducts in high concentrations. These photoproducts interfere with the effective resolutio n of tryptophan triplet relaxation. The low threshold for photo damage is also a concern. Future studies that overcome these challenges by the possible use of flow techniques to avoid tryptophan photo -damage and optimized concentrations to minimize quenching and radical creation could enable a better understanding of one of the fundamental events in the folding process. Conclusions We have characterized and optimized a transient spectroscopy system capable of acquiring multi -wavelength spectral data at s trategically chosen time delays on the submillisecond timescale with nanosecond time resolution. We have characterized the relaxation of our probe molecule, the triplet state of tryptophan in wavelength and time. We have been able to observe the triple t peak at 450 nm, as well as its relaxation on the microsecond timescale. We have also been able to observe kinetics for the relaxation of the tryptophan triplet that is fit well by a power law. This relaxation corresponds to a triplet -triplet annihilation mechanism. We have also seen the effect of oxygen quenching on triplet relaxation time, and

PAGE 115

115 note that relaxation of the tryptophan triplet is more sensitive to the bimolecular annihilation reaction than oxygen quenching. This result is valid only for high concentrations of tryptophan triplet ([Trp*] >> 50 M). Previous studies of diffusion limited contact formation probing tryptophan triplet relaxation have employed kinetic information at one single wavelength of interest. Our multi wavelength studies highlight the rich complexity of tryptophan photochemistry and the possibility that the reported values of tryptophan relaxation do not consider the different contributions from the photoproducts of tryptophan photolysis. Our attempts at accurately mea suring the rate of loop formation by probing tryptophan triplet transfer have been challenged by difficulties in resolving signal due to different tryptophan photoproducts. Future work in this direction would involve surmounting these difficulties by optim ization of experimental conditions to prevent tryptophan photo -damage and allow removal of photoproducts that interfere with our analysis.

PAGE 116

116 CHAPTER 6 FUTURE DIRECTIONS AND CONCLUSIONS We have enhanced the capabilities of a nanosecond laser pulse triggered temperature jump system, so that we can acquire multi -wavelength spectra synchronously with kinetic data on submicrosecond timescales. This system has been used to investigate the kinetics of inhibition of the enzyme YPrA in yeast by an endogenous inhibit or IA3. We have been able to gain a unique insight into the sequence of events that lead to the folding of the natively unstructured IA3 while it binds to YPrA. This folding and binding inhibition reaction is essential for the protection of cellular protei ns from damage by YPrA. We have undertaken one of the few studies of the kinetics of coupled folding and binding in natively unfolded proteins, and the first study of this kind for the IA3YPrA system. Our studies have revealed that the unstructured IA3 pe ptide first makes non -specific contacts with YPrA, which stabilize the helically folded state of IA3. Our study has not considered the role of the C terminus of the unstructured peptide IA3 in the binding interaction with YPrA. Previous NMR studies of the IA3YPrA interaction have shown that certain residues of the C terminus of IA3 away from the binding site actually interact with the YPrA. The inhibitory ability of IA3 peptides with and without the C -terminus are very similar, so this interaction does n ot affect the equilibrium inhibition constant. One interesting possibility is that the C -terminus forms fleeting contacts with YPrA leading to the formation of the encounter complex that has an unfolded N terminus, while the IA3 is bound to YPrA by the C-terminal contacts. Studies of folding kinetics of truncated IA3 peptides with only N terminus in the future could elucidate the role of the C terminus in the interaction with YPrA. Another interesting possibility for future work is identification of key r esidues involved in this non specific binding by experiments and simulations with mutants of IA3.

PAGE 117

117 Our study of the folding dynamics of the tryptophan zipper in different solvent viscosities investigates its navigation of the energetic landscape under dif ferent experimental conditions. Previous theoretical studies have indicated a heterogeneous folding behavior of tryptophan zipper folding at low temperatures due to a rough energy landscape. It has been argued that at high temperatures, the landscape is dominated by one global minimum, leading to one folding pathway. Our kinetic studies reveal the presence of two folding pathways at temperatures less than 35C, with distinguishably different relaxation rates. This result has several repercussions for the rate limiting events leading to secondary structure formation in the tryptophan zipper. While our studies look at the kinetics of folding pathways accessible to the protein, the actual structural features of the protein formed along these routes are still s ubject to speculation. Studies in simulation and experiment with systems like tryptophan zippers are needed for a complete view of the initial events in the folding process. We have designed and characterized a transient spectroscopy system capable of acq uiring nanosecond time resolved multi -wavelength spectral data on events occurring on the ns -ms timescale. We have been able to study the photochemistry of tryptophan after flash photolysis by a nanosecond pulse of ultra violet light. This study revealed the rich spectral and kinetic complexity of photoproducts of tryptophan. This system holds promise for experiments that probe early events in the protein folding process (loop and turn formation) with molecular probes whose photochemistry is well characte rized.

PAGE 118

118 APPENDIX A NUMERICAL METHODS Singular Value D ecomposition of D ata for R elaxation of IA3 A fter a T -jump We use singular value decomposition in chapter 4 to understand the folding transition of IA3 during its association with YPrA. Time resolved fl uorescence data is collected for (figure A 1) wavelengths from 290725 nm, over timescales from nanoseconds to tens of microseconds. Singular value decomposition is a matrix technique that can take this multi -wavelength kinetic dataset and separate the sig nals in order of increasing importance from the noise. Figure A 1 : time resolved fluorescence spectra of double mutant N2W -K16C IA3 peptide recorded on CCD camera, over wavelengths 250725 nm over microsecond timescales. The negative times indicate pre trigger times, from 7 s to the time after the IR pulse thermally triggers the protein unfolding. The pre -trigger fluorescence intensity is markedly different from the post trigger fluorescence intensity.

PAGE 119

119 For a typical laser triggered time resolved fl uorescence experiment, we have a dataset of fluorescence we call F, with n rows (number of wavelengths) and nt columns (number of time points per wavelength). The SVD algorithm resolves F into its eigenvectors U in wavelength and V in time, with eigenvalu es that are specified in the diagonal matrix S. Tt V S U t F )) ( ( ) ( ) ( [A 1] U and V are orthonormal vectors in the wavelength and time basis such that T T T T T T T TV S V V S U U S V F F U S U U S V V S U F F . . . .2 2 [A 2 ] The matrix FFT has an overlap of kinetic vectors for all wavelengths. So, the U vector contains all spectral contributions to the signal, that are mutually independent. The matrix FTF similarly contains the overlap of all spectral vectors for all ti mes; V contains the kinetic information of each spectrum in U. The importance of each spectrum and its corresponding relaxation is given by the eigenvalues in S. We can select the most important contributions to F from elements of U, S and V. noise V S U Fi T i i i [A 3 ] This effectively filters all the noise, and gives us the relevant signal in the least number of basis vectors in the different parametric basis. Notice that the traditional method of resolv ing the signals from a given dataset involves separately solving the eigenvalues equation for FFT and FTF. In our experiments with IA3, we observe three contributions to our signal. The first component has a spectrum similar to the average spectrum of our sample, and a relaxation that looks like a step function. This is the mean fluorescence of our sample that is reduced after a rise in temperature due to the intrinsic negative temperature dependence of fluorescence. The second

PAGE 120

120 component has a spectrum th at resembles the first derivative of the mean spectrum of emission with respect to wavelength, or the shift in wavelength. Its relaxation is fit by a single exponential decay. The third component has a spectrum that resembles a broadening of the spectrum, and is fit well by the same single exponential that fits the second component. All the other lower ranked component of the signal matrix are on the level of the noise, so we have (figure A 2) ) ( / ). ( ) ( ) / ).( ( ) ( ). (3 11 33 3 2 11 22 2 1 1t V S S U t V S S U t V U F [A 4] Figure A 2 : Results of singular value decomposition of dataset shown in previous figure, with three components above the level of noise shown by the semi -logarithmic plot of S1 to S10. The spectral eigenvectors in the wavelength basis is given by U1 to U3 while the correspon ding relaxation kinetics is given by weighted SV1 to SV3 where SV = v*s. We simultaneously fit the weighted components SV2 and SV3 to a single exponential 22 1 21 3 33 12 1 11 2 22) / exp( ) / exp( a t a V S a t a V S [A 5]

PAGE 121

121 This relaxation is the signature of the folding transition of the unstructured peptide IA3 The fall of the second component in the red region, and its rise in the blue region of the spectrum indicate a rise in tryptophan fluorescence concurrent with the fall in the dans yl fluorescence. This would imply a reduced transfer of energy to the dansyl moiety, our FRET acceptor, from tryptophan. This would arise only when the peptide melts pushing the chromophores apart at the higher temperature. Figure A 3 : Single exponentia l fit for both SV2 and SV3 in time, providing an observed lifetime ~ 500 ns. This estimate of can be used with the estimated G (T, TFE) from our fits of CD data for helix -coil transition of IA3 in TFE to determine the folding and unfolding rate (kF, kU) of IA3 at this temperature T and TFE concentration. This procedure is repeated for multiple measurements at various temperatures and TFE concentrations.

PAGE 122

122 LIST OF REFERENCES 1. Dill, K. A., Ozkan, S. B., Shell, M. S., and Weikl, T. R. (2008) The Protein Folding Problem. Annu. Rev. Biophys. 37, 289316. 2. Fersht, A. R. (2008) From the First Protein Structures to our Current Knowledge of Protein Folding: Delights and Scepticisms. Nat. Rev. Mol. Cell Biol. 3. Finkelstein, A. V., and Ptitsyn, O. B. (2002) Protein Physics : A Course of Lectures. Academic Press, Amsterdam ; Boston. 4. Anfinsen, C. B. (1973) Principles that Govern the Folding of Protein Chains. Science. 181 223230. 5. Anfinsen, C. B., and Scheraga, H. A. (1975) Exper imental and Theoretical Aspects of Protein Folding. Adv. Protein Chem. 29, 205300. 6. Bryngelson, J. D., Onuchic, J. N., Socci, N. D., and Wolynes, P. G. (1995) Funnels, Pathways, and the Energy Landscape of Protein Folding: A Synthesis. Proteins. 21, 16 7 195. 7. Onuchic, J. N., Luthey -Schulten, Z., and Wolynes, P. G. (1997) Theory of Protein Folding: The Energy Landscape Perspective. Annu. Rev. Phys. Chem. 48, 545600. 8. Dill, K. A., and Chan, H. S. (1997) From Levinthal to Pathways to Funnels. Nat. S truct. Biol. 4 10 19. 9. Ma, B., Kumar, S., Tsai, C. J., and Nussinov, R. (1999) Folding Funnels and Binding Mechanisms. Protein Eng. 12, 713720. 10. Eaton, W. A., Munoz, V., Hagen, S. J., Jas, G. S., Lapidus, L. J., Henry, E. R., and Hofrichter, J. (2 000) Fast Kinetics and Mechanisms in Protein Folding. Annu. Rev. Biophys. Biomol. Struct. 29 327359. 11. Jones, C. M., Henry, E. R., Hu, Y., Chan, C. K., Luck, S. D., Bhuyan, A., Roder, H., Hofrichter, J., and Eaton, W. A. (1993) Fast Events in Protein Folding Initiated by Nanosecond Laser Photolysis. Proc. Natl. Acad. Sci. U. S. A. 90, 1186011864. 12. Bieri, O., and Kiefhaber, T. (1999) Elementary Steps in Protein Folding. Biol. Chem. 380, 923929.

PAGE 123

123 13. Creighton, T. E. (1997) Protein Folding: Does Di ffusion Determine the Folding Rate? Curr. Biol. 7 R3803. 14. Friedman, B., and Oshaughnessy, B. (1993) Theory of Intramolecular Reactions in Polymeric Liquids. Macromolecules. 26, 48884898. 15. Chan, H. S., and Dill, K. A. (1991) Polymer Principles in Protein Structure and Stability. Annu. Rev. Biophys. Biophys. Chem. 20, 447490. 16. Hyeon, C., and Thirumalai, D. (2006) Kinetics of Interior Loop Formation in Semiflexible Chains. J. Chem. Phys. 124, 104905. 17. Zhou, H. X. (2004) Polymer Models of Pr otein Stability, Folding, and Interactions. Biochemistry. 43 21412154. 18. Szabo, A., Schulten, K., and Schulten, Z. (1980) 1st Passage Time Approach to Diffusion Controlled Reactions. J. Chem. Phys. 72 43504357. 19. Chan, H. S., and Dill, K. A. (199 0) The Effects of Internal Constraints on the Configurations of Chain Molecules. J. Chem. Phys. 92 31183135. 20. Doucet, D., Roitberg, A., and Hagen, S. J. (2007) Kinetics of Internal Loop Formation in Polypeptide Chains: A Simulation Study. Biophys. J. 92, 22812289. 21. Hagen, S. J., Hofrichter, J., Szabo, A., and Eaton, W. A. (1996) DiffusionLimited Contact Formation in Unfolded Cytochrome c: Estimating the Maximum Rate of Protein Folding. Proc. Natl. Acad. Sci. U. S. A. 93, 1161511617. 22. Lapidu s, L. J., Eaton, W. A., and Hofrichter, J. (2000) Measuring the Rate of Intramolecular Contact Formation in Polypeptides. Proc. Natl. Acad. Sci. U. S. A. 97, 7220-7225. 23. Krieger, F., Fierz, B., Bieri, O., Drewello, M., and Kiefhaber, T. (2003) Dynamics of Unfolded Polypeptide Chains as Model for the Earliest Steps in Protein Folding. J. Mol. Biol. 332, 265274. 24. Bieri, O., Wirz, J., Hellrung, B., Schutkowski, M., Drewello, M., and Kiefhaber, T. (1999) The Speed Limit for Protein Folding Measured by Triplet Triplet Energy Transfer. Proc. Natl. Acad. Sci. U. S. A. 96, 95979601. 25. Kramers, H. A. (1940) Brownian Motion in a Field of Force and the Diffusion Model of Chemical Reactions. Physica. 7 284304.

PAGE 124

124 26. Hanggi, P., Talkner, P., and Borkovec, M (1990) ReactionRate Theory 50 Years After Kramers. Reviews of Modern Physics. 62, 251341. 27. Ansari, A., Jones, C. M., Henry, E. R., Hofrichter, J., and Eaton, W. A. (1992) The Role of Solvent Viscosity in the Dynamics of Protein Conformational Cha nges. Science. 256 17961798. 28. Hagen, S. J., Qiu, L. L., and Pabit, S. A. (2005) Diffusional Limits to the Speed of Protein Folding: Fact Or Friction? Journal of Physics -Condensed Matter. 17, S1503 -S1514. 29. Qiu, L., and Hagen, S. J. (2004) A Limiti ng Speed for Protein Folding at Low Solvent Viscosity. J. Am. Chem. Soc. 126, 33983399. 30. Snow, C. D., Qiu, L., Du, D., Gai, F., Hagen, S. J., and Pande, V. S. (2004) Trp Zipper Folding Kinetics by Molecular Dynamics and Temperature Jump Spectroscopy. Proc. Natl. Acad. Sci. U. S. A. 101, 40774082. 31. Jacob, M., Schindler, T., Balbach, J., and Schmid, F. X. (1997) Diffusion Control in an Elementary Protein Folding Reaction. Proc. Natl. Acad. Sci. U. S. A. 94, 56225627. 32. Jacob, M., and Schmid, F. X. (1999) Protein Folding as a Diffusional Process. Biochemistry. 38, 1377313779. 33. Jacob, M., Geeves, M., Holtermann, G., and Schmid, F. X. (1999) Diffusional Barrier Crossing in a Two -State Protein Folding Reaction. Nat. Struct. Biol. 6 923-926. 34. Plaxco, K. W., and Baker, D. (1998) Limited Internal Friction in the Rate Limiting Step of a Two -State Protein Folding Reaction. Proc. Natl. Acad. Sci. U. S. A. 95, 1359113596. 35. Bhattacharyya, R. P., and Sosnick, T. R. (1999) Viscosity Dependence of the Folding Kinetics of a Dimeric and Monomeric Coiled Coil. Biochemistry. 38 2601-2609. 36. Qiu, L. L., and Hagen, S. J. (2005) Internal Friction in the Ultrafast Folding of the Tryptophan Cage (Vol 307, Pg 243, 2004). Chem. Phys. 312, 325+. 37. Pabi t, S. A., Roder, H., and Hagen, S. J. (2004) Internal Friction Controls the Speed of Protein Folding from a Compact Configuration. Biochemistry. 43 1253212538. 38. Hagen, S. J., Qiu, L. L., and Pabit, S. A. (2005) Diffusional Limits to the Speed of Prot ein Folding: Fact Or Friction? Journal of Physics -Condensed Matter. 17, S1503 -S1514.

PAGE 125

125 39. Klimov, D. K., and Thirumalai, D. (1997) Viscosity Dependence of the Folding Rates of Proteins. Phys. Rev. Lett. 79 317 320. 40. Zagrovic, B., and Pande, V. (2003) Solvent Viscosity Dependence of the Folding Rate of a Small Protein: Distributed Computing Study. J. Comput. Chem. 24, 14321436. 41. Rhee, Y. M., and Pande, V. S. (2008) Solvent Viscosity Dependence of the Protein Folding Dynamics. J Phys Chem B. 112, 62 216227. 42. Dyson, H. J., and Wright, P. E. (2005) Intrinsically Unstructured Proteins and their Functions. Nat. Rev. Mol. Cell Biol. 6 197208. 43. Dyson, H. J., and Wright, P. E. (2004) Unfolded Proteins and Protein Folding Studied by NMR. Chem. Rev. 104, 36073622. 44. Uversky, V. N., Gillespie, J. R., and Fink, A. L. (2000) Why are "Natively Unfolded" Proteins Unstructured Under Physiologic Conditions? Proteins. 41, 415427. 45. Fink, A. L. (2005) Natively Unfolded Proteins. Curr. Opin. Struct. Bi ol. 15, 35 41. 46. Dunker, A. K., Lawson, J. D., Brown, C. J., Williams, R. M., Romero, P., Oh, J. S., Oldfield, C. J., Campen, A. M., Ratliff, C. M., Hipps, K. W., Ausio, J., Nissen, M. S., Reeves, R., Kang, C., Kissinger, C. R., Bailey, R. W., Griswold, M. D., Chiu, W., Garner, E. C., and Obradovic, Z. (2001) Intrinsically Disordered Protein. J. Mol. Graph. Model. 19, 26 59. 47. Le Gall, T., Romero, P. R., Cortese, M. S., Uversky, V. N., and Dunker, A. K. (2007) Intrinsic Disorder in the Protein Data Ba nk. J. Biomol. Struct. Dyn. 24, 325-342. 48. Sickmeier, M., Hamilton, J. A., LeGall, T., Vacic, V., Cortese, M. S., Tantos, A., Szabo, B., Tompa, P., Chen, J., Uversky, V. N., Obradovic, Z., and Dunker, A. K. (2007) DisProt: The Database of Disordered Pro teins. Nucleic Acids Res. 35 D78693. 49. Wright, P. E., and Dyson, H. J. (1999) Intrinsically Unstructured Proteins: Re -Assessing the Protein Structure Function Paradigm. J. Mol. Biol. 293, 321331. 50. Haynes, C., Oldfield, C. J., Ji, F., Klitgord, N., Cusick, M. E., Radivojac, P., Uversky, V. N., Vidal, M., and Iakoucheva, L. M. (2006) Intrinsic Disorder is a Common Feature of Hub Proteins from Four Eukaryotic Interactomes. PLoS Comput. Biol. 2, e100. 51. Dunker, A. K., Brown, C. J., Lawson, J. D., I akoucheva, L. M., and Obradovic, Z. (2002) Intrinsic Disorder and Protein Function. Biochemistry. 41 65736582.

PAGE 126

126 52. Uversky, V. N. (2002) Natively Unfolded Proteins: A Point Where Biology Waits for Physics. Protein Sci. 11, 739756. 53. Gunasekaran, K., Tsai, C. J., Kumar, S., Zanuy, D., and Nussinov, R. (2003) Extended Disordered Proteins: Targeting Function with Less Scaffold. Trends Biochem. Sci. 28, 81 85. 54. Iakoucheva, L. M., Brown, C. J., Lawson, J. D., Obradovic, Z., and Dunker, A. K. (2002) In trinsic Disorder in Cell -Signaling and Cancer -Associated Proteins. J. Mol. Biol. 323 573584. 55. Liu, J., Perumal, N. B., Oldfield, C. J., Su, E. W., Uversky, V. N., and Dunker, A. K. (2006) Intrinsic Disorder in Transcription Factors. Biochemistry. 45 68736888. 56. Vacic, V., Oldfield, C. J., Mohan, A., Radivojac, P., Cortese, M. S., Uversky, V. N., and Dunker, A. K. (2007) Characterization of Molecular Recognition Features, MoRFs, and their Binding Partners. J. Proteome Res. 6 23512366. 57. Uvers ky, V. N., Oldfield, C. J., and Dunker, A. K. (2005) Showing Your ID: Intrinsic Disorder as an ID for Recognition, Regulation and Cell Signaling. J. Mol. Recognit. 18, 343384. 58. Uversky, V. N., and Narizhneva, N. V. (1998) Effect of Natural Ligands on the Structural Properties and Conformational Stability of Proteins. Biochemistry (Mosc). 63 420433. 59. Uversky, V. N., Gillespie, J. R., Millett, I. S., Khodyakova, A. V., Vasilenko, R. N., Vasiliev, A. M., Rodionov, I. L., Kozlovskaya, G. D., Dolgikh, D. A., Fink, A. L., Doniach, S., Permyakov, E. A., and Abramov, V. M. (2000) Zn(2+) Mediated Structure Formation and Compaction of the "Natively Unfolded" Human Prothymosin Alpha. Biochem. Biophys. Res. Commun. 267, 663668. 60. Engel, J., Taylor, W., Pa ulsson, M., Sage, H., and Hogan, B. (1987) Calcium Binding Domains and Calcium Induced Conformational Transition of SPARC/BM 40/osteonectin, an Extracellular Glycoprotein Expressed in Mineralized and Nonmineralized Tissues. Biochemistry. 26, 69586965. 61. Baskakov, I. V., Kumar, R., Srinivasan, G., Ji, Y. S., Bolen, D. W., and Thompson, E. B. (1999) Trimethylamine N Oxide Induced Cooperative Folding of an Intrinsically Unfolded Transcription -Activating Fragment of Human Glucocorticoid Receptor. J. Biol. C hem. 274, 1069310696. 62. Dyson, H. J., and Wright, P. E. (2002) Coupling of Folding and Binding for Unstructured Proteins. Curr. Opin. Struct. Biol. 12, 54 60.

PAGE 127

127 63. Dyson, H. J., and Wright, P. E. (1998) Equilibrium NMR Studies of Unfolded and Partially Folded Proteins. Nat. Struct. Biol. 5 Suppl 499503. 64. Verkhivker, G. M., Bouzida, D., Gehlhaar, D. K., Rejto, P. A., Freer, S. T., and Rose, P. W. (2003) Simulating Disorder -Order Transitions in Molecular Recognition of Unstructured Proteins: Where F olding Meets Binding. Proc. Natl. Acad. Sci. U. S. A. 100, 51485153. 65. Shoemaker, B. A., Portman, J. J., and Wolynes, P. G. (2000) Speeding Molecular Recognition by using the Folding Funnel: The Fly-Casting Mechanism. Proc. Natl. Acad. Sci. U. S. A. 97 88688873. 66. Lu, H. P. (2005) Probing Single -Molecule Protein Conformational Dynamics. Acc. Chem. Res. 38 557565. 67. Lu, Q., Lu, H. P., and Wang, J. (2007) Exploring the Mechanism of Flexible Biomolecular Recognition with Single Molecule Dynamics. Phys. Rev. Lett. 98 128105. 68. Sugase, K., Dyson, H. J., and Wright, P. E. (2007) Mechanism of Coupled Folding and Binding of an Intrinsically Disordered Protein. Nature. 447, 10211025. 69. Onitsuka, M., Kamikubo, H., Yamazaki, Y., and Kataoka, M. (2008) Mechanism of Induced Folding: Both Folding before Binding and Binding before Folding can be Realized in Staphylococcal Nuclease Mutants. Proteins. 70. Meszaros, B., Tompa, P., Simon, I., and Dosztanyi, Z. (2007) Molecular Principles of the Interactio ns of Disordered Proteins. J. Mol. Biol. 372, 549561. 71. Romero, P., Obradovic, Z., Li, X., Garner, E. C., Brown, C. J., and Dunker, A. K. (2001) Sequence Complexity of Disordered Protein. Proteins. 42 38 48. 72. Williams, R. M., Obradovi, Z., Mathura V., Braun, W., Garner, E. C., Young, J., Takayama, S., Brown, C. J., and Dunker, A. K. (2001) The Protein NonFolding Problem: Amino Acid Determinants of Intrinsic Order and Disorder. Pac. Symp. Biocomput. 89 100. 73. Gunasekaran, K., Tsai, C. J., and Nussinov, R. (2004) Analysis of Ordered and Disordered Protein Complexes Reveals Structural Features Discriminating between Stable and Unstable Monomers. J. Mol. Biol. 341, 13271341. 74. Mohan, A., Oldfield, C. J., Radivojac, P., Vacic, V., Cortese, M. S., Dunker, A. K., and Uversky, V. N. (2006) Analysis of Molecular Recognition Features (MoRFs). J. Mol. Biol. 362 10431059.

PAGE 128

128 75. Zhou, P., Lugovskoy, A. A., McCarty, J. S., Li, P., and Wagner, G. (2001) Solution Structure of DFF40 and DFF45 N Terminal D omain Complex and Mutual Chaperone Activity of DFF40 and DFF45. Proc. Natl. Acad. Sci. U. S. A. 98, 60516055. 76. Li, M., Phylip, L. H., Lees, W. E., Winther, J. R., Dunn, B. M., Wlodawer, A., Kay, J., and Gustchina, A. (2000) The Aspartic Proteinase from Saccharomyces Cerevisiae Folds its Own Inhibitor into a Helix. Nat. Struct. Biol. 7 113117. 77. Halperin, I., Ma, B. Y., Wolfson, H., and Nussinov, R. (2002) Principles of Docking: An Overview of Search Algorithms and a Guide to Scoring Functions. Pro teins. 47 409443. 78. Levy, Y., Cho, S. S., Onuchic, J. N., and Wolynes, P. G. (2005) A Survey of Flexible Protein Binding Mechanisms and their Transition States using Native Topology Based Energy Landscapes. J. Mol. Biol. 346, 11211145. 79. Gupta, N. and Irback, A. (2004) Coupled Folding -Binding Versus Docking: A Lattice Model Study. J. Chem. Phys. 120, 39833989. 80. Tsai, C. J., Xu, D., and Nussinov, R. (1998) Protein Folding Via Binding and Vice Versa. Fold. Des. 3 R71 80. 81. Muralidhara, B. K ., Rathinakumar, R., and WittungStafshede, P. (2006) Folding of Desulfovibrio Desulfuricans Flavodoxin is Accelerated by Cofactor Fly Casting. Arch. Biochem. Biophys. 451, 51 58. 82. Turjanski, A. G., Gutkind, J. S., Best, R. B., and Hummer, G. (2008) Bi nding Induced Folding of a Natively Unstructured Transcription Factor. PLoS Comput. Biol. 4, e1000060. 83. Gall, C., Xu, H., Brickenden, A., Ai, X., and Choy, W. Y. (2007) The Intrinsically Disordered TC 1 Interacts with Chibby Via Regions with High Helic al Propensity. Protein Sci. 16, 25102518. 84. Dreyer, T., Valler, M. J., Kay, J., Charlton, P., and Dunn, B. M. (1985) The Selectivity of Action of the Aspartic -Proteinase Inhibitor IA3 from Yeast (Saccharomyces Cerevisiae). Biochem. J. 231, 777779. 85. Green, T. B., Ganesh, O., Perry, K., Smith, L., Phylip, L. H., Logan, T. M., Hagen, S. J., Dunn, B. M., and Edison, A. S. (2004) IA3, an Aspartic Proteinase Inhibitor from Saccharomyces Cerevisiae, is Intrinsically Unstructured in Solution. Biochemistry. 43, 40714081.

PAGE 129

129 86. Biedermann, K., Montali, U., Martin, B., Svendsen, I. B., and Ottesen, M. (1980) The Amino -Acid -Sequence of Proteinase a Inhibitor 3 from Bakers Yeast. Carlsberg Res. Commun. 45, 225235. 87. Lenney, J. F., and Dalbec, J. M. (1967) Purification and Properties of Two Proteinases from Saccharomyces Cerevisiae. Arch. Biochem. Biophys. 120, 42 48. 88. Phylip, L. H., Lees, W. E., Brownsey, B. G., Bur, D., Dunn, B. M., Winther, J. R., Gustchina, A., Li, M., Copeland, T., Wlodawer, A., and K ay, J. (2001) The Potency and Specificity of the Interaction between the IA3 Inhibitor and its Target Aspartic Proteinase from Saccharomyces Cerevisiae. J. Biol. Chem. 276, 20232030. 89. Winterburn, T. J., Wyatt, D. M., Phylip, L. H., Bur, D., Harrison, R. J., Berry, C., and Kay, J. (2007) Key Features Determining the Specificity of Aspartic Proteinase Inhibition by the HelixForming IA3 Polypeptide. J. Biol. Chem. 282, 65086516. 90. Winterburn, T. J., Phylip, L. H., Bur, D., Wyatt, D. M., Berry, C., an d Kay, J. (2007) N Terminal Extension of the Yeast IA3 Aspartic Proteinase Inhibitor Relaxes the Strict Intrinsic Selectivity. FEBS J. 274 36853694. 91. Davies, D. R. (1990) The Structure and Function of the Aspartic Proteinases. Annu. Rev. Biophys. Bio phys. Chem. 19, 189215. 92. Pearl, L., and Blundell, T. (1984) The Active Site of Aspartic Proteinases. FEBS Lett. 174, 96101. 93. Dreyer, T., Halkier, B., Svendsen, I., and Ottesen, M. (1986) Primary Structure of the Aspartic Proteinase a from Sacchar omyces Cerevisiae. Carlsberg Res. Commun. 51, 27 41. 94. Aguilar, C. F., Cronin, N. B., Badasso, M., Dreyer, T., Newman, M. P., Cooper, J. B., Hoover, D. J., Wood, S. P., Johnson, M. S., and Blundell, T. L. (1997) The Three Dimensional Structure at 2.4 A Resolution of Glycosylated Proteinase A from the Lysosome Like Vacuole of Saccharomyces Cerevisiae. J. Mol. Biol. 267, 899915. 95. Roder, H., Maki, K., and Cheng, H. (2006) Early Events in Protein Folding Explored by Rapid Mixing Methods. Chem. Rev. 106 18361861. 96. Eaton, W. A., Munoz, V., Thompson, P. A., Chan, C. K., and Hofrichter, J. (1997) Submillisecond Kinetics of Protein Folding. Curr. Opin. Struct. Biol. 7 10-14.

PAGE 130

130 97. Eaton, W. A., Thompson, P. A., Chan, C. K., Hage, S. J., and Hofrichter, J. (1996) Fast Events in Protein Folding. Structure. 4, 11331139. 98. Gruebele, M. (2005) Downhill Protein Folding: Evolution Meets Physics. C. R. Biol. 328 701712. 99. Yang, W. Y., and Gruebele, M. (2004) Folding Lambda Repressor at its Speed Limit. Biophys. J. 87, 596608. 100. Zhu, Y., Alonso, D. O., Maki, K., Huang, C. Y., Lahr, S. J., Daggett, V., Roder, H., DeGrado, W. F., and Gai, F. (2003) Ultrafast Folding of alpha3D: A De Novo Designed Three Helix Bundle Protein. Proc. Natl. Acad. Sci. U. S. A. 100, 1548615491. 101. Ballew, R. M., Sabelko, J., and Gruebele, M. (1996) Observation of Distinct Nanosecond and Microsecond Protein Folding Events. Nat. Struct. Biol. 3 923 926. 102. Ballew, R. M., Sabelko, J., and Gruebele, M. (1996) Direct Obser vation of Fast Protein Folding: The Initial Collapse of Apomyoglobin. Proc. Natl. Acad. Sci. U. S. A. 93, 57595764. 103. Ervin, J., Sabelko, J., and Gruebele, M. (2000) Submicrosecond Real Time Fluorescence Sampling: Application to Protein Folding. J. Ph otochem. Photobiol. B. 54, 1 15. 104. Yang, W. Y., and Gruebele, M. (2003) Folding at the Speed Limit. Nature. 423, 193197. 105. Gruebele, M. (1999) The Fast Protein Folding Problem. Annu. Rev. Phys. Chem. 50, 485516. 106. Roder, H., and Colon, W. (1997) Kinetic Role of Early Intermediates in Protein Folding. Curr. Opin. Struct. Biol. 7, 1528. 107. Roder, H., and Shastry, M. R. (1999) Methods for Exploring Early Events in Protein Folding. Curr. Opin. Struct. Biol. 9 620626. 108. Du, D., Tucker, M. J., and Gai, F. (2006) Understanding the Mechanism of Beta Hairpin Folding Via Phi -Value Analysis. Biochemistry. 45 26682678. 109. Dyer, R. B., Maness, S. J., Peterson, E. S., Franzen, S., Fesinmeyer, R. M., and Andersen, N. H. (2004) The Mechanism of Beta Hairpin Formation. Biochemistry. 43 1156011566. 110. Klimov, D. K., and Thirumalai, D. (2000) Mechanisms and Kinetics of Beta Hairpin Formation. Proc. Natl. Acad. Sci. U. S. A. 97, 25442549.

PAGE 131

131 111. Roder, H. (2004) Stepwise Helix Formation and Chai n Compaction during Protein Folding. Proc. Natl. Acad. Sci. U. S. A. 101, 17931794. 112. Nolting, B. (1999) Protein Folding Kinetics : Biophysical Methods. Springer, Berlin ; New York. 113. Creighton, T. E. (1994) The Energetic Ups and Downs of Protein Folding. Nat. Struct. Biol. 1 135138. 114. Cantor, C. R., and Schimmel, P. R. (1980) Biophysical Chemistry. Freeman, San Francisco. 115. Demchenko, A. P. (1986) Ultraviolet Spectroscopy of Proteins. Rev. and enl. translation of the Russian ed., Spring er -Verlag, Berlin ; New York. 116. Lakowicz, J. R., and NetLibrary, I. (2002) Topics in Fluorescence Spectroscopy. Kluwer Academic, New York. 117. Forster, T. (1948) *Zwischenmolekulare Energiewanderung Und Fluoreszenz. Annalen Der Physik. 2 55 75. 118. Parr, C. L., Keates, R. A., Bryksa, B. C., Ogawa, M., and Yada, R. Y. (2007) The Structure and Function of Saccharomyces Cerevisiae Proteinase A. Yeast. 24, 467480. 119. Yang, W. Y., Pitera, J. W., Swope, W. C., and Gruebele, M. (2004) Heterogeneous Folding of the Trpzip Hairpin: Full Atom Simulation and Experiment. J. Mol. Biol. 336, 241251. 120. Banishev, A. A., Shirshin, E. A., and Fadeev, V. V. (2008) Determination of Photophysical Parameters of Tryptophan Molecules by Methods of Laser Fluorimetry Quantum Electronics. 38, 77 81. 121. Strambini, G. B., Kerwin, B. A., Mason, B. D., and Gonnelli, M. (2004) The Triplet -State Lifetime of Indole Derivatives in Aqueous Solution. Photochem. Photobiol. 80, 462470. 122. Bent, D. V., and Hayon, E. (1975) Excited State Chemistry of Aromatic Amino Acids and Related Peptides. III. Tryptophan. J. Am. Chem. Soc. 97 26122619. 123. Gonnelli, M., and Strambini, G. B. (1995) Phosphorescence Lifetime of Tryptophan in Proteins. Biochemistry. 34 1384713857. 124. Gonnelli, M., and Strambini, G. B. (2005) Intramolecular Quenching of Tryptophan Phosphorescence in Short Peptides and Proteins. Photochem. Photobiol. 81, 614622.

PAGE 132

132 125. Li, Z., Lee, W. E., and Galley, W. C. (1989) Distance Dependence of the TryptophanDi sulfide Interaction at the Triplet Level from Pulsed Phosphorescence Studies on a Model System. Biophys. J. 56, 361367. 126. Tsentalovich, Y. P., Snytnikova, O. A., and Sagdeev, R. Z. (2004) Properties of Excited States of Aqueous Tryptophan. Journal of Photochemistry and Photobiology A -Chemistry. 162 371379. 127. Stevenson, K. L., Papadantonakis, G. A., and LeBreton, P. R. (2000) Nanosecond UV Laser Photoionization of Aqueous Tryptophan: Temperature Dependence of Quantum Yield, Mechanism, and Kinetics of Hydrated Electron Decay. Journal of Photochemistry and Photobiology A -Chemistry. 133 159167. 128. Fischer, C. J., Gafni, A., Steel, D. G., and Schauerte, J. A. (2002) The Triplet State Lifetime of Indole in Aqueous and Viscous Environments: Signific ance to the Interpretation of Room Temperature Phosphorescence in Proteins. J. Am. Chem. Soc. 124, 1035910366. 129. Banks, D. D., and Kerwin, B. A. (2004) A Deoxygenation System for Measuring Protein Phosphorescence. Anal. Biochem. 324, 106114. 130. Ev ans, R. F., Ghiron, C. A., Volkert, W. A., Kuntz, R. R., Santus, R., and Bazin, M. (1976) Flash -Photolysis of N -Acetyl L Tryptophanamide Evidence for Radical Production without Hydrated Electron Formation. Chemical Physics Letters. 42 39 42. 131. Bryan t, F. D., Santus, R., and Grossweiner, L. I. (1975) Laser Flash-Photolysis of Aqueous Tryptophan. J. Phys. Chem. 79 27112716. 132. Evans, R. F., Ghiron, C. A., Kuntz, R. R., and Volkert, W. A. (1976) Flash -Photolysis of N Acetyl L Tryptophanamide Prot on Quenching of Radical and Electron Production. Chemical Physics Letters. 42 415418. 133. Baugher, J. F., and Grossweiner, L. I. (1977) Photolysis Mechanism of Aqueous Tryptophan. J. Phys. Chem. 81 13491354. 134. Roder, H., Maki, K., Cheng, H., and Shastry, M. C. (2004) Rapid Mixing Methods for Exploring the Kinetics of Protein Folding. Methods. 34, 15 27. 135. Shastry, M. C., Luck, S. D., and Roder, H. (1998) A Continuous Flow Capillary Mixing Method to Monitor Reactions on the Microsecond Time Sca le. Biophys. J. 74, 27142721.

PAGE 133

133 136. Fierz, B., Joder, K., Krieger, F., and Kiefhaber, T. (2007) Using Triplet Triplet Energy Transfer to Measure Conformational Dynamics in Polypeptide Chains. Methods Mol. Biol. 350, 169187. 137. Fierz, B., and Kiefhaber T. (2007) End-to -End Vs Interior Loop Formation Kinetics in Unfolded Polypeptide Chains. J. Am. Chem. Soc. 129, 672679. 138. Fierz, B., Satzger, H., Root, C., Gilch, P., Zinth, W., and Kiefhaber, T. (2007) Loop Formation in Unfolded Polypeptide Chains on the Picoseconds to Microseconds Time Scale. Proc. Natl. Acad. Sci. U. S. A. 104, 21632168. 139. Englander, S. W., Calhoun, D. B., and Englander, J. J. (1987) Biochemistry without Oxygen. Anal. Biochem. 161, 300306. 140. Eigen, M., and Hammes, G. G. (1960) Kinetic Studies of Adp Reactions with the Temperature Jump Method. J. Am. Chem. Soc. 82 59515952. 141. Ballew, R. M., Sabelko, J., Reiner, C., and Gruebele, M. (1996) A Single -Sweep, Nanosecond Time Resolution Laser Temperature Jump Apparatus. Re v. Sci. Instrum. 67, 36943699. 142. Ballew, R. M., Sabelko, J., and Gruebele, M. (1996) Observation of Distinct Nanosecond and Microsecond Protein Folding Events. Nat. Struct. Biol. 3 923 926. 143. Williams, S., Causgrove, T. P., Gilmanshin, R., Fang, K. S., Callender, R. H., Woodruff, W. H., and Dyer, R. B. (1996) Fast Events in Protein Folding: Helix Melting and Formation in a Small Peptide. Biochemistry. 35 691697. 144. Munoz, V., Thompson, P. A., Hofrichter, J., and Eaton, W. A. (1997) Folding Dynamics and Mechanism of Beta Hairpin Formation. Nature. 390, 196199. 145. Ganesh, O. K., Green, T. B., Edison, A. S., and Hagen, S. J. (2006) Characterizing the Residue Level Folding of the Intrinsically Unstructured IA3. Biochemistry. 45 1358513596. 146. Buck, M. (1998) Trifluoroethanol and Colleagues: Cosolvents Come of Age. Recent Studies with Peptides and Proteins. Q. Rev. Biophys. 31, 297355. 147. Kentsis, A., and Sosnick, T. R. (1998) Trifluoroethanol Promotes Helix Formation by Destabilizing B ackbone Exposure: Desolvation rather than Native Hydrogen Bonding Defines the Kinetic Pathway of Dimeric Coiled Coil Folding. Biochemistry. 37 1461314622.

PAGE 134

134 148. Thompson, P. A., Eaton, W. A., and Hofrichter, J. (1997) Laser Temperature Jump Study of the Helixcoil Kinetics of an Alanine Peptide Interpreted with a 'Kinetic Zipper' Model. Biochemistry. 36 92009210. 149. Dyer RB. Helix Folding Kinetics. Analytical biochemistry. 150. Luo, P., and Baldwin, R. L. (1997) Mechanism of Helix Induction by Trifluoroethanol: A Framework for Extrapolating the Helix -Forming Properties of Peptides from trifluoroethanol/water Mixtures Back to Water. Biochemistry. 36 84138421. 151. Jasanoff, A., and Fersht, A. R. (1994) Quantitative Determination of Helical Propensit ies from Trifluoroethanol Titration Curves. Biochemistry. 33 21292135. 152. Zimm, B. H., and Bragg, J. K. (1959) Theory of the Phase Transition between Helix and Random Coil in Polypeptide Chains. J. Chem. Phys. 31 526535. 153. Schwarz, G. (1965) On Kinetics of Helix Coil Transition of Polypeptides in Solution. J. Mol. Biol. 11 64 -&. 154. Sosnick, T. R., Jackson, S., Wilk, R. R., Englander, S. W., and DeGrado, W. F. (1996) The Role of Helix Formation in the Folding of a Fully Alpha Helical Coiled Co il. Proteins. 24, 427432. 155. Galzitskaya, O. V., Higo, J., and Finkelstein, A. V. (2002) Alpha Helix and Beta Hairpin Folding from Experiment, Analytical Theory and Molecular Dynamics Simulations. Curr. Protein Peptide Sci. 3, 191200. 156. Huang, C. Y., Getahun, Z., Zhu, Y., Klemke, J. W., DeGrado, W. F., and Gai, F. (2002) Helix Formation Via Conformation Diffusion Search. Proc. Natl. Acad. Sci. U. S. A. 99 27882793. 157. Gai, F., Du, D., and Xu, Y. (2007) Infrared Temperature Jump Study of the Fo lding Dynamics of Alpha Helices and Beta Hairpins. Methods Mol. Biol. 350, 1 20. 158. Dumont, C., Matsumura, Y., Kim, S. J., Li, J., Kondrashkina, E., Kihara, H., and Gruebele, M. (2006) Solvent Tuning the Collapse and Helix Formation Time Scales of Lambd a(6 85)*. Protein Sci. 15, 25962604. 159. Chan, H. S., and Dill, K. A. (1998) Protein Folding in the Landscape Perspective: Chevron Plots and Non-Arrhenius Kinetics. Proteins. 30 2 33.

PAGE 135

135 160. Dinner, A. R., Sali, A., Smith, L. J., Dobson, C. M., and Karp lus, M. (2000) Understanding Protein Folding Via Free -Energy Surfaces from Theory and Experiment. Trends Biochem. Sci. 25, 331339. 161. Udgaonkar, J. B. (2008) Multiple Routes and Structural Heterogeneity in Protein Folding. Annu. Rev. Biophys. 37, 4895 10. 162. Dobson, C. M., and Karplus, M. (1999) The Fundamentals of Protein Folding: Bringing Together Theory and Experiment. Curr. Opin. Struct. Biol. 9, 92 101. 163. Dobson, C. M., and Evans, P. A. (1988) Protein Structure .Trapping Folding Intermediate s. Nature. 335, 666667. 164. Jas, G. S., Eaton, W. A., and Hofrichter, J. (2001) Effect of Viscosity on the Kinetics of Alpha Helix and Beta Hairpin Formation. J Phys Chem B. 105, 261272. 165. Cochran, A. G., Skelton, N. J., and Starovasnik, M. A. (2001) Tryptophan Zippers: Stable, Monomeric Beta Hairpins. Proc. Natl. Acad. Sci. U. S. A. 98, 55785583. 166. Hauser, K., Krejtschi, C., Huang, R., Wu, L., and Keiderling, T. A. (2008) Site Specific Relaxation Kinetics of a Tryptophan Zipper Hairpin Peptide using Temperature Jump IR Spectroscopy and Isotopic Labeling. J. Am. Chem. Soc. 130, 29842992. 167. Grishina, I. B., and Woody, R. W. (1994) Contributions of Tryptophan Side Chains to the Circular Dichroism of Globular Proteins: Exciton Couplets and Coupled Oscillators. Faraday Discuss. (99) 245262. 168. Yang, W. Y., and Gruebele, M. (2004) DetectionDependent Kinetics as a Probe of Folding Landscape Microstructure. J. Am. Chem. Soc. 126, 77587759. 169. Wilemski, G., and Fixman, M. (1974) DiffusionControlled Intrachain Reactions of Polymers .2. Results for a Pair of Terminal Reactive Groups. J. Chem. Phys. 60 878890. 170. Huang, F., Hudgins, R. R., and Nau, W. M. (2004) Primary and Secondary Structure Dependence of Peptide Flexibility Assessed b y Fluorescence Based Measurement of Endto End Collision Rates. J. Am. Chem. Soc. 126, 1666516675. 171. Wang, X. J., and Nau, W. M. (2004) Kinetics of Endto -End Collision in Short Single Stranded Nucleic Acids. J. Am. Chem. Soc. 126, 808813.

PAGE 136

136 172. Fier z, B., and Kiefhaber, T. (2007) End-to -End Vs Interior Loop Formation Kinetics in Unfolded Polypeptide Chains. J. Am. Chem. Soc. 129, 672679. 173. Buscaglia, M., Schuler, B., Lapidus, L. J., Eaton, W. A., and Hofrichter, J. (2003) Kinetics of Intramolecu lar Contact Formation in a Denatured Protein. J. Mol. Biol. 332, 9 12. 174. Lapidus, L. J., Eaton, W. A., and Hofrichter, J. (2001) Dynamics of Intramolecular Contact Formation in Polypeptides: Distance Dependence of Quenching Rates in a Room Temperature Glass. Phys. Rev. Lett. 87 258101.

PAGE 137

137 BIOGRAPHICAL SKETCH Ranjani Narayanan was born in 1980 in Chidambaram, a small town in the state of Tamil Nadu, India. She spent her childhood years in Durgapur, a steel township in the state of West Bengal. She attended Carmel School and DAV Model School (Durgapur), where she was the high school valedictorian. She was active in academic and extra -curricular activities. After completing her schooling in 1998, she moved north to the Indian capital c ity, New Delhi to pursue an undergraduate degree in electronics from Sri Venkateswara College. She then joined the prestigious Indian Institute of Technology Madras, or IITM in Chennai, (then Madras) for her Masters degree in Physics (Ranked first in t he competitive entrance examination.) It is here that she discovered her passion for experimental research after working with Dr. Vijayan on non -linear optics. She was the recipient of a merit scholarship for both years of her study at IITM. She was als o active in extra curricular events at IITM, volunteering for the prestigious cultural festival Saarang and NGO Child Relief and You. The physics department at the University of Florida, Gainesville was the next stop for Ranjani, who decided to work on pr otein folding kinetics for her doctoral dissertation under the tutelage of Dr. Stephen J. Hagen. She learned to study the folding of proteins with nanosecond lasers and honed her experimental skills by enhancing the capabilities of a temperature jump spec troscopy system. She coauthored a paper on coupled folding and binding in the unstructured IA3 peptide, appearing in the prestigious Journal of the American Chemical Society. She also presented her work in the American Physical Society meeting at New Or leans, and won the student travel award, awarded by the Division of Biological Physics. She met her husband, Balaji (Another Gator) during her stay in Gainesville and married him in December 2006. She has volunteered for the Alachua County Library as a Homework Helper. She is an avid Gator football and basketball fan.